Build AI methods with your coding agent, then run them anywhere: as an MCP for chatbots, as a webapp for people, or via API for your software.
Feature/epic docs and examples (#87)
Louis Choquel68b95e8f976c
diff --git a/.cursor/rules/pydantic.mdc b/.cursor/rules/base_models.mdcsimilarity index 100%rename from .cursor/rules/pydantic.mdcrename to .cursor/rules/base_models.mdcdiff --git a/.cursor/rules/docs.mdc b/.cursor/rules/docs.mdcnew file mode 100644index 000000000..31461aa94--- /dev/null+++ b/.cursor/rules/docs.mdc@@ -0,0 +1,8 @@+---+description:+globs: docs/**/*.md+alwaysApply: false+---+Write docs and answer questions about writing docs.++We use Material for MkDocs. All markdown in our docs must be compatible with Material for MkDocs and done using best practices to get the best results with Material for MkDocs.diff --git a/.cursor/rules/llms.mdc b/.cursor/rules/llms.mdcindex 5951a1ab4..6de8f9f08 100644--- a/.cursor/rules/llms.mdc+++ b/.cursor/rules/llms.mdc@@ -5,8 +5,10 @@ alwaysApply: false---# Rules to choose LLM models used in PipeLLMs.+## LLM Handles+In order to use it in a pipe, an LLM is referenced by its llm_handle and possibly by an llm_preset.-Both llm_handles and llm_presets are defined in this toml config file: [config_pipelex_llm_deck.toml](mdc:pipelex/config_pipelex_llm_deck.toml)+Both llm_handles and llm_presets are defined in this toml config file: [base_llm_deck.toml](mdc:pipelex/libraries/llm_deck/base_llm_deck.toml)## LLM Handles@@ -15,19 +17,21 @@ An llm_handle matches the handle (an id of sorts) with the full specification of- llm_version- llm_platform_choice-The declaration looks like this in toml syntax:+The declaration of llm_handleslooks like this in toml syntax:```toml-[cogt.llm_config.llm_deck.llm_handle_to_llm_engine_blueprint.gpt-4o-2024-05-13]-llm_name = "gpt-4o"-llm_version = "2024-05-13"-llm_platform_choice = "openai"+[llm_handles]+gpt-4o-2024-08-06 = { llm_name = "gpt-4o", llm_version = "2024-08-06" }```In mosty cases, we only want to use version "latest" and llm_platform_choice "default" in which case the declaration is simply a match of the llm_handle to the llm_name, like this:```toml-gemini-2-5-pro = "gemini-2.5-pro"+best-claude = "claude-4-opus"+best-gemini = "gemini-2.5-pro"+best-mistral = "mistral-large"```+And of course, llm_handles are automatically assigned for all models by their name, with version "latest" and llm_platform_choice "default".+## Using an LLM Handle in a PipeLLMHere is an example of using an llm_handle to specify which LLM to use in a PipeLLM:@@ -59,7 +63,7 @@ The interest is that these presets can be used to set the LLM choice in a PipeLL```toml[pipe.extract_invoice]PipeLLM = "Extract invoice information from an invoice text transcript"-input = "InvoiceText"+inputs = { invoice_text = "InvoiceText" }output = "Invoice"llm = "llm_to_extract_invoice"prompt_template = """@@ -73,3 +77,6 @@ The category of this invoice is: $invoice_details.category.The setting here `llm = "llm_to_extract_invoice"` works because "llm_to_extract_invoice" has been declared as an llm_preset in the deck.You must not use an LLM preset in a PipeLLM that does not exist in the deck. If needed, you can add llm presets.+++You can override the predefined llm presets in [overrides.toml](mdc:pipelex/libraries/llm_deck/overrides.toml).diff --git a/.cursor/rules/pipes.mdc b/.cursor/rules/pipes.mdcindex 4d6b56166..a32b3d3bc 100644--- a/.cursor/rules/pipes.mdc+++ b/.cursor/rules/pipes.mdc@@ -1,6 +1,6 @@---description:-globs: **/pipelex_libraries/pipelines/toml/**/*.toml+globs: **/pipelex/libraries/pipelines/**/*.tomlalwaysApply: false---This rule explains how to build pipes.@@ -9,7 +9,7 @@ This rule explains how to build pipes.## The pipelines/ directory-- Pipelines and structures are defined in the `pipelex/libraries/` directory+- Pipelines and structures are defined in the `pipelex/libraries/pipelines` directorypipelex/libraries└── pipelines@@ -40,6 +40,11 @@ pipelex/libraries- Never include the usage context in the concept. The concept indicates what the stuff is in itself, it's not "for something", it just is something. e.g. don't define "TextToSummarize": it's just "Text". If you want to refine the concept for instance you can define "Essay".- In particular, never define a concept as a plural form: if the context of the pipeline execution makes it multiple, it will just be handled using a ListContent (see below). e.g. don't define a concept for "Stories", just define "Story".- Also, avoid including adjectives in concepts, e.g. don't define "LargeText", it's just "Text".+- Don't redefine the native concepts from [concept_native.py](mdc:pipelex/core/concept_native.py)++⚠️ Important ⚠️++A Concept MUST NEVER be a plural noun and you should never create a SomeConceptList: lists and arrays are implicitly handled by Pipelex according to the context. Just define SomeConcept.- Define concepts in one of two ways:@@ -56,14 +61,14 @@ ConceptName = "Description of the concept"[concept.ConceptName]Concept = "Description of the concept"structure = "StructureName"-refines = ["ParentConcept"] # Optional, for concept inheritance+refines = "ParentConcept" # Optional, for concept inheritance```About the `structure` field:- It's Optional- It's the name of the Python BaseModel class used for the concept- The class must be a subclass of StuffContent-- The class must be defined in a python module placed inside the `pipelex/pipelex_libraries/pipelines/structures/` directory+- The class must be defined in a python module placed inside the `pipelex/libraries/pipelines/` directory- If the `structure` field is omitted but a class with the same name as the concept is defined in the structures directory, then it's implicitly applied to the conceptAbout the `refines` field:@@ -83,7 +88,7 @@ Pipelex provides the following structures natively:- MermaidContent- LLMPromptContent-The native structures are implied when using the native concepts: "native.Text", "native.Number", "native.Image"... do you don't have the state it.+The native structures are implied when using the native concepts: "Text", "Number", "Image", "PDF",... do you don't have the state it.Some subclasses of StuffContent exist which you should never use directly:- ListContent: this is used internally to manipulate a list of stuff, but you should NEVER define a plural concept@@ -130,7 +135,7 @@ Write a poem about an AI that meets a Software and they fall in l0ve.# Example of a PipeLLM that uses @ prefix to insert a block of text[pipe.process_text]PipeLLM = "Process input text using an LLM"-input = "Text"+inputs = { text = "Text" }output = "Text"llm = "llm_to_summarize_text"system_prompt = """@@ -143,10 +148,10 @@ Summarize the following text:"""-# Example of a PipeLLM that uses @ prefix to insert a block of text but also a $ prefix to insert text inline in a sentence (that is teh case for the $topic)+# Example of a PipeLLM that uses '@' prefix to insert a block of text but also a '$' prefix to insert text inline in a sentence (that is teh case for the $topic)[pipe.summarize_topic]PipeLLM = "Summarize a dense text with of focus on a specific topic."-input = "Topic"+inputs = { topic = "Topic", text = "Text" }output = "Summary"prompt_template = """Your goal is to summarize everything related to $topic in the provided text:@@ -160,7 +165,7 @@ Your summary should not be longer than 2 sentences.# Example of a PipeLLM with image vision processing by a VLM[pipe.get_html_table_from_image]PipeLLM = "Convert table screenshot to HTML"-input = "TableScreenshot"+inputs = { table_screenshot = "TableScreenshot" }output = "HtmlTable"images = ["table_screenshot"]system_prompt = """@@ -178,7 +183,7 @@ llm = "llm_to_extract_tables"# Example of a PipeSequence[pipe.answer_question_with_instructions]PipeSequence = "Answer a question with instructions"-input = "Question"+inputs = { question = "Question" }output = "FormattedAnswer"steps = [{ pipe = "enrich_instructions", result = "instructions", },@@ -189,7 +194,7 @@ steps = [# Example of a PipeParallel[pipe.extract_expense_report]PipeParallel = "Extract useful information from an expense report"-input = "ExpenseReportText"+inputs = { expense_report = "ExpenseReportText" }output = "Composite"parallels = [{ pipe = "extract_employee_from_expense_report", result = "employee" },@@ -199,7 +204,7 @@ parallels = [# Example of a PipeCondition[pipe.expense_conditional_validation]PipeCondition = "Choose the rules to apply"-input = "Expense"+inputs = { expense = "Expense" }output = "RulesToApply"expression = "expense_category.category"```diff --git a/.cursor/rules/structures.mdc b/.cursor/rules/structures.mdcindex e064b5bbb..9671eef39 100644--- a/.cursor/rules/structures.mdc+++ b/.cursor/rules/structures.mdc@@ -1,6 +1,6 @@---description:-globs: **/pipelex_libraries/pipelines/structures/**/*.py+globs: pipelex/libraries/pipelines/**/*.pyalwaysApply: false---Rules to write structure classes for concepts used in pipes with structured generations.@@ -10,11 +10,14 @@ In particular, these structures are generated by PipeLLM using structured genera## Model Location and Registration-- Create models for structured generations related to "some_domain" in `pipelex/pipelex_libraries/pipelines/structures/<some_domain>.py`+- Create models for structured generations related to "some_domain" in `pipelex_libraries/pipelines/<some_domain>.py`- Models must inherit from `StructuredContent` or appropriate content type## Model Structure+Concepts and their structure classes are meant to indicate an idea.+A Concept MUST NEVER be a plural noun and you should never create a SomeConceptList: lists and arrays are implicitly handled by Pipelex according to the context. Just define SomeConcept.+```pythonfrom datetime import datetimefrom typing import List, Optional@@ -41,6 +44,12 @@ class YourModel(StructuredContent):return v.replace(tzinfo=None)return v```+## Usage++Structures are meant to indicate what class to use for a particular Concept. In general they use the same name as the concept.++Structure classes defined within `pipelex_libraries/pipelines/` are automatically loaded into the class_registry when setting up Pipelex, no need to do it manually.+## Best Practices for structures@@ -49,4 +58,4 @@ class YourModel(StructuredContent):- Use `Field` declaration and write the description- Use Pydantic validators for data cleaning/validation- Remove timezone info from datetime fields-- Respect rules of [pydantic.mdc](mdc:.cursor/rules/pydantic.mdc)+- Respect rules of @base_models.diff --git a/.github/workflows/deploy-doc.yml b/.github/workflows/deploy-doc.ymlnew file mode 100644index 000000000..f8fa04801--- /dev/null+++ b/.github/workflows/deploy-doc.yml@@ -0,0 +1,31 @@+# .github/workflows/docs-preview.yml+name: Preview MkDocs on GitHub Pages++on:+ push:+ branches: [codex/update-documentation-for-installation-module]++jobs:+ preview:+ runs-on: ubuntu-latest+ steps:+ - uses: actions/checkout@v4++ - uses: actions/setup-python@v5+ with:+ python-version: '3.x'++ - name: Build docs+ run: |+ pip install mkdocs==1.6.1 mkdocs-material==9.6.14 mkdocs-glightbox==0.4.0 mkdocs-meta-manager==1.1.0+ mkdocs build --strict # generates ./site++ - name: Publish to *gh-pages-preview*+ uses: peaceiris/actions-gh-pages@v4+ with:+ github_token: ${{ secrets.GITHUB_TOKEN }}+ publish_dir: ./site+ publish_branch: gh-pages-preview # <─ test branch+ # put each branch in its own folder so multiple previews can coexist+ destination_dir: ${{ github.ref_name }}+ force_orphan: truediff --git a/.github/workflows/doc-check.yml b/.github/workflows/doc-check.ymlnew file mode 100644index 000000000..1ebb886f6--- /dev/null+++ b/.github/workflows/doc-check.yml@@ -0,0 +1,39 @@+name: Documentation Check++on:+ push:+ branches:+ - main+ - dev+ - "release/v[0-9]+.[0-9]+.[0-9]+"+ paths:+ - 'docs/**'+ - 'mkdocs.yml'+ pull_request:+ branches:+ - main+ - dev+ - "release/v[0-9]+.[0-9]+.[0-9]+"+ paths:+ - 'docs/**'+ - 'mkdocs.yml'++jobs:+ doc-check:+ runs-on: ubuntu-latest+ steps:+ - uses: actions/checkout@v4++ - name: Set up Python+ uses: actions/setup-python@v5+ with:+ python-version: '3.11'+ cache: 'pip'++ - name: Install mkdocs+ run: |+ python -m pip install --upgrade pip+ pip install mkdocs==1.6.1 mkdocs-material==9.6.14 mkdocs-glightbox==0.4.0 mkdocs-meta-manager==1.1.0++ - name: Check documentation build+ run: mkdocs build --strict\ No newline at end of filediff --git a/.github/workflows/guard-branches.yml b/.github/workflows/guard-branches.ymlindex acb6a7285..0e3289fb1 100644--- a/.github/workflows/guard-branches.yml+++ b/.github/workflows/guard-branches.yml@@ -37,7 +37,7 @@ jobs:exit 0fiif [[ ! "$HEAD" =~ ^(fix|feature|refactor|chore|doc|ci-cd|changelog|codex)\/[A-Za-z0-9._\/\<\>\=\-]+$ ]]; then- echo "::error::Branch must start with fix/, feature/, refactor/, chore/, doc/, or ci-cd/."+ echo "::error::Branch must start with fix/, feature/, refactor/, chore/, docs/, or ci-cd/."exit 1fidiff --git a/.gitignore b/.gitignoreindex 7b05d685e..8d5bc3845 100644--- a/.gitignore+++ b/.gitignore@@ -10,6 +10,9 @@ dist/build/*.egg-info/+# mkdocs+site/+# Unit test / coverage reportshtmlcov/.pytest_cache/diff --git a/.vscode/settings.json b/.vscode/settings.jsonindex f06faf615..563688e44 100644--- a/.vscode/settings.json+++ b/.vscode/settings.json@@ -19,5 +19,6 @@"tests"],"python.testing.unittestEnabled": false,- "python.testing.pytestEnabled": true+ "python.testing.pytestEnabled": true,+ "djlint.showInstallError": false}\ No newline at end of filediff --git a/MANIFESTO.md b/MANIFESTO.mdnew file mode 100644index 000000000..fd04a996e--- /dev/null+++ b/MANIFESTO.md@@ -0,0 +1,61 @@+### This is our manifesto, this is why we build [Pipelex](https://pipelex.com){:target="_blank"}.++First published on our [blog](https://www.pipelex.com/post/repeatable-ai-workflows-knowledge-pipelines){:target="_blank"} on June 3, 2025.++---+# The Knowledge Pipeline Manifesto++## Agents are brilliant but they're hopeless at repeatability++AI agents are getting impressive at novel tasks: research, planning, coding from scratch. But here's what we're missing: lots of knowledge work isn't novel. It's the same patterns, repeated thousands of times.++Ask an agent to review 10,000 expense reports. Watch it improvise a new approach for each one. Different reasoning paths, different output formats, different edge case handling. Run it twice on the same report? Different results. The intelligence is there, but the consistency isn't. And **someone’s paying for those improv tokens.**++Agents are explorers. But if you know the path, you don't need an expedition. You need a bullet train.++## Enter the knowledge pipeline++The solution should be obvious: delegate the repetitive tasks to deterministic AI workflows. But here's the deeper problem: AI workflows today are handcrafted, not engineered. We prompt, tweak, test, and pray.++Without proper tools for capturing repeatable workflows, every company recreates the same solutions from scratch. The expense categorization that took your team weeks to perfect? The next company starts at zero. We're in the pre-SQL era of AI: brilliant capabilities trapped in proprietary silos, no way to reuse what actually works.++We need a way to capture proven AI workflows and make them as reliable and reusable as any other software component. We call this a **knowledge pipeline.**++A knowledge pipeline is built from modular components called pipes. Each pipe is a knowledge transformer with a simple contract: **knowledge in, knowledge out**. Unlike data pipelines that move bytes or ML pipelines that train models, knowledge pipes transform *meaning*.++This isn't about dumbing down AI. Each pipe leverages full AI intelligence to handle variation in content while guaranteeing its output structure. You get **deterministic structure with adaptive intelligence:** consistency without rigidity.++## The architecture of understanding++### **Knowledge pipes compose like LEGO blocks for repeatable AI workflows:**++- Connect them sequentially for step-by-step transformations+- Run them in parallel to process multiple perspectives+- Feed multiple outputs into a single pipe for synthesis+- Call sub-pipes conditionally based on categorization++Why does this matter? Because real knowledge work isn't linear.++A pipe analyzing contracts might need context from step 2, results from steps 5-7, and external compliance data. The architecture mirrors how knowledge actually flows: not as a conveyor belt, but as a network of understanding.++## The method becomes the artifact++When every pipe guarantees its output structure, you can compose them fearlessly. When each pipe uses AI intelligently, the pipeline adapts without breaking. **When you capture your method as actual code, you can version it, test it, and scale it like software.**++With knowledge pipelines, your process becomes something you can hold, examine, and refine. **Swap out language models to find the sweet spot between cost and quality.** Adjust prompts until they sing. Reconfigure the flow when you discover a shortcut. Each iteration teaches you something new because you're working with stable, measurable outcomes.++Remember when ChatGPT launched and suddenly everyone was sharing prompts? That same collaborative instinct is waiting to be unleashed on knowledge pipelines. A pipeline for extracting customer insights gets forked for user research. A financial audit workflow becomes the skeleton for compliance reviews. The building blocks translate across industries.++These pipeline artifacts are meant to be shared, forked, and composed. Period.++## The path forward++We've seen this movie before. Docker transformed deployment by giving us a simple way to package and share what used to live in scattered shell scripts and tribal knowledge. Suddenly, complex configurations became artifacts you could version, test, and run anywhere. Better yet, Docker made it possible to build on each other's work rather than starting from scratch each time.++Today's AI workflows are ready for the same breakthrough.++This is why we built [**Pipelex**](https://github.com/Pipelex/pipelex){:target="_blank"} as an open-source standard for repeatable AI workflows. Not just another tool, but a foundation for the community to build on.++The manifesto is simple: **When you discover a method that works, capture it as a knowledge pipeline. When you build a pipeline that roars, share it with the world!**++Join us in building the knowledge infrastructure of tomorrow. One pipe at a time.diff --git a/Makefile b/Makefileindex da72125ed..9514b78e7 100644--- a/Makefile+++ b/Makefile@@ -13,6 +13,7 @@ VENV_RUFF := $(VIRTUAL_ENV)/bin/ruffVENV_PYRIGHT := $(VIRTUAL_ENV)/bin/pyrightVENV_MYPY := $(VIRTUAL_ENV)/bin/mypyVENV_PIPELEX := $(VIRTUAL_ENV)/bin/pipelex+VENV_MKDOCS := $(VIRTUAL_ENV)/bin/mkdocsUV_MIN_VERSION = $(shell grep -m1 'required-version' pyproject.toml | sed -E 's/.*= *"([^<>=, ]+).*/\1/')@@ -44,6 +45,7 @@ make install - Create local virtualenv & install all dependencimake update - Upgrade dependencies via uvmake validate - Run the setup sequence to validate the config and librariesmake build - Build the wheels+make doc - Serve documentation locally with mkdocsmake format - format with ruff formatmake lint - lint with ruff check@@ -99,7 +101,7 @@ export HELPrun-all-tests run-manual-trigger-gha-tests run-gha_disabled-tests \validate v check c cc \merge-check-ruff-lint merge-check-ruff-format merge-check-mypy merge-check-pyright \- li check-unused-imports fix-unused-imports check-uv check-TODOs+ li check-unused-imports fix-unused-imports check-uv check-TODOs doc doc-checkall help:@echo "$$HELP"@@ -388,3 +390,16 @@ check-TODOs: envfix-unused-imports: env$(call PRINT_TITLE,"Fixing unused imports")$(VENV_RUFF) check --select=F401 --fix -v .++doc: env+ $(call PRINT_TITLE,"Serving documentation with mkdocs")+ $(VENV_MKDOCS) serve++doc-check: env+ $(call PRINT_TITLE,"Checking documentation build with mkdocs")+ $(VENV_MKDOCS) build --strict++doc-deploy: env+ $(call PRINT_TITLE,"Deploying documentation with mkdocs")+ $(VENV_MKDOCS) gh-deploy+\ No newline at end of filediff --git a/README.md b/README.mdindex 32beb7973..3529a2a7f 100644--- a/README.md+++ b/README.md@@ -6,7 +6,7 @@<div><a href="https://www.pipelex.com/demo"><strong>Demo</strong></a> -- <a href="https://github.com/Pipelex/pipelex/blob/main/doc/Documentation.md"><strong>Documentation</strong></a> -+ <a href="https://github.com/Pipelex/pipelex/blob/main/docs/Documentation.md"><strong>Documentation</strong></a> -<a href="https://github.com/Pipelex/pipelex/issues"><strong>Report Bug</strong></a> -<a href="https://github.com/Pipelex/pipelex/discussions"><strong>Feature Request</strong></a></div>@@ -18,10 +18,12 @@alt="PyPI – latest release"><br/><br/>+ <a href="https://discord.gg/SReshKQjWt"><img src="https://img.shields.io/badge/Discord-5865F2?logo=discord&logoColor=white" alt="Discord"></a><a href="https://www.youtube.com/@PipelexAI"><img src="https://img.shields.io/badge/YouTube-FF0000?logo=youtube&logoColor=white" alt="YouTube"></a><a href="https://pipelex.com"><img src="https://img.shields.io/badge/Homepage-03bb95?logo=google-chrome&logoColor=white&style=flat" alt="Website"></a>- <a href="https://github.com/Pipelex/pipelex-cookbook"><img src="https://img.shields.io/badge/Cookbook-03bb95?logo=github&logoColor=white&style=flat" alt="Cookbook"></a>- <a href="https://discord.gg/SReshKQjWt"><img src="https://img.shields.io/badge/Discord-5865F2?logo=discord&logoColor=white" alt="Discord"></a>+ <a href="https://github.com/Pipelex/pipelex-cookbook"><img src="https://img.shields.io/badge/Cookbook-5a0dad?logo=github&logoColor=white&style=flat" alt="Cookbook"></a>+ <a href="https://docs.pipelex.com/"><img src="https://img.shields.io/badge/Docs-03bb95?logo=read-the-docs&logoColor=white&style=flat" alt="Documentation"></a>+ <a href="https://docs.pipelex.com/changelog/"><img src="https://img.shields.io/badge/Changelog-03bb95?logo=git&logoColor=white&style=flat" alt="Changelog"></a><br/><br/></div>@@ -81,7 +83,7 @@ Pipelex is an **open-source Python library** with a hosted API launching soon. I# 🚀 Quick start-> :books: Note that you can check out the [Pipelex Documentation](doc/Documentation.md) for more information and clone the [Pipelex Cookbook](https://github.com/Pipelex/pipelex-cookbook) repository for ready-to-run samples.+> :books: Note that you can check out the [Pipelex Documentation](docs/Documentation.md) for more information and clone the [Pipelex Cookbook](https://github.com/Pipelex/pipelex-cookbook) repository for ready-to-run samples.Follow these steps to get started:@@ -92,7 +94,12 @@ Follow these steps to get started:- Python ≥3.10- [pip](https://pip.pypa.io/en/stable/), [poetry](https://python-poetry.org/), or [uv](https://github.com/astral-sh/uv) package manager-### Install the package+### Option #1: Run examples++Visit the+[](https://github.com/Pipelex/pipelex-cookbook/): you can clone it, fork it, play with it++### Option #2: Install the package```bash# Using pip@@ -136,6 +143,130 @@ Using `uv`:uv pip install "pipelex[anthropic,google,mistralai,bedrock,fal]"```+---++## Example: optimizing a tweet in 2 steps++### 1. Define the pipeline in TOML++```toml+domain = "tech_tweet"+definition = "A pipeline for optimizing tech tweets using Twitter/X best practices"++[concept]+DraftTweet = "A draft version of a tech tweet that needs optimization"+OptimizedTweet = "A tweet optimized for Twitter/X engagement following best practices"+TweetAnalysis = "Analysis of the tweet's structure and potential improvements"+WritingStyle = "A style of writing"++[pipe]+[pipe.analyze_tweet]+PipeLLM = "Analyze the draft tweet and identify areas for improvement"+inputs = { draft_tweet = "DraftTweet" }+output = "TweetAnalysis"+llm = "llm_for_writing_analysis"+system_prompt = """+You are an expert in social media optimization, particularly for tech content on Twitter/X.+Your role is to analyze tech tweets and check if they display typical startup communication pitfalls.+"""+prompt_template = """+Evaluate the tweet for these key issues:++**Fluffiness** - Overuse of buzzwords without concrete meaning (e.g., "synergizing disruptive paradigms")++**Cringiness** - Content that induces secondhand embarrassment (overly enthusiastic, trying too hard to be cool, excessive emoji use)++**Humblebragginess** - Disguising boasts as casual updates or false modesty ("just happened to close our $ 10M round 🤷")++**Vagueness** - Failing to clearly communicate what the product/service actually does++For each criterion, provide:+1. A score (1-5) where 1 = not present, 5 = severely present+2. If the problem is not present, no comment. Otherwise, explain of the issue and give concise guidance on fixing it, without providing an actual rewrite++@draft_tweet++"""++[pipe.optimize_tweet]+PipeLLM = "Optimize the tweet based on the analysis"+inputs = { draft_tweet = "DraftTweet", tweet_analysis = "TweetAnalysis", writing_style = "WritingStyle" }+output = "OptimizedTweet"+llm = "llm_for_social_post_writing"+system_prompt = """+You are an expert in writing engaging tech tweets that drive meaningful discussions and engagement.+Your goal is to rewrite tweets to be impactful and avoid the pitfalls identified in the analysis.+"""+prompt_template = """+Rewrite this tech tweet to be more engaging and effective, based on the analysis:++Original tweet:+@draft_tweet++Analysis:+@tweet_analysis++Requirements:+- Include a clear call-to-action+- Make it engaging and shareable+- Use clear, concise language++### Reference style example++@writing_style++### Additional style instructions++No hashtags.+Minimal emojis.+Keep the core meaning of the original tweet.+"""++[pipe.optimize_tweet_sequence]+PipeSequence = "Analyze and optimize a tech tweet in sequence"+inputs = { draft_tweet = "DraftTweet", writing_style = "WritingStyle" }+output = "OptimizedTweet"+steps = [+ { pipe = "analyze_tweet", result = "tweet_analysis" },+ { pipe = "optimize_tweet", result = "optimized_tweet" },+]+```++### 2. Run the pipeline++Here is the flowchart generated during this run:+```mermaid+---+config:+ layout: dagre+ theme: base+---+flowchart LR+ subgraph "optimize_tweet_sequence"+ direction LR+ FGunn["draft_tweet:<br>**Draft tweet**"]+ EWhtJ["tweet_analysis:<br>**Tweet analysis**"]+ 65Eb2["optimized_tweet:<br>**Optimized tweet**"]+ i34D5["writing_style:<br>**Writing style**"]+ end+class optimize_tweet_sequence sub_a;++ classDef sub_a fill:#e6f5ff,color:#333,stroke:#333;++ classDef sub_b fill:#fff5f7,color:#333,stroke:#333;++ classDef sub_c fill:#f0fff0,color:#333,stroke:#333;+ FGunn -- "Analyze tweet" ----> EWhtJ+ FGunn -- "Optimize tweet" ----> 65Eb2+ EWhtJ -- "Optimize tweet" ----> 65Eb2+ i34D5 -- "Optimize tweet" ----> 65Eb2+```+++### 3. wait… no, there is not 3, you're done!++---+## 🤝 ContributingWe welcome contributions! Please see our [Contributing Guidelines](CONTRIBUTING.md) for details on how to get started, including development setup and testing information.@@ -150,7 +281,7 @@ Join our vibrant Discord community to connect with other developers, share your- **GitHub Issues**: For bug reports and feature requests- **Discussions**: For questions and community discussions-- [**Documentation**](doc/Documentation.md)+- [**Documentation**](docs/Documentation.md)## ⭐ Star Us!diff --git a/doc/Documentation.md b/doc/Documentation.mddeleted file mode 100644index 337f0bad1..000000000--- a/doc/Documentation.md+++ /dev/null@@ -1,24 +0,0 @@-# Documentation--Pipelex is a Python framework for defining and running LLM-based knowledge pipelines with high reliability and replicability.--## Table of Contents--[Installation](pages/Installation/Installation.md)--[Quick-start](pages/Quick-start/Quick-start.md)--[Pipelex paradigm](pages/Pipelex%20paradigm/Pipelex%20paradigm.md)--[Libraries](pages/Libraries/libraries.md)--[Concepts](pages/Concepts/Concepts.md)--[Pipes](pages/Pipes/Pipes.md)--------"Pipelex" is a trademark of Evotis S.A.S.--© 2025 Evotis S.A.S.diff --git a/doc/pages/Concepts/Concepts.md b/doc/pages/Concepts/Concepts.mddeleted file mode 100644index 3ebd06184..000000000--- a/doc/pages/Concepts/Concepts.md+++ /dev/null@@ -1,184 +0,0 @@-# Concepts--A concept **defines** **what is processed** through pipes: they represent the **knowledge** that is processed as input and generated as output in Pipelex pipelines.--**Rigorous concept definitions are the key to ensure controlled, reliable outputs in pipelines.**--Concepts are defined both with **natural language** (so non-developers and LLMs can understand and write them) and optionally also **with code** (to ensure deterministic validation with software).--Based on concepts, pipes embed **validation** on their inputs and outputs, so **we make sure of the knowledge we process and produce**.--## Concept definition--Concepts are defined using a **`toml` syntax** and (optional) **Python code ([BaseModel](https://docs.pydantic.dev/latest/concepts/models/)).**--- The `toml` part is meant to be **written in a library `toml` file.**-- - 💡*In the quick-start example (character generator) this is the role of `character.toml`[.](https://www.notion.so/Pipelex-Documentation-1b34fbe82c898062af44d70f6aa0f461?pvs=21)*--- The python code related to the Concept definition **must be saved in a `pipelex_libraries` folder at the root of your project.**--As seen in the quick-start, we use Python to instantiate concepts into Stuff objects.----_Concepts can be defined both with toml and code. Stuff are instantiated from Concepts with Python code._--## Concept with text--In the 1st quick-start example, `character.toml` includes a text-based definition of a Concept (Character).--It is a good example of how to define a Concept. Just describe what the stuff is.--```toml-# Library header-domain = "character"-definition = "Tools for creating characters"--[concept]-Character = "A detailed description of a character"-```--## Structured concepts with text and code--Once you are familiar with defining Concepts with text, you may be interested in using a BaseModel to improve LLM reliability by structuring the knowledge you manipulate.--**_This is mandatory if you want your outputs to have a controlled structure and benefit from Pipe validation_**--For that purpose, you will need both `toml` and Python.--In this example, we define a `Summary` structured concept. It consists of a text and a creation date.--```toml-# summarize.toml--domain = "summarize"-definition = "Summarize text using LLMs."--[concept]-Summary = "A concise and organized rewriting of a large and dense text."-```--```python-# summarize_models.py--from datetime import datetime-from pydantic import Field-from pipelex.core.stuff_content import StructuredContent--# Class definition, inherits from StructuredContent-class Summary(StructuredContent):- text: str- created_at: datetime = Field(default_factory=datetime.now)-```--:bulb: We take benefit of the `Field` pydantic class to automatically set the `created_at` attribute to the current date and time.--**Super important: How do you bind Models and Concepts?**--:right_arrow: They have the same name! `Summary` and `Summary`--💡 When you define a BaseModel for one of your concepts, **make sure it inherits from a StructuredContent class.**--## Fields description--Defining a list of concepts is done in the `toml` file. Just define a list of concepts with their definition.--```toml-[concept] # Required-Concept1 = "Concept1 definition with Natural Language" # Required, str-Concept2 = "Concept2 definition with Natural Language" # Required, str-...-ConceptN = "ConceptN definition with Natural Language" # Required, str-```--- `Concept` _(required, str)_: The Concept definition.--## Usage in code--A concept instance is called a `Stuff`. Once instantiated **it can be loaded into the working memory and processed by Pipes.**--### Stuff overview--**A stuff object** has:--- **metadata** (code, domain, concept_code, …)-- **a content object** (that stores the data)--For your information, this is what the Stuff and StuffContent classes look like:--- **Stuff class (optional attributes removed)**--```python-class Stuff(BaseModel):- concept_code: str # To indicate what concept to instantiate: <domain>.<concept_name>- stuff_name: str # To identify the stuff- content: StuffContent # To store the data-```--- **StuffContent classes:**--Stuff content objects are instances of a subclass of the StuffContent class. They are very useful when you want to manipulate specific objects and render them.--```python-# Base Classes-class StuffContent(ABC, BaseModel):- """Base class for all content types. Provides methods to render content in different formats- (plain text, HTML, Markdown, JSON, spreadsheet)."""-```--### Make stuffs for text-only Concepts--In this example we load a contract content and make a stuff from it. We assume that a legal.Contract concept already exists.--```python-from pipelex.core.stuff_factory import StuffFactory--with open("contract.txt", "r") as file:- contract_content = file.read()--question_stuff = StuffFactory.make_from_str(- concept_code="legal.Contract", # <domain>.<concept_name>- str_value=contract_content, # The str content to add to the stuff- stuff_name="contract", # The name of the stuff-)-```--### Make stuffs for structured Concepts--Make stuff easily using the StuffFactory class.--```python-from pipelex.core.stuff_factory import StuffFactory-from pipelex_libraries.character_model import Character---character = Character(- name="Elias",- age=38,- gender="man",- occupation="unemployed",- description="""Elias Varrin is a 38-year-old man, standing at approximately 1.85 meters tall, with a lean,-weathered frame shaped by decades of travel through remote and often unforgiving landscapes.-His name, though not widely known, carries weight among historians, explorers, and those who trade in whispered legends.-Elias has piercing storm-gray eyes that scan every environment with sharp precision, and his ash-blond hair—flecked with-early streaks of grey—is usually tucked beneath a wide-brimmed, timeworn hat.His hands are etched with fine scars and stained-with ink, each mark a silent record of years spent charting unrecorded lands and handling fragile relics of lost civilizations.-He moves with quiet purpose and speaks with a calm, thoughtful cadence that suggests he’s always listening for more than just what’s said.""",-)-character_stuff = StuffFactory.make_stuff(- concept_code="character.Character",- stuff_name="character",- content=character,-)-```--:arrow_left: [**Back to Libraries**](../Libraries/libraries.md)--:arrow_right: [**Next section: Pipes**](../Pipes/Pipes.md)-------"Pipelex" is a trademark of Evotis S.A.S.--© 2025 Evotis S.A.S.diff --git a/doc/pages/Concepts/concept_and_stuff.png b/doc/pages/Concepts/concept_and_stuff.pngdeleted file mode 100644index e91ef9c26..000000000Binary files a/doc/pages/Concepts/concept_and_stuff.png and /dev/null differdiff --git a/doc/pages/Installation/Installation.md b/doc/pages/Installation/Installation.mddeleted file mode 100644index e69ee5fb9..000000000--- a/doc/pages/Installation/Installation.md+++ /dev/null@@ -1,42 +0,0 @@-# Installation--## 📋 Prerequisites--Pipelex requires `python ≥3.10`, and access to LLM api keys, if you want to run pipelines locally, or an API key provided by Pipelex to use our cloud-hosted service.--## 🧭 Getting Started--Everything you need to discover and test Pipelex is available in our **Getting Started repository**.-Along with [our Documentation](../Quick-start/Quick-start.md), we recommend you review it before any further usage: [Cookbook](https://github.com/Pipelex/pipelex-cookbook).--- **Install Pipelex using**--```bash-pip install pipelex-```--- **Make sure you have a .env** file at the root of your project that contains the following fields--```bash-OPENAI_API_KEY=sk_...-```--- **Make sure your project looks like this:**--```bash-<your_repo>-├── pipelex.toml # the config file for pipelex-├── .env # the .env file to put your API keys and other pipelex variables in-├── pipelex_libraries # a directory to store your pipelex related code-│ ├── __init__.py-```--💡 _Any troubles? Have a look at our [Cookbook](https://github.com/Pipelex/pipelex-cookbook)!_--:arrow_right: [**Next section: Quick-start**](../Quick-start/Quick-start.md)-------"Pipelex" is a trademark of Evotis S.A.S.--© 2025 Evotis S.A.S.diff --git a/doc/pages/Libraries/libraries.md b/doc/pages/Libraries/libraries.mddeleted file mode 100644index fc092b866..000000000--- a/doc/pages/Libraries/libraries.md+++ /dev/null@@ -1,58 +0,0 @@-# Libraries--Pipelex proposes a Python framework along with a syntax for defining and running knowledge processing workflows.--## Scripts--What we call _scripts_ are the `TOML` files containing the concept and pipe definitions written using the Pipelex syntax.--## `pipelex_libraries` folder--Concepts, Pipelines and code are meant to be stored in a `pipelex_libraries` folder at the root of your project.--## Pipelex Libraries--Within a consistent **domain** (domain is the Pipelex name for topic), scripts constitute a library.--A library is a set of scripts along with their corresponding Python code (used for structuring LLM outputs through structured Concepts).--## Library structure--A library is defined by a `toml` file in a `pipelex_libraries` folder.--The `toml` file contains the concepts and pipes definitions.--It looks like this:----## Fields--### Domain definition--As a library `toml` file header, you need to define the domain of the library.--```toml-domain = "domain_name"-definition = "Definition of the domain"-```--:bulb: Optional global settings (such as `system_prompt`) can be defined at the library level, in the header.--### Concepts--See [Concepts](../Concepts/Concepts.md)--### Pipes--See [Pipes](../Pipes/Pipes.md)--:arrow_left: [**Back to Quick-start**](../Quick-start/Quick-start.md)--:arrow_right: [**Next section: Concepts**](../Concepts/Concepts.md)-------"Pipelex" is a trademark of Evotis S.A.S.--© 2025 Evotis S.A.S.diff --git a/doc/pages/Libraries/library_structure.png b/doc/pages/Libraries/library_structure.pngdeleted file mode 100644index 93a543691..000000000Binary files a/doc/pages/Libraries/library_structure.png and /dev/null differdiff --git a/doc/pages/Pipelex paradigm/Pipelex paradigm.md b/doc/pages/Pipelex paradigm/Pipelex paradigm.mddeleted file mode 100644index a777d6308..000000000--- a/doc/pages/Pipelex paradigm/Pipelex paradigm.md+++ /dev/null@@ -1,73 +0,0 @@-# Pipelex paradigm--Pipelex is a **Python framework** for **defining** and **running** **LLM-based knowledge pipelines** with high **reliability** and **replicability**.--As an **LLM scaffolding middleware**, Pipelex introduces relevant mechanisms to leverage the capabilities of these ai models while ensuring their **production readiness.**--**_Before going further with Pipelex, we suggest you have a look at the most important mechanisms._**--## TLDR--The Pipelex framework **processes knowledge through pipelines**, which consist of **sequential knowledge processing units called pipes**. Pipes access other pipes' outputs via the **Working Memory** system, making **processed knowledge available throughout the pipeline**. To ensure **reliability**, we implement **deterministic validation**: all pipe inputs and outputs (defined as Stuffs) **conform to specific concepts defined by the user (or by AI)**.--## Knowledge--Knowledge refers to information **extracted**, **processed**, or **organized** from various data sources such as documents, PDFs, images, or information output by our pipelines. It has been structured in a way that enables **understanding**, **interpretation**, and **actionable insights**.--## Pipeline--A pipeline is the Pipelex term for "Workflow". **It aims at processing knowledge using organized, consistent, validated steps.**----:bulb: **Remember the Quick-start chapter?** We defined a pipeline (in toml, using the [pipe] section) to generate a character!--## Pipe--A pipe is an **elementary step in a pipeline**. A pipeline is made of pipes.-Each pipe can integrate **both LLM-based** or software-based knowledge processing.----:bulb: **Remember the Quick-start chapter?** We defined a pipe (in toml, using the [pipe.create_character] section) to generate a character. This was a very short pipeline as it comprised only one pipe.--## Working Memory--A pipe has access to all inputs as well as all the knowledge that was processed along the pipeline: **everything is stored in the Pipeline's Working Memory**. So can then **use all that stuff at any point in the pipeline.**---_Working Memory: Pipe C can use Pipe A and Pipe B outputs. Pipe D can use Pipe C and Pipe B outputs, Pipe E …_--:bulb: **Remember the Quick-start chapter?** We used the working memory to store a character before processing it (and extract metadata from it).--## Concepts--A concept **defines** **what is processed** through pipes.--**Rigorous concept definitions are the key to ensure controlled, reliable outputs.**--Concepts are defined both with **natural language** (so non-developer and LLM can understand and write them) and **with code** (to validate them deterministically with software).--Pipes embed **validation** of their outputs, so **we make sure of the knowledge we process and produce**.-----:bulb: **Remember the Quick-start chapter?** We defined a `Character` concept and a `CharacterMetadata` concept (in toml, using the [concept] section and with Python BaseModels) to define the input and output of the pipe!--:bulb: Soon we will also add input validation for pipes.---## Stuff--On the code side, **when a concept is instantiated** (to be processed or added to the working memory), **we call it a Stuff**.--:arrow_left: [**Back to Quick-start**](../Quick-start/Quick-start.md)--:arrow_right: [**Next section: Libraries**](../Libraries/libraries.md)-------"Pipelex" is a trademark of Evotis S.A.S.--© 2025 Evotis S.A.S.diff --git a/doc/pages/Pipelex paradigm/output_validation.png b/doc/pages/Pipelex paradigm/output_validation.pngdeleted file mode 100644index 6a4e68605..000000000Binary files a/doc/pages/Pipelex paradigm/output_validation.png and /dev/null differdiff --git a/doc/pages/Pipelex paradigm/pipe.png b/doc/pages/Pipelex paradigm/pipe.pngdeleted file mode 100644index f564f0cc1..000000000Binary files a/doc/pages/Pipelex paradigm/pipe.png and /dev/null differdiff --git a/doc/pages/Pipelex paradigm/pipeline.png b/doc/pages/Pipelex paradigm/pipeline.pngdeleted file mode 100644index c8f90f084..000000000Binary files a/doc/pages/Pipelex paradigm/pipeline.png and /dev/null differdiff --git a/doc/pages/Pipelex paradigm/working_memory.png b/doc/pages/Pipelex paradigm/working_memory.pngdeleted file mode 100644index bf5f43b6f..000000000Binary files a/doc/pages/Pipelex paradigm/working_memory.png and /dev/null differdiff --git a/doc/pages/Pipes/Pipes.md b/doc/pages/Pipes/Pipes.mddeleted file mode 100644index fb68c658b..000000000--- a/doc/pages/Pipes/Pipes.md+++ /dev/null@@ -1,357 +0,0 @@-# Pipes--A pipe is a **Pipeline step**.-It can integrate **both LLM-based** or software-based knowledge processing.--:bulb: **Remember the Quick-start chapter?** We defined a pipe (in toml, using the [pipe.create_character] section) to generate a character. It constituted a one-pipe-long pipeline.--## Define pipes--Like concepts, Pipes are defined using a **`toml` syntax.**--- This part is meant to be **written in a library `toml` file, in the same one as concepts** (see [Libraries](../Libraries/library.md)).- 💡*In the quick-start example (text summary generator) this is the role of `summarize.toml`.*--### General case--This is how to define a Pipe using the Pipelex `toml` syntax.----#### **Fields definition**--```toml-[pipe]-[pipe.<pipe_name>]-Pipe<Type> = "Pipe definition" # required, str-input = "InputConcept" # required, str-output = "OutputConcept" # required, str-... then come the Pipe specific fields-```--- `Pipe<Type>`(str, required) The presence of this field indicates what type of pipe your are defining and adds a specific definition for it (explaining what it does).-- Productive Pipes:- - `PipeLLM`: A call to an LLM whihc generates either text or a structured object, or a list of those.- - `PipeJinja2`: Application of a Jinja2 template to generate Text.-- Control Pipes, to orchestrate the flow of the pipeline:-- - `PipeSequence`: Sequence of Pipes that share a common working memory- - `PipeParallel`: Executes several pipes in parallel (asynchronously) and output a list of the results- - `PipeBatch`: Executes the same pipe on each item of a list and output a list of the results- - `PipeCondition`: Evaluate a test expression and choose which pipe to run according to the test's result--- `input` (str, required) The expected input Concept (PascalCase)-- `output` (str, required) The expected output Concept (PascalCase)-- `additional specific fields` depending on the kind of Pipe you define.--### Type 1: Pipe LLM--If you specifically want to call an LLM and benefit from Pipelex scaffolding, PipeLLM is made for you.--Pipe LLM embeds all you need to run reliable LLM calls in your pipeline.--#### **PipeLLM fields description**--```toml-[pipe.<pipe_name>]-PipeLLM = "Pipe definition" # required, str-input = "InputConcept" # required, str-output = "OutputConcept" # required, str-llm = "llm_preset_name" # optional, str-llm_to_structure = "llm_preset_name" # optional, str-llm_to_structure_direct = "llm_preset_name" # optional, str-system_prompt = """system prompt""" # optional, str-prompt_template = """jinja2 prompt template""" # required, str-images = [] # optional-```--- `PipeLLM`(str, required) The presence of this field indicates that your pipe is a PipeLLM and stores a definition of it (explaining what it does).-- `input` (str, required) The expected input Concept (PascalCase)-- `output` (str, required) The expected output Concept (PascalCase)-- `llm` (str, optional) What llm_preset to use for text generation-- `llm_to_structure` (str, optional) What llm_preset to use for structured object generation-- `llm_to_structure_direct` (str, optional) What llm_preset to use for structured object list generation-- `system_prompt` (str, optional) system prompt used in the LLM call-- `prompt_template` (str, optional) prompt used in the LLM call. This can be a jinja template. See #Jinja-- `images` (List[str], optional) A list of stuff names that contains images.--#### **Example**--This pipe outputs an html table from a table screenshot. It basically converts an image of a table into the html representation of it.--```toml-[pipe.get_html_table_from_image]-PipeLLM = "Get an HTML table"-input = "TableScreenshot"-output = "HtmlTable"-images = ["table_screenshot"]-system_prompt = """-You are a vision-based table extractor.-"""-prompt_template = """-You are given a picture (screenshot) of a table, taken from a PDF document. Your goal is to extract the table from the image **in html**.-Make sure you do not forget any text. Make sure you do not invent any text. Make sure your merge is consistent. Make sure you replicate the formatting (borders, text formatting, colors, text alignment...)-"""-llm = "llm_to_extract_tables"-llm_to_structure = "llm_to_extract_tables"-```--### Type 2: PipeSequence--PipeSequence embeds all you need to run multi-steps processing with a sequence of pipes.--#### **PipeSequence fields description**--- `PipeSequence` (str, required) The presence of this field indicates that your pipe is a PipeSequence and stores a definition of it (explaining what it does).-- `input` (str, required) The expected input Concept (PascalCase)-- `output` (str, required) The expected output Concept (PascalCase)-- `steps` (List[Dict], required) The steps in the Pipe- - `result` (str, required) The name of the output stuff- - `pipe` (str, required) The `pipe_code` of the Pipe used for this step.--```toml-[pipe.<pipe_name>]-PipeSequence = "Pipe definition" # required, str-input = "InputConcept" # required, str-output = "OutputConcept" # required, str-steps = [ # required, List[Dict]- { result = "pipe_1_output_name", pipe = "pipe_1" },- { result = "pipe_2_output_name", pipe = "pipe_2" },- ...,- {result = "pipe_n_output_name", pipe = "pipe_n" },-]-```--#### **PipeSequence example**--```toml-[pipe.answer_question_with_instructions]-PipeSequence = "Answer a question with instructions"-input = "Question"-output = "FormattedAnswer"-steps = [-{ result = "instructions", pipe = "enrich_instructions" },-{ result = "answer", pipe = "answer_enriched_question" },-{ result = "formatted_answer", pipe = "format_answer" },-]-```--#### **PipeParallel example**--```toml-[pipe.extract_expense_report]-PipeParallel = "Extract useful information from an expense report"-input = "ExpenseReportText"-output = "Composite"-parallels = [- { pipe = "extract_employee_from_expense_report", result = "employee" },- { pipe = "extract_expenses_from_expense_report", result = "expenses" },-]-```--#### **PipeCondition example**--```toml-[pipe.expense_conditional_validation]-PipeCondition = "Choose the rules to apply"-input = "Expense"-output = "RulesToApply"-expression = "expense_category.category"-```--## Working Memory----In a pipeline, processed Stuff are stored in the **Working** **Memory.** The working memory is accessible from any Pipe in the pipeline.--Basically, the Working Memory is a wrapper on a Dict of Stuff objects.--```python-StuffDict = Dict[str, Stuff]--class WorkingMemory(BaseModel):- root: StuffDict = Field(default_factory=dict)-```--**You can easily preload the memory with the dedicated Factory**--```python-from pipelex.core.stuff_factory import StuffFactory-from pipelex.core.working_memory_factory import WorkingMemoryFactory--# Here is a Stuff object-table_screenshot_stuff = StuffFactory.make_from_str(- name="table_screenshot",- concept_code="TableScreenshot",- str_value=table_screenshot,-)--# And we load it in the memory-working_memory = WorkingMemoryFactory.make_from_single_stuff(- table_screenshot_stuff,-)-```--### Access memory in prompts--You can access working memory stuffs directly in prompts using the jinja2 syntax.--You just need to call them by their name.--_PS: what comes after the | (stuff("extract") for instance) will be explained later._--```toml-[pipe.get_answer_with_extract]-PipeLLM = "Answer the question with extract"-input = "QuestionWithExtract"-output = "AnswerToAQuestionWithExtract"-prompt_template = """-I am asking you to read an extract and answer a question about it.-{{ question_with_extract|tag("extract") }} # question_with_extract-{{ question_with_extract|tag("question") }}-Please return your answer in english.-"""-```--## Run Pipes--Running pipes requires:--### **A Pipe**--```toml-[concept]-[concept.TableImage]-Concept = "An image of a table, in the sense of a data structure used to organize information in rows and columns."-structure = "ImageContent"-refines = ["Image"]--[concept.HtmlTable]-Concept = "An HTML table"-structure = "HtmlTable"--[pipe]-[pipe.get_html_table_from_image]-PipeLLM = "Get an HTML table"-input = "TableImage"-output = "HtmlTable"-images = ["table_image"]-system_prompt = """You are a vision-based table extractor."""-prompt_template = """You are given a picture (screenshot) of a table,-taken from a PDF document.-Your goal is to extract the table from the image **in html**.-Make sure you do not forget any text.-Make sure you do not invent any text.-Make sure your merge is consistent.-Make sure you replicate the formatting (borders, text formatting, colors, text alignment...)-"""--[pipe.review_html_table]-PipeLLM = "Review an HTML table"-input = "HtmlTable"-output = "HtmlTable"-images = ["table_image"]-system_prompt = "You are a vision-based html checker."-prompt_template = """Your role is to correct an html_table to make sure that it matches the one in the provided image.-Here is the html table:{{ html_table|tag("html_table") }}-Pay attention to the text and formatting (color, borders, ...).-Rewrite the entire html table with your potential corrections.-Make sure you do not forget any text.-"""--[pipe.get_html_table_and_review]-PipeSequence = "Get an HTML table and review it"-input = "TableScreenshot"-output = "HtmlTable"-steps = [- { pipe = "get_html_table_from_image", result = "html_table" },- { pipe = "review_html_table", result = "reviewed_html_table" },-]-```--### Loading the Working Memory with the required Pipes and running it--TODO--## Run pipes with any LLM--How to use a specific LLM for a specific pipe?--### **`.env`**--Make sure you have the appropriate API key(s) in your .env.--| **LLM provider** | **API KEY NAME** |-| ---------------- | ----------------------- |-| Anthropic | `ANTHROPIC_API_KEY` |-| Azure OpenAI | `AZURE_OPENAI_API_KEY` |-| AWS | `AWS_ACCESS_KEY_ID` |-| AWS | `AWS_SECRET_ACCESS_KEY` |-| Mistral AI | `MISTRAL_API_KEY` |-| OpenAI | `OPENAI_API_KEY` |--### **`config_<your_workdir_name>_llm_presets.toml`**--Create LLM presets to specify your model settings:--```python-[cogt.llm_config.llm_deck.llm_presets]-llm_for_reviewing_code = { llm_handle = "gpt-4-turbo", prompting_target = "openai", temperature = 0.1, max_tokens = "auto" }-```--- `llm_preset_<your_name_for_this_llm_preset>`(optional): Define LLM presets to store your LLM settings and use them when running pipes- - `llm_handle` (required, str) the codename for the LLM-- `prompting_target` (required, str) indicates what prompting style to apply when processing your input prompt_templates.-- `temperature`(required, float) LLM temperature-- `max_tokens` (required, int or "auto") auto means maximum value provided by the LLM provider--#### Available LLMs--- **Anthropic**--| **Model Family** | **Model** | **LLM handle** |-| ---------------- | -------------------- | ---------------------- |-| claude-3 | Claude 3 Haiku | `claude-3-haiku` |-| claude-3 | Claude 3 Opus | `claude-3-opus` |-| claude-3.5 | Claude 3.5 Sonnet | `claude-3-5-sonnet` |-| claude-3.5 | Claude 3.5 Sonnet v2 | `claude-3-5-sonnet-v2` |-| claude-3.7 | Claude 3.7 Sonnet | `claude-3-7-sonnet` |--- **Mistral AI**--| **Model Family** | **Model** | **LLM handle** |-| ---------------- | ------------- | ------------------- |-| Ministral | Ministral 3B | `ministral-3b` |-| Ministral | Ministral 8B | `ministral-8b` |-| Mistral | Mistral 7B | `mistral-7b` |-| Mistral | Mixtral 8x7B | `mistral-8x7b` |-| Mistral | Codestral | `mistral-codestral` |-| Mistral | Mistral Large | `mistral-large` |-| Mistral | Mistral Small | `mistral-small` |-| Pixtral | Pixtral 12B | `pixtral-12b` |-| Pixtral | Pixtral Large | `pixtral-large` |--- **Open AI**--| **Model Family** | **Model** | **LLM handle** |-| ---------------- | ---------------------- | ------------------------ |-| GPT-4 | GPT-4 Turbo | `gpt-4-turbo` |-| GPT-4 | GPT-4o (May 2024) | `gpt-4o-2024-05-13` |-| GPT-4 | GPT-4o (Aug 2024) | `gpt-4o-2024-08-06` |-| GPT-4 | GPT-4o (Nov 2024) | `gpt-4o-2024-11-20` |-| GPT-4 | GPT-4o | `gpt-4o` |-| GPT-4 | GPT-4o Mini (Jul 2024) | `gpt-4o-mini-2024-07-18` |-| GPT-4 | GPT-4o Mini | `gpt-4o-mini` |-| GPT-4.5 | GPT-4.5 Preview | `gpt-4.5-preview` |-| o | o1-mini | `o1-mini` |-| o | o1 | `o1` |-| o | o3-mini | `o3-mini` |--:arrow_left: [**Back to Concepts**](../Concepts/Concepts.md)-------"Pipelex" is a trademark of Evotis S.A.S.--© 2025 Evotis S.A.S.diff --git a/doc/pages/Pipes/working_memory_cloud.png b/doc/pages/Pipes/working_memory_cloud.pngdeleted file mode 100644index 085960451..000000000Binary files a/doc/pages/Pipes/working_memory_cloud.png and /dev/null differdiff --git a/doc/templates/pipes.toml b/doc/templates/pipes.tomldeleted file mode 100644index edea6cfba..000000000--- a/doc/templates/pipes.toml+++ /dev/null@@ -1,78 +0,0 @@---domain = "template"-definition = "Template file demonstrating different pipe types in Pipelex"--[concept]-InputText = "A text input for processing"-OutputText = "A processed text output"-TableScreenshot = "An image of a table"-HtmlTable = "An HTML table representation"-Question = "A question to be answered"-FormattedAnswer = "A formatted answer to a question"-ExpenseReportText = "Text content of an expense report"-Composite = "A composite output containing multiple pieces of information"-Expense = "An expense record"-RulesToApply = "Rules to be applied to an expense"--[pipe]-# Example of a PipeLLM-[pipe.process_text]-PipeLLM = "Process input text using an LLM"-input = "Text"-output = "Text"-llm = "llm_to_summarize_text"-system_prompt = """-You are an expert in text summarization.-"""-prompt_template = """-Summarize the following text:-@text-"""--# Example of a PipeLLM with image processing-[pipe.get_html_table_from_image]-PipeLLM = "Convert table screenshot to HTML"-input = "TableScreenshot"-output = "HtmlTable"-images = ["table_screenshot"]-system_prompt = """-You are a vision-based table extractor.-"""-prompt_template = """-You are given a picture (screenshot) of a table, taken from a PDF document.-Your goal is to extract the table from the image **in html**.-Make sure you do not forget any text. Make sure you do not invent any text.-Make sure your merge is consistent. Make sure you replicate the formatting-(borders, text formatting, colors, text alignment...)-"""-llm = "llm_to_extract_tables"-llm_to_structure = "llm_to_extract_tables"--# Example of a PipeSequence-[pipe.answer_question_with_instructions]-PipeSequence = "Answer a question with instructions"-input = "Question"-output = "FormattedAnswer"-steps = [- { result = "instructions", pipe = "enrich_instructions" },- { result = "answer", pipe = "answer_enriched_question" },- { result = "formatted_answer", pipe = "format_answer" },-]--# Example of a PipeParallel-[pipe.extract_expense_report]-PipeParallel = "Extract useful information from an expense report"-input = "ExpenseReportText"-output = "Composite"-parallels = [- { pipe = "extract_employee_from_expense_report", result = "employee" },- { pipe = "extract_expenses_from_expense_report", result = "expenses" },-]--# Example of a PipeCondition-[pipe.expense_conditional_validation]-PipeCondition = "Choose the rules to apply"-input = "Expense"-output = "RulesToApply"-expression = "expense_category.category"diff --git a/docs/.meta.yml b/docs/.meta.ymlnew file mode 100644index 000000000..e3cf63b09--- /dev/null+++ b/docs/.meta.yml@@ -0,0 +1,3 @@+image: https://d2cinlfp2qnig1.cloudfront.net/docs/pipelex_banner_1200x630.png?v=1+description: Docs for Pipelex, open-source language for repeatable AI workflows.+keywords: [pipelex, ai, llm, workflows, knowledge pipelines, pipelines, deterministic, repeatable, framework, language]diff --git a/docs/CNAME b/docs/CNAMEnew file mode 100644index 000000000..84b574cb1--- /dev/null+++ b/docs/CNAME@@ -0,0 +1 @@+docs.pipelex.com\ No newline at end of filediff --git a/docs/CODE_OF_CONDUCT.md b/docs/CODE_OF_CONDUCT.mdnew file mode 100644index 000000000..a9abffea9--- /dev/null+++ b/docs/CODE_OF_CONDUCT.md@@ -0,0 +1,5 @@+---+title: Code of Conduct+---++--8<-- "../CODE_OF_CONDUCT.md"\ No newline at end of filediff --git a/docs/changelog.md b/docs/changelog.mdnew file mode 100644index 000000000..ece2e41b0--- /dev/null+++ b/docs/changelog.md@@ -0,0 +1,5 @@+---+title: Changelog+---++--8<-- "CHANGELOG.md"\ No newline at end of filediff --git a/docs/contributing.md b/docs/contributing.mdnew file mode 100644index 000000000..ec207150d--- /dev/null+++ b/docs/contributing.md@@ -0,0 +1,5 @@+---+title: Contributing to Pipelex+---++--8<-- "CONTRIBUTING.md"diff --git a/docs/images/pipelex-favicon-32.png b/docs/images/pipelex-favicon-32.pngnew file mode 100644index 000000000..9498c80a9Binary files /dev/null and b/docs/images/pipelex-favicon-32.png differdiff --git a/docs/index.md b/docs/index.mdnew file mode 100644index 000000000..fe81ccce6--- /dev/null+++ b/docs/index.md@@ -0,0 +1,22 @@+---+title: "Open-source AI workflows"+---++++# Easily build repeatable AI workflows++Pipelex is an open-source language specifically designed to build reliable, repeatable workflows. Stop wrestling with fragile workflows and start building AI systems you can trust.++## Ready to build something amazing?++**Just want to see it in action?** Get up and running in minutes with our quick-start guide.++[:material-rocket-launch: Install & Quick Start](pages/quick-start/index.md){ .md-button .md-button--primary .equal-width }++**Want to understand the bigger picture?** Discover why Pipelex changes everything about AI development.++[:material-book-open: Read the Manifesto](manifesto.md){ .md-button .equal-width }+[:material-lightbulb: Explore the Paradigm](pages/pipelex-paradigm-for-repeatable-ai-workflows/index.md){ .md-button .equal-width }++diff --git a/docs/license.md b/docs/license.mdnew file mode 100644index 000000000..3ddc5d7b5--- /dev/null+++ b/docs/license.md@@ -0,0 +1,11 @@+---+title: License+hide:+- feedback+---++# License++```+--8<-- "LICENSE"+```\ No newline at end of filediff --git a/docs/manifesto.md b/docs/manifesto.mdnew file mode 100644index 000000000..a5cc3de90--- /dev/null+++ b/docs/manifesto.md@@ -0,0 +1,5 @@+---+title: The Knowledge Pipeline Manifesto+---++--8<-- "MANIFESTO.md"diff --git a/docs/overrides/main.html b/docs/overrides/main.htmlnew file mode 100644index 000000000..5378d5f05--- /dev/null+++ b/docs/overrides/main.html@@ -0,0 +1,19 @@+{# overrides/main.html #}+{% extends "base.html" %}++{% block extrahead %}+{{ super() }} {# keep anything Material already outputs #}+{% if page and page.title and page.meta %}+<meta property="og:type" content="article">+<meta property="og:title" content="{{ page.meta.title or page.title }}">+<meta property="og:description" content="{{ page.meta.description }}">+<meta property="og:image" content="{{ page.meta.image }}">+<meta property="og:image:width" content="1200">+<meta property="og:image:height" content="630">+<meta property="og:url" content="{{ page.canonical_url }}">+<meta name="twitter:card" content="summary_large_image">+<meta name="twitter:title" content="{{ page.meta.title or page.title }}">+<meta name="twitter:description" content="{{ page.meta.description }}">+<meta name="twitter:image" content="{{ page.meta.image }}">+{% endif %}+{% endblock %}\ No newline at end of filediff --git a/docs/pages/advanced-customization/activity-manager-injection.md b/docs/pages/advanced-customization/activity-manager-injection.mdnew file mode 100644index 000000000..700663965--- /dev/null+++ b/docs/pages/advanced-customization/activity-manager-injection.md@@ -0,0 +1,3 @@+# Activity Manager Injection++⚠️ Under constructiondiff --git a/docs/pages/advanced-customization/content-generator-injection.md b/docs/pages/advanced-customization/content-generator-injection.mdnew file mode 100644index 000000000..e6724ce7c--- /dev/null+++ b/docs/pages/advanced-customization/content-generator-injection.md@@ -0,0 +1,3 @@+# Content Generator Injection++⚠️ Under constructiondiff --git a/docs/pages/advanced-customization/index.md b/docs/pages/advanced-customization/index.mdnew file mode 100644index 000000000..a30846b36--- /dev/null+++ b/docs/pages/advanced-customization/index.md@@ -0,0 +1,141 @@+# Dependency Injection++## Overview++Pipelex uses dependency injection to manage service dependencies and make components more modular and testable. The system allows you to customize and extend Pipelex's functionality by injecting your own implementations of various services.++## Injection Methods++There are two main ways to inject custom implementations:++### 1. During Initialization++```python+from pipelex import Pipelex++pipelex = Pipelex(+ template_provider=MyTemplateProvider(),+ llm_model_provider=MyLLMProvider(),+ plugin_manager=MyPluginManager(),+ inference_manager=MyInferenceManager(),+ pipeline_tracker=MyPipelineTracker(),+ activity_manager=MyActivityManager(),+ reporting_delegate=MyReportingDelegate()+)++pipelex.setup(+ secrets_provider=MySecretsProvider(),+ content_generator=MyContentGenerator(),+ pipe_router=MyPipeRouter()+)+```++### 2. Through the Hub++```python+from pipelex.hub import PipelexHub++hub = PipelexHub()+hub.set_template_provider(MyTemplateProvider())+hub.set_llm_models_provider(MyLLMProvider())+hub.set_plugin_manager(MyPluginManager())+# ... and so on for other components+```++## NoOp Implementations++Some components have "NoOp" (No Operation) implementations that are used when the feature is disabled:++- `ReportingNoOp`: Used when reporting is disabled+- `PipelineTrackerNoOp`: Used when pipeline tracking is disabled+- `ActivityManagerNoOp`: Used when activity tracking is disabled++These NoOp implementations implement the same protocol but do nothing, allowing the system to function without the specific feature.++## Protocol Compliance++All custom implementations MUST:++1. Implement ALL methods defined in their respective protocols+2. Match the exact method signatures (parameter names and types)+3. Follow the protocol's documented behavior+4. Handle errors appropriately+5. Clean up resources when needed++## Feature Flags++Some components are controlled by feature flags in the configuration:++- `is_reporting_enabled`: Controls Reporting system+- `is_pipeline_tracking_enabled`: Controls Pipeline Tracking+- `is_activity_tracking_enabled`: Controls Activity Tracking++When a feature is disabled, the corresponding NoOp implementation is used automatically.++## Available Injectable Components++Pipelex supports injection of the following components:++1. **Template Provider** (`TemplateLibrary`)++ - Protocol: `TemplateLibraryProtocol`+ - Default: `TemplateLibrary`+ - [Details](template-provider-injection.md)++2. **LLM Model Provider** (`LLMModelLibrary`)++ - Protocol: `LLMModelLibraryProtocol`+ - Default: `LLMModelLibrary`+ - [Details](llm-model-provider-injection.md)++3. **Plugin Manager** (`PluginManager`)++ - Protocol: `PluginManagerProtocol`+ - Default: `PluginManager`+ - [Details](plugin-manager-injection.md)++4. **Inference Manager** (`InferenceManager`)++ - Protocol: `InferenceManagerProtocol`+ - Default: `InferenceManager`+ - [Details](inference-manager-injection.md)++5. **Reporting Delegate** (`ReportingManager`)++ - Protocol: `ReportingProtocol`+ - Default: `ReportingManager` or `ReportingNoOp` if disabled+ - [Details](reporting-delegate-injection.md)++6. **Pipeline Tracker** (`PipelineTracker`)++ - Protocol: `PipelineTrackerProtocol`+ - Default: `PipelineTracker` or `PipelineTrackerNoOp` if disabled+ - [Details](pipeline-tracker-injection.md)++7. **Activity Manager** (`ActivityManager`)++ - Protocol: `ActivityManagerProtocol`+ - Default: `ActivityManager` or `ActivityManagerNoOp` if disabled+ - [Details](activity-manager-injection.md)++8. **Secrets Provider** (`EnvSecretsProvider`)++ - Protocol: `SecretsProviderProtocol`+ - Default: `EnvSecretsProvider`+ - [Details](secrets-provider-injection.md)++9. **Content Generator** (`ContentGenerator`)++ - Protocol: `ContentGeneratorProtocol`+ - Default: `ContentGenerator`+ - [Details](content-generator-injection.md)++10. **Pipe Router** (`PipeRouter`)++ - Protocol: `PipeRouterProtocol`+ - Default: `PipeRouter`+ - [Details](pipe-router-injection.md)++## Best Practices++⚠️ Under constructiondiff --git a/docs/pages/advanced-customization/inference-manager-injection.md b/docs/pages/advanced-customization/inference-manager-injection.mdnew file mode 100644index 000000000..d051dbc05--- /dev/null+++ b/docs/pages/advanced-customization/inference-manager-injection.md@@ -0,0 +1,3 @@+# Inference Manager Injection++⚠️ Under constructiondiff --git a/docs/pages/advanced-customization/llm-model-provider-injection.md b/docs/pages/advanced-customization/llm-model-provider-injection.mdnew file mode 100644index 000000000..bc63eeb2a--- /dev/null+++ b/docs/pages/advanced-customization/llm-model-provider-injection.md@@ -0,0 +1,3 @@+# LLM Model Provider Injection++⚠️ Under constructiondiff --git a/docs/pages/advanced-customization/pipe-router-injection.md b/docs/pages/advanced-customization/pipe-router-injection.mdnew file mode 100644index 000000000..7c8332535--- /dev/null+++ b/docs/pages/advanced-customization/pipe-router-injection.md@@ -0,0 +1,3 @@+# Pipe Router Injection++⚠️ Under constructiondiff --git a/docs/pages/advanced-customization/pipeline-tracker-injection.md b/docs/pages/advanced-customization/pipeline-tracker-injection.mdnew file mode 100644index 000000000..4c99287d8--- /dev/null+++ b/docs/pages/advanced-customization/pipeline-tracker-injection.md@@ -0,0 +1,3 @@+# Pipeline Tracker Injection++⚠️ Under constructiondiff --git a/docs/pages/advanced-customization/plugin-manager-injection.md b/docs/pages/advanced-customization/plugin-manager-injection.mdnew file mode 100644index 000000000..4811fb0c5--- /dev/null+++ b/docs/pages/advanced-customization/plugin-manager-injection.md@@ -0,0 +1,3 @@+# Plugin Manager Injection++⚠️ Under constructiondiff --git a/docs/pages/advanced-customization/reporting-delegate-injection.md b/docs/pages/advanced-customization/reporting-delegate-injection.mdnew file mode 100644index 000000000..8326a126d--- /dev/null+++ b/docs/pages/advanced-customization/reporting-delegate-injection.md@@ -0,0 +1,3 @@+# Reporting Delegate Injection++⚠️ Under constructiondiff --git a/docs/pages/advanced-customization/secrets-provider-injection.md b/docs/pages/advanced-customization/secrets-provider-injection.mdnew file mode 100644index 000000000..cfeb5cde5--- /dev/null+++ b/docs/pages/advanced-customization/secrets-provider-injection.md@@ -0,0 +1,3 @@+# Secrets Provider Injection++⚠️ Under constructiondiff --git a/docs/pages/advanced-customization/template-provider-injection.md b/docs/pages/advanced-customization/template-provider-injection.mdnew file mode 100644index 000000000..79fc31655--- /dev/null+++ b/docs/pages/advanced-customization/template-provider-injection.md@@ -0,0 +1,3 @@+# Template Provider Injection++⚠️ Under constructiondiff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/adapt-to-llm-prompting-style-openai-anthropic-mistral.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/adapt-to-llm-prompting-style-openai-anthropic-mistral.mdnew file mode 100644index 000000000..2d17890c5--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/adapt-to-llm-prompting-style-openai-anthropic-mistral.md@@ -0,0 +1,57 @@+# Prompting Configuration++The `PromptingConfig` class controls how Pipelex handles prompting styles for different LLM targets.++## Configuration Options++```python+class PromptingConfig(ConfigModel):+ default_prompting_style: PromptingStyle+ prompting_styles: Dict[str, PromptingStyle]+```++### Fields++- `default_prompting_style`: The default prompting style to use when none is specified+- `prompting_styles`: Dictionary mapping LLM targets to their specific prompting styles++## Prompting Styles++Each prompting style defines how prompts are formatted and presented to the LLM. The style can be customized per LLM target to optimize performance and ensure compatibility.++## Example Configuration++```toml+[pipelex.prompting_config]+default_prompting_style = "chat"++[pipelex.prompting_config.prompting_styles]+gpt4 = "chat"+claude = "instruction"+llama = "completion"+```++## Usage++The configuration provides a method to get the appropriate prompting style:++```python+def get_prompting_style(self, prompting_target: Optional[LLMPromptingTarget] = None) -> Optional[PromptingStyle]:+ if prompting_target:+ return self.prompting_styles.get(prompting_target, self.default_prompting_style)+ else:+ return None+```++This allows for:++- Target-specific prompting styles+- Fallback to default style when no specific style is defined+- Optional prompting when no target is specified++## Best Practices++- Define a sensible default prompting style+- Configure specific styles for LLMs with unique requirements+- Test prompting styles with each LLM target+- Document any special formatting requirementsdiff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/ai-plugins-for-multi-llm-workflows.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/ai-plugins-for-multi-llm-workflows.mdnew file mode 100644index 000000000..e07b9484f--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/ai-plugins-for-multi-llm-workflows.md@@ -0,0 +1,238 @@+# Plugins Configuration++The Plugins Configuration manages all external service integrations in Pipelex, including various LLM providers and image generation services.++## Overview++```toml+[pipelex.plugins]+# Plugin sections+[pipelex.plugins.anthropic_config]+[pipelex.plugins.azure_openai_config]+[pipelex.plugins.bedrock_config]+[pipelex.plugins.vertexai_config]+[pipelex.plugins.mistral_config]+[pipelex.plugins.openai_config]+[pipelex.plugins.perplexity_config]+[pipelex.plugins.xai_config]+[pipelex.plugins.custom_endpoint_config]+[pipelex.plugins.fal_config]+```++## Common Authentication Methods++Most plugins support two authentication methods:++- `ENV`: Read credentials from environment variables+- `SECRET_PROVIDER`: Read credentials from a secrets provider++## Plugin-Specific Configurations++### 1. Anthropic Configuration++```toml+[pipelex.plugins.anthropic_config]+# Use 8192 for better streaming/timeout handling, or "unlimited" for full 32/64K tokens (Opus/Sonnet)+claude_4_reduced_tokens_limit = 8192+api_key_method = "env" # or "secret_provider"+```++Environment Variables:++- `ANTHROPIC_API_KEY`: API key for Anthropic services++### 2. Azure OpenAI Configuration++```toml+[pipelex.plugins.azure_openai_config]+api_key_method = "env" # or "secret_provider"+```++Environment Variables:++- `AZURE_OPENAI_API_KEY`: API key+- `AZURE_OPENAI_API_ENDPOINT`: API endpoint URL+- `AZURE_OPENAI_API_VERSION`: API version++### 3. AWS Bedrock Configuration++```toml+[pipelex.plugins.bedrock_config]+client_method = "aioboto3" # or "boto3"+```++Environment Variables:++- `AWS_REGION`: AWS region for Bedrock services++### 4. Google Vertex AI Configuration++```toml+[pipelex.plugins.vertexai_config]+api_key_method = "env" # or "secret_provider"+```++Environment Variables:++- `GCP_PROJECT_ID`: Google Cloud project ID+- `GCP_REGION`: Google Cloud region+- `GCP_CREDENTIALS_FILE_PATH`: Path to service account credentials file++Dependencies:++- Requires `google-auth-oauthlib` package (`pip install pipelex[google]`)++### 5. Mistral Configuration++```toml+[pipelex.plugins.mistral_config]+api_key_method = "env" # or "secret_provider"+```++Environment Variables:++- `MISTRAL_API_KEY`: API key for Mistral services++### 6. OpenAI Configuration++```toml+[pipelex.plugins.openai_config]+image_output_compression = 100 # 1-100+api_key_method = "env" # or "secret_provider"+```++Environment Variables:++- `OPENAI_API_KEY`: API key for OpenAI services++### 7. Perplexity Configuration++```toml+[pipelex.plugins.perplexity_config]+api_key_method = "env" # or "secret_provider"+```++Environment Variables:++- `PERPLEXITY_API_KEY`: API key+- `PERPLEXITY_API_ENDPOINT`: API endpoint URL++### 8. XAI Configuration++```toml+[pipelex.plugins.xai_config]+api_key_method = "env" # or "secret_provider"+```++Environment Variables:++- `XAI_API_KEY`: API key+- `XAI_API_ENDPOINT`: API endpoint URL++### 9. Custom Endpoint Configuration++For custom OpenAI-compatible endpoints (e.g., Ollama, LM Studio):++```toml+[pipelex.plugins.custom_endpoint_config]+api_key_method = "env" # or "secret_provider"+```++Environment Variables:++- `CUSTOM_ENDPOINT_API_KEY`: Optional API key+- `CUSTOM_ENDPOINT_BASE_URL`: Base URL for the custom endpoint++### 10. FAL Configuration++Configuration for FAL image generation services:++```toml+[pipelex.plugins.fal_config]+# Quality to steps mapping for Flux model+flux_map_quality_to_steps = {+ "low" = 14,+ "medium" = 28,+ "high" = 56+}++# Quality to steps mapping for SDXL Lightning model+sdxl_lightning_map_quality_to_steps = {+ "low" = 2,+ "medium" = 4,+ "high" = 8+}+```++## Error Handling++Each plugin has its own error types:++- `AnthropicCredentialsError`+- `AzureOpenAICredentialsError`+- `BedrockCredentialsError`+- `VertexAICredentialsError`+- `MistralCredentialsError`+- `OpenAICredentialsError`+- `PerplexityCredentialsError`+- `XaiCredentialsError`+- `CustomEndpointCredentialsError`++## Best Practices++1. **Credentials Management**:++ - Use environment variables for local development+ - Use secrets provider for production environments+ - Never commit credentials to version control++2. **Error Handling**:++ - Always handle credential errors appropriately+ - Implement proper fallbacks when using multiple providers+ - Check for required dependencies (especially for Google services)++3. **Configuration**:++ - Set appropriate quality levels for image generation+ - Configure retry limits and timeouts+ - Use appropriate client methods for async/sync operations++## Example Complete Configuration++```toml+[pipelex.plugins]++[pipelex.plugins.anthropic_config]+claude_4_reduced_tokens_limit = 8192 # Use "unlimited" for full 32/64K tokens+api_key_method = "env"++[pipelex.plugins.azure_openai_config]+api_key_method = "env"++[pipelex.plugins.bedrock_config]+client_method = "aioboto3"++[pipelex.plugins.vertexai_config]+api_key_method = "env"++[pipelex.plugins.mistral_config]+api_key_method = "env"++[pipelex.plugins.openai_config]+image_output_compression = 100+api_key_method = "env"++[pipelex.plugins.perplexity_config]+api_key_method = "env"++[pipelex.plugins.xai_config]+api_key_method = "env"++[pipelex.plugins.custom_endpoint_config]+api_key_method = "env"++[pipelex.plugins.fal_config]+flux_map_quality_to_steps = { "low" = 14, "medium" = 28, "high" = 56 }+sdxl_lightning_map_quality_to_steps = { "low" = 2, "medium" = 4, "high" = 8 }+```diff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/configure-ai-llm-to-optimize-workflows.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/configure-ai-llm-to-optimize-workflows.mdnew file mode 100644index 000000000..cd431f7ac--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/configure-ai-llm-to-optimize-workflows.md@@ -0,0 +1,114 @@+# LLM Settings Guide++## Overview++Pipelex provides a flexible way to configure and manage your LLM (Large Language Model) integrations through three main concepts:++- LLM Handles+- LLM Presets+- LLM Deck++Those configuration are present in the `pipelex_libraries/llm_deck` directory.++## LLM Handles++An LLM handle is a unique identifier that maps to a specific LLM configuration. It defines:++- The LLM name (e.g., `gpt-4o-mini`, `claude-4-sonnet`, `mistral-large`, `gemini-2.5-flash`)+- The model version+- The platform-specific settings (OpenAI, Anthropic, etc...)++### Example Handle Configurations++```toml+[llm_handles]+gpt-4o-2024-08-06 = { llm_name = "gpt-4o", llm_version = "2024-08-06" }+```++There's a much simpler syntax if you want a handle to the latest version and default platform:++```toml+[llm_handles]+best-claude = "claude-4-opus"+```++💡 Defining a llm_handle is alway meant to describe what model it is. Never define a llm_handle to describe what it is for or what it's good at. LLM Settings are for that.++## LLM Settings & LLM Presets++LLM Settings combine an LLM handle with specific parameters optimized for particular tasks. They help maintain consistency across similar operations and make it easier to switch between different configurations.++An LLM Preset is simply a name for a LLM Settings that you have predefined in order to use it in various places.++### Example LLM Preset definitions++```toml+[llm_presets]++llm_to_reason = {+ llm_handle = "gpt-4-turbo",+ temperature = 0.7,+ max_tokens = "auto"+}++llm_to_extract = {+ llm_handle = "claude-4-sonnet",+ temperature = 0.1,+ max_tokens = "auto"+}+```++### Using LLM Settings in Pipelines++Here's how to use these configurations in your pipelines:++```toml+[pipe.generate_response]+PipeLLM = "Generate a creative response"+inputs = { question = "Question" }+output = "Response"+llm = {+ llm_handle = "gpt-4-turbo", # Using inline LLM settings+ temperature = 0.8,+ max_tokens = "auto",+}+prompt = """+Generate a creative response to this question:++@question+"""++[pipe.extract_weather_data]+PipeLLM = "Extract structured weather data from text"+inputs = { text = "Text" }+output = "WeatherData"+llm = "llm_to_extract" # Using a preset+prompt = """+Extract the weather data from this text:++@text+"""+```++## LLM Deck++The LLM deck is your central configuration hub for all LLM-related settings. It's stored in the `pipelex_libraries/llm_deck` directory and consists of:++- `base_llm_deck.toml`: Core LLM configurations+- `overrides.toml`: Custom overrides for specific use cases++### Directory Structure++```bash+pipelex_libraries/+└── llm_deck/+ ├── base_llm_deck.toml+ └── overrides.toml+```+++## Best Practices++1. **Consistent Naming**: Use clear, descriptive names for handles and presets+1. **Task-Specific Presets**: Create presets optimized for specific skills and tasks+1. **Cost Management**: Consider using different models based on task complexity and cost requirementsdiff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/define_your_concepts.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/define_your_concepts.mdnew file mode 100644index 000000000..c5201c47f--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/define_your_concepts.md@@ -0,0 +1,296 @@+# Defining Your Concepts++Concepts are the foundation of reliable knowledge pipelines. They define what flows through your pipes—not just as data types, but as meaningful pieces of knowledge with clear boundaries and validation rules.++## Writing Concept Definitions++Every concept starts with a natural language definition. This definition serves two audiences: developers who build with your pipeline, and the LLMs that process your knowledge.++### Basic Concept Definition++The simplest way to define a concept is with a descriptive sentence:++```toml+[concept]+Invoice = "A commercial document issued by a seller to a buyer"+Employee = "A person employed by an organization"+ProductReview = "A customer's evaluation of a product or service"+```++**Key principles for concept definitions:**++1. **Define what it is, not what it's for**+ ```toml+ # ❌ Wrong: includes usage context+ TextToSummarize = "Text that needs to be summarized"++ # ✅ Right: defines the essence+ Article = "A written composition on a specific topic"+ ```++2. **Use singular forms**+ ```toml+ # ❌ Wrong: plural form+ Invoices = "Commercial documents from sellers"++ # ✅ Right: singular form+ Invoice = "A commercial document issued by a seller to a buyer"+ ```++3. **Avoid unnecessary adjectives**+ ```toml+ # ❌ Wrong: includes subjective qualifier+ LongArticle = "A lengthy written composition"++ # ✅ Right: neutral definition+ Article = "A written composition on a specific topic"+ ```++### Organizing Related Concepts++Group concepts that naturally belong together in the same domain. A domain acts as a namespace for a set of related concepts and pipes, helping you organize and reuse your pipeline components. You can learn more about them in [Kick off a Knowledge Pipeline Project](kick-off-a-knowledge-pipeline-project.md#what-are-domains).++```toml+# pipelex_libraries/pipelines/finance.toml+domain = "finance"+description = "Financial document processing"++[concept]+Invoice = "A commercial document issued by a seller to a buyer"+Receipt = "Proof of payment for goods or services"+PurchaseOrder = "A buyer's formal request to purchase goods or services"+PaymentTerms = "Conditions under which payment is to be made"+LineItem = "An individual item or service listed in a financial document"+```++## Adding Structure with Python Models++While text definitions help LLMs understand your concepts, Python models ensure structured, validated outputs. This combination gives you the best of both worlds: AI flexibility with software reliability.++**Important**: If you don't create a Python class for a concept, it defaults to text-based content. Only create Python models when you need structured output with specific fields.++### Creating Your First Structured Model++For each concept that needs structured output, create a corresponding Python class:++```python+# pipelex_libraries/pipelines/finance.py+from datetime import datetime+from typing import List, Optional+from pydantic import Field+from pipelex.core.stuff_content import StructuredContent++class Invoice(StructuredContent):+ invoice_number: str+ issue_date: datetime+ due_date: datetime+ vendor_name: str+ customer_name: str+ total_amount: float = Field(ge=0, description="Total invoice amount")+ currency: str = Field(default="USD", description="Three-letter currency code")+ line_items: List[str] = Field(default_factory=list)+```++The model name must match the concept name exactly: `Invoice` concept → `Invoice` class.++### Basic Validation Examples++Use Pydantic's validation features to ensure data quality:++```python+from pydantic import field_validator+from pipelex.core.stuff_content import StructuredContent++class Employee(StructuredContent):+ name: str+ email: str+ department: str+ years_of_experience: int = Field(ge=0, le=50, description="Years of work experience")++ @field_validator("email")+ @classmethod+ def validate_email(cls, v: str) -> str:+ if "@" not in v:+ raise ValueError("Invalid email format")+ return v.lower()++class ProductReview(StructuredContent):+ product_name: str+ reviewer_name: str+ rating: int = Field(ge=1, le=5, description="Rating from 1 to 5 stars")+ review_text: str+ verified_purchase: bool = False+```++### Working with Optional Fields++Not all data is always available. Use Optional fields with sensible defaults:++```python+from typing import Optional+from datetime import datetime+from pipelex.core.stuff_content import StructuredContent++class Meeting(StructuredContent):+ title: str+ scheduled_date: datetime+ duration_minutes: int = Field(ge=15, le=480, description="Meeting duration")+ location: Optional[str] = None+ attendees: List[str] = Field(default_factory=list)+ notes: Optional[str] = None+ is_recurring: bool = False+```++### Linking Concepts to Models++The connection between TOML definitions and Python models happens automatically through naming:++```toml+# pipelex_libraries/pipelines/hr.toml+domain = "hr"++[concept]+Employee = "A person employed by an organization"+Meeting = "A scheduled gathering of people for discussion"+PerformanceReview = "An evaluation of an employee's work performance"+Department = "An organizational unit within a company" # No Python model => text-based+```++```python+# pipelex_libraries/pipelines/hr.py+from pipelex.core.stuff_content import StructuredContent+from datetime import datetime+from typing import List, Optional++# Only define models for concepts that need structure+class Employee(StructuredContent):+ name: str+ email: str+ department: str+ hire_date: datetime++class Meeting(StructuredContent):+ title: str+ scheduled_date: datetime+ duration_minutes: int+ attendees: List[str]++class PerformanceReview(StructuredContent):+ employee_name: str+ review_period: str+ rating: int = Field(ge=1, le=5)+ strengths: List[str]+ areas_for_improvement: List[str]++# Note: Department concept has no Python model, so it's text-based+```++## Concept Refinement and Inheritance++Sometimes concepts build on each other. A `Contract` is a kind of `Document`. A `NonCompeteClause` is a specific part of a `Contract`. Pipelex lets you express these relationships.++### Declaring Concept Refinement++Use the `refines` field to indicate when one concept is a more specific version of another:++```toml+[concept]+Document = "A written or printed record"++[concept.Contract]+Concept = "A legally binding agreement between parties"+refines = "Document"++[concept.EmploymentContract]+Concept = "A contract between an employer and employee"+refines = "Contract"++[concept.NonCompeteClause]+Concept = "A contract clause restricting competitive activities"+refines = "ContractClause"+```++### Why Refinement Matters++Concept refinement helps in two ways:++1. **Semantic clarity**: Makes relationships between concepts explicit+2. **Pipeline flexibility**: Pipes accepting general concepts can work with refined ones++For example, a pipe that processes `Document` can also process `Contract` or `EmploymentContract`:++```toml+[pipe.extract_key_points]+PipeLLM = "Extract main points from any document"+inputs = { doc = "Document" } # Can accept Document, Contract, or EmploymentContract+output = "KeyPoints"+```++### Practical Refinement Example++Here's a complete example showing concept refinement in action:++```toml+# pipelex_libraries/pipelines/content.toml+domain = "content"++[concept]+Text = "Written content in natural language"++[concept.Article]+Concept = "A written composition on a specific topic"+refines = "Text"++[concept.NewsArticle]+Concept = "An article reporting current events"+refines = "Article"++[concept.OpinionPiece]+Concept = "An article expressing personal views"+refines = "Article"++[pipe.summarize_text]+PipeLLM = "Create a summary of any text"+inputs = { content = "Text" } # Works with Text, Article, NewsArticle, etc.+output = "Summary"++[pipe.extract_facts]+PipeLLM = "Extract factual claims from news"+inputs = { article = "NewsArticle" } # Specifically requires news articles+output = "FactualClaims"+```++### Structure Inheritance in Python++While TOML refinement is primarily semantic, you can mirror these relationships in Python when both concepts need structure:++```python+# pipelex_libraries/pipelines/content.py+from pipelex.core.stuff_content import StructuredContent+from datetime import datetime+from typing import Optional, List++# Base Article with structure+class Article(StructuredContent):+ title: str+ content: str+ author: str+ word_count: int = Field(ge=1)++# NewsArticle inherits and extends Article's structure+class NewsArticle(Article):+ publication_date: datetime+ news_category: str+ sources: List[str] = Field(default_factory=list)+ breaking_news: bool = False++# OpinionPiece also inherits and extends+class OpinionPiece(Article):+ opinion_type: str # e.g., "editorial", "column", "review"+ disclaimer: Optional[str] = None++# Note: Text, Summary, and FactualClaims have no Python models,+# so they remain text-based concepts+```++With well-defined concepts—both in natural language and code—your pipelines gain clarity, reliability, and maintainability. Next, we'll see how to build pipes that transform these concepts.diff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/design_and_run_pipelines.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/design_and_run_pipelines.mdnew file mode 100644index 000000000..c2e1782ac--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/design_and_run_pipelines.md@@ -0,0 +1,157 @@+# Designing and Running Pipelines++In Pipelex, a pipeline is not just a rigid sequence of steps; it's a dynamic and intelligent workflow built by composing individual, reusable components called **pipes**. This approach allows you to break down complex AI tasks into manageable, testable, and reliable units.++This guide provides an overview of how to design your pipelines and execute them.++## The Building Blocks: Pipes++A pipeline is composed of pipes. There are two fundamental types of pipes you will use to build your workflows:++* **[Pipe Operators](pipe-operators/index.md)**: These are the "workers" of your pipeline. They perform concrete actions like calling an LLM (`PipeLLM`), extracting text from a document (`PipeOcr`), or running a Python function (`PipeFunc`). Each operator is a specialized tool designed for a specific task.+* **[Pipe Controllers](pipe-controllers/index.md)**: These are the "managers" of your pipeline. They don't perform tasks themselves but orchestrate the execution flow of other pipes. They define the logic of your workflow, such as running pipes in sequence (`PipeSequence`), in parallel (`PipeParallel`), or based on a condition (`PipeCondition`).++## Designing a Pipeline: Composition in TOML++The most common way to design a pipeline is by defining and composing pipes in a `.toml` configuration file. This provides a clear, declarative way to see the structure of your workflow.++Each pipe, whether it's an operator or a controller, is defined in its own `[pipe.<pipe_name>]` table. The `<pipe_name>` becomes the unique identifier for that pipe.++Let's look at a simple example. Imagine we want a workflow that:+1. Takes a product description.+2. Generates a short, catchy marketing tagline for it.++We can achieve this with a `PipeLLM` operator.++```toml+# Filename: marketing_pipeline.toml++domain = "marketing"+definition = "Marketing content generation domain"++# 1. Define the concepts used in our pipes+[concept]+ProductDescription = "A description of a product's features and benefits"+Tagline = "A catchy marketing tagline"++# 2. Define the pipe that does the work+[pipe.generate_tagline]+PipeLLM = "Generate a catchy tagline for a product"+inputs = { description = "ProductDescription" }+output = "Tagline"+prompt_template = """+Product Description:+@description++Generate a catchy tagline based on the above description.+The tagline should be memorable, concise, and highlight the key benefit.+"""+```++This defines a single-step pipeline. The pipe `generate_tagline` takes a `ProductDescription` as input and outputs a `Tagline`.++To create a multi-step workflow, you use a controller. The `PipeSequence` controller is the most common one. It executes a series of pipes in a specific order.++```toml+# Filename: marketing_pipeline.toml++domain = "marketing"+definition = "Marketing content generation domain"++# 1. Define concepts+[concept]+ProductDescription = "A description of a product's features and benefits"+Keyword = "A keyword extracted from a text"+Tagline = "A catchy marketing tagline"++# 2. Define operator pipes+[pipe.extract_keywords]+PipeLLM = "Extract keywords from a product description"+inputs = { description = "ProductDescription" }+output = "Keyword"+multiple_output = true+prompt_template = """+Please extract the most relevant keywords from the following product description:++@description++Focus on features, benefits, and unique selling points.+"""++[pipe.generate_tagline_from_keywords]+PipeLLM = "Generate a tagline from keywords"+inputs = { keywords = "Keyword" }+output = "Tagline"+prompt_template = """+Here are the key product keywords:+@keywords++Generate a catchy marketing tagline based on these keywords.+The tagline should be memorable, concise (under 10 words), and highlight the main benefit.+"""++# 3. This controller pipe defines the two-step pipeline+[pipe.description_to_tagline]+PipeSequence = "From product description to tagline"+inputs = { description = "ProductDescription" }+output = "Tagline"+steps = [+ { pipe = "extract_keywords", result = "extracted_keywords" },+ { pipe = "generate_tagline_from_keywords", result = "tagline" },+]+```++## Data Flow: The Working Memory++How does data get from `extract_keywords` to `generate_tagline_from_keywords`? This is handled by the **Working Memory**.++The Working Memory is a temporary storage space that exists for the duration of a single pipeline run.++1. When a pipe in a sequence executes, its output is given a name using the `result` key (e.g., `result = "extracted_keywords"`).+2. This named result is placed into the Working Memory.+3. Subsequent pipes can then reference this data by its name in their `inputs` field (e.g., `inputs = { keywords = "Keyword" }`).++This mechanism allows you to chain pipes together, creating a flow of information through your pipeline.++## Running a Pipeline++Once your pipes are defined, you can execute them from your Python code. Pipelex provides two main functions for this: `start_pipeline` and `execute_pipeline`.++To run the `description_to_tagline` pipeline we defined above, you would call it by its unique name:++```python+import asyncio+from pipelex.core.working_memory import WorkingMemory+from pipelex.core.working_memory_factory import WorkingMemoryFactory+from pipelex.pipelex import Pipelex+from pipelex.pipeline.execute import execute_pipeline++async def main():+ # First, initialize Pipelex (this loads all pipeline definitions)+ Pipelex.make()++ # Create working memory with the pipeline's input+ working_memory = WorkingMemoryFactory.make_from_text(+ text="EcoClean Pro is a revolutionary biodegradable cleaning solution that removes 99.9% of germs while being completely safe for children and pets. Made from plant-based ingredients.",+ concept_code="ProductDescription",+ name="description"+ )++ # Execute the pipeline and wait for the result+ pipe_output, _ = await execute_pipeline(+ pipe_code="description_to_tagline",+ working_memory=working_memory,+ )++ # Get the final output+ tagline = pipe_output.main_stuff_as(content_type=str)+ print(f"Generated tagline: {tagline}")++if __name__ == "__main__":+ asyncio.run(main())+```++- `execute_pipeline`: Runs the specified pipe and waits for it to complete, returning the final output. This is useful for simple, synchronous-style interactions.+- `start_pipeline`: Immediately returns a `pipeline_run_id` and an `asyncio.Task`. This allows you to run pipelines in the background and manage them asynchronously, which is essential for complex, long-running, or parallel workflows.++By combining declarative TOML definitions with a powerful Python execution model, Pipelex gives you a robust framework for building and running reliable AI workflows.diff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/kick-off-a-knowledge-pipeline-project.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/kick-off-a-knowledge-pipeline-project.mdnew file mode 100644index 000000000..574a6ac4a--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/kick-off-a-knowledge-pipeline-project.md@@ -0,0 +1,154 @@+# Kicking off a Knowledge Pipeline Project++## Creating Your First Pipeline++A pipeline in Pipelex is a collection of related concepts and pipes. Start by creating a TOML file in the `pipelines` directory:++```toml+# pipelex_libraries/pipelines/tutorial.toml++domain = "tutorial"+definition = "My first Pipelex library"+system_prompt = "You are a helpful assistant."++[concept]+Question = "A question that needs to be answered"+Answer = "A response to a question"++[pipe]+[pipe.answer_question]+PipeLLM = "Answer a question"+inputs = { question = "tutorial.Question" }+output = "tutorial.Answer"+prompt_template = """+Please answer the following question:++@question++Provide a clear and concise answer.+"""+```++This creates a simple Q&A pipeline with:+- A domain called "tutorial"+- Two concepts: Question and Answer+- One pipe that transforms a Question into an Answer++The `domain` property is the most important part of your pipeline file. It groups all your concepts and pipes into a single, easy to read bundle.++## What Are Domains?++A domain in Pipelex represents a topic or area of functionality within your pipeline. Every pipeline file must specify its domain, which helps organize and categorize your pipes and concepts.++### Domain in Practice++When you create a pipeline file (`.toml`), you always start by declaring its domain:++```toml+domain = "finance" # The domain name for this file+definition = "Financial document processing" # Optional description+system_prompt = "You are an expert financial analyst." # Optional system prompt for all PipeLLM in this domain+```++A domain consists of:++1. **Pipeline File** (`.toml`)+ ```toml+ domain = "finance"++ [concept]+ Invoice = "A commercial document for a sale of products or services"+ InvoiceSummary = "A summary of an invoice with key details"++ [pipe]+ [pipe.summarize_invoice]+ PipeLLM = "Summarize an invoice to extract key information"+ inputs = { invoice = "finance.Invoice" }+ output = "finance.InvoiceSummary"+ ```++2. **Python Models** (`.py`)+ ```python+ from pipelex.core.stuff_content import StructuredContent+ from pydantic import Field+ from typing import List+ from datetime import date++ class Invoice(StructuredContent):+ invoice_number: str+ vendor: str+ customer: str+ total_amount: float = Field(ge=0)+ issue_date: date+ line_items: List[str]++ class InvoiceSummary(StructuredContent):+ vendor: str+ total_amount: float+ is_overdue: bool+ ```++### Best Practices++1. **Naming**+ - Use clear, descriptive domain names.+ - Keep names lowercase and simple.+ - Use names that reflect the purpose (e.g., "finance", "legal", "content_creation").++2. **Organization**+ - One domain per topic/functionality.+ - Match Python file names with domain names (`finance.toml` -> `finance.py`).+ - Keep related concepts within the same domain.++3. **Documentation**+ - Always add a description to your domain.+ - Document concepts clearly.+ - Include examples where helpful.++### Using Domains++When using a domain in your code, you refer to concepts with `domain.ConceptName`:++```python+from pipelex.core.stuff_factory import StuffFactory++# The concept_code combines domain and concept names+invoice_stuff = StuffFactory.make_stuff(+ concept_str="finance.Invoice", # domain.ConceptName+ name="invoice_123",+ content=invoice_data # dictionary or Invoice object+)+```++## File Naming Conventions++Consistent naming makes your pipeline code discoverable and maintainable:++### TOML Files+- Use lowercase with underscores: `legal_contracts.toml`, `customer_service.toml`+- Match the domain name when possible: domain "legal" → `legal.toml`+- For multi-word domains, use underscores: domain "customer_service" → `customer_service.toml`++### Python Model Files+- It is recommended to match the TOML filename exactly: `legal.toml` → `legal.py`+- But in any case, Pipelex will load models from all python modules in the `pipelines` directory or its subdirectories.++## Project Structure++Every Pipelex project follows a simple directory structure that keeps your knowledge pipelines organized and maintainable:++```+your-project/+├── pipelex_libraries/ # All your pipeline code lives here+│ ├── pipelines/ # Pipeline definitions and models+│ │ ├── __init__.py+│ │ ├── characters.toml # Domain definitions+│ │ └── characters.py # Python models for concepts+│ ├── templates/ # Reusable prompt templates+│ ├── llm_integrations/ # LLM provider configurations+│ └── llm_deck/ # LLM model presets+├── main.py # Your application code+└── requirements.txt # Python dependencies+```++The `pipelex_libraries` directory is where Pipelex looks for your pipeline definitions. This standardized structure means you can share libraries between projects, version control them separately, and maintain clean separation between your pipeline logic and application code.diff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/llm-structured-generation-config.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/llm-structured-generation-config.mdnew file mode 100644index 000000000..6b0220916--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/llm-structured-generation-config.md@@ -0,0 +1,92 @@+# Structure Configuration++The `StructureConfig` class controls how Pipelex handles structural processing of content, particularly in LLM-based pipes.++## Configuration Options++```python+class StructureConfig(ConfigModel):+ is_default_text_then_structure: bool+```++### Fields++- `is_default_text_then_structure`: When true, uses a two-step LLM process where text is generated first, then structured into a JSON object++## Example Configuration++```toml+[pipelex.structure_config]+is_default_text_then_structure = true+```++## Processing Flow++The `is_default_text_then_structure` flag determines how LLM pipes generate structured content:++### Two-Step Process (When `true`)++1. **First LLM Call**: Generates preliminary text+ - Uses the pipe's main prompt template+ - Produces natural language text about the subject++2. **Second LLM Call**: Converts text to structure+ - Uses a specialized system prompt: "You are a data modeling expert specialized in extracting structure from text"+ - Uses a template that instructs the LLM to extract structured data:+ ```+ Your job is to extract and structure information from a text.+ Here is the text:+ {preliminary_text}++ Now generate the JSON in the required format.+ Do not create information that is not in the text.+ ```+ - Produces a JSON object matching the required concept structure++### Single-Step Process (When `false`)++- Single LLM call that directly generates the structured output+- Uses the pipe's main prompt template+- Produces structured JSON directly++## Use Cases++### When to Use Two-Step Process (`true`)++- When you want more natural and fluid content generation+- When the structure is complex and needs careful extraction+- When you want to ensure all generated content is properly structured+- When you need better control over the extraction process+- When you want to debug or inspect the intermediate text++### When to Use Single-Step Process (`false`)++- When the structure is simple and straightforward+- When you need faster processing (avoids second LLM call)+- When the output format is well-defined and easy to generate+- When you want to reduce API costs+- When the prompt is already optimized for structured output++## Technical Details++The two-step process uses:++1. The pipe's configured prompts for initial text generation+2. A specialized structure extraction prompt that can be customized:+ - Through domain configuration (`domain.prompt_template_to_structure`)+ - Through pipe configuration (`pipe.prompt_template_to_structure`)+ - Falls back to base template if not specified++## Example Flow++```mermaid+graph TD+ A[Input] --> B{is_default_text_then_structure?}+ B -->|true| C[First LLM Call]+ C --> D[Generate Text]+ D --> E[Second LLM Call]+ E --> F[Extract Structure]+ F --> H[Final JSON Output]+ B -->|false| G[Direct Structure Generation]+ G --> H+```diff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeBatch.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeBatch.mdnew file mode 100644index 000000000..6c1c57eb3--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeBatch.md@@ -0,0 +1,67 @@+# PipeBatch++The `PipeBatch` controller provides a powerful "map" operation for your pipelines. It takes a list of items as input and runs the same pipe on each item in the list, executing them all in parallel for maximum efficiency.++This is the ideal controller for processing collections of documents, images, or any other data records where the same logic needs to be applied to each one independently.++## How it works++`PipeBatch` orchestrates a parallel, per-item execution of a single "branch pipe".++1. **Input List**: It identifies an input list from the working memory.+2. **Branching**: For each item in the input list, it creates a new, isolated execution branch.+3. **Isolation & Injection**: Each branch gets a deep copy of the `WorkingMemory`. The specific item for that branch is injected into this memory with a defined name.+4. **Concurrent Execution**: The specified `branch_pipe_code` is executed in all branches simultaneously. Each branch pipe operates only on its own item.+5. **Aggregation**: After all branches have completed, `PipeBatch` collects the individual output from each one and aggregates them into a single new list. This list becomes the final output of the `PipeBatch` pipe.++## Configuration++`PipeBatch` is configured in your pipeline's `.toml` file.++### TOML Parameters++| Parameter | Type | Description | Required |+| ------------------ | ------------ | ------------------------------------------------------------------------------------------------------------------------------------------------ | -------- |+| `PipeBatch` | string | A descriptive name for the batch operation. | Yes |+| `inputs` | dictionary | The input concept(s) for the batch operation, as a dictionary mapping input names to concept codes. | Yes |+| `output` | string | The output concept produced by the batch operation. | Yes |+| `branch_pipe_code` | string | The name of the single pipe to execute for each item in the input list. | Yes |+| `batch_params` | table (dict) | An optional table to provide more specific names for the batch operation. | No |++### Batch Parameters (`batch_params`)++| Key | Type | Description | Required |+| ----------------------- | ------ | ------------------------------------------------------------------------------------------------------------------------------------- | -------- |+| `input_list_stuff_name` | string | The name of the list in the `WorkingMemory` to iterate over. If not provided, it defaults to the name of the `PipeBatch`'s main `input`. | No |+| `input_item_stuff_name` | string | The name that an individual item from the list will have inside its execution branch. This is how the branch pipe finds its input. | Yes |++### Example: Summarizing a list of articles++Suppose you have a list of articles and you want to generate a summary for each one.++```toml+# The pipe that knows how to summarize one article+[pipe.summarize_one_article]+PipeLLM = "Summarize a single article"+inputs = { article = "ArticleText" }+output = "ArticleSummary"+prompt_template = "Please provide a one-sentence summary of the following article:\n\n@article_text"++# The PipeBatch definition+[pipe.summarize_all_articles]+PipeBatch = "Summarize a batch of articles in parallel"+inputs = { articles = "ArticleList" } # This is the list to iterate over+output = "SummaryList" # This will be the list of summaries+branch_pipe_code = "summarize_one_article"++[pipe.summarize_all_articles.batch_params]+input_item_stuff_name = "ArticleText" # Name of an item within the branch+```++How this works:+1. The `summarize_all_articles` pipe receives an `ArticleList`. Let's say it contains 10 articles.+2. `PipeBatch` creates 10 parallel branches.+3. In branch #1, it takes the first article from `ArticleList`, puts it into the branch's isolated working memory, and gives it the name `ArticleText` (as specified by `input_item_stuff_name`).+4. The `summarize_one_article` pipe is then executed in branch #1. It looks for an input named `ArticleText`, finds the injected article, and produces a summary.+5. Steps 3 and 4 happen simultaneously for all 10 articles in their respective branches.+6. Once all `summarize_one_article` pipes are done, `PipeBatch` collects the 10 `ArticleSummary` outputs and bundles them into a single `SummaryList`. This list is the final result.\ No newline at end of filediff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeCondition.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeCondition.mdnew file mode 100644index 000000000..702bb6d13--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeCondition.md@@ -0,0 +1,107 @@+# PipeCondition++The `PipeCondition` controller adds branching logic to your pipelines. It evaluates an expression and, based on the string result, chooses which subsequent pipe to execute from a map of possibilities.++## How it works++`PipeCondition` is a routing mechanism. Its execution flow is as follows:++1. **Evaluate an Expression**: It takes an expression and renders it using Jinja2, with the full `WorkingMemory` available as context. This evaluation results in a simple string.+2. **Look Up in Pipe Map**: The resulting string is used as a key to find a corresponding pipe name in the `pipe_map`.+3. **Use Default (Optional)**: If the key is not found in the `pipe_map`, it will use the `default_pipe_code` if one is provided. If there's no match and no default, an error is raised.+4. **Execute Chosen Pipe**: The chosen pipe is then executed. It receives the exact same `WorkingMemory` and inputs that were passed to the `PipeCondition` operator. The output of the chosen pipe becomes the output of the `PipeCondition` itself.++## Configuration++`PipeCondition` is configured in your pipeline's `.toml` file.++### TOML Parameters++| Parameter | Type | Description | Required |+| ------------------------------ | -------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------ |+| `PipeCondition` | string | A descriptive name for the condition. | Yes |+| `inputs` | dictionary | The input concept(s) for the condition, as a dictionary mapping input names to concept codes. | Yes |+| `output` | string | The output concept produced by the selected pipe. | Yes |+| `expression` | string | A simple Jinja2 expression. `{{ ... }}` are automatically added. Good for simple variable access like `"my_var.category"`. | Yes (or `expression_jinja2`) |+| `expression_jinja2` | string | A full Jinja2 template string. Use this for more complex logic, like `{% if my_var.value > 10 %}high{% else %}low{% endif %}`. | Yes (or `expression`) |+| `pipe_map` | table (dict) | A mapping where keys are the possible string results of the expression, and values are the names of the pipes to execute. | Yes |+| `default_pipe_code` | string | The name of a pipe to execute if the expression result does not match any key in `pipe_map`. | No |+| `add_alias_from_expression_to` | string | An advanced feature. If provided, the string result of the expression evaluation is added to the working memory as an alias with this name. | No |++### Example: Routing based on document type++Imagine a pipeline that needs to process invoices and receipts differently.++```toml+[pipe.classify_document_type]+PipeLLM = "Classify the document as 'invoice' or 'receipt'"+inputs = { document = "DocumentText" }+output = "DocumentClassification" # A structure with a 'type' field+prompt_template = """+Classify the document as 'invoice' or 'receipt'++@document+"""++[pipe.process_invoice]+# ... pipe definition ...+output = "ProcessedDocument"++[pipe.process_receipt]+# ... pipe definition ...+output = "ProcessedDocument"++[pipe.process_other]+# ... pipe definition ...+output = "ProcessedDocument"++# The PipeCondition definition+[pipe.route_by_doc_type]+PipeCondition = "Route document based on its classified type"+inputs = { classification = "DocumentClassification" }+output = "ProcessedDocument"+expression = "doc_classification.type" # Assumes input from a previous step was named 'doc_classification'++[pipe.route_by_doc_type.pipe_map]+invoice = "process_invoice"+receipt = "process_receipt"++[pipe.route_by_doc_type]+default_pipe_code = "process_other"+```++How this works:+1. A previous step, `classify_document_type`, runs and its output is named `doc_classification`. This output is an object with a `type` attribute (e.g., `"invoice"`).+2. `PipeCondition` evaluates the `expression`: `"doc_classification.type"`. Let's say it results in the string `"invoice"`.+3. It looks up `"invoice"` in the `pipe_map` and finds the corresponding pipe: `"process_invoice"`.+4. The `process_invoice` pipe is executed.+5. If the classification had been `"letter"`, it would not match any key in the `pipe_map`, so the `default_pipe_code` `"process_other"` would be executed instead.++## Expression Types++### Simple Expression+```python+expression = "product_or_services_category.category"+```++- Direct access to working memory variables+- No template syntax needed+- Good for simple field access+- Access to Jinja2 filters and functions++## Features++### Default Routing+```python+default_pipe_code = "process_unknown"+```++- Fallback pipe when no match is found++### Expression Aliasing+```python+add_alias_from_expression_to = "category_type"+```++- Creates an alias from the expression result+- Makes the result available in working memorydiff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeParallel.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeParallel.mdnew file mode 100644index 000000000..32994fdeb--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeParallel.md@@ -0,0 +1,78 @@+# PipeParallel++The `PipeParallel` controller executes multiple pipes simultaneously. This is highly effective for running pipes concurrently in isolated branches.++## How it works++`PipeParallel` runs a list of sub-pipes in concurrent branches.++1. **Isolation**: Before execution, `PipeParallel` creates a deep copy of the current `WorkingMemory` for each branch. This means every parallel pipe starts with the exact same state, but they run in complete isolation—a change in one branch will not affect another.+2. **Concurrent Execution**: All specified pipes are executed at the same time using `asyncio.gather`.+3. **Output Handling**: After all parallel tasks have finished, their results are collected and added back to the main working memory. You can control how this happens with two parameters:+ - `add_each_output`: If `true`, the individual result of each branch is added to the working memory under the name specified in its `result` key.+ - `combined_output`: If you provide an output concept here, the results of all branches are bundled together into a single structured object. The field names of this object correspond to the `result` names of the branches.++You must use `add_each_output`, `combined_output`, or both.++## Configuration++`PipeParallel` is configured in your pipeline's `.toml` file.++### TOML Parameters++| Parameter | Type | Description | Required |+| ----------------- | ------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------- |+| `PipeParallel` | string | A descriptive name for the parallel operation. | Yes |+| `inputs` | dictionary | The input concept(s) for the parallel operation, as a dictionary mapping input names to concept codes. | Yes |+| `output` | string | The output concept produced by the parallel operation. | Yes |+| `parallels` | array of tables| An array defining the pipes to run in parallel. Each table is a sub-pipe definition. | Yes |+| `add_each_output` | boolean | If `true`, adds the output of each parallel pipe to the working memory individually. Defaults to `false`. | No |+| `combined_output` | string | The name of a concept to use for a single, combined output object. The structure of this concept must have fields that match the `result` names from the `parallels` array. | No |++### Parallel Step Configuration++Each entry in the `parallels` array is a table with the following keys:++| Key | Type | Description | Required |+| -------- | ------ | ---------------------------------------------------------------------------------------- | -------- |+| `pipe` | string | The name of the pipe to execute for this branch. | Yes |+| `result` | string | The name for this branch's output. Must be unique within the `PipeParallel` definition. | Yes |++### Example: Extracting different details from a text++Imagine you have a product description and you want to extract the product features and the product sentiment at the same time.++```toml+[concept.ProductAnalysis]+structure = "ProductAnalysis" # A Pydantic model with 'features' and 'sentiment' fields++[pipe.extract_features]+PipeLLM = "Extract features from text"+inputs = { description = "ProductDescription" }+output = "ProductFeatures"++[pipe.analyze_sentiment]+PipeLLM = "Analyze sentiment of text"+inputs = { description = "ProductDescription" }+output = "ProductSentiment"++# The PipeParallel definition+[pipe.analyze_product_in_parallel]+PipeParallel = "Extract features and sentiment at the same time"+inputs = { description = "ProductDescription" }+output = "ProductAnalysis" # This name is for the combined output+add_each_output = false+combined_output = "ProductAnalysis"+parallels = [+ { pipe = "extract_features", result = "features" },+ { pipe = "analyze_sentiment", result = "sentiment" },+]+```++How this works:+1. The `analyze_product_in_parallel` pipe starts. It receives a `ProductDescription`.+2. Two parallel branches are created, both with access to the `ProductDescription`.+3. The `extract_features` pipe runs in one branch, and the `analyze_sentiment` pipe runs in the other, simultaneously.+4. After both complete, `PipeParallel` collects the results. The output from `extract_features` is named `features`, and the output from `analyze_sentiment` is named `sentiment`.+5. Because `combined_output` is set to `ProductAnalysis`, a new structured object of type `ProductAnalysis` is created. This object is populated with the results, like `{"features": ..., "sentiment": ...}`.+6. This single `ProductAnalysis` object becomes the final output of the `analyze_product_in_parallel` pipe.\ No newline at end of filediff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeSequence.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeSequence.mdnew file mode 100644index 000000000..1fcd5d2bf--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeSequence.md@@ -0,0 +1,64 @@+# PipeSequence++The `PipeSequence` controller is used to execute a series of pipes one after another. It is the fundamental building block for creating linear workflows where the output of one step becomes the input for the next.++## How it works++A `PipeSequence` defines a list of `steps`. Each step calls another pipe and gives a name to its output. The working memory is passed from one step to the next, accumulating results along the way.++- The `input` of the `PipeSequence` is passed to the first pipe in the sequence.+- The `output` of each intermediate step is named via the `result` key and becomes available in the working memory for all subsequent steps.+- The final `output` of the `PipeSequence` is the output produced by the very last step in the sequence.++## Configuration++`PipeSequence` is configured in your pipeline's `.toml` file.++### TOML Parameters++| Parameter | Type | Description | Required |+| ---------- | --------------- | -------------------------------------------------------------------------------------------------------------- | -------- |+| `PipeSequence` | string | A descriptive name for the sequence. | Yes |+| `inputs` | dictionary | The input concept(s) for the *first* pipe in the sequence, as a dictionary mapping input names to concept codes. | No |+| `output` | string | The output concept produced by the *last* pipe in the sequence. | Yes |+| `steps` | array of tables | An ordered list of the pipes to execute. Each table in the array defines a single step. | Yes |++### Step Configuration++Each entry in the `steps` array is a table with the following keys:++| Key | Type | Description | Required |+| -------- | ------ | ------------------------------------------------------------------ | -------- |+| `pipe` | string | The name of the pipe to execute for this step. | Yes |+| `result` | string | The name to give to the output of this step in the working memory. | Yes |++### Example++Let's imagine a pipeline that first extracts text from an image, then summarizes that text, and finally translates the summary into French.++```toml+[pipe.extract_text_from_image]+PipeOcr = "..." # (definition of the OCR pipe)+output = "Text"++[pipe.summarize_text]+PipeLLM = "..." # (definition of the summarization pipe)+inputs = { text = "Text" }+output = "Text"++[pipe.translate_to_french]+PipeLLM = "..." # (definition of the translation pipe)+inputs = { text = "Text" }+output = "Text"+++[pipe.image_to_french_summary]+PipeSequence = "Extract, summarize, and translate text from an image"+inputs = { image = "source.Image" }+output = "target.FrenchText"+steps = [+ { pipe = "extract_text_from_image", result = "extracted_text" },+ { pipe = "summarize_text", result = "english_summary" },+ { pipe = "translate_to_french", result = "french_summary" },+]+```\ No newline at end of filediff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/index.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/index.mdnew file mode 100644index 000000000..48b56f70b--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/index.md@@ -0,0 +1,44 @@+# Pipe Controllers++Pipe controllers are the orchestrators of a Pipelex pipeline. While [Pipe Operators](../pipe-operators/index.md) perform the work, pipe controllers define the workflow and manage the execution logic. They allow you to run other pipes in sequence, in parallel, or conditionally.++## Core Controllers++Here are the primary pipe controllers available in Pipelex:++- [**`PipeSequence`**](./PipeSequence.md): The most fundamental controller. It runs a series of pipes one after another, passing the results from one step to the next.+- [**`PipeParallel`**](./PipeParallel.md): Executes multiple independent pipes at the same time, significantly speeding up workflows where tasks don't depend on each other.+- [**`PipeBatch`**](./PipeBatch.md): Performs a "map" operation. It takes a list of items and runs the same pipe on every single item in parallel.+- [**`PipeCondition`**](./PipeCondition.md): Adds branching logic (`if/else`) to your pipeline. It evaluates an expression and chooses which pipe to run next based on the result.++## Overview++Pipelex provides the following pipe operators:++- `PipeSequence`: For chaining multiple pipes in sequence+- `PipeParallel`: For running different pipes in parallel+- `PipeBatch`: For running one pipe over a batch of inputs+- `PipeCondition`: For conditional execution based on input validation++## PipeSequence++Run multiple pipes in sequence.++### Key Features++- Sequential execution+- Working memory management+- Sub-pipe handling+- Pipeline composition++## PipeCondition++Enables conditional execution based on input validation.++### Key Features++- Expression-based routing+- Default fallback paths+- Jinja2 template support+- Input validation+- Error handlingdiff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeFunc.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeFunc.mdnew file mode 100644index 000000000..4e0ba1875--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeFunc.md@@ -0,0 +1,75 @@+# PipeFunc++The `PipeFunc` operator provides an essential escape hatch, allowing you to execute arbitrary Python code from within a pipeline. This is useful for custom data manipulation, complex logic, or integrating with external services not supported by other operators.++## How it works++`PipeFunc` operates by calling a Python function that has been registered with Pipelex's central function registry.++1. **Function Registration**: First, you must define a standard Python function and register it with a unique name. This is done once when your application starts up.+2. **Function Signature**: The registered function **must** accept a single argument: `working_memory: WorkingMemory`. This object provides access to all the data currently available in the pipeline.+3. **Execution**: When the `PipeFunc` pipe is executed, it looks up your function by its registered name and calls it, passing in the current `working_memory`.+4. **Returning Data**: The function can return data, which `PipeFunc` will then place back into the working memory, associated with the pipe's `output` concept.++### Return values++Your Python function can return one of the following:+- A `StuffContent` object (e.g., `TextContent`, `ImageContent`, or a custom `StructuredContent` model).+- A `list` of `StuffContent` objects.+- A simple Python `str`, which will be automatically converted to a `TextContent`.++## How to Register a Function++To make a Python function available to `PipeFunc`, you must register it using the global `func_registry`.++Here is an example of a function and its registration:++```python+# in a file like my_custom_functions.py++from pipelex.core.working_memory import WorkingMemory+from pipelex.core.stuff_content import TextContent+from pipelex.tools.func_registry import func_registry++def concatenate_texts(working_memory: WorkingMemory) -> TextContent:+ """+ Retrieves two text stuffs, concatenates them, and returns a new text stuff.+ """+ # Get data from working memory using the names from previous steps+ text1 = working_memory.get_stuff_as_str("text_a")+ text2 = working_memory.get_stuff_as_str("text_b")++ concatenated = f"{text1} -- {text2}"++ return TextContent(text=concatenated)++def register_my_functions():+ """This function should be called at application startup."""+ func_registry.register_function(concatenate_texts, name="combine_two_texts")++```++You would then call `register_my_functions()` when your Pipelex application initializes.++## Configuration++Once the function is registered, you can use it in your `.toml` file.++### TOML Parameters++| Parameter | Type | Description | Required |+| --------------- | ------ | --------------------------------------------------------------------------- | -------- |+| `PipeFunc` | string | A descriptive name for the pipe's function. | Yes |+| `function_name` | string | The unique name used to register the Python function (e.g., "combine_two_texts"). | Yes |+| `output` | string | The concept to associate with the function's return value. | Yes |++### Example++This TOML snippet shows how to use the `combine_two_texts` function defined above. It assumes two previous pipes have produced outputs named `text_a` and `text_b`.++```toml+[pipe.combine_them]+PipeFunc = "Combine two text inputs using a custom Python function"+function_name = "combine_two_texts"+output = "ConcatenatedText"+```diff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeImgGen.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeImgGen.mdnew file mode 100644index 000000000..6e859c7cd--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeImgGen.md@@ -0,0 +1,68 @@+# PipeImgGen++The `PipeImgGen` operator is used to generate images from a text prompt using a specified image generation model.++## How it works++`PipeImgGen` takes a text prompt and a set of parameters, then calls an underlying image generation service (like DALL-E 3 or Stable Diffusion) to create one or more images.++The prompt can be provided in two ways:+1. As a static string directly in the pipe's TOML definition using the `imgg_prompt` parameter.+2. As a dynamic input by connecting a concept that is or refines `native.Text`.++The pipe can be configured to generate a single image or a list of images.++## Configuration++`PipeImgGen` is configured in your pipeline's `.toml` file.++### TOML Parameters++| Parameter | Type | Description | Required |+| ----------------------- | --------------- | ----------------------------------------------------------------------------------------------------------------------------- | -------- |+| `PipeImgGen` | string | A descriptive name for the image generation operation. | Yes |+| `inputs` | dictionary | The input concept(s) for the image generation operation, as a dictionary mapping input names to concept codes. | Yes |+| `output` | string | The output concept produced by the image generation operation. | Yes |+| `imgg_prompt` | string | A static text prompt for image generation. Use this *or* `input`. | No |+| `output_multiplicity` | integer | The number of images to generate. If omitted, a single image is generated. | No |+| `imgg_handle` | string | The handle for the image generation model to use (e.g., `"dall-e-3"`). Defaults to the model specified in the global config. | No |+| `aspect_ratio` | string | The desired aspect ratio of the image (e.g., `"16:9"`, `"1:1"`). | No |+| `quality` | string | The quality of the generated image (e.g., `"standard"`, `"hd"`). | No |+| `seed` | integer or "auto" | A seed for the random number generator to ensure reproducibility. `"auto"` uses a random seed. | No |+| `nb_steps` | integer | For diffusion models, the number of steps to run. More steps can increase detail but take longer. | No |+| `guidance_scale` | float | How strictly the model should adhere to the prompt. Higher values mean closer adherence. | No |+++### Example: Generating a single image from a static prompt++This pipe generates one image of a futuristic car without requiring any input.++```toml+[pipe.generate_car_image]+PipeImgGen = "Generate a futuristic car image"+output = "Image"+imgg_prompt = "A sleek, futuristic sports car driving on a neon-lit highway at night."+imgg_handle = "dall-e-3"+aspect_ratio = "16:9"+quality = "hd"+```++### Example: Generating multiple images from a dynamic prompt++This pipe takes a text prompt as input and generates three variations of the image.++```toml+[concept]+ImagePrompt = "A text prompt for generating an image"++[pipe.generate_logo_variations]+PipeImgGen = "Generate three logo variations from a prompt"+inputs = { prompt = "images.ImgGenPrompt" }+output = "Image"+nb_output = 3+imgg_handle = "stable-diffusion"+aspect_ratio = "1:1"+```++To use this pipe, you would first load a text prompt (e.g., "A minimalist logo for a coffee shop, featuring a stylized mountain and a coffee bean") into the "prompt" stuff (`ImgGenPrompt` concept). After running, the output will be a list containing three generated `ImageContent` objects.+diff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeJinja2.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeJinja2.mdnew file mode 100644index 000000000..d09c9dd08--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeJinja2.md@@ -0,0 +1,65 @@+# PipeJinja2++The `PipeJinja2` operator is a powerful utility for rendering [Jinja2 templates](https://jinja.palletsprojects.com/). It allows you to dynamically generate text by combining data from your pipeline's working memory with a template. This is ideal for creating formatted reports, HTML content, or constructing complex, multi-part prompts for LLMs.++## How it works++`PipeJinja2` takes all the data currently in the `WorkingMemory` and uses it as the context for rendering a Jinja2 template. The resulting text is then saved back to the working memory as a new `Text` output.++The template itself can be provided in one of two ways:+1. **By Name**: Referring to a template file that has been loaded into Pipelex's template provider. This is the most common and maintainable method.+2. **Inline**: Providing the template as a multi-line string directly in the `.toml` file.++### Templating Context++The Jinja2 template has access to all the "stuffs" currently in the working memory. You can access them by the names they were given in previous pipeline steps. For example, if a previous step produced an output named `user_profile`, you can access its attributes in the template like `{{ user_profile.name }}` or `{{ user_profile.email }}`.++## Configuration++`PipeJinja2` is configured in your pipeline's `.toml` file.++### TOML Parameters++| Parameter | Type | Description | Required |+| --------------- | --------------- | --------------------------------------------------------------------------------------------------------- | --------------------------- |+| `PipeJinja2` | string | A descriptive name for the pipe's function. | Yes |+| `output` | string | The concept for the output. Defaults to `native.Text`. | No |+| `jinja2_name` | string | The name of a pre-loaded template file. | Yes (or `jinja2`) |+| `jinja2` | string | An inline Jinja2 template string. | Yes (or `jinja2_name`) |+| `extra_context` | table (dict) | A table of key-value pairs to add to the rendering context, making them available as variables in the template. | No |++### Example: Generating a report from a template file++This example assumes you have a user profile object and a list of activities in the working memory, and a template file named `weekly_report.md` has been registered.++**`weekly_report.md` template file:**+```jinja+# Weekly Report for {{ user.name }}++Hello {{ user.first_name }},++Here is a summary of your activity this week:++{% for activity in activities %}+- {{ activity.timestamp }}: {{ activity.description }}+{% else %}+- No activity logged this week.+{% endfor %}++Report generated on: {{ report_date }}+```++**Pipeline TOML definition:**+```toml+[pipe.generate_weekly_report]+PipeJinja2 = "Generate a formatted weekly report for a user"+output = "WeeklyReportText"+jinja2_name = "weekly_report.md"+extra_context = { report_date = "2023-10-27" }+```++In this scenario:+- `PipeJinja2` will load the `weekly_report.md` template.+- It will use the `user` and `activities` objects from the working memory.+- It will add `report_date` to the context from the `extra_context` table.+- The rendered markdown text will be saved as the `WeeklyReportText` concept.\ No newline at end of filediff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeLLM.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeLLM.mdnew file mode 100644index 000000000..a6991b3b8--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeLLM.md@@ -0,0 +1,108 @@+# PipeLLM++`PipeLLM` is the core operator in Pipelex for leveraging Large Language Models (LLMs). It can be used for a wide range of tasks, including text generation, summarization, classification, and structured data extraction.++## How it works++At its core, `PipeLLM` constructs a detailed prompt from various inputs and templates, sends it to a specified LLM, and processes the output. It can produce simple text or complex structured data (in the form of Pydantic models).++For structured data output, `PipeLLM` employs two main strategies:++1. **Direct Mode**: The LLM is prompted to directly generate a JSON object that conforms to the target Pydantic model's schema. This is fast but relies on the LLM's ability to generate well-formed JSON.+2. **Preliminary Text Mode**: This is a more robust two-step process:+ a. First, the LLM generates a free-form text based on the initial prompt.+ b. Second, another LLM call is made with a specific prompt designed to extract and structure the information from the generated text into the target Pydantic model.++## Configuration++`PipeLLM` is configured in your pipeline's `.toml` file.++### TOML Parameters++| Parameter | Type | Description | Required |+| --------------------------- | ------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- |+| `PipeLLM` | string | A descriptive name for the LLM operation. | Yes |+| `inputs` | dictionary | The input concept(s) for the LLM operation, as a dictionary mapping input names to concept codes. | Yes |+| `output` | string | The output concept produced by the LLM operation. | Yes |+| `llm` | string or table | Specifies the LLM preset(s) to use. Can be a single preset or a table mapping different presets for different generation modes (e.g., `main`, `object_direct`). | No |+| `system_prompt` | string | A system-level prompt to guide the LLM's behavior (e.g., "You are a helpful assistant"). Can be inline text or a reference to a template file (`"file:path/to/prompt.md"`). | No |+| `prompt` | string | A simple, static user prompt. Use this when you don't need to inject any variables. | No |+| `prompt_template` | string | A template for the user prompt. Use `$` for inline variables (e.g., `$topic`) and `@` to insert the content of an entire input (e.g., `@text_to_summarize`). | No |+| `images` | list of strings | For Vision Language Models (VLMs), specifies which input variables are images. | No |+| `structuring_method` | string | The method for generating structured output. Can be `direct` or `preliminary_text`. Defaults to the global configuration. | No |+| `prompt_template_to_structure` | string | The prompt template for the second step in `preliminary_text` mode. | No |+| `output_multiplicity` | string or integer | Defines the number of outputs. Use `"list"` for a variable-length list, or an integer (e.g., `3`) for a fixed-size list. | No |+++### Simple Text Generation Example++This pipe takes no input and writes a poem.++```toml+[pipe.write_poem]+PipeLLM = "Write a short poem"+output = "Text"+llm = "llm_for_creative_writing"+prompt = """+Write a four-line poem about pipes.+"""+```++### Text-to-Text Example++This pipe summarizes an input text, using a `prompt_template` to inject the input.++```toml+[pipe.summarize_text]+PipeLLM = "Summarize a text"+inputs = { text = "TextToSummarize" }+output = "TextSummary"+prompt_template = """+Please provide a concise summary of the following text:++@text++The summary should be no longer than 3 sentences.+"""+```++### Vision (VLM) Example++This pipe takes an image of a table and uses a VLM to extract the content as an HTML table.++```toml+[pipe.extract_table_from_image]+PipeLLM = "Extract table data from an image"+inputs = { image = "TableScreenshot" }+output = "TableData"+images = ["image"]+prompt_template = """+Extract the table data from this image and format it as a structured table.+"""+```++### Structured Data Extraction Example++This pipe extracts a list of `Expense` items from a block of text.++```toml+[concept.Expense]+structure = "Expense" # Assumes a Pydantic model 'Expense' is defined++[pipe.process_expense_report]+PipeLLM = "Process an expense report"+inputs = { report = "ExpenseReport" }+output = "ProcessedExpenseReport"+prompt_template = """+Analyze this expense report and extract the following information:+- Total amount+- Date+- Vendor+- Category+- Line items++@report+"""+```++In this example, `Pipelex` will instruct the LLM to return a list of objects that conform to the `Expense` structure.diff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeOcr.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeOcr.mdnew file mode 100644index 000000000..53195bdfc--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeOcr.md@@ -0,0 +1,53 @@+# PipeOcr++The `PipeOcr` operator performs Optical Character Recognition (OCR) on images and PDF documents. It extracts text, embedded images, and provides full-page renderings.++## How it works++`PipeOcr` takes a single input, which must be either an `Image` or a `Pdf`. It processes the document page by page and produces a list of `PageContent` objects. Each `PageContent` object encapsulates all the information extracted from a single page.++The output is always a list, even if the input is a single image (in which case the list will contain just one item).++### The `PageContent` Structure++The `PageContent` object has the following structure:++- `text_and_images`: Contains the main extracted content.+ - `text`: The recognized text from the page as a `TextContent`.+ - `images`: A list of any images found embedded within the page.+- `page_view`: An `ImageContent` object representing a full visual rendering of the page.++## Configuration++`PipeOcr` is configured in your pipeline's `.toml` file.++### TOML Parameters++| Parameter | Type | Description | Required |+| --------------------------- | ------- | ---------------------------------------------------------------------------------------------------------------------------------------- | -------- |+| `PipeOcr` | string | A descriptive name for the OCR operation. | Yes |+| `inputs` | dictionary | The input concept(s) for the OCR operation, as a dictionary mapping input names to concept codes. | Yes |+| `output` | string | The output concept produced by the OCR operation. | Yes |+| `should_include_images` | boolean | If `true`, any images found within the document pages will be extracted and included in the output. Defaults to `false`. | No |+| `should_include_page_views` | boolean | If `true`, a high-fidelity image of each page will be included in the `page_view` field. Defaults to `false`. | No |+| `page_views_dpi` | integer | The resolution (in Dots Per Inch) for the generated page views when processing a PDF. Defaults to `150`. | No |+| `should_caption_images` | boolean | If `true`, the OCR service may attempt to generate captions for the images found. *Note: This feature depends on the OCR provider.* | No |++### Example: Processing a PDF++This example defines a pipe that takes a PDF, extracts text and full-page images, and outputs them as a list of pages.++```toml+[concept]+ScannedDocument = "A document that has been scanned as a PDF"+ExtractedPages = "A list of pages extracted from a document by OCR"++[pipe.extract_text_from_document]+PipeOcr = "Extract text from a scanned document"+inputs = { document = "ScannedDocument" }+output = "ExtractedText"+should_include_page_views = true+page_views_dpi = 200+```++To use this pipe, you would first need to load a PDF into the `ScannedDocument` concept. After the pipe runs, the `ExtractedPages` concept will contain a list of `PageContent` objects, where each object has the extracted text and a 200 DPI image of the corresponding page.diff --git a/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/index.md b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/index.mdnew file mode 100644index 000000000..fac87270f--- /dev/null+++ b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/index.md@@ -0,0 +1,78 @@+# Pipe Operators++Pipe operators are the fundamental building blocks in Pipelex, representing a single, focused task. They are the "verbs" of your pipeline that perform the actual work.++Each operator specializes in a specific kind of action, from interacting with Large Language Models to executing custom Python code. You combine these operators using [Pipe Controllers](../pipe-controllers/index.md) to create complex workflows.++## Core Operators++Here are the primary pipe operators available in Pipelex:++- [**`PipeLLM`**](./PipeLLM.md): The core operator for all interactions with Large Language Models (LLMs), including text generation, structured data extraction, and vision tasks.+- [**`PipeOcr`**](./PipeOcr.md): Performs Optical Character Recognition (OCR) on images and PDF documents to extract text and embedded images.+- [**`PipeImgGen`**](./PipeImgGen.md): Generates images from a text prompt using models like DALL-E 3 or Stable Diffusion.+- [**`PipeFunc`**](./PipeFunc.md): An escape hatch that allows you to execute any custom Python function, giving you maximum flexibility.+- [**`PipeJinja2`**](./PipeJinja2.md): Renders a Jinja2 template using data from the working memory, perfect for creating formatted reports or complex prompts.++## Overview++Pipelex provides the following pipe operators:++- `PipeLLM`: For LLM-based text generation and processing+- `PipeOcr`: For optical character recognition and document processing+- `PipeFunc`: For executing custom functions+- `PipeImgGen`: For image generation and manipulation++## PipeLLM++Core operator for LLM-based text generation and processing.++### Key Features++- Text generation+- Structured output generation+- Multiple output modes+- System prompt customization+- LLM configuration++### Key Features++- Sequential execution+- Working memory management+- Sub-pipe handling+- Pipeline composition++## PipeOcr++Processes images and PDFs using Optical Character Recognition.++### Key Features++- PDF processing+- Image processing+- Text extraction+- Image extraction+- Page view generation++## PipeFunc++Executes custom functions within the pipeline.++### Key Features++- Custom function execution+- Working memory integration+- Multiple output types+- Function registry integration++## PipeImgGen++Generates and manipulates images.++### Key Features++- Image generation+- Quality control+- Multiple output formats+- Batch processing+- Parameter customizationdiff --git a/doc/pages/Pipes/pipe_toml.png b/docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/pipe_toml.pngsimilarity index 100%rename from doc/pages/Pipes/pipe_toml.pngrename to docs/pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/pipe_toml.pngdiff --git a/docs/pages/configuration/config-advanced/feature-config.md b/docs/pages/configuration/config-advanced/feature-config.mdnew file mode 100644index 000000000..1997cab6a--- /dev/null+++ b/docs/pages/configuration/config-advanced/feature-config.md@@ -0,0 +1,76 @@+# Feature Configuration++The `FeatureConfig` class controls which features are enabled in Pipelex.++## Configuration Options++```python+class FeatureConfig(ConfigModel):+ is_pipeline_tracking_enabled: bool+ is_activity_tracking_enabled: bool+ is_reporting_enabled: bool+```++### Fields++- `is_pipeline_tracking_enabled`: When true, enables pipeline tracking functionality+- `is_activity_tracking_enabled`: When true, enables activity tracking functionality+- `is_reporting_enabled`: When true, enables the reporting system++## Impact on Dependency Injection++The feature flags directly affect which implementation is used for certain components:++| Feature Flag | When True | When False |+|--------------|-----------|------------|+| `is_pipeline_tracking_enabled` | `PipelineTracker` | `PipelineTrackerNoOp` |+| `is_activity_tracking_enabled` | `ActivityManager` | `ActivityManagerNoOp` |+| `is_reporting_enabled` | `ReportingManager` | `ReportingNoOp` |++## Feature Details++### Pipeline Tracking++```toml+is_pipeline_tracking_enabled = false+```++- Controls whether pipeline execution tracking is enabled+- When enabled, tracks the flow and execution of pipelines using by default mermaid chart:+ - View and edit charts at [Mermaid Live Editor](https://mermaid.live)+- Useful for debugging and monitoring pipeline behavior+- Default: `false`++### Activity Tracking++```toml+is_activity_tracking_enabled = false+```++- Controls whether activity tracking is enabled+- When enabled, tracks detailed information about system activities+- Default: `false`++### Reporting++```toml+is_reporting_enabled = true+```++- Controls whether reporting functionality is enabled+- When enabled, generates the cost report of the pipelex execution (LLM costs, OCR costs, etc...)+- Default: `true`++## Example Configuration++```toml+[pipelex.feature_config]+# Enable pipeline tracking for debugging+is_pipeline_tracking_enabled = true++# Disable activity tracking for performance+is_activity_tracking_enabled = false++# Enable reporting for cost monitoring+is_reporting_enabled = true+```diff --git a/docs/pages/configuration/config-pipeline-validation/dry-run-config.md b/docs/pages/configuration/config-pipeline-validation/dry-run-config.mdnew file mode 100644index 000000000..6cb4998d8--- /dev/null+++ b/docs/pages/configuration/config-pipeline-validation/dry-run-config.md@@ -0,0 +1,70 @@+# Dry Run Configuration++The `DryRunConfig` class controls how Pipelex behaves during dry runs.++## Configuration Options++```python+class DryRunConfig(ConfigModel):+ apply_to_jinja2_rendering: bool+ text_gen_truncate_length: int+```++### Fields++- `apply_to_jinja2_rendering`: When true, simulates Jinja2 template rendering during dry runs+- `text_gen_truncate_length`: Maximum length of generated text during dry runs++## Example Configuration++```toml+[pipelex.dry_run_config]+apply_to_jinja2_rendering = true+text_gen_truncate_length = 100+```++## Dry Run Behavior++### Template Rendering++When `apply_to_jinja2_rendering` is true:++- Templates are processed but not actually rendered+- Variables are validated+- Template syntax is checked+- No actual content is generated++### Text Generation++The `text_gen_truncate_length` controls:++- Maximum length of simulated text output+- Helps prevent excessive resource usage during testing+- Makes dry run output more manageable++## Use Cases++1. **Testing Pipeline Logic**++ - Validate pipeline structure+ - Check template syntax+ - Verify variable references++2. **Resource Estimation**++ - Estimate processing time+ - Calculate potential costs+ - Plan resource allocation++3. **Debugging**++ - Trace execution paths+ - Identify potential issues+ - Test error handling++## Best Practices++- Use dry runs for testing before production+- Set appropriate truncation lengths+- Enable template validation when testing templates+- Review dry run logs for potential issuesdiff --git a/docs/pages/configuration/config-pipeline-validation/static-validation-config.md b/docs/pages/configuration/config-pipeline-validation/static-validation-config.mdnew file mode 100644index 000000000..371ca1c72--- /dev/null+++ b/docs/pages/configuration/config-pipeline-validation/static-validation-config.md@@ -0,0 +1,56 @@+# Static Validation Configuration++The `StaticValidationConfig` class controls how Pipelex handles validation errors during static analysis.++## Configuration Options++```python+class StaticValidationReaction(StrEnum):+ RAISE = "raise" # Raise an exception+ LOG = "log" # Log the error but continue+ IGNORE = "ignore" # Silently ignore the error++class StaticValidationConfig(ConfigModel):+ default_reaction: StaticValidationReaction+ reactions: Dict[StaticValidationErrorType, StaticValidationReaction]+```++### Fields++- `default_reaction`: Default reaction for validation errors not specifically configured+- `reactions`: Dictionary mapping specific error types to their reactions++## Error Types and Reactions++Each validation error type can be configured to have one of three reactions:++- `RAISE`: Stops execution and raises an exception+- `LOG`: Logs the error but allows execution to continue+- `IGNORE`: Silently ignores the error++## Example Configuration++```toml+[pipelex.static_validation_config]+default_reaction = "raise"++[pipelex.static_validation_config.reactions]+missing_input_variable = "log"+extraneous_input_variable = "log"+inadequate_input_concept = "log"+```++## Validation Process++1. When a validation error is detected, the system looks up the error type in the `reactions` dictionary+2. If the error type is found, the corresponding reaction is used+3. If the error type is not found, the `default_reaction` is used+4. The reaction determines how the error is handled++## Best Practices++- Use `RAISE` for critical errors that should never be ignored+- Use `LOG` for warnings that should be addressed but aren't critical+- Use `IGNORE` sparingly and only for known, harmless cases+- Configure specific reactions for known error types+- Use a reasonable `default_reaction` for unexpected error typesdiff --git a/docs/pages/configuration/config-practical/logging-config.md b/docs/pages/configuration/config-practical/logging-config.mdnew file mode 100644index 000000000..971a983f0--- /dev/null+++ b/docs/pages/configuration/config-practical/logging-config.md@@ -0,0 +1,192 @@+# Logging Configuration++Configuration section: `[pipelex.log_config]`++## Overview++The logging configuration controls how Pipelex handles log messages, their formatting, and their display. The configuration is split into general logging settings and Rich console output settings.++## General Settings++### Log Levels++```toml+default_log_level = "INFO"+```++- Sets the default logging level for all loggers+- Valid values: `"VERBOSE"`, `"DEBUG"`, `"DEV"`, `"INFO"`, `"WARNING"`, `"ERROR"`, `"CRITICAL"`, `"OFF"`++### Package-Specific Log Levels++```toml+[pipelex.log_config.package_log_levels]+anthropic = "INFO"+asyncio = "INFO"+botocore = "INFO"+openai = "INFO"+pipelex = "INFO"+```++- Override log levels for specific packages+- Use `-` instead of `.` in package names (e.g., `urllib3-connectionpool`)++### Console Output++```toml+is_console_logging_enabled = true+```++- Enable/disable console logging+- Default: `true`++### JSON Formatting++```toml+json_logs_indent = 4+presentation_line_width = 120+```++- `json_logs_indent`: Indentation for JSON log output+- `presentation_line_width`: Maximum line width for formatted output++### Caller Information++```toml+is_caller_info_enabled = false+caller_info_template = "file_line"+```++Available templates:++- `"file_line"`: "file.py:123"+- `"file_line_func"`: "file.py:123 function_name"+- `"func"`: "function_name"+- `"file_func"`: "file.py function_name"+- `"func_line"`: "function_name 123"+- `"func_module"`: "function_name module"+- `"func_module_line"`: "function_name module 123"++### Problem Silencing++```toml+silenced_problem_ids = ["azure_openai_no_stream_options"]+```++- List of problem IDs to silence+- Prevents specific warnings from being logged++### Poor Loggers++```toml+generic_poor_logger = "#sandbox"+poor_loggers = [+ "kajson.decoder.sandbox",+ "kajson.encoder.sandbox",+ "class_registry.sandbox"+]+```++- Loggers that use simplified formatting without fancy features+- Useful for sandbox environments or when simple logging is needed++## Rich Console Configuration++Configuration section: `[pipelex.log_config.rich_log_config]`++### Display Options++```toml+is_show_time = false+is_show_level = true+is_link_path_enabled = true+```++- `is_show_time`: Show timestamp in logs+- `is_show_level`: Show log level+- `is_link_path_enabled`: Make file paths clickable++### Syntax Highlighting++```toml+highlighter_name = "json" # or "repr"+is_markup_enabled = false+```++- `highlighter_name`: Choose between JSON or repr highlighting+- `is_markup_enabled`: Enable Rich markup syntax in log messages++### Traceback Settings++```toml+is_rich_tracebacks = true+is_tracebacks_word_wrap = true+is_tracebacks_show_locals = false+tracebacks_suppress = []+```++- Control how Python tracebacks are displayed+- Enable/disable word wrapping and local variable display+- Suppress specific traceback patterns++### Keyword Highlighting++```toml+keywords_to_hilight = []+```++- List of keywords to highlight in log messages+- Useful for emphasizing important terms++## Example Configuration++```toml+[pipelex.log_config]+default_log_level = "INFO"+is_console_logging_enabled = true+json_logs_indent = 4+presentation_line_width = 120+is_caller_info_enabled = true+caller_info_template = "file_line_func"++silenced_problem_ids = []+generic_poor_logger = "#sandbox"+poor_loggers = []++[pipelex.log_config.package_log_levels]+pipelex = "INFO"+openai = "WARNING"+anthropic = "INFO"++[pipelex.log_config.rich_log_config]+is_show_time = false+is_show_level = true+is_link_path_enabled = true+highlighter_name = "json"+is_markup_enabled = false+is_rich_tracebacks = true+is_tracebacks_word_wrap = true+is_tracebacks_show_locals = false+tracebacks_suppress = []+keywords_to_hilight = ["error", "warning", "failed"]+```++## Best Practices++1. **Development Environment**:++ - Enable caller info for better debugging+ - Use verbose logging levels+ - Enable local variables in tracebacks++2. **Production Environment**:++ - Disable caller info for performance+ - Use INFO or higher log levels+ - Disable local variables in tracebacks++3. **Package Log Levels**:++ - Set noisy third-party packages to WARNING+ - Keep pipelex at INFO for important updates+ - Use VERBOSE only when debugging specific issuesdiff --git a/docs/pages/configuration/config-practical/pipe-run-config.md b/docs/pages/configuration/config-practical/pipe-run-config.mdnew file mode 100644index 000000000..13acc214f--- /dev/null+++ b/docs/pages/configuration/config-practical/pipe-run-config.md@@ -0,0 +1,34 @@+# Pipe Run Configuration++The `PipeRunConfig` class controls execution parameters for pipes in Pipelex.++## Configuration Options++```python+class PipeRunConfig(ConfigModel):+ pipe_stack_limit: int+```++### Fields++- `pipe_stack_limit`: Maximum depth of nested pipe executions allowed++## Example Configuration++```toml+[pipelex.pipe_run_config]+pipe_stack_limit = 100+```++## Stack Limit++The `pipe_stack_limit` prevents infinite recursion in pipe execution by:++- Limiting the depth of nested pipe calls+- Throwing an exception when the limit is exceeded+- Protecting against accidental circular dependencies++## Best Practices++- Set a reasonable stack limit based on your pipeline complexity+- Monitor stack usage in complex pipelinesdiff --git a/docs/pages/configuration/config-practical/reporting-config.md b/docs/pages/configuration/config-practical/reporting-config.mdnew file mode 100644index 000000000..954b046b3--- /dev/null+++ b/docs/pages/configuration/config-practical/reporting-config.md@@ -0,0 +1,89 @@+# Reporting Configuration++Configuration section: `[pipelex.reporting_config]`++## Overview++The reporting configuration controls how Pipelex generates and manages cost reports for various operations (LLM usage, OCR processing, etc.). This configuration allows you to customize where and how cost reports are generated.++## Configuration Options++### Console Logging++```toml+is_log_costs_to_console = false+```++- Controls whether costs are logged to the console in **real-time**+- When enabled, displays token cost reports after each LLM operation (and other operations with costs)+- Default: `false`++### Cost Report File Generation++```toml+is_generate_cost_report_file_enabled = true+```++- Controls whether cost report files are generated after the execution of the pipeline+- When enabled, creates detailed reports in the xlsx format in the directory `reports/` by default.+- Default: `true`++### Report File Location++```toml+cost_report_dir_path = "reports"+```++- Directory where cost report files will be saved+- Path is relative to your project root+- Directory will be created if it doesn't exist.+- Default: `"reports"`++### Report File Naming++```toml+cost_report_base_name = "cost_report"+cost_report_extension = "xlsx"+```++- Base name for report files+- File extension for report files (e.g., "xlsx" for Excel files)+- Final filename will include an incremental number (e.g., `cost_report_1.xlsx`, `cost_report_2.xlsx`)+- Default: Base name = `"cost_report"`, Extension = `"xlsx"`++### Cost Unit Scaling++```toml+cost_report_unit_scale = 1.0+```++- Scaling factor applied to cost values in reports+- Use 1.0 for actual costs+- Use different values to scale costs (e.g., 1000.0 for thousands)+- The currency is in USD.+- Default: `1.0`++## Example Configuration++```toml+[pipelex.reporting_config]+# Enable console logging for development+is_log_costs_to_console = true++# Generate report files+is_generate_cost_report_file_enabled = true++# Store reports in a custom directory+cost_report_dir_path = "analytics/cost_reports"++# Customize report file naming+cost_report_base_name = "llm_costs"+cost_report_extension = "xlsx"++# Show costs in thousands+cost_report_unit_scale = 1000.0+```++## Best Practices++⚠️ Under constructiondiff --git a/docs/pages/configuration/config-practical/tracker-config.md b/docs/pages/configuration/config-practical/tracker-config.mdnew file mode 100644index 000000000..902eebf02--- /dev/null+++ b/docs/pages/configuration/config-practical/tracker-config.md@@ -0,0 +1,130 @@+# Tracker Configuration++The Tracker Configuration allows you to customize how your pipeline execution is tracked and visualized. The tracker creates a Mermaid flowchart visualization of your pipeline's execution, showing how data flows through different pipes and transformations.++## Overview++The pipeline tracker visualizes:++- Stuff nodes (data objects) and their transformations+- Pipe execution steps+- Batch processing branches+- Condition nodes and choices+- Aggregation steps++## Configuration Options++### Basic Settings++- `is_debug_mode` (bool): Enable or disable debug mode for tracking+ - When enabled, shows additional information like:+ - Node codes and internal identifiers+ - Extended comments for repeated nodes+ - Pipe codes in edge captions with comments++- `is_include_text_preview` (bool): Whether to include text previews in the tracking interface+ - When enabled, shows first 100 characters of text content in stuff nodes+ - Only applies to text-based stuff objects++- `is_include_interactivity` (bool): Enable or disable interactive features in the tracking interface++### Visual Settings++- `theme` (str | "auto"): The visual theme to use for the Mermaid flowchart++ - Use "auto" for automatic theme selection based on context+ - Or specify a custom theme name++- `layout` (str | "auto"): The layout algorithm to use for graph visualization++ - Use "auto" for automatic layout selection+ - Or specify a custom layout algorithm+ - Affects how nodes and edges are arranged in the flowchart++- `wrapping_width` (int | "auto"): Text wrapping width for node labels++ - Use "auto" for automatic width adjustment+ - Or specify a fixed width in characters+ - Helps control the visual width of node content++- `nb_items_limit` (int | "unlimited"): Maximum number of items to display++ - Use "unlimited" for no limit+ - Or specify a maximum number of items+ - Helps manage visualization of large pipelines++### Graph Styling++- `sub_graph_colors` (List[str]): List of colors to use for sub-graphs++ - Colors are used to visually distinguish different pipeline layers+ - Example: `["#1f77b4", "#ff7f0e", "#2ca02c"]`++- Edge Styles (all str type):++ - `pipe_edge_style`: Style for regular pipe transformation edges+ - `branch_edge_style`: Style for batch processing branch edges+ - `aggregate_edge_style`: Style for aggregation step edges+ - `condition_edge_style`: Style for condition evaluation edges+ - `choice_edge_style`: Style for condition choice result edges++## Example Configuration++```toml+[tracker]+is_debug_mode = false+is_include_text_preview = true+is_include_interactivity = true+theme = "auto"+layout = "auto"+wrapping_width = "auto"+nb_items_limit = "unlimited"+sub_graph_colors = ["#1f77b4", "#ff7f0e", "#2ca02c"]+pipe_edge_style = "solid"+branch_edge_style = "dashed"+aggregate_edge_style = "dotted"+condition_edge_style = "dashdot"+choice_edge_style = "solid"+```++## Property Accessors++The TrackerConfig class provides convenient property accessors that handle the "auto" and "unlimited" special values:++- `applied_theme`: Returns None for "auto", otherwise returns the theme name+- `applied_layout`: Returns None for "auto", otherwise returns the layout name+- `applied_wrapping_width`: Returns None for "auto", otherwise returns the width as an integer+- `applied_nb_items_limit`: Returns None for "unlimited", otherwise returns the limit as an integer++These properties make it easy to work with the configuration values in your code while maintaining the flexibility of automatic settings.++## Visualization Features++The tracker generates Mermaid flowcharts with the following features:++1. Node Types:++ - Start node (special)+ - Stuff nodes (data objects)+ - Condition nodes (for pipeline branching)++2. Edge Types:++ - Pipe edges (regular transformations)+ - Branch edges (batch processing)+ - Aggregate edges (combining results)+ - Condition edges (with expressions)+ - Choice edges (condition results)++3. Node Content:++ - Concept type and name+ - Content preview (for text)+ - Debug information (when debug mode is enabled)+ - Comments and descriptions++4. Visual Organization:++ - Sub-graphs for different pipeline layers+ - Color coding for visual distinction+ - Different edge styles for different types of connectionsdiff --git a/docs/pages/configuration/config-technical/aws-config.md b/docs/pages/configuration/config-technical/aws-config.mdnew file mode 100644index 000000000..4f9826247--- /dev/null+++ b/docs/pages/configuration/config-technical/aws-config.md@@ -0,0 +1,54 @@+# AWS Configuration++Configuration section: `[pipelex.aws_config]`++## Overview++The AWS configuration controls how Pipelex authenticates with AWS services. It supports two authentication methods: environment variables or a secret provider.++## Authentication Methods++```toml+[pipelex.aws_config]+api_key_method = "env" # or "secret_provider"+```++### Environment Variables Method (`"env"`)+When using `api_key_method = "env"`, Pipelex expects the following environment variables:+- `AWS_ACCESS_KEY_ID`: Your AWS access key ID+- `AWS_SECRET_ACCESS_KEY`: Your AWS secret access key+- `AWS_REGION`: Your AWS region++Example `.env` file:+```env+AWS_ACCESS_KEY_ID=your_access_key_id+AWS_SECRET_ACCESS_KEY=your_secret_access_key+AWS_REGION=us-east-1+```++### Secret Provider Method (`"secret_provider"`)+When using `api_key_method = "secret_provider"`, Pipelex will:+1. Connect to your configured secret provider+2. Look for the same keys as environment variables:+ - `AWS_ACCESS_KEY_ID`+ - `AWS_SECRET_ACCESS_KEY`+ - `AWS_REGION`++⚠️ **Important**: To use the secret provider method, you must:+1. Configure a secret provider in your project using the `SecretsProviderProtocol`: See more in the [Secrets](../../advanced-customization/secrets-provider-injection.md) documentation.+2. Store your AWS credentials in your secret provider+3. Ensure your secret provider is properly authenticated++## Dependency Injection++Pipelex uses dependency injection to manage AWS clients and credentials. You can:++- Inject custom AWS client implementations+- Override default credential providers+- Mock AWS services for testing++For detailed information about dependency injection, including examples and best practices, see the [Dependency Injection](../../advanced-customization/index.md) documentation.++## Best Practices++⚠️ Under constructiondiff --git a/docs/pages/configuration/config-technical/cogt-config.md b/docs/pages/configuration/config-technical/cogt-config.mdnew file mode 100644index 000000000..02d703d0b--- /dev/null+++ b/docs/pages/configuration/config-technical/cogt-config.md@@ -0,0 +1,197 @@+# Cognitive Tools (Cogt) Configuration++The Cogt configuration manages all cognitive tools in Pipelex, including LLM (Language Models), IMGG (Image Generation), and OCR (Optical Character Recognition) capabilities.++## Overview++```toml+[pipelex.cogt]+# Main Cogt configuration sections+[pipelex.cogt.inference_manager_config]+[pipelex.cogt.llm_config]+[pipelex.cogt.imgg_config]+[pipelex.cogt.ocr_config]+```++## Inference Manager Configuration++Controls automatic setup of various cognitive tools:++```toml+[pipelex.cogt.inference_manager_config]+is_auto_setup_preset_llm = true+is_auto_setup_preset_imgg = true+is_auto_setup_preset_ocr = true+```++## LLM Configuration++Configuration for Language Model interactions:++```toml+[pipelex.cogt.llm_config]+default_max_images = 4 # Maximum number of images in prompts++# Platform preferences for different LLMs+[pipelex.cogt.llm_config.preferred_platforms]+gpt-4 = "openai"+claude-3-opus = "anthropic"++# Job configuration+[pipelex.cogt.llm_config.llm_job_config]+is_streaming_enabled = true+max_retries = 3 # Between 1 and 10++# Instructor settings+[pipelex.cogt.llm_config.instructor_config]+is_openai_structured_output_enabled = true+```++### LLM Job Parameters++When configuring LLM jobs, you can set:++- `temperature` (float, 0-1): Controls randomness in outputs+- `max_tokens` (optional int): Maximum tokens in response+- `seed` (optional int): For reproducible outputs++## Image Generation (IMGG) Configuration++Configuration for image generation capabilities:++```toml+[pipelex.cogt.imgg_config]+default_imgg_handle = "stable_diffusion"+imgg_handles = ["stable_diffusion", "dall_e"]++[pipelex.cogt.imgg_config.imgg_job_config]+is_sync_mode = true++# Default parameters for image generation+[pipelex.cogt.imgg_config.imgg_param_defaults]+aspect_ratio = "square" # Options: square, landscape_4_3, landscape_3_2, landscape_16_9, landscape_21_9,+ # portrait_4_3, portrait_2_3, portrait_9_16, portrait_9_21+background = "auto" # Options: transparent, opaque, auto+quality = "high" # Options: low, medium, high+nb_steps = 50 # Number of diffusion steps+guidance_scale = 7.5 # Controls adherence to prompt+is_moderated = true # Enable content moderation+safety_tolerance = 3 # Safety level (1-6)+is_raw = false # Raw output mode+output_format = "png" # Options: png, jpg, webp+seed = "auto" # "auto" or specific integer+```++### IMGG Job Parameters++Image generation jobs support these parameters:++- **Dimensions**:++ - `aspect_ratio`: Predefined ratios for image dimensions+ - `background`: Background handling mode++- **Quality Control**:++ - `quality`: Output quality level+ - `nb_steps`: Number of generation steps+ - `guidance_scale`: How closely to follow the prompt++- **Safety**:++ - `is_moderated`: Enable content moderation+ - `safety_tolerance`: Safety check strictness (1-6)++- **Output**:++ - `is_raw`: Raw output mode+ - `output_format`: Image format (PNG/JPG/WEBP)+ - `seed`: For reproducible generation++## OCR Configuration++Configuration for Optical Character Recognition:++```toml+[pipelex.cogt.ocr_config]+ocr_handles = ["tesseract", "azure_ocr"]+page_output_text_file_name = "page_text.txt"+```++## Validation Rules++### LLM Configuration+- Temperature must be between 0 and 1+- Max tokens must be positive+- Max retries must be between 1 and 10+- Seeds must be non-negative++### IMGG Configuration+- Guidance scale must be positive+- Safety tolerance must be between 1 and 6+- Number of steps must be positive+- Strict validation for enums (aspect ratio, background, quality, output format)++## Best Practices++1. **LLM Settings**:++ - Start with lower temperatures (0.1-0.3) for consistent outputs+ - Use streaming for better user experience+ - Set appropriate retry limits based on your use case++2. **IMGG Settings**:++ - Enable moderation for production use+ - Use appropriate aspect ratios for your use case+ - Balance quality and performance with step count++3. **General**:++ - Enable auto-setup for easier initialization+ - Use platform preferences to ensure consistent model selection+ - Configure OCR handles based on your accuracy needs++## Example Complete Configuration++```toml+[pipelex.cogt]+[pipelex.cogt.inference_manager_config]+is_auto_setup_preset_llm = true+is_auto_setup_preset_imgg = true+is_auto_setup_preset_ocr = true++[pipelex.cogt.llm_config]+default_max_images = 4+preferred_platforms = { "gpt-4" = "openai", "claude-3-opus" = "anthropic" }++[pipelex.cogt.llm_config.llm_job_config]+is_streaming_enabled = true+max_retries = 3++[pipelex.cogt.llm_config.instructor_config]+is_openai_structured_output_enabled = true++[pipelex.cogt.imgg_config]+default_imgg_handle = "stable_diffusion"+imgg_handles = ["stable_diffusion", "dall_e"]++[pipelex.cogt.imgg_config.imgg_job_config]+is_sync_mode = true++[pipelex.cogt.imgg_config.imgg_param_defaults]+aspect_ratio = "square"+background = "auto"+quality = "high"+nb_steps = 50+guidance_scale = 7.5+is_moderated = true+safety_tolerance = 3+is_raw = false+output_format = "png"+seed = "auto"++[pipelex.cogt.ocr_config]+ocr_handles = ["tesseract", "azure_ocr"]+page_output_text_file_name = "page_text.txt"+```diff --git a/docs/pages/configuration/config-technical/library-config.md b/docs/pages/configuration/config-technical/library-config.mdnew file mode 100644index 000000000..e621067ec--- /dev/null+++ b/docs/pages/configuration/config-technical/library-config.md@@ -0,0 +1,133 @@+# Library Configuration++The Library Configuration manages how Pipelex organizes, loads, and handles libraries in your project. Libraries in Pipelex include pipelines, LLM integrations, LLM decks, and templates.++## Directory Structure++The library system uses two main root directories:++- Internal library root (`pipelex/libraries`): Contains the base libraries shipped with Pipelex+- Exported library root (`pipelex_libraries`): Contains your project's libraries, including copies of base libraries++### Standard Paths++```+pipelex_libraries/ # Exported library root+├── pipelines/ # Pipeline definitions+│ ├── base_library/ # Base pipelines from Pipelex+│ └── your_pipelines/ # Your custom pipelines+├── llm_integrations/ # LLM integration configurations+├── llm_deck/ # LLM model configurations+└── templates/ # Template files+```++## Library Loading Process++1. **Domain Loading**:++ - Loads domain definitions first+ - Each domain must be defined exactly once+ - Supports system prompts and structure templates++2. **Concept Loading**:++ - Loads native concepts first+ - Loads custom concepts from TOML files+ - Validates concept definitions and relationships++3. **Pipe Loading**:++ - Loads pipe definitions after concepts+ - Validates pipe configurations+ - Links pipes with their respective domains++## Configuration Options++### Path Configuration++All paths are configurable through class variables in `LibraryConfig`:++```python+package_name = "pipelex"+internal_library_root = "libraries"+exported_library_root = "pipelex_libraries"+exported_pipelines_path = "pipelex_libraries/pipelines"+exported_llm_integrations_path = "pipelex_libraries/llm_integrations"+exported_llm_deck_path = "pipelex_libraries/llm_deck"+exported_templates_path = "pipelex_libraries/templates"+```++### Library Initialization++Use the CLI command to initialize libraries:+```bash+pipelex init-libraries+```++This will:++1. Create the necessary directory structure+2. Copy base libraries to your project+3. Set up initial configuration files++### Library Export Options++When exporting libraries to your project:++- Use `overwrite=True` to force update existing files+- Preserve custom overrides in `llm_deck/overrides.toml`+- Maintain directory structure and initialization files++## Validation++The library manager performs several validation steps:++1. **LLM Deck Validation**:++ - Ensures LLM configurations are complete+ - Validates model settings++2. **Concept Library Validation**:++ - Checks concept relationships+ - Validates concept definitions++3. **Pipe Library Validation**:++ - Verifies pipe configurations+ - Checks domain relationships++4. **Domain Library Validation**:++ - Ensures domain completeness+ - Validates domain relationships++## Error Handling++The library system includes specific error types:++- `LibraryError`: Base error for library issues+- `LibraryParsingError`: For TOML parsing issues+- `ConceptLibraryError`: For concept-related issues+- `PipeLibraryError`: For pipe-related issues+- `LLMDeckNotFoundError`: For missing LLM configurations++## Best Practices++1. **Organization**:++ - Keep related concepts and pipes in the same TOML file+ - Use meaningful domain names+ - Structure complex libraries using subdirectories++2. **Validation**:++ - Run `pipelex validate` after making changes+ - Check for domain consistency+ - Verify concept relationships++3. **Customization**:++ - Use `overrides.toml` for local LLM settings+ - Keep custom pipelines separate from base library+ - Document domain-specific configurationsdiff --git a/docs/pages/configuration/index.md b/docs/pages/configuration/index.mdnew file mode 100644index 000000000..3777eec51--- /dev/null+++ b/docs/pages/configuration/index.md@@ -0,0 +1,73 @@+# Configuration++## Overview++Pipelex uses a TOML-based configuration system. The main configuration file `pipelex.toml` must be located at the root of your project. You can create this file by running:++```bash+pipelex init-config+```++💡 **Important Notes**:++1. `pipelex init-config` creates a **template** configuration file with sample settings. It does not include all possible configuration options - it's meant as a starting point.+2. Using `pipelex init-config --reset` will **overwrite** your existing `pipelex.toml` file without warning. Make sure to backup your configuration before using this flag.++For a complete list of all possible configuration options, refer to the configuration group documentation below.++## Configuration Structure++The configuration is organized into three main sections:++1. `[pipelex]` - Core Pipelex settings+2. `[cogt]` - Cognitive tools and LLM settings+3. `[plugins]` - Plugin-specific configurations++Each section contains multiple subsections for specific features and functionalities.++## Configuration Override System++Pipelex uses a sophisticated configuration override system that loads and merges configurations in a specific order. This allows for fine-grained control over settings in different environments and scenarios.++The exact loading sequence is:++1. Base configuration from the installed Pipelex package (`pipelex.toml`)+2. Your project's base configuration (`pipelex.toml` in your project root)+3. Local overrides (`pipelex_local.toml`)+4. Environment-specific overrides (`pipelex_{environment}.toml`)+ - Example environments: dev, staging, prod -> based on the environment variable `ENV` in your .env file+5. Run mode overrides (`pipelex_{run_mode}.toml`)+ - Example run modes: normal, unit_test+6. Super user overrides (`pipelex_super.toml`) (recommanded to put in .gitignore)++Each subsequent configuration file in this sequence can override settings from the previous ones. This means:+- Settings in `pipelex_local.toml` override the base configuration+- Environment-specific settings override local settings+- Run mode settings override environment settings+- Super user settings override all previous settings++### Override File Naming++- Base config: `pipelex.toml`+- Local overrides: `pipelex_local.toml`+- Environment overrides: `pipelex_dev.toml`, `pipelex_staging.toml`, `pipelex_prod.toml`, etc.+- Run mode overrides: `pipelex_normal.toml`, `pipelex_unit_test.toml`, etc.+- Super user overrides: `pipelex_super.toml`++NB: The run_mode unit_test is used for testing purposes.++### Best Practices for Overrides++1. Use the base `pipelex.toml` for default settings+2. Use `pipelex_local.toml` for machine-specific settings+3. Use environment files for environment-specific settings (dev, staging, prod)+4. Use run mode files for normal or unit_test configurations+5. Use `pipelex_super.toml` sparingly, only for temporary overrides (add to .gitignore)++## Best Practices++1. **Version Control**: Include your base `pipelex.toml` in version control+2. **Environment Overrides**: Use environment-specific files for sensitive or environment-dependent settings+3. **Documentation**: Comment any custom settings for team reference+4. **Validation**: Run `pipelex validate` after making configuration changes+5. **Gitignore**: Add local and sensitive override files to `.gitignore`diff --git a/docs/pages/cookbook-examples/extract-dpe.md b/docs/pages/cookbook-examples/extract-dpe.mdnew file mode 100644index 000000000..744209d65--- /dev/null+++ b/docs/pages/cookbook-examples/extract-dpe.md@@ -0,0 +1,82 @@+# Example: DPE Extraction++This example demonstrates how to extract information from a French "Diagnostic de Performance Énergétique" (DPE) document. This is a specialized document, and the pipeline is tailored to its specific structure.++## Get the code++[**➡️ View on GitHub: examples/extract_dpe.py**](https://github.com/Pipelex/pipelex-cookbook/blob/main/examples/extract_dpe.py)++## The Pipeline Explained++The pipeline `power_extractor_dpe` is designed to recognize and extract the key information from a DPE document. The result is a structured `Dpe` object.++```python+async def extract_dpe(pdf_url: str) -> Dpe:+ working_memory = WorkingMemoryFactory.make_from_pdf(+ pdf_url=pdf_url,+ concept_str="PDF",+ name="pdf",+ )+ pipe_output, _ = await execute_pipeline(+ pipe_code="power_extractor_dpe",+ working_memory=working_memory,+ )+ working_memory = pipe_output.working_memory+ dpe: Dpe = working_memory.get_list_stuff_first_item_as(name="dpe", item_type=Dpe)+ return dpe+```++This example shows how Pipelex can be used for very specific document extraction tasks by creating custom pipelines and data models.++## The Data Structure: `Dpe` Model++The pipeline extracts a `Dpe` object, which is structured to hold the specific information found in a French "Diagnostic de Performance Énergétique". It even uses a custom `IndexScale` enum for the energy efficiency classes.++```python+class IndexScale(StrEnum):+ A = "A"+ B = "B"+ C = "C"+ D = "D"+ E = "E"+ F = "F"+ G = "G"+++class Dpe(StructuredContent):+ address: Optional[str] = None+ date_of_issue: Optional[datetime] = None+ date_of_expiration: Optional[datetime] = None+ energy_efficiency_class: Optional[IndexScale] = None+ per_year_per_m2_consumption: Optional[float] = None+ co2_emission_class: Optional[IndexScale] = None+ per_year_per_m2_co2_emissions: Optional[float] = None+ yearly_energy_costs: Optional[float] = None+```++## The Pipeline Definition: `extract_dpe.toml`++The pipeline uses a `PipeLLM` with a very specific prompt to extract the information from the document. The combination of the image and the OCR text allows the LLM to accurately capture all the details.++```toml+[pipe.write_markdown_from_page_content_dpe]+PipeLLM = "Write markdown from page content of a 'Diagnostic de Performance Energetique'"+inputs = { page_content = "Page" }+output = "Dpe" # The output is structured as a Dpe object+images = ["page_content.page_view"]+llm = "llm_for_img_to_text"+structuring_method = "preliminary_text"+system_prompt = """You are a multimodal LLM, expert at converting images into perfect markdown."""+prompt_template = """+You are given an image of a French 'Diagnostic de Performance Energetique'.+Your role is to convert the image into perfect markdown.++To help you do so, you are given the text extracted from the page by an OCR model.+@page_content.text_and_images.text.text++- It is very important that you collect every element, especially if they are related to the energy performance of the building.+- We value letters like "A, B, C, D, E, F, G" as they are energy performance classes.+# ... (prompt continues)+"""+```+This is a great example of how to create a highly specialized extraction pipeline by combining a custom data model with a detailed, guiding prompt.\ No newline at end of filediff --git a/docs/pages/cookbook-examples/extract-gantt.md b/docs/pages/cookbook-examples/extract-gantt.mdnew file mode 100644index 000000000..9b0290e6f--- /dev/null+++ b/docs/pages/cookbook-examples/extract-gantt.md@@ -0,0 +1,214 @@+# Example: Gantt Chart Extraction++This example showcases the ability of Pipelex to extract structured information from images. In this case, it processes an image of a Gantt chart and extracts the tasks, dates, and dependencies.++## Get the code++[**➡️ View on GitHub: examples/extract_gantt.py**](https://github.com/Pipelex/pipelex-cookbook/blob/main/examples/extract_gantt.py)++## The Pipeline Explained++The pipeline takes an image as input, creates a working memory, and then executes the `extract_gantt_by_steps` pipeline to produce a structured `GanttChart` object.++```python+async def extract_gantt(image_url: str) -> GanttChart:+ # Create Working Memory+ working_memory = WorkingMemoryFactory.make_from_image(+ image_url=image_url,+ concept_str="gantt.GanttImage",+ name="gantt_chart_image",+ )++ # Run the pipe+ pipe_output, _ = await execute_pipeline(+ pipe_code="extract_gantt_by_steps",+ working_memory=working_memory,+ )++ # Output the result+ return pipe_output.main_stuff_as(content_type=GanttChart)+```++This is a powerful demonstration of multi-modal capabilities, combining vision and language understanding.++## The Data Structure: `GanttChart` Model++The final output is a `GanttChart` object, which contains lists of tasks and milestones. These are themselves structured objects, ensuring the data is clean and easy to work with.++```python+class GanttTaskDetails(StructuredContent):+ """Do not include timezone in the dates."""+ name: str+ start_date: Optional[datetime] = None+ end_date: Optional[datetime] = None+ # ...++class Milestone(StructuredContent):+ name: str+ date: Optional[datetime]+ # ...++class GanttChart(StructuredContent):+ tasks: Optional[List[GanttTaskDetails]]+ milestones: Optional[List[Milestone]]+```++## The Pipeline Definition: `gantt.toml`++The `extract_gantt_by_steps` pipeline is a sequence of smaller, focused pipes. This is a great example of building a complex workflow from simple, reusable components.++```toml+[pipe.extract_gantt_by_steps]+PipeSequence = "Extract all details from a gantt chart"+inputs = { gantt_chart_image = "GanttChartImage" }+output = "GanttChart"+steps = [+ # First, figure out the timescale of the chart+ { pipe = "extract_gantt_timescale", result = "gantt_timescale" },+ # Then, get the names of all the tasks+ { pipe = "extract_gantt_task_names", result = "gantt_task_names" },+ # Then, for each task, extract the details+ { pipe = "extract_details_of_task", batch_as = "gantt_task_name", result = "details_of_all_tasks" },+ # Finally, assemble everything into a single GanttChart object+ { pipe = "gather_in_a_gantt_chart", result = "gantt_chart" },+]++# This is the pipe that extracts the details for a single task+[pipe.extract_details_of_task]+PipeLLM = "Extract the precise dates of the task, start_date and end_date"+inputs = { gantt_chart_image = "GanttChartImage", gantt_timescale = "GanttTimescaleDescription", gantt_task_name = "GanttTaskName" }+output = "GanttTaskDetails" # The output is structured as a GanttTaskDetails object+structuring_method = "preliminary_text"+images = ["gantt_chart_image"]+llm = "llm_to_extract_diagram"+prompt_template = """+I am sharing an image of a Gantt chart.+Please analyse the image and for a given task name (and only this task), extract the information of the task, if relevant.++Be careful, the time unit is this:+@gantt_timescale++If the task is a milestone, then only output the start_date.++Here is the name of the task you have to extract the dates for:+@gantt_task_name+"""+```+This demonstrates the "divide and conquer" approach that Pipelex encourages. By breaking down a complex problem into smaller steps, each step can be handled by a specialized pipe, making the overall workflow more robust and easier to debug.++## Flowchart++```mermaid+---+config:+ layout: dagre+ theme: base+---+flowchart LR+ subgraph "extract_gantt_by_steps"+ direction LR+ ZJNLe["gantt_chart_image:<br>**Gantt image**"]+ Bh9Ab["gantt_timescale:<br>**Gantt timescale description**"]+ R2oLH["gantt_task_names:<br>**List of [Gantt task name]**"]+ end+ subgraph "extract_details_of_task"+ direction LR+ R2oLH-branch-0["**Gantt task name** #1"]+ QFEdmVvYVHdL6DsCRtwqHq-branch-0["**Gantt task details** #1"]+ R2oLH-branch-1["**Gantt task name** #2"]+ QFEdmVvYVHdL6DsCRtwqHq-branch-1["**Gantt task details** #2"]+ R2oLH-branch-2["**Gantt task name** #3"]+ QFEdmVvYVHdL6DsCRtwqHq-branch-2["**Gantt task details** #3"]+ R2oLH-branch-3["**Gantt task name** #4"]+ QFEdmVvYVHdL6DsCRtwqHq-branch-3["**Gantt task details** #4"]+ R2oLH-branch-4["**Gantt task name** #5"]+ QFEdmVvYVHdL6DsCRtwqHq-branch-4["**Gantt task details** #5"]+ R2oLH-branch-5["**Gantt task name** #6"]+ QFEdmVvYVHdL6DsCRtwqHq-branch-5["**Gantt task details** #6"]+ R2oLH-branch-6["**Gantt task name** #7"]+ QFEdmVvYVHdL6DsCRtwqHq-branch-6["**Gantt task details** #7"]+ R2oLH-branch-7["**Gantt task name** #8"]+ QFEdmVvYVHdL6DsCRtwqHq-branch-7["**Gantt task details** #8"]+ R2oLH-branch-8["**Gantt task name** #9"]+ QFEdmVvYVHdL6DsCRtwqHq-branch-8["**Gantt task details** #9"]+ R2oLH-branch-9["**Gantt task name** #10"]+ QFEdmVvYVHdL6DsCRtwqHq-branch-9["**Gantt task details** #10"]+ R2oLH-branch-10["**Gantt task name** #11"]+ QFEdmVvYVHdL6DsCRtwqHq-branch-10["**Gantt task details** #11"]+ R2oLH-branch-11["**Gantt task name** #12"]+ QFEdmVvYVHdL6DsCRtwqHq-branch-11["**Gantt task details** #12"]+ nCmpx["details_of_all_tasks:<br>**List of [Gantt task details]**"]+ NuMVc["gantt_chart:<br>**Gantt chart**"]+ end+class extract_gantt_by_steps sub_a;+class extract_details_of_task sub_b;++ classDef sub_a fill:#e6f5ff,color:#333,stroke:#333;++ classDef sub_b fill:#fff5f7,color:#333,stroke:#333;++ classDef sub_c fill:#f0fff0,color:#333,stroke:#333;+ ZJNLe -- "Extract gantt timescale" ----> Bh9Ab+ ZJNLe -- "Extract gantt task names" ----> R2oLH+ ZJNLe -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-0+ ZJNLe -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-1+ ZJNLe -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-2+ ZJNLe -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-3+ ZJNLe -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-4+ ZJNLe -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-5+ ZJNLe -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-6+ ZJNLe -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-7+ ZJNLe -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-8+ ZJNLe -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-9+ ZJNLe -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-10+ ZJNLe -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-11+ Bh9Ab -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-0+ Bh9Ab -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-1+ Bh9Ab -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-2+ Bh9Ab -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-3+ Bh9Ab -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-4+ Bh9Ab -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-5+ Bh9Ab -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-6+ Bh9Ab -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-7+ Bh9Ab -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-8+ Bh9Ab -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-9+ Bh9Ab -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-10+ Bh9Ab -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-11+ R2oLH -...- R2oLH-branch-0+ R2oLH -...- R2oLH-branch-1+ R2oLH -...- R2oLH-branch-2+ R2oLH -...- R2oLH-branch-3+ R2oLH -...- R2oLH-branch-4+ R2oLH -...- R2oLH-branch-5+ R2oLH -...- R2oLH-branch-6+ R2oLH -...- R2oLH-branch-7+ R2oLH -...- R2oLH-branch-8+ R2oLH -...- R2oLH-branch-9+ R2oLH -...- R2oLH-branch-10+ R2oLH -...- R2oLH-branch-11+ R2oLH-branch-0 -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-0+ QFEdmVvYVHdL6DsCRtwqHq-branch-0 -...- nCmpx+ R2oLH-branch-1 -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-1+ QFEdmVvYVHdL6DsCRtwqHq-branch-1 -...- nCmpx+ R2oLH-branch-2 -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-2+ QFEdmVvYVHdL6DsCRtwqHq-branch-2 -...- nCmpx+ R2oLH-branch-3 -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-3+ QFEdmVvYVHdL6DsCRtwqHq-branch-3 -...- nCmpx+ R2oLH-branch-4 -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-4+ QFEdmVvYVHdL6DsCRtwqHq-branch-4 -...- nCmpx+ R2oLH-branch-5 -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-5+ QFEdmVvYVHdL6DsCRtwqHq-branch-5 -...- nCmpx+ R2oLH-branch-6 -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-6+ QFEdmVvYVHdL6DsCRtwqHq-branch-6 -...- nCmpx+ R2oLH-branch-7 -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-7+ QFEdmVvYVHdL6DsCRtwqHq-branch-7 -...- nCmpx+ R2oLH-branch-8 -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-8+ QFEdmVvYVHdL6DsCRtwqHq-branch-8 -...- nCmpx+ R2oLH-branch-9 -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-9+ QFEdmVvYVHdL6DsCRtwqHq-branch-9 -...- nCmpx+ R2oLH-branch-10 -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-10+ QFEdmVvYVHdL6DsCRtwqHq-branch-10 -...- nCmpx+ R2oLH-branch-11 -- "Extract details of task" ----> QFEdmVvYVHdL6DsCRtwqHq-branch-11+ QFEdmVvYVHdL6DsCRtwqHq-branch-11 -...- nCmpx+ nCmpx -- "Gather in a gantt chart" ----> NuMVc+```\ No newline at end of filediff --git a/docs/pages/cookbook-examples/extract-generic.md b/docs/pages/cookbook-examples/extract-generic.mdnew file mode 100644index 000000000..0331c8a97--- /dev/null+++ b/docs/pages/cookbook-examples/extract-generic.md@@ -0,0 +1,55 @@+# Example: Generic Document Extraction++This example demonstrates a powerful and generic pipeline for extracting content from complex PDF documents. It can handle documents that contain both text and images, and it merges the extracted content into a single, coherent output.++## Get the code++[**➡️ View on GitHub: examples/extract_generic.py**](https://github.com/Pipelex/pipelex-cookbook/blob/main/examples/extract_generic.py)++## The Pipeline Explained++The `power_extractor` pipeline is at the heart of this example. After its execution, a custom function `merge_markdown_and_images` is used to combine the text (converted to Markdown) and the images from all pages.++```python+async def extract_generic(pdf_url: str) -> TextAndImagesContent:+ working_memory = WorkingMemoryFactory.make_from_pdf(+ pdf_url=pdf_url,+ concept_str="PDF",+ name="pdf",+ )+ pipe_output, _ = await execute_pipeline(+ pipe_code="power_extractor",+ working_memory=working_memory,+ )+ working_memory = pipe_output.working_memory+ markdown_and_images: TextAndImagesContent = merge_markdown_and_images(working_memory)+ return markdown_and_images+```++The `merge_markdown_and_images` function is a great example of how you can add your own Python code to a Pipelex workflow to perform custom processing.++```python+def merge_markdown_and_images(working_memory: WorkingMemory) -> TextAndImagesContent:+ # Pages extracted from the PDF by PipeOCR+ page_contents_list = working_memory.get_stuff_as_list(item_type=PageContent, name="page_contents")+ # Markdown text extracted from the Pages by PipeLLM+ page_markdown_list = working_memory.get_stuff_as_list(item_type=TextContent, name="markdowns")++ # ... (check for length equality)++ # Concatenate the markdown text+ concatenated_markdown_text: str = "\\n".join([page_markdown.text for page_markdown in page_markdown_list.items])++ # Aggregate the images from the page contents+ image_contents: List[ImageContent] = []+ for page_content in page_contents_list.items:+ if page_content.text_and_images.images:+ image_contents.extend(page_content.text_and_images.images)++ return TextAndImagesContent(+ text=TextContent(text=concatenated_markdown_text),+ images=image_contents,+ )+```++This example shows the flexibility of Pipelex in handling complex, multi-modal documents and allowing for custom logic.\ No newline at end of filediff --git a/docs/pages/cookbook-examples/extract-proof-of-purchase.md b/docs/pages/cookbook-examples/extract-proof-of-purchase.mdnew file mode 100644index 000000000..0bea51594--- /dev/null+++ b/docs/pages/cookbook-examples/extract-proof-of-purchase.md@@ -0,0 +1,81 @@+# Example: Proof of Purchase Extraction++This example demonstrates a pipeline designed to extract structured data from a proof of purchase, such as a receipt or an invoice.++## Get the code++[**➡️ View on GitHub: examples/extract_proof_of_purchase.py**](https://github.com/Pipelex/pipelex-cookbook/blob/main/examples/extract_proof_of_purchase.py)++## The Pipeline Explained++The pipeline `power_extractor_proof_of_purchase` is specifically designed to handle receipts and invoices. It extracts key information and returns a structured `ProofOfPurchase` object.++```python+async def extract_proof_of_purchase(pdf_url: str) -> ProofOfPurchase:+ working_memory = WorkingMemoryFactory.make_from_pdf(+ pdf_url=pdf_url,+ concept_str="PDF",+ name="pdf",+ )+ pipe_output, _ = await execute_pipeline(+ pipe_code="power_extractor_proof_of_purchase",+ working_memory=working_memory,+ )+ working_memory = pipe_output.working_memory+ proof_of_purchase: ProofOfPurchase = working_memory.get_list_stuff_first_item_as(name="proof_of_purchase", item_type=ProofOfPurchase)+ return proof_of_purchase+```++This is a great starting point for building more complex expense processing or accounting automation pipelines.++## The Data Structure: `ProofOfPurchase` Model++The pipeline is designed to extract a `ProofOfPurchase` object, which is a structured model that includes a list of `Products`.++```python+class Products(StructuredContent):+ name: Optional[str] = None+ quantity: Optional[int] = None+ unit_price: Optional[float] = None+ total_price: Optional[float] = None+++class ProofOfPurchase(StructuredContent):+ date_of_purchase: Optional[datetime] = None+ amount_paid: Optional[float] = None+ currency: Optional[str] = None+ payment_method: Optional[str] = None+ purchase_number: Optional[str] = None+ products: Optional[List[Products]] = None+```+This demonstrates how you can create nested data structures to accurately model your data.++## The Pipeline Definition: `extract_proof_of_purchase.toml`++The pipeline uses a powerful `PipeLLM` to extract the structured data from the document. The prompt is carefully engineered to guide the LLM.++```toml+[pipe.write_markdown_from_page_content_proof_of_purchase]+PipeLLM = "Write markdown from page content"+inputs = { page_content = "Page" }+output = "ProofOfPurchase" # The LLM is forced to output a ProofOfPurchase object+images = ["page_content.page_view"] # The LLM receives the image of the page+llm = "llm_for_img_to_text"+structuring_method = "preliminary_text"+system_prompt = """You are a multimodal LLM, expert at converting images into perfect markdown."""+prompt_template = """+You are given an image of a proof of purchase.+Your role is to convert the image into perfect markdown.++To help you do so, you are given the text extracted from the page by an OCR model.+@page_content.text_and_images.text.text++- Ensure you collect every title, number, and currency from the proof of purchase.+- Pay attention to the text alignment, it might have been misaligned by the OCR.+- The OCR extraction may be highly incomplete. It is your job to complete the text and add the missing information using the image.+- Output only the markdown, nothing else. No need for "```markdown" or "```".+- You can use HTML if it helps you.+- You can use tables if it is relevant.+"""+```+The combination of a detailed prompt, the OCR text, and the document image allows the LLM to accurately extract the required information and structure it as a `ProofOfPurchase` object.diff --git a/docs/pages/cookbook-examples/extract-table.md b/docs/pages/cookbook-examples/extract-table.mdnew file mode 100644index 000000000..a74bf4e68--- /dev/null+++ b/docs/pages/cookbook-examples/extract-table.md@@ -0,0 +1,89 @@+# Example: Table Extraction from Image++This example shows how to extract a table from an image and convert it into a structured HTML format. This is a common requirement when dealing with scanned documents or reports where data is presented in tabular form.++## Get the code++[**➡️ View on GitHub: examples/extract_table.py**](https://github.com/Pipelex/pipelex-cookbook/blob/main/examples/extract_table.py)++## The Pipeline Explained++The pipeline `extract_html_table_and_review` takes an image of a table, processes it, and returns an `HtmlTable` object.++```python+async def extract_table(table_screenshot: str) -> HtmlTable:+ working_memory = WorkingMemoryFactory.make_from_image(+ image_url=table_screenshot,+ concept_str="tables.TableScreenshot",+ name="table_screenshot",+ )+ pipe_output, _ = await execute_pipeline(+ pipe_code="extract_html_table_and_review",+ working_memory=working_memory,+ )+ html_table = pipe_output.main_stuff_as(content_type=HtmlTable)+ return html_table+```++This is another example of Pipelex's multi-modal capabilities, turning visual information into structured data.++## The Data Structure: `HtmlTable` Model++The pipeline outputs an `HtmlTable` object. This is a great example of a "smart" data model. It uses a `pydantic.model_validator` to parse the generated HTML with BeautifulSoup and validate its structure, ensuring the LLM has produced valid and well-formed HTML.++```python+class HtmlTable(StructuredContent):+ title: str+ inner_html_table: str+ allowed_tags: ClassVar[set[str]] = { "br", "table", "thead", "tbody", "tr", "th", "td" }++ @model_validator(mode="after")+ def validate_html_table(self) -> Self:+ soup = BeautifulSoup(self.inner_html_table, "html.parser")+ # Check if there's exactly one table element+ tables = soup.find_all("table")+ if len(tables) != 1:+ raise ValueError(f"HTML must contain exactly one table element...")++ # Validate that only allowed table-related tags are present+ all_tags = {tag.name for tag in soup.find_all()}+ invalid_tags = all_tags - self.allowed_tags+ if invalid_tags:+ raise ValueError(f"Invalid HTML tags found: {invalid_tags}")++ # ... more validation+ return self+```++## The Pipeline Definition: `table.toml`++The pipeline uses a two-step "extract and review" pattern. The first pipe does the initial extraction, and the second pipe reviews the generated HTML against the original image to correct any errors. This is a powerful pattern for increasing the reliability of LLM outputs.++```toml+[pipe.extract_html_table_and_review]+PipeSequence = "Get an HTML table and review it"+inputs = { table_screenshot = "TableScreenshot" }+output = "HtmlTable"+steps = [+ # First, do an initial extraction of the table from the image+ { pipe = "extract_html_table_from_image", result = "html_table" },+ # Then, ask an LLM to review the extracted table against the image and correct it+ { pipe = "review_html_table", result = "reviewed_html_table" },+]++[pipe.review_html_table]+PipeLLM = "Review an HTML table"+inputs = { table_screenshot = "TableScreenshot", html_table = "HtmlTable" }+output = "HtmlTable"+images = ["table_screenshot"]+prompt_template = """+Your role is to correct an html_table to make sure that it matches the one in the provided image.++@html_table++Pay attention to the text and formatting (color, borders, ...).+Rewrite the entire html table with your potential corrections.+Make sure you do not forget any text.+"""+```+This self-correction pattern is a key technique for building robust and reliable AI workflows with Pipelex.\ No newline at end of filediff --git a/docs/pages/cookbook-examples/hello-world.md b/docs/pages/cookbook-examples/hello-world.mdnew file mode 100644index 000000000..c1f7dcddc--- /dev/null+++ b/docs/pages/cookbook-examples/hello-world.md@@ -0,0 +1,28 @@+# Example: Hello World++This is the "Hello World" of Pipelex. A simple pipeline that demonstrates the basic concepts of Pipelex.++It's the perfect starting point to verify your installation and get a first taste of how Pipelex works.++## Get the code++You can find the complete code for this example in the Pipelex Cookbook repository.++[**➡️ View on GitHub: quick_start/hello_world.py**](https://github.com/Pipelex/pipelex-cookbook/blob/main/quick_start/hello_world.py)++## How to run++1. Clone the cookbook repository:+ ```bash+ git clone https://github.com/Pipelex/pipelex-cookbook.git+ cd pipelex-cookbook+ ```+2. Install dependencies:+ ```bash+ make install+ ```+3. Set up your environment variables by copying `.env.example` to `.env` and adding your API keys.+4. Run the example:+ ```bash+ python quick_start/hello_world.py+ ```\ No newline at end of filediff --git a/docs/pages/cookbook-examples/index.md b/docs/pages/cookbook-examples/index.mdnew file mode 100644index 000000000..3c9c82ae0--- /dev/null+++ b/docs/pages/cookbook-examples/index.md@@ -0,0 +1,49 @@+# Pipelex Cookbook Examples++Welcome to the Pipelex Cookbook!++[](https://github.com/Pipelex/pipelex-cookbook/)++This is your go-to resource for practical examples and ready-to-use recipes to build powerful and reliable AI workflows with Pipelex. Whether you're a beginner looking to get started or an experienced user searching for advanced patterns, you'll find something useful here.++## Philosophy++Our examples are designed to be:++- **Practical**: Solving real-world problems.+- **Readable**: Easy to understand and adapt.+- **A Starting Point**: A foundation for your own custom pipelines.++We strongly encourage you to clone our [Pipelex Cookbook repository](https://github.com/Pipelex/pipelex-cookbook/), play with the examples, and tailor them to your specific needs.++## Contribute to the Community!++Have you built a cool pipeline with Pipelex? Share it with the community! We welcome contributions. Check out our [contribution guidelines](../../contributing.md) and help us grow this collection of examples.++## Available Examples++Here are some of the examples you can find in the cookbook, organized by category:++### Getting Started++* [**Hello World**](./hello-world.md): A simple start to your Pipelex journey.++### Document Processing++* [**Simple OCR**](./simple-ocr.md): A basic OCR pipeline to extract text from a PDF.+* [**Generic Document Extraction**](./extract-generic.md): A powerful pipeline to extract text and images from complex documents.+* [**Invoice Extractor**](./invoice-extractor.md): A complete workflow for processing invoices, including reporting.+* [**Proof of Purchase Extraction**](./extract-proof-of-purchase.md): A targeted pipeline for extracting data from receipts.++### Graphical Extraction++* [**Gantt Chart Extraction**](./extract-gantt.md): Extract tasks, dates, and dependencies from a Gantt chart image.+* [**Table Extraction from Image**](./extract-table.md): Convert a table from an image into structured HTML.+* [**DPE Extraction**](./extract-dpe.md): A specialized pipeline for extracting information from French energy performance certificates.++### Text Generation++* [**Screenplay Generator (wip)**](./write-screenplay.md): A creative pipeline that generates a screenplay from a simple pitch.+* [**Tweet Optimizer (wip)**](./write-tweet.md): A style-transfer pipeline that optimizes a draft tweet based on a given writing style.++Dive in and happy piping! 🚀diff --git a/docs/pages/cookbook-examples/invoice-extractor.md b/docs/pages/cookbook-examples/invoice-extractor.mdnew file mode 100644index 000000000..c4e1267b5--- /dev/null+++ b/docs/pages/cookbook-examples/invoice-extractor.md@@ -0,0 +1,124 @@+# Example: Invoice Extractor++This example provides a comprehensive pipeline for processing invoices. It takes a PDF invoice, extracts key information, and returns a structured `Invoice` object. It also demonstrates how to generate reports and track pipeline execution.++## Get the code++[**➡️ View on GitHub: examples/invoice_extractor.py**](https://github.com/Pipelex/pipelex-cookbook/blob/main/examples/invoice_extractor.py)++## The Pipeline Explained++The `process_invoice` pipeline is a complete workflow for invoice processing.++```python+async def process_expense_report() -> ListContent[Invoice]:+ invoice_pdf_path = "assets/invoice_extractor/invoice_1.pdf"++ # Create Stuff objects+ working_memory = WorkingMemoryFactory.make_from_pdf(+ pdf_url=invoice_pdf_path,+ name="invoice_pdf",+ )+ pipe_output, _ = await execute_pipeline(+ pipe_code="process_invoice",+ working_memory=working_memory,+ )++ return pipe_output.main_stuff_as_list(item_type=Invoice)+```++This example also showcases some of the powerful observability features of Pipelex. After the pipeline runs, it generates a cost report and a flowchart of the execution.++```python+# Print the cost reporting+get_report_delegate().generate_report()++# Print the flowchart url of the pipeline.+get_pipeline_tracker().output_flowchart()+```+This is invaluable for understanding the cost and the execution flow of your pipelines.++## The Data Structure: `Invoice` Model++The pipeline's output is a structured `Invoice` object. This is defined using Pydantic's `BaseModel`, which allows for clear, typed, and validated data.++```python+class Invoice(StructuredContent):+ """Invoice information extracted from text, supporting both formal bills and receipts"""++ invoice_id: Optional[str] = Field(None, description="Unique identifier for the invoice")+ invoice_number: Optional[str] = Field(None, description="Invoice number as shown on the document")+ date: Optional[datetime] = Field(None, description="Date when the invoice was issued")++ amount_incl_tax: Optional[float] = Field(None, description="Total amount including taxes")++ vendor: Optional[str] = Field(None, description="Name of the vendor/seller")++ # ... other fields+```++## The Pipeline Definition: `invoice.toml`++The entire workflow is defined in a TOML file. This declarative approach makes the pipeline easy to understand and modify. Here's a snippet from `invoice.toml`:++```toml+# The main pipeline, a sequence of steps+[pipe.process_invoice]+PipeSequence = "Process relevant information from an invoice"+inputs = { invoice_pdf = "PDF" }+output = "Invoice"+steps = [+ # First, run OCR on the PDF+ { pipe = "extract_text_from_image", result = "invoice_pages" },+ # Then, run the invoice extraction on each page+ { pipe = "extract_invoice", batch_over = "invoice_pages", batch_as = "invoice_page", result = "invoice" },+]++# A sub-pipeline that uses an LLM to extract the data+[pipe.extract_invoice_data]+PipeLLM = "Extract invoice information from an invoice text transcript"+inputs = { invoice_page = "Page", invoice_details = "InvoiceDetails" }+output = "Invoice"+# The output is constrained to the "Invoice" model+llm = "llm_to_extract_invoice"+images = ["invoice_page.page_view"]+prompt_template = """+Extract invoice information from this invoice:++The category of this invoice is: $invoice_details.category.++@invoice_page.text_and_images.text.text+"""+```+This shows how a complex workflow, including OCR and LLM calls, can be defined in a simple, readable format. The `llm = "llm_to_extract_invoice"` line is particularly powerful, as it tells the LLM to structure its output according to the `Invoice` model.++## The Pipeline Flowchart++```mermaid+---+config:+ layout: dagre+ theme: base+---+flowchart LR+ subgraph "extract_invoice"+ direction LR+ ZynbH-branch-0["invoice_page:<br>**Page**"]+ RRYZF["invoice_details:<br>**Invoice details**"]+ RzjEzwGpkk5dXnrK3HXQJx-branch-0["invoice:<br>**Invoice**"]+ ZynbH["invoice_pages:<br>**List of [Page]**"]+ 5SXqJ["invoice:<br>**List of [Invoice]**"]+ end+class extract_invoice sub_a;++ classDef sub_a fill:#e6f5ff,color:#333,stroke:#333;++ classDef sub_b fill:#fff5f7,color:#333,stroke:#333;++ classDef sub_c fill:#f0fff0,color:#333,stroke:#333;+ ZynbH-branch-0 -- "Analyze invoice" ----> RRYZF+ ZynbH-branch-0 -- "Extract invoice data" ----> RzjEzwGpkk5dXnrK3HXQJx-branch-0+ RRYZF -- "Extract invoice data" ----> RzjEzwGpkk5dXnrK3HXQJx-branch-0+ RzjEzwGpkk5dXnrK3HXQJx-branch-0 -...- 5SXqJ+ ZynbH -...- ZynbH-branch-0+```\ No newline at end of filediff --git a/docs/pages/cookbook-examples/simple-ocr.md b/docs/pages/cookbook-examples/simple-ocr.mdnew file mode 100644index 000000000..1cbc24eca--- /dev/null+++ b/docs/pages/cookbook-examples/simple-ocr.md@@ -0,0 +1,30 @@+# Example: Simple OCR++This example demonstrates a basic OCR (Optical Character Recognition) pipeline. It takes a PDF file as input, extracts the text from each page, and saves the content.++This is a fundamental building block for many document processing workflows.++## Get the code++[**➡️ View on GitHub: examples/simple_ocr.py**](https://github.com/Pipelex/pipelex-cookbook/blob/main/examples/simple_ocr.py)++## The Pipeline Explained++The core of this example is a simple function that creates a "working memory" from a PDF and then executes a pre-defined pipeline called `extract_page_contents_from_pdf`.++```python+async def simple_ocr(pdf_url: str):+ working_memory = WorkingMemoryFactory.make_from_pdf(+ pdf_url=pdf_url,+ concept_str="PDF",+ name="pdf",+ )+ pipe_output, _ = await execute_pipeline(+ pipe_code="extract_page_contents_from_pdf",+ working_memory=working_memory,+ )+ page_content_list: ListContent[PageContent] = pipe_output.main_stuff_as_list(item_type=PageContent)+ return page_content_list+```++This showcases how easy it is to kick off a complex process with just a few lines of code.\ No newline at end of filediff --git a/docs/pages/cookbook-examples/write-screenplay.md b/docs/pages/cookbook-examples/write-screenplay.mdnew file mode 100644index 000000000..78de6497c--- /dev/null+++ b/docs/pages/cookbook-examples/write-screenplay.md@@ -0,0 +1,112 @@+# Example: Screenplay Generator++This example demonstrates how to use Pipelex for creative text generation. It takes a simple pitch and generates a full screenplay.++## Get the code++[**➡️ View on GitHub: examples/wip/write_screen_play.py**](https://github.com/Pipelex/pipelex-cookbook/blob/main/examples/wip/write_screen_play.py)++## The Pipeline Explained++The `generate_screenplay` function takes a pitch as a string, creates a `Stuff` object with the `screenplay.Pitch` concept, and then runs the `generate_screenplay` pipeline.++```python+async def generate_screenplay(pitch: str):+ """Generate a screenplay from a pitch using the pipeline."""++ # Create Stuff object for the pitch+ pitch_stuff = StuffFactory.make_from_str(+ str_value=pitch,+ concept_str="screenplay.Pitch",+ name="pitch",+ )++ # Create Working Memory+ working_memory = WorkingMemoryFactory.make_from_single_stuff(pitch_stuff)++ # Run the pipe+ pipe_output, _ = await execute_pipeline(+ pipe_code="generate_screenplay",+ working_memory=working_memory,+ )+ pretty_print(pipe_output, title="Pipe Output")+```++This example shows how a simple text input can be used to kick off a complex, creative workflow.++## The Data Structures: `Screenplay` Models++This pipeline uses a rich set of Pydantic models to represent the different components of a screenplay, from the initial `Pitch` to the final `FormattedScreenplay`. This ensures that each step of the pipeline has clear, structured inputs and outputs.++```python+class DetailedPitch(StructuredContent):+ """A detailed pitch with character ideas and synopsis."""+ original_pitch: str+ expanded_pitch: str+ character_ideas: List[str]+ synopsis: str++class Character(StructuredContent):+ """A character in the screenplay."""+ name: str+ age: int+ description: str+ # ...++class Scene(StructuredContent):+ """A scene in the screenplay."""+ index: int+ title: str+ location: str+ script: str+ # ...++class Chapter(StructuredContent):+ """A chapter in the screenplay."""+ title: str+ description: str+ scenes: List[Scene]++class Screenplay(StructuredContent):+ """A complete screenplay with characters and scenes."""+ title: str+ pitch: DetailedPitch+ characters: "CharacterList"+ chapters: "ChapterList"+```++## The Pipeline Definition: `screenplay_writer.toml`++The `generate_screenplay` pipeline is a master `PipeSequence` that orchestrates a series of smaller, specialized pipes. This is a perfect example of how to build a complex, creative workflow by breaking it down into manageable steps.++```toml+[pipe.generate_screenplay]+PipeSequence = "Generate a complete screenplay from a pitch"+inputs = { pitch = "Pitch" }+output = "FormattedScreenplay"+steps = [+ # 1. Analyze the initial pitch to expand it.+ { pipe = "analyze_pitch", result = "detailed_pitch" },+ # 2. Create detailed character profiles.+ { pipe = "create_characters", result = "characters" },+ # 3. Break the story down into chapters.+ { pipe = "create_chapters", result = "chapters" },+ # 4. For each chapter, create the scenes. This is a nested sequence!+ { pipe = "create_scenes_sequence", batch_over = "chapters", batch_as = "chapter", result = "chapters_with_scenes" },+ # 5. Format the entire screenplay.+ { pipe = "create_formatted_screenplay", result = "formatted_screenplay" }+]++# This is the nested sequence that creates all the scenes for a single chapter.+[pipe.create_scenes_sequence]+PipeSequence = "Create all scenes for a chapter sequentially"+inputs = { chapter = "Chapter", characters = "CharacterList", detailed_pitch = "DetailedPitch" }+output = "Chapter"+steps = [+ # First, create the initial scene outlines for the chapter.+ { pipe = "create_initial_scenes", result = "initial_scenes" },+ # Then, for each scene outline, develop the full scene content.+ { pipe = "create_scene_sequence", batch_over = "initial_scenes", batch_as = "scene", result = "developed_scenes" }+]+```+This modular, step-by-step process, guided by the structured data models, allows Pipelex to generate a complete, well-structured screenplay from a single, simple pitch. It's a powerful demonstration of building complex, creative AI agents.\ No newline at end of filediff --git a/docs/pages/cookbook-examples/write-tweet.md b/docs/pages/cookbook-examples/write-tweet.mdnew file mode 100644index 000000000..265ddf644--- /dev/null+++ b/docs/pages/cookbook-examples/write-tweet.md@@ -0,0 +1,94 @@+# Example: Tweet Optimizer++This example demonstrates how to create a pipeline that takes a draft of a tweet and a desired writing style, and then generates an optimized tweet. This is a practical example of "style transfer" for text.++## Get the code++[**➡️ View on GitHub: examples/wip/write_tweet.py**](https://github.com/Pipelex/pipelex-cookbook/blob/main/examples/wip/write_tweet.py)++## The Pipeline Explained++The `optimize_tweet` function is the core of this example. It takes two strings, `draft_tweet_str` and `writing_style_str`, creates two `Stuff` objects with the concepts `tech_tweet.DraftTweet` and `tech_tweet.WritingStyle`, and then runs the `optimize_tweet_sequence` pipeline.++```python+async def optimize_tweet(draft_tweet_str: str, writing_style_str: str) -> OptimizedTweet:+ # Create the draft tweet stuff+ draft_tweet = StuffFactory.make_stuff(+ concept_str="tech_tweet.DraftTweet",+ content=TextContent(text=draft_tweet_str),+ name="draft_tweet",+ )+ writing_style = StuffFactory.make_stuff(+ concept_str="tech_tweet.WritingStyle",+ content=TextContent(text=writing_style_str),+ name="writing_style",+ )++ # Create working memory+ working_memory = WorkingMemoryFactory.make_from_multiple_stuffs(+ [+ draft_tweet,+ writing_style,+ ]+ )++ # Run the sequence pipe+ pipe_output, _ = await execute_pipeline(+ pipe_code="optimize_tweet_sequence",+ working_memory=working_memory,+ )++ # Get the optimized tweet+ optimized_tweet = pipe_output.main_stuff_as(content_type=OptimizedTweet)+ return optimized_tweet+```++This example shows how to use multiple inputs to guide the generation process and produce text that adheres to a specific style.++## The Data Structure: `OptimizedTweet` Model++The data model for this pipeline is very simple, as the final output is just a piece of text. However, the pipeline uses several concepts internally to manage the workflow, such as `DraftTweet`, `TweetAnalysis`, and `WritingStyle`.++```python+class OptimizedTweet(TextContent):+ """A tweet optimized for Twitter/X engagement following best practices."""+ pass+```++## The Pipeline Definition: `tech_tweet.toml`++This pipeline uses a two-step "analyze and optimize" sequence. The first pipe analyzes the draft tweet for common pitfalls, and the second pipe rewrites the tweet based on the analysis and a provided writing style. This is a powerful pattern for refining generated content.++```toml+[pipe.optimize_tweet_sequence]+PipeSequence = "Analyze and optimize a tech tweet in sequence"+inputs = { draft_tweet = "DraftTweet", writing_style = "WritingStyle" }+output = "OptimizedTweet"+steps = [+ # First, analyze the draft tweet for issues like "fluffiness" and "vagueness".+ { pipe = "analyze_tweet", result = "tweet_analysis" },+ # Then, optimize the tweet based on the analysis and the desired writing style.+ { pipe = "optimize_tweet", result = "optimized_tweet" },+]++# This is the pipe that analyzes the draft tweet.+[pipe.analyze_tweet]+PipeLLM = "Analyze the draft tweet and identify areas for improvement"+inputs = { draft_tweet = "DraftTweet" }+output = "TweetAnalysis"+system_prompt = """+You are an expert in social media optimization, particularly for tech content on Twitter/X.+Your role is to analyze tech tweets and check if they display typical startup communication pitfalls.+"""+prompt_template = """+Evaluate the tweet for these key issues:++**Fluffiness** - Overuse of buzzwords without concrete meaning...+**Cringiness** - Content that induces secondhand embarrassment...+**Humblebragginess** - Disguising boasts as casual updates...+**Vagueness** - Failing to clearly communicate what the product/service actually does...++@draft_tweet+"""+```+This "analyze and refine" pattern is a great way to build more reliable and sophisticated text generation workflows. The first step provides a structured critique, and the second step uses that critique to improve the final output.\ No newline at end of filediff --git a/docs/pages/installation/index.md b/docs/pages/installation/index.mdnew file mode 100644index 000000000..618dffb6a--- /dev/null+++ b/docs/pages/installation/index.md@@ -0,0 +1,64 @@+# Installation++## Prerequisites++Pipelex requires `python` version `3.10` or above, and access to an LLM, via an API key or a custom endpoint.++## Getting Started++Along with [our Documentation](../quick-start/index.md), we recommend you review it before any further usage: [Cookbook](https://github.com/Pipelex/pipelex-cookbook).++- **Create a virtual environment** (recommended)++```bash+python3 -m venv .venv && source .venv/bin/activate+```++ - **Install Pipelex**++Pipelex can be installed from PyPI. We encourage the use of [uv](https://github.com/astral-sh/uv) for faster installs and dependency management:++```bash+uv pip install pipelex+```++Otherwise use pip:+```bash+pip install pipelex+```++- **Make sure you have a .env** file at the root of your project that contains the following fields++```bash+OPENAI_API_KEY=sk_...+```++All the secret keys used by `pipelex` are specified in the `.env.example` file. However, by default, only the `OPENAI_API_KEY` is required.+++- **Make sure you run the init commands:**++In order to set the pipelex configuration files, you need to run 2 commands using the CLI (we recommend to run it at the root of your project):++- `pipelex init-libraries`: This will create a `pipelex_libraries` folder, with the base llm configuration and the base pipelines.+This is the directory where you should add your pipelines.++The structure is like this:++```bash+├── pipelex_libraries+│ ├── __init__.py+│ ├── pipelines/ # The pipelines and the structured output are stored here+│ │ ├── __init__.py+│ │ └── base_library/ # The base library with basic pipelines+│ ├── templates/ # Those are template prompt libraries+│ ├── llm_deck/ # A llm deck is a simple way to name a llm and its configuration.+│ └── llm_integrations/ # This directory regroups the configuration of the different models+```++Learn more about pipelex_libraries in our [Libraries documentation](../build-reliable-ai-workflows-with-pipelex/kick-off-a-knowledge-pipeline-project.md)++- `pipelex init-config`: This cli command will create a `pipelex.toml` file at the root of the project, with basic configuration. This configuration file gathers all configuration for feature flags, logging, cost reporting, and so on... Learn more in our [Configuration documentation](../configuration/index.md)+++💡 _Any troubles? Have a look at our [Cookbook](https://github.com/Pipelex/pipelex-cookbook)! and come ask for help on our [Discord](https://discord.gg/SReshKQjWt)_diff --git a/docs/pages/pipelex-paradigm-for-repeatable-ai-workflows/index.md b/docs/pages/pipelex-paradigm-for-repeatable-ai-workflows/index.mdnew file mode 100644index 000000000..aa97bf2a7--- /dev/null+++ b/docs/pages/pipelex-paradigm-for-repeatable-ai-workflows/index.md@@ -0,0 +1,44 @@+# The Pipelex Paradigm++Pipelex is an **open-source Python framework** for defining and running **repeatable AI workflows**.++Here's what we've learned: LLMs are powerful, but asking them to do everything in one prompt is like asking a brilliant colleague to solve ten problems while juggling. The more complexity you pack into a single prompt, the more reliability drops. You've seen it: the perfect prompt that works 90% of the time until it doesn't.++The solution is straightforward: break complex tasks into focused steps. But without proper tooling, you end up with spaghetti code and prompts scattered across your codebase.++Pipelex introduces **knowledge pipelines**: a way to capture these workflow steps as **composable pipes**. Each pipe follows one rule: **knowledge in, knowledge out**. Unlike rigid templates, each pipe uses AI's full intelligence to handle variation while guaranteeing consistent output structure. You get **deterministic structure with adaptive intelligence**, the reliability of software with the flexibility of AI.++## Working with Knowledge and Using Concepts to Make Sense++Knowledge refers to information you input from various data sources such as documents, PDFs, images, or information output by our pipes. There are different kinds of knowledge.++In traditional programming, we work with data types: strings, integers, booleans (true/false), etc. But **knowledge work operates at a higher level.** Take for instance a *"non-compete clause from a contract"* and a *"description of a flower"*: they're both text, both are stored as strings, but they represent fundamentally different concepts. You can ask AI to extract the duration in months from a non-compete clause. You can ask AI to render a flower description as a Monet-style painting. But if you try the reverse, it doesn't make sense.++This is why Pipelex introduces **Concepts: typing with meaning attached.**++**Thanks to Pipelex's conceptual level of abstraction, knowledge pipelines can guarantee they make sense.**++## How *Making Sense* Translates to *Reliability*++Concretely, in Pipelex, a piece of knowledge is an object in the working memory. It could be anything, so we call that "stuff", and our Python class for it is `Stuff`.++Stuff can be pretty basic, like plain text or an image, but it can also be structured with attributes, comprise lists, include other nested stuff, you get it... it's content is actually a `Pydantic BaseModel`. Also, each Stuff knows what concept it belongs to, e.g., `Text` or `NonCompeteClause` or `FlowerDescription`.++And with that we are fully equipped for Lego-style plug-n-play:++- Each pipe declares what inputs it uses, indicating the expected concept(s)+- Each pipe also declares what it outputs++So when you connect two pipes, Pipelex systematically checks that they are compatible. Actually it's even more powerful than that: imagine you have a sequence of 5 pipes, maybe pipe #4 takes two inputs, one given from the output of previous pipe #3 and the other one from the output of pipe #1. Pipelex checks that running the previous steps will have generated the required stuff and with the required concepts.++What if you have a pipe that requires a `Text` input, to translate it to Spanish for instance, then it should also accept a `FlowerDescription`, right? Yes, the greater includes the lesser. This is expressed in Pipelex by indicating that concept `FlowerDescription` **refines** concept `Text`.++One problem that developers face when working with LLMs is that LLMs always try to answer your queries, and it pushes them to hallucinate. For instance, if you're trying to extract structured purchase details from an invoice, and there's a bug in your system so instead of receiving an invoice as input, the LLM received a picture of a flower, or nothing at all, there's a good chance it will generate a 100% hallucinated mock invoice. Pipelex prevents this kind of bugs, first by making sure the pipeline makes sense a priori, but it can also check any input for compatibility with the expected concept.++**By letting you clearly assign concepts to the input and output stuff of your pipes, Pipelex makes sure that your pipeline makes sense, that it works in theory and detects any failure of meaning at the earliest stage possible in practice.**++## Who Defines the Concepts?++***You* define *your* concepts!**++Apart from a few very basic concepts (`Text`, `Image`, etc.), you can and you should define the concepts you work with. And that's a big win: the concepts you define make up the glossary of the domain you are working on. That's where you remove any ambiguity, because the same word can mean different things, and even the same concept can imply different things in various settings.diff --git a/doc/pages/Quick-start/character_sheet.png b/docs/pages/quick-start/character_sheet.pngsimilarity index 100%rename from doc/pages/Quick-start/character_sheet.pngrename to docs/pages/quick-start/character_sheet.pngdiff --git a/doc/pages/Quick-start/extracted_character_metadata.png b/docs/pages/quick-start/extracted_character_metadata.pngsimilarity index 100%rename from doc/pages/Quick-start/extracted_character_metadata.pngrename to docs/pages/quick-start/extracted_character_metadata.pngdiff --git a/doc/pages/Quick-start/Quick-start.md b/docs/pages/quick-start/index.mdsimilarity index 56%rename from doc/pages/Quick-start/Quick-start.mdrename to docs/pages/quick-start/index.mdindex 9436bb063..2bd481875 100644--- a/doc/pages/Quick-start/Quick-start.md+++ b/docs/pages/quick-start/index.md@@ -1,28 +1,25 @@# Quick-start-In this section, we introduce the basics of Pipelex for the simplest use-cases: LLM calling and structured outputs.+This guide shows the basics of Pipelex for the simplest use-cases: LLM calling and structured outputs.-You can **run the following examples** directly from the [Getting-started repository](https://github.com/Pipelex/ev-getting-started).-No need to create files in the getting-started repo, they are already there.+You can **find more powerful examples** in the [Cookbook Examples](../cookbook-examples/index.md) section of the docs or dive directly into the [Cookbook repository](https://github.com/Pipelex/pipelex-cookbook). [](https://github.com/Pipelex/pipelex-cookbook/)++---## Your first LLM call with Pipelex-Let's start by running your very first LLM call using the Pipelex framework.-For illustration purposes, let's build **a character generator.**+Let's start by running your very first LLM call using Pipelex.+For illustration purposes, let's build **a character generator**. Each example relies on asynchronous execution and typed models for reliable prompts.-### **🖊️ Write your first Pipelex script:**+### Write your first pipeline-You have to create a `TOML` library file that will store your Pipelex script.-If you are using Pipelex elsewhere than in the Getting Started repository, make sure you add this file to the `pipelex_libraries` folder.+You have to create a `.toml` library file in the `pipelex_libraries/pipelines` directory that will store your pipe definition.+Run `pipelex init-libraries` to create this directory if it doesn't exist. For now, keep all your pipeline definitions inside that folder only.+`character.toml````toml-# character.toml--# Library definitiondomain = "characters"-definition = "Tools for creating characters"-# Pipe definition[pipe][pipe.create_character]PipeLLM = "Creates a character."@@ -31,32 +28,32 @@ prompt_template = """You are a book writer. Your task is to create a character.Think of it and then output the character description."""```-### **🏃 Run your first Pipelex script:**+### Run your first Pipelex script-You have to create a `python` file to run your script. You can save it anywhere in your repository.+You have to create a `.py` python file to run your script. You can save it anywhere in your repository.+`character.py````python-# character.py-import asynciofrom pipelex.pipeline.execute import execute_pipeline+from pipelex.pipelex import Pipelex-async def create_character() -> str:+async def create_character():# Run the script with execute_pipe- pipe_output, pipeline_run_id = await execute_pipeline(+ pipe_output, _ = await execute_pipeline(pipe_code="create_character",)# Print the output- print(pipe_output.main_stuff_as_text)+ print(pipe_output.main_stuff_as_str)-# Setup required to initialize the Pipelex framework and load the pipeline libraries+# Initialize pipelex to load your pipeline librariesPipelex.make()# Run using asyncio because our APIs are all asyncasyncio.run(create_character())```-### **🎉 Get your first Pipelex result!**+### Get your first Pipelex result```bashpython character.py@@ -64,48 +61,49 @@ python character.py-## How to use a specific LLM+## How to use a specific LLM or LLM provider-### **🖊️ Indicate your LLM selection explicitly using the `llm` attribute:**+### Indicate your LLM selection explicitly using the `llm` attribute```toml[pipe.create_character]PipeLLM = "Create a character."output = "Text"+llm = { llm_handle = "gpt-4o-mini", temperature = 0.9, max_tokens = "auto" }prompt_template = """You are a book writer. Your task is to create a character.Think of it and then output the character description."""-llm = { llm_handle = "gpt-4o-mini", temperature = 0.9, max_tokens = "auto" }```-### **🖊️ Or use an LLM preset:**+### Or use an LLM preset from the LLM deck```toml[pipe.create_character]PipeLLM = "Create a character."output = "Text"+llm = "llm_for_creative_writing"prompt_template = """You are a book writer. Your task is to create a character.Think of it and then output the character description."""-llm = "llm_for_creative_writing"-# The llm preset above is defined in `pipelex_libraries/llm_deck/base_llm_deck.toml` as:+# The llm deck above is defined in `pipelex_libraries/llm_deck/base_llm_deck.toml` as:# llm_for_creative_writing = { llm_handle = "best-claude", temperature = 0.9 }-# it's a base preset but you can add your own+# it's a base preset that we provide. you can add your own presets, too.```-💡We have a lot of [LLM presets available](https://github.com/Pipelex/pipelex/tree/main/pipelex/libraries/llm_deck/base_llm_deck.toml).+💡 We have a lot of [LLM presets available by default](https://github.com/Pipelex/pipelex/tree/main/pipelex/libraries/llm_deck/base_llm_deck.toml).Make sure you have credentials for the underlying LLM provider (and added your API key to the `.env`) and select the one you want!-### **Generate a structured output**+Learn more about LLM presets, LLM handles and LLM deck in our [LLM Configuration Guide](../build-reliable-ai-workflows-with-pipelex/configure-ai-llm-to-optimize-workflows.md)-Let's say that we no longer want plain text as output but a rigorous, structured Character object.+### Generate a structured output-### **🖊️ Define the model**+Let's say that we no longer want plain text as output but a rigorously structured Character object.-Using the [Pydantic Basemodel](https://docs.pydantic.dev/latest/) syntax, define your object structure as a Python class.+### Define the model-```python-# pipelex_libraries/pipelines/characters.py+Using the [Pydantic Basemodel](https://docs.pydantic.dev/latest/) syntax, define your object structure as a Python class, in the `pipelex_libraries/pipelines` directory:+`pipelex_libraries/pipelines/characters.py`+```pythonfrom pipelex.core.stuff_content import StructuredContent# Define the structure of your output here@@ -117,34 +115,32 @@ class Character(StructuredContent):description: str```-### **🖊️ Write the script**+ℹ️ We'll soon make it possible to define your structure directly in the `.toml` file, without having to write any python code-It's time to specify that your output should be a `Character` instance. Use the `output` field for that purpose.+### Improve the pipeline++It's time to specify that your output be a `Character` instance. Use the `output` field for that purpose.💡 Here, the concept name matches the class name (ie. `Character`), the `Character` class will automatically be considered as the structure to output.+`pipelex_libraries/pipelines/characters.toml````tomldomain = "characters"-definition = "Tools for creating characters"[concept]-[concept.Character]-Concept = "A character is a fiction story"-# Concept matches class name. Structure is automatically used for structuring the output+Character = "A character is a fiction story" # <- Define here your output concept so that it is linked to the class name[pipe][pipe.create_character]PipeLLM = "Create a character. Get a structured result."-output = "Character"+output = "Character" # <- This is the output concept for your pipeprompt_template = """You are a book writer. Your task is to create a character.Think of it and then output the character description."""```💡 Defining the `Character` concept as "A character is a fiction story" might seem obvious but… think of it: "character" can also mean a letter or symbol in a text. Defining concepts is the best way to avoid any ambiguity and make sure the LLMs understand what you mean.-### **🏃 Run**--+### Run your pipelineAs you can see, the output is a `Character` instance.@@ -153,17 +149,17 @@ As you can see, the output is a `Character` instance.## Generate using information in a prompt template-What if you want to integrate some data into prompts?+What if you want to pass some data into a prompt?You can do that using a prompt template.In this example, we no longer want to generate characters. We want to process existing ones, especially their description attributes.We want to extract structured information from the description field. Thus we have a `Character` input and a `CharacterMetadata` output.-### **Define the output structure**+### Define the output structure```python-# pipelex_libraries/character_model.py+# pipelex_libraries/pipelines/character_model.pyfrom pipelex.core.stuff_content import StructuredContent# input class@@ -181,9 +177,9 @@ class CharacterMetadata(StructuredContent):height: float```-### **Let's use a template to fill prompts with data:**+### **Let's use a template to fill prompts with data**-💡Our template syntax is based on [Jinja2 syntax](https://jinja.palletsprojects.com/en/stable/) so you can include a variable using the classic {{ double.curly.braces }} and to make it simpler, we've added the possibility to just prefix your variable with the "@" symbol:+💡 Our template syntax is based on [Jinja2 syntax](https://jinja.palletsprojects.com/en/stable/). You can include a variable using the **classic** `{{ double.curly.braces }}` and, to make it simpler, we've added the possibility to just prefix your variable with the `@` symbol (recommended). Pipes now declare their required inputs explicitly with the `inputs` table:```toml[concept]@@ -193,7 +189,7 @@ CharacterMetadata = "Metadata regarding a character."[pipe][pipe.extract_character_1]PipeLLM = "Get character information from a description."-input = "Character"+inputs = { character = "Character" } # <- These are the inputs of your pipe, usable in the prompt_templateoutput = "CharacterMetadata"prompt_template = """You are given a text description of a character.@@ -203,12 +199,22 @@ Your task is to extract specific data from the following description."""```+💡 `@character.description` is substituted by grabbing the stuff named `character`in the working memory and using its `description`attribute++Learn more about how we use Jinja in the [PipeLLM documentation](../build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeLLM.md).+### **This is how you do it from the code side**```python+import asyncio+from pipelex.core.stuff_factory import StuffFactory+from pipelex.core.working_memory_factory import WorkingMemoryFactory+from pipelex.pipelex import Pipelexfrom pipelex.pipeline.execute import execute_pipeline-from pipelex_libraries.pipeline.characters import Character, CharacterMetadata++from pipelex_libraries.pipelines.screenplay import Character, CharacterMetadata+async def process_existing_character():# Your existing data@@ -216,17 +222,18 @@ async def process_existing_character():name="Elias",age=38,gender="man",- description = """Elias Varrin is a 38-year-old man, standing at approximately 1.85 meters tall, with a lean,+ occupation="explorer",+ description="""Elias Varrin is a 38-year-old man, standing at approximately 1.85 meters tall, with a lean,weathered frame shaped by decades of travel through remote and often unforgiving landscapes.His name, though not widely known, carries weight among historians, explorers, and those who trade in whispered legends.Elias has piercing storm-gray eyes that scan every environment with sharp precision, and his ash-blond hair—flecked withearly streaks of grey—is usually tucked beneath a wide-brimmed, timeworn hat.His hands are etched with fine scars and stainedwith ink, each mark a silent record of years spent charting unrecorded lands and handling fragile relics of lost civilizations.- He moves with quiet purpose and speaks with a calm, thoughtful cadence that suggests he's always listening for more than just what's said."""+ He moves with quiet purpose and speaks with a calm, thoughtful cadence that suggests he's always listening for more than just what's said.""",)# Wrap it into a stuff objectcharacter_stuff = StuffFactory.make_stuff(- concept_code="character.Character",+ concept_code="character.Character", # <- `character` is the domain, `Character` is the concept namename="character",content=character,)@@ -241,24 +248,15 @@ async def process_existing_character():)# Get the result as a porperly typed instance- extracted_metadata = pipe_output.main_stuff_as(content_type=CharacterMetadata)-- print(CharacterMetadata)+ extracted_metadata = pipe_output.main_stuff_as(content_type=CharacterMetadata) # <- This is the output of your pipe, properly typed++ print(extracted_metadata)+Pipelex.make()-asyncio.run(create_character())+asyncio.run(process_existing_character())```### **Get result**---:arrow_left: [**Back to Installation**](../Installation/Installation.md)--:arrow_right: [**Next section: Pipelex paradigm**](../Pipelex%20paradigm/Pipelex%20paradigm.md)-------"Pipelex" is a trademark of Evotis S.A.S.--© 2025 Evotis S.A.S.+diff --git a/doc/pages/Quick-start/structured_character_sheet.png b/docs/pages/quick-start/structured_character_sheet.pngsimilarity index 100%rename from doc/pages/Quick-start/structured_character_sheet.pngrename to docs/pages/quick-start/structured_character_sheet.pngdiff --git a/doc/pages/Quick-start/structured_character_sheet_json.png b/docs/pages/quick-start/structured_character_sheet_json.pngsimilarity index 100%rename from doc/pages/Quick-start/structured_character_sheet_json.pngrename to docs/pages/quick-start/structured_character_sheet_json.pngdiff --git a/docs/pages/tools/cli.md b/docs/pages/tools/cli.mdnew file mode 100644index 000000000..15fb31849--- /dev/null+++ b/docs/pages/tools/cli.md@@ -0,0 +1,79 @@+# Pipelex CLI Documentation++The Pipelex CLI provides a command-line interface for managing and interacting with your Pipelex projects. This document outlines all available commands and their usage.++## Available Commands++### `pipelex init-libraries`++Initialize Pipelex libraries in the current directory.++```bash+pipelex init-libraries [--overwrite/-o]+```++**Options:**+- `--overwrite`, `-o`: If set, existing library files will be overwritten. Otherwise, only missing files will be created.++### `pipelex init-config`++Initialize Pipelex configuration in the current directory.++```bash+pipelex init-config [--reset/-r]+```++**Options:**+- `--reset`, `-r`: If set, existing configuration file (pipelex.toml) will be overwritten. Otherwise, the command will warn if the file already exists.++### `pipelex validate`++Run the setup sequence to validate your Pipelex configuration and pipelines.++```bash+pipelex validate+```++This command:++1. Exports libraries+2. Validates the configuration+3. Ensures all pipelines are properly set up++### `pipelex show-config`++Display the current Pipelex configuration.++```bash+pipelex show-config+```++Shows the complete configuration for your project, including all settings and parameters.+See more in our [Configuration documentation](../configuration/index.md)++### `pipelex list-pipes`++List all available pipes in your project.++```bash+pipelex list-pipes+```++Displays a table of pipes organized by domain, showing:++- Code: The unique identifier for each pipe+- Definition: Description of the pipe's purpose+- Input: Required input parameters and their concept codes+- Output: The output concept code++The output is formatted as tables grouped by domain, with concept codes simplified when they belong to the current domain.++## Usage Tips++1. Always run `pipelex validate` after making changes to your configuration or pipelines+2. Use `pipelex show-config` to debug configuration issues+3. When initializing a new project:++ - Start with `pipelex init-config`+ - Then run `pipelex init-libraries`+ - Finally, validate your setup with `pipelex validate`diff --git a/docs/pages/tools/logging.md b/docs/pages/tools/logging.mdnew file mode 100644index 000000000..05597db59--- /dev/null+++ b/docs/pages/tools/logging.md@@ -0,0 +1,123 @@+# Pipelex Logging System++## Overview++Pipelex provides a sophisticated logging system that extends Python's standard logging with additional features like emoji support, rich formatting, and custom log levels.++## Log Levels++In addition to standard Python log levels, Pipelex introduces custom levels:++| Level | Value | Description |+|-------|-------|-------------|+| VERBOSE | 5 | Most detailed logging, below DEBUG |+| DEBUG | 10 | Standard debug information |+| DEV | 15 | Development-specific logging (between DEBUG and INFO) |+| INFO | 20 | General informational messages |+| WARNING | 30 | Warning messages |+| ERROR | 40 | Error messages |+| CRITICAL | 50 | Critical errors |+| OFF | 999 | Disable logging |++## Logging Features++### 1. Rich Formatting++- Beautiful console output with color coding+- Syntax highlighting for JSON and other data structures+- Support for clickable file paths+- Word wrapping for better readability++### 2. Emoji Support++Built-in emoji indicators for different components:++- 🧠 Pipelex core messages+- ⚪️ OpenAI-related logs+- 🌀 Google-related logs+- ⚡️ Network connections+- 📡 Web server (Werkzeug)+- *️⃣ JSON processing+- 🧿 Sandbox operations++### 3. Caller Information++Optional inclusion of caller information in logs:++- File name and line number+- Function name+- Module name+- Customizable format templates++### 4. Structured Data Logging++Intelligent handling of different data types:++- Pretty-printing for dictionaries and lists+- JSON formatting with customizable indentation+- Special handling for None values+- Exception traceback integration++### 5. Log Dispatch System++Smart routing of log messages:++- Automatic detection of log origin+- Separate handlers for different logging needs+- Sandbox-aware logging behavior+- Support for poor loggers (simplified logging)++## Using the Logger++```python+from pipelex import log++# Basic logging+log.info("Simple message")++# Logging with title+log.info("Detailed message", title="Process Status")++# Logging with inline title+log.info("Quick update", inline="Status")++# Logging structured data+data = {"key": "value", "nested": {"data": True}}+log.debug(data, title="Configuration")++# Warning with problem ID+log.warning("API rate limit approaching", problem_id="rate_limit_warning")++# Error with exception traceback+log.error("Failed to process", include_exception=True)++# Development logging+log.dev("Testing new feature")++# Verbose logging+log.verbose("Detailed debug information")+```++## Best Practices++1. **Log Level Selection**:++ - Use VERBOSE for detailed debugging+ - Use DEBUG for general debugging+ - Use DEV for development-specific logging+ - Use INFO for general progress+ - Use WARNING for potential issues+ - Use ERROR for actual errors+ - Use CRITICAL for system-critical issues++2. **Structured Data**:++ - Log complex data structures directly+ - Use titles for context+ - Include problem IDs for trackable issues++3. **Exception Handling**:++ - Use `include_exception=True` for error context+ - Include relevant data in error logs+ - Use appropriate log levels for exceptionsdiff --git a/docs/stylesheets/general.css b/docs/stylesheets/general.cssnew file mode 100644index 000000000..46c984aff--- /dev/null+++ b/docs/stylesheets/general.css@@ -0,0 +1,8 @@+/* Custom styles for Pipelex documentation */++/* Equal width buttons */+.equal-width {+ width: 100%;+ max-width: 300px;+ display: inline-block;+}\ No newline at end of filediff --git a/docs/stylesheets/mermaid.css b/docs/stylesheets/mermaid.cssnew file mode 100644index 000000000..d9052745b--- /dev/null+++ b/docs/stylesheets/mermaid.css@@ -0,0 +1,36 @@+/* docs/stylesheets/mermaid.css+ Tweak the colours Material exposes for Mermaid */++.mermaid {+ --md-mermaid-edge-color: #000000;+ --md-mermaid-label-bg-color: #ffdbfa;+ --md-mermaid-label-fg-color: #000000;++ --md-mermaid-node-bg-color: #fff9db;+ --md-mermaid-node-fg-color: #000000;+}++/* Give the diagram itself a code-block like backdrop so it stands out+ on both light and dark themes. */+.md-typeset .mermaid {+ padding: 0.3rem;+ border-radius: 0.3rem;+ overflow: auto;+}++/* --- Light color scheme ---------------------------------------------- */++[data-md-color-scheme="default"] .md-typeset .mermaid {+ background-color: hsla(140, 27%, 98%, 0.983);+}++/* --- Dark color scheme ------------------------------------------------ */++[data-md-color-scheme="slate"] .md-typeset .mermaid {+ background-color: hsla(140, 27%, 98%, 0.983);+}++/* Ensure text inside nodes and labels is black */+.mermaid text {+ fill: #000 !important;+}\ No newline at end of filediff --git a/mkdocs.yml b/mkdocs.ymlnew file mode 100644index 000000000..67c537bed--- /dev/null+++ b/mkdocs.yml@@ -0,0 +1,164 @@+site_name: Pipelex Docs+site_url: https://docs.pipelex.com/+site_description: "Official documentation for Pipelex, a framework and language for building deterministic, repeatable AI workflows and knowledge pipelines."+docs_dir: docs+repo_url: "https://github.com/Pipelex/pipelex"+repo_name: "Pipelex on GitHub"+watch: [README.md, CONTRIBUTING.md, CHANGELOG.md, MANIFESTO.md, LICENSE]++theme:+ name: material+ custom_dir: docs/overrides+ favicon: images/pipelex-favicon-32.png+ logo: https://pipelex-web.s3.amazonaws.com/logo/Pipelex-logo-wot-mono-560x176.png+ features:+ - navigation.footer+ - navigation.tabs+ - navigation.tabs.sticky+ - navigation.sections+ - navigation.top+ palette:+ - scheme: default # light+ primary: teal+ accent: deep purple+ toggle:+ icon: material/weather-night+ name: Switch to dark mode+ - scheme: slate # dark+ primary: teal+ accent: purple+ toggle:+ icon: material/weather-sunny+ name: Switch to light mode++copyright: "© 2025 Evotis S.A.S. <br/>Pipelex is a trademark of Evotis S.A.S."++extra:+ social:+ - icon: fontawesome/brands/github+ link: https://github.com/Pipelex/pipelex+ name: Pipelex on GitHub+ - icon: fontawesome/brands/twitter+ link: https://x.com/PipelexAI+ name: Pipelex on X+ - icon: fontawesome/brands/discord+ link: https://discord.gg/SReshKQjWt+ name: Pipelex on Discord+ generator: false++plugins:+ - meta-manager+ - glightbox:+ touchNavigation: true+ loop: false+ effect: zoom+ slide_effect: slide+ width: 100%+ height: auto+ zoomable: true+ draggable: true+ auto_caption: false+ caption_position: bottom++markdown_extensions:+ - meta+ - attr_list+ - md_in_html+ - pymdownx.blocks.caption+ - pymdownx.highlight+ - pymdownx.superfences:+ custom_fences:+ - name: mermaid+ class: mermaid+ format: !!python/name:pymdownx.superfences.fence_code_format+ - pymdownx.snippets:+ base_path: .+ check_paths: true+ - pymdownx.emoji:+ emoji_index: !!python/name:material.extensions.emoji.twemoji+ emoji_generator: !!python/name:material.extensions.emoji.to_svg++nav:+ - Home:+ - Welcome: index.md+ - Installation: pages/installation/index.md+ - Quick-start: pages/quick-start/index.md+ - Understand Pipelex:+ - The Knowledge Pipeline Manifesto: manifesto.md+ - The Pipelex Paradigm: pages/pipelex-paradigm-for-repeatable-ai-workflows/index.md+ - Cookbook Examples:+ - Overview: pages/cookbook-examples/index.md+ - Hello World: pages/cookbook-examples/hello-world.md+ - Document Processing:+ - Simple OCR: pages/cookbook-examples/simple-ocr.md+ - Generic Document Extraction: pages/cookbook-examples/extract-generic.md+ - Invoice Extractor: pages/cookbook-examples/invoice-extractor.md+ - Proof of Purchase Extraction: pages/cookbook-examples/extract-proof-of-purchase.md+ - Graphical Extraction:+ - Gantt Chart Extraction: pages/cookbook-examples/extract-gantt.md+ - Table Extraction from Image: pages/cookbook-examples/extract-table.md+ - DPE Extraction: pages/cookbook-examples/extract-dpe.md+ - Text Generation:+ - Screenplay Generator (wip): pages/cookbook-examples/write-screenplay.md+ - Tweet Optimizer (wip): pages/cookbook-examples/write-tweet.md+ - Build Reliable AI Workflows:+ - Kick off a Pipeline Project: pages/build-reliable-ai-workflows-with-pipelex/kick-off-a-knowledge-pipeline-project.md+ - Define Your Concepts: pages/build-reliable-ai-workflows-with-pipelex/define_your_concepts.md+ - Design and Run Pipelines: pages/build-reliable-ai-workflows-with-pipelex/design_and_run_pipelines.md+ - Pipe Operators:+ - Overview: pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/index.md+ - PipeLLM: pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeLLM.md+ - PipeOcr: pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeOcr.md+ - PipeImgGen: pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeImgGen.md+ - PipeJinja2: pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeJinja2.md+ - PipeFunc: pages/build-reliable-ai-workflows-with-pipelex/pipe-operators/PipeFunc.md+ - Pipe Controllers:+ - Overview: pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/index.md+ - PipeSequence: pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeSequence.md+ - PipeParallel: pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeParallel.md+ - PipeBatch: pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeBatch.md+ - PipeCondition: pages/build-reliable-ai-workflows-with-pipelex/pipe-controllers/PipeCondition.md+ - Optimize Cost & Quality: pages/build-reliable-ai-workflows-with-pipelex/configure-ai-llm-to-optimize-workflows.md+ - LLM Structured Generation: pages/build-reliable-ai-workflows-with-pipelex/llm-structured-generation-config.md+ - LLM Prompting Style: pages/build-reliable-ai-workflows-with-pipelex/adapt-to-llm-prompting-style-openai-anthropic-mistral.md+ - LLM Providers & Models: pages/build-reliable-ai-workflows-with-pipelex/ai-plugins-for-multi-llm-workflows.md+ - Configuration:+ - Overview: pages/configuration/index.md+ - Pipeline Validation Configuration:+ - Dry Run: pages/configuration/config-pipeline-validation/dry-run-config.md+ - Static Validation: pages/configuration/config-pipeline-validation/static-validation-config.md+ - Practical Configuration:+ - Logging: pages/configuration/config-practical/logging-config.md+ - Pipe Run: pages/configuration/config-practical/pipe-run-config.md+ - Reporting: pages/configuration/config-practical/reporting-config.md+ - Tracker: pages/configuration/config-practical/tracker-config.md+ - Technical Configuration:+ - AWS: pages/configuration/config-technical/aws-config.md+ - Cogt: pages/configuration/config-technical/cogt-config.md+ - Library: pages/configuration/config-technical/library-config.md+ - Feature: pages/configuration/config-advanced/feature-config.md+ - Development:+ - Changelog: changelog.md+ - Tools:+ - CLI: pages/tools/cli.md+ - Logging: pages/tools/logging.md+ - Advanced Customizations:+ - Overview: pages/advanced-customization/index.md+ - Secrets Provider: pages/advanced-customization/secrets-provider-injection.md+ - Reporting Delegate: pages/advanced-customization/reporting-delegate-injection.md+ - Pipeline Tracker: pages/advanced-customization/pipeline-tracker-injection.md+ - Activity Manager: pages/advanced-customization/activity-manager-injection.md+ - Template Provider: pages/advanced-customization/template-provider-injection.md+ - Content Generator: pages/advanced-customization/content-generator-injection.md+ - LLM Model Provider: pages/advanced-customization/llm-model-provider-injection.md+ - Pipe Router: pages/advanced-customization/pipe-router-injection.md+ - Plugin Manager: pages/advanced-customization/plugin-manager-injection.md+ - Inference Manager: pages/advanced-customization/inference-manager-injection.md+ - License: license.md+ - Contributing: contributing.md+ - Code of Conduct: CODE_OF_CONDUCT.md++extra_css:+ - stylesheets/mermaid.css+ - stylesheets/general.css+\ No newline at end of filediff --git a/pipelex/cogt/inference/cost_registry.py b/pipelex/cogt/inference/cost_registry.pyindex c8fc821cc..69a27e96a 100644--- a/pipelex/cogt/inference/cost_registry.py+++ b/pipelex/cogt/inference/cost_registry.py@@ -34,7 +34,10 @@ class CostRegistry(RootModel[CostRegistryRoot]):cost_report_file_path: Optional[str] = None,):if not llm_tokens_usages:- log.warning(f"No report to generate for pipeline '{pipeline_run_id}'")+ if pipeline_run_id != "untitled":+ log.warning(f"No report to generate for pipeline '{pipeline_run_id}'")+ else:+ log.verbose(f"No report to generate for pipeline '{pipeline_run_id}'")returncost_registry = CostRegistry()for llm_tokens_usage in llm_tokens_usages:diff --git a/pipelex/core/concept.py b/pipelex/core/concept.pyindex e5ba8f457..cf74c6fd3 100644--- a/pipelex/core/concept.py+++ b/pipelex/core/concept.py@@ -6,6 +6,7 @@ from pydantic import BaseModel, ConfigDict, Field, field_validator, model_validafrom typing_extensions import Selffrom pipelex import log+from pipelex.core.concept_native import NativeConceptfrom pipelex.core.stuff_content import StuffContentfrom pipelex.exceptions import ConceptCodeError, ConceptDomainError, ConceptError, StructureClassErrorfrom pipelex.tools.misc.string_utils import pascal_case_to_sentence@@ -56,14 +57,30 @@ class Concept(BaseModel):@field_validator("refines")@classmethoddef validate_refines(cls, value: List[str]) -> List[str]:- for code in value:- if not cls.concept_str_contains_domain(code):- raise ConceptCodeError(f"Each inherited code must contain a dot (.), got: {code}")-- domain, code = Concept.extract_domain_and_concept_from_str(concept_str=code)- cls.validate_concept_code_syntax(code=code, concept_code=code, domain_field=code)- cls.validate_domain_syntax(domain=domain, code=code, domain_field=code)- return value+ validated_refines: List[str] = []++ for refine_code in value:+ # Handle NativeConcept values directly without importing ConceptCodeFactory to avoid circular import+ if not cls.concept_str_contains_domain(refine_code):+ # Check if it's a valid NativeConcept name+ if refine_code in NativeConcept.names():+ native_concept = NativeConcept(refine_code)+ full_code = native_concept.code+ validated_refines.append(full_code)+ continue+ else:+ raise ConceptCodeError(f"Each refine code must contain a single dot (.), got: {refine_code}")+ else:+ # Already has domain, validate it directly+ full_code = refine_code+ validated_refines.append(full_code)++ # Validate the domain and concept syntax for the full code+ domain, code = cls.extract_domain_and_concept_from_str(concept_str=full_code)+ cls.validate_concept_code_syntax(code=code, concept_code=full_code, domain_field=domain)+ cls.validate_domain_syntax(domain=domain, code=full_code, domain_field=domain)++ return validated_refines@field_validator("structure_class_name")@classmethod@@ -81,12 +98,6 @@ class Concept(BaseModel):log.warning(f"Concept class '{structure_class_name}' is registered but it's not a subclass of StuffContent")return False- @classmethod- def check_possible_concept_from_str(cls, concept_str: str):- parts = concept_str.split(".")- if len(parts) > 2:- raise ConceptError(f"Concept code '{concept_str}' contains more than one dot")-@classmethoddef extract_domain_and_concept_from_str(cls, concept_str: str) -> Tuple[str, str]:if "." in concept_str:diff --git a/pipelex/core/concept_code_factory.py b/pipelex/core/concept_code_factory.pynew file mode 100644index 000000000..4a533b39f--- /dev/null+++ b/pipelex/core/concept_code_factory.py@@ -0,0 +1,25 @@+from typing import Optional++from pipelex.core.concept import Concept+from pipelex.core.concept_native import NativeConcept+from pipelex.exceptions import ConceptFactoryError+++class ConceptCodeFactory:+ @classmethod+ def make_concept_code(cls, domain: str, code: str) -> str:+ if "." in code:+ return code+ return f"{domain}.{code}"++ @classmethod+ def make_concept_code_from_str(cls, concept_str: str, fallback_domain: Optional[str] = None) -> str:+ if Concept.concept_str_contains_domain(concept_str=concept_str):+ return concept_str+ elif concept_str in NativeConcept.names():+ native_concept = NativeConcept(concept_str)+ return native_concept.code+ elif fallback_domain:+ return cls.make_concept_code(domain=fallback_domain, code=concept_str)+ else:+ raise ConceptFactoryError(f"Concept '{concept_str}' does not contain a domain and no fallback domain was provided")diff --git a/pipelex/core/concept_factory.py b/pipelex/core/concept_factory.pyindex 80f130647..de37aebc9 100644--- a/pipelex/core/concept_factory.py+++ b/pipelex/core/concept_factory.py@@ -1,10 +1,13 @@from inspect import getsource-from typing import Any, Dict, List, Optional, Type+from typing import Any, Dict, List, Optional, Type, Unionfrom kajson.class_registry import class_registryfrom pydantic import BaseModel, ConfigDict, Field, ValidationErrorfrom pipelex.core.concept import Concept+from pipelex.core.concept_code_factory import ConceptCodeFactory+from pipelex.core.concept_native import NativeConcept+from pipelex.core.domain import SpecialDomainfrom pipelex.core.stuff_content import TextContentfrom pipelex.exceptions import ConceptFactoryError, StructureClassError@@ -14,26 +17,23 @@ class ConceptBlueprint(BaseModel):definition: strstructure: Optional[str] = None- refines: List[str] = Field(default_factory=list)+ refines: Union[str, List[str]] = Field(default_factory=list)domain: Optional[str] = Noneclass ConceptFactory:@classmethod- def make_refines(cls, domain: str, refines: List[str]) -> List[str]:+ def make_refines(cls, domain: str, refines: Union[str, List[str]]) -> List[str]:+ if isinstance(refines, str):+ concept_str_list = [refines]+ else:+ concept_str_list = refinesnew_refines: List[str] = []- for ref in refines:- if not Concept.concept_str_contains_domain(ref):- ref = ConceptFactory.make_concept_code(domain=domain, code=ref)- new_refines.append(ref)+ for concept_str in concept_str_list:+ concept_code = ConceptCodeFactory.make_concept_code_from_str(concept_str=concept_str, fallback_domain=domain)+ new_refines.append(concept_code)return new_refines- @classmethod- def make_concept_code(cls, domain: str, code: str) -> str:- if "." in code:- return code- return f"{domain}.{code}"-@classmethoddef make_from_details_dict_if_possible(cls,@@ -51,7 +51,7 @@ class ConceptFactory:# legacy formatdetails_dict["domain"] = domaindetails_dict["refines"] = ConceptFactory.make_refines(domain=domain, refines=details_dict.pop("refines", []))- details_dict["code"] = ConceptFactory.make_concept_code(domain, code)+ details_dict["code"] = ConceptCodeFactory.make_concept_code(domain, code)try:the_concept = Concept.model_validate(details_dict)except ValidationError as e:@@ -94,7 +94,7 @@ class ConceptFactory:# TODO: don't wall make_concept_code from the factory, the code received here must already be a valid concept codeconcept_dict = {"domain": domain_code,- "code": ConceptFactory.make_concept_code(domain_code, code),+ "code": ConceptCodeFactory.make_concept_code(domain_code, code),"definition": definition,"structure_class_name": structure_class_name,}@@ -125,14 +125,59 @@ class ConceptFactory:else:structure_class_name = TextContent.__name__+ refines_list: List[str]+ if isinstance(concept_blueprint.refines, str):+ refines_list = [concept_blueprint.refines]+ else:+ refines_list = concept_blueprint.refines+return Concept(- code=ConceptFactory.make_concept_code(domain, code),+ code=ConceptCodeFactory.make_concept_code(domain, code),domain=domain,definition=concept_blueprint.definition,structure_class_name=structure_class_name,- refines=concept_blueprint.refines,+ refines=refines_list,+ )++ @classmethod+ def make_native_concept(cls, native_concept: NativeConcept) -> Concept:+ definition: str+ match native_concept:+ case NativeConcept.TEXT:+ definition = "A text"+ case NativeConcept.IMAGE:+ definition = "An image"+ case NativeConcept.PDF:+ definition = "A PDF"+ case NativeConcept.TEXT_AND_IMAGES:+ definition = "A text and an image"+ case NativeConcept.NUMBER:+ definition = "A number"+ case NativeConcept.LLM_PROMPT:+ definition = "A prompt for an LLM"+ case NativeConcept.DYNAMIC:+ definition = "A dynamic concept"+ case NativeConcept.PAGE:+ definition = "The content of a page of a document, comprising text and linked images as well as an optional page view image"+ case NativeConcept.ANYTHING:+ raise RuntimeError("NativeConcept.ANYTHING cannot be used as a concept")++ return Concept(+ code=native_concept.code,+ domain=SpecialDomain.NATIVE,+ definition=definition,+ structure_class_name=native_concept.content_class_name,)+ @classmethod+ def list_native_concepts(cls) -> List[Concept]:+ concepts: List[Concept] = []+ for native_concept in NativeConcept:+ if native_concept == NativeConcept.ANYTHING:+ continue+ concepts.append(cls.make_native_concept(native_concept=native_concept))+ return concepts+@classmethoddef get_concept_class_source_code(cls, concept_name: str, base_class: Type[Any]) -> str:if not class_registry.has_class(concept_name):diff --git a/pipelex/core/concept_native.py b/pipelex/core/concept_native.pyindex 0a1dab12d..d0c6f6e83 100644--- a/pipelex/core/concept_native.py+++ b/pipelex/core/concept_native.py@@ -1,8 +1,6 @@from enum import Enumfrom typing import List-from pipelex.core.concept import Concept-from pipelex.core.concept_factory import ConceptFactoryfrom pipelex.core.domain import SpecialDomainfrom pipelex.types import StrEnum@@ -19,7 +17,7 @@ class NativeConceptClass(StrEnum):# Exceptionally, we use an Enum here (and not our usual StrEnum) to avoid confusion with-# the concept_code with must have the form "native.ConceptName"+# the concept_code which must have the form "native.ConceptName"class NativeConcept(Enum):ANYTHING = "Anything"DYNAMIC = "Dynamic"@@ -37,7 +35,9 @@ class NativeConcept(Enum):@propertydef code(self) -> str:- return ConceptFactory.make_concept_code(SpecialDomain.NATIVE, self.value)+ if "." in self.value:+ return self.value+ return f"{SpecialDomain.NATIVE}.{self.value}"@propertydef content_class_name(self) -> NativeConceptClass:@@ -60,41 +60,3 @@ class NativeConcept(Enum):return NativeConceptClass.PAGEcase NativeConcept.ANYTHING:raise RuntimeError("NativeConcept.ANYTHING cannot be used as a content class name")-- def make_concept(self) -> Concept:- definition: str- match self:- case NativeConcept.TEXT:- definition = "A text"- case NativeConcept.IMAGE:- definition = "An image"- case NativeConcept.PDF:- definition = "A PDF"- case NativeConcept.TEXT_AND_IMAGES:- definition = "A text and an image"- case NativeConcept.NUMBER:- definition = "A number"- case NativeConcept.LLM_PROMPT:- definition = "A prompt for an LLM"- case NativeConcept.DYNAMIC:- definition = "A dynamic concept"- case NativeConcept.PAGE:- definition = "The content of a page of a document, comprising text and linked images as well as an optional page view image"- case NativeConcept.ANYTHING:- raise RuntimeError("NativeConcept.ANYTHING cannot be used as a concept")-- return Concept(- code=self.code,- domain=SpecialDomain.NATIVE,- definition=definition,- structure_class_name=self.content_class_name,- )-- @classmethod- def all_concepts(cls) -> List[Concept]:- concepts: List[Concept] = []- for code in cls:- if code == cls.ANYTHING:- continue- concepts.append(code.make_concept())- return conceptsdiff --git a/pipelex/core/pipe_blueprint.py b/pipelex/core/pipe_blueprint.pyindex 5992279c3..60488a4a0 100644--- a/pipelex/core/pipe_blueprint.py+++ b/pipelex/core/pipe_blueprint.py@@ -1,9 +1,9 @@from typing import Any, Dict, Optional, Protocol, TypeVar-from pydantic import ConfigDict, field_validator, model_validator+from pydantic import ConfigDict, model_validatorfrom typing_extensions import Self, runtime_checkable-from pipelex.core.concept_native import NativeConcept+from pipelex.core.concept_code_factory import ConceptCodeFactoryfrom pipelex.core.pipe_abstract import PipeAbstractfrom pipelex.core.stuff_content import StructuredContent@@ -18,34 +18,18 @@ class PipeBlueprint(StructuredContent):@model_validator(mode="after")def add_domain_prefix(self) -> Self:- # if self.input and "." not in self.input:- # self.input = f"{self.domain}.{self.input}"if self.inputs:for input_name, input_concept_code in self.inputs.items():- if "." not in input_concept_code:- self.inputs[input_name] = f"{self.domain}.{input_concept_code}"- if self.output and "." not in self.output:- self.output = f"{self.domain}.{self.output}"+ self.inputs[input_name] = ConceptCodeFactory.make_concept_code_from_str(+ concept_str=input_concept_code,+ fallback_domain=self.domain,+ )+ self.output = ConceptCodeFactory.make_concept_code_from_str(+ concept_str=self.output,+ fallback_domain=self.domain,+ )return self- @classmethod- def _add_native_prefix_if_needed(cls, value: str) -> str:- if value in NativeConcept.names():- return f"native.{value}"- return value-- @field_validator("inputs")- @classmethod- def validate_inputs(cls, value: Optional[Dict[str, str]]) -> Optional[Dict[str, str]]:- if value:- return {name: cls._add_native_prefix_if_needed(concept_code) for name, concept_code in value.items()}- return value-- @field_validator("output")- @classmethod- def validate_output(cls, value: str) -> str:- return cls._add_native_prefix_if_needed(value)-PipeBlueprintType = TypeVar("PipeBlueprintType", bound="PipeBlueprint", contravariant=True)diff --git a/pipelex/core/pipe_input_details.py b/pipelex/core/pipe_input_details.pyindex d3814d9e9..94e0ccd4b 100644--- a/pipelex/core/pipe_input_details.py+++ b/pipelex/core/pipe_input_details.py@@ -3,8 +3,7 @@ from typing import Callable, Dict, List, Set, Tuplefrom pydantic import Field, RootModel, field_validatorfrom pipelex import log-from pipelex.core.concept import Concept-from pipelex.exceptions import ConceptError, PipeInputSpecError+from pipelex.core.concept_code_factory import ConceptCodeFactoryPipeInputDetailsRoot = Dict[str, str]@@ -27,16 +26,13 @@ class PipeInputDetails(RootModel[PipeInputDetailsRoot]):log.warning(f"Sub-attribute {required_input} detected, using {transformed_key} as variable name")# Validate concept_code- try:- Concept.check_possible_concept_from_str(concept_str=concept_str)- except ConceptError as exc:- raise PipeInputSpecError(f"Invalid concept code: {concept_str}") from exc+ concept_code = ConceptCodeFactory.make_concept_code_from_str(concept_str=concept_str)- if transformed_key in transformed_dict and transformed_dict[transformed_key] != concept_str:+ if transformed_key in transformed_dict and transformed_dict[transformed_key] != concept_code:log.warning(f"Variable {transformed_key} already exists with a different concept code: {transformed_dict[transformed_key]} -> {concept_str}")- transformed_dict[transformed_key] = concept_str+ transformed_dict[transformed_key] = concept_codereturn transformed_dictdiff --git a/pipelex/core/pipe_input_spec.py b/pipelex/core/pipe_input_spec.pyindex e41a68625..00466df5b 100644--- a/pipelex/core/pipe_input_spec.py+++ b/pipelex/core/pipe_input_spec.py@@ -3,8 +3,8 @@ from typing import Callable, Dict, List, Set, Tuplefrom pydantic import Field, RootModel, field_validatorfrom pipelex import log-from pipelex.core.concept import Concept-from pipelex.exceptions import ConceptError, PipeInputNotFoundError, PipeInputSpecError+from pipelex.core.concept_code_factory import ConceptCodeFactory+from pipelex.exceptions import PipeInputNotFoundErrorPipeInputSpecRoot = Dict[str, str]@@ -32,16 +32,13 @@ class PipeInputSpec(RootModel[PipeInputSpecRoot]):log.warning(f"Sub-attribute {required_input} detected, using {transformed_key} as variable name")# Validate concept_code- try:- Concept.check_possible_concept_from_str(concept_str=concept_str)- except ConceptError as exc:- raise PipeInputSpecError(f"Invalid concept code: {concept_str}") from exc+ concept_code = ConceptCodeFactory.make_concept_code_from_str(concept_str=concept_str)- if transformed_key in transformed_dict and transformed_dict[transformed_key] != concept_str:+ if transformed_key in transformed_dict and transformed_dict[transformed_key] != concept_code:log.warning(f"Variable {transformed_key} already exists with a different concept code: {transformed_dict[transformed_key]} -> {concept_str}")- transformed_dict[transformed_key] = concept_str+ transformed_dict[transformed_key] = concept_codereturn transformed_dictdiff --git a/pipelex/core/stuff.py b/pipelex/core/stuff.pyindex 48e0fbeb9..3ef47b43f 100644--- a/pipelex/core/stuff.py+++ b/pipelex/core/stuff.py@@ -87,7 +87,7 @@ class Stuff(CustomBaseModel):@overridedef __str__(self) -> str:- return f"{self.title}\n{self.content.rendered_plain()}"+ return f"{self.title}\n{self.content.rendered_json()}"@propertydef is_list(self) -> bool:diff --git a/pipelex/core/stuff_content.py b/pipelex/core/stuff_content.pyindex 237f724af..4e32ebeb2 100644--- a/pipelex/core/stuff_content.py+++ b/pipelex/core/stuff_content.py@@ -38,7 +38,7 @@ class StuffContent(ABC, CustomBaseModel):@overridedef __str__(self) -> str:- return self.rendered_plain()+ return self.rendered_json()def rendered_str(self, text_format: TextFormat = TextFormat.PLAIN) -> str:match text_format:diff --git a/pipelex/core/stuff_factory.py b/pipelex/core/stuff_factory.pyindex 412eb28d8..64071b3a0 100644--- a/pipelex/core/stuff_factory.py+++ b/pipelex/core/stuff_factory.py@@ -5,7 +5,8 @@ from kajson.class_registry import class_registryfrom pydantic import BaseModel, Fieldfrom pipelex.config import get_config-from pipelex.core.concept import Concept+from pipelex.core.concept import Concept, ConceptError+from pipelex.core.concept_code_factory import ConceptCodeFactoryfrom pipelex.core.concept_native import NativeConceptfrom pipelex.core.stuff import Stuff, StuffCreationRecordfrom pipelex.core.stuff_content import StuffContent, StuffContentInitableFromStr@@ -31,15 +32,18 @@ class StuffFactory:@classmethoddef make_stuff(cls,- concept_code: str,+ concept_str: str,content: StuffContent,name: Optional[str] = None,code: Optional[str] = None,creation_record: Optional[StuffCreationRecord] = None,pipelex_session_id: Optional[str] = None,) -> Stuff:- if not Concept.concept_str_contains_domain(concept_str=concept_code):- raise StuffFactoryError(f"Concept '{concept_code}' does not contain a domain")+ try:+ concept_code = ConceptCodeFactory.make_concept_code_from_str(concept_str=concept_str)+ except ConceptError:+ stuff_ref = name or code or "unnamed"+ raise StuffFactoryError(f"Concept '{concept_str}' does not contain a domain, could not make stuff '{stuff_ref}'")if not name:name = cls.make_stuff_name(concept_code)return Stuff(@@ -85,7 +89,7 @@ class StuffFactory:@classmethoddef make_from_blueprint(cls, blueprint: StuffBlueprint) -> "Stuff":the_stuff = cls.make_from_str(- concept_code=blueprint.concept_code,+ concept_str=blueprint.concept_code,str_value=blueprint.value,name=blueprint.name,pipelex_session_id="blueprint",@@ -97,11 +101,14 @@ class StuffFactory:cls,str_value: str,name: Optional[str] = None,- concept_code: str = NativeConcept.TEXT.code,+ concept_str: str = NativeConcept.TEXT.code,pipelex_session_id: Optional[str] = None,) -> Stuff:- if not Concept.concept_str_contains_domain(concept_code):- raise StuffFactoryError(f"Concept '{concept_code}' does not contain a domain")+ try:+ concept_code = ConceptCodeFactory.make_concept_code_from_str(concept_str=concept_str)+ except ConceptError:+ stuff_ref = name or "unnamed"+ raise StuffFactoryError(f"Concept '{concept_str}' does not contain a domain, could not make stuff '{stuff_ref}'")the_concept = get_required_concept(concept_code=concept_code)the_subclass_name = the_concept.structure_class_namethe_subclass = class_registry.get_class(name=the_subclass_name) or eval(the_subclass_name)@@ -110,10 +117,10 @@ class StuffFactory:stuff_content: StuffContent = the_subclass.make_from_str(str_value)if not name:- name = cls.make_stuff_name(concept_code)+ name = cls.make_stuff_name(concept_str)return Stuff(- concept_code=concept_code,+ concept_code=concept_str,content=stuff_content,stuff_name=name,stuff_code=shortuuid.uuid()[:5],@@ -127,7 +134,7 @@ class StuffFactory:Make multiple stuffs from a dictionary of strings.It is implied that each string value should be associated with a native.Text concept."""- return [cls.make_from_str(concept_code=NativeConcept.TEXT.code, str_value=str_value, name=name) for name, str_value in str_text_dict.items()]+ return [cls.make_from_str(concept_str=NativeConcept.TEXT.code, str_value=str_value, name=name) for name, str_value in str_text_dict.items()]@classmethoddef make_multiple_stuff_from_str(cls, str_stuff_and_concepts_dict: Dict[str, Tuple[str, str]]) -> List[Stuff]:@@ -137,7 +144,7 @@ class StuffFactory:"""result: List[Stuff] = []for name, (concept_code, str_value) in str_stuff_and_concepts_dict.items():- stuff = cls.make_from_str(concept_code=concept_code, str_value=str_value, name=name)+ stuff = cls.make_from_str(concept_str=concept_code, str_value=str_value, name=name)result.append(stuff)return result@@ -151,7 +158,7 @@ class StuffFactory:the_subclass = class_registry.get_required_subclass(name=the_subclass_name, base_class=StuffContent)the_stuff_content = the_subclass.model_validate(obj=stuff_contents)return cls.make_stuff(- concept_code=concept_code,+ concept_str=concept_code,content=the_stuff_content,name=name,)diff --git a/pipelex/core/working_memory_factory.py b/pipelex/core/working_memory_factory.pyindex a95765ca4..d6ea5c447 100644--- a/pipelex/core/working_memory_factory.py+++ b/pipelex/core/working_memory_factory.py@@ -16,11 +16,11 @@ class WorkingMemoryFactory(BaseModel):def make_from_text(cls,text: str,- concept_code: str = NativeConcept.TEXT.code,+ concept_str: str = NativeConcept.TEXT.code,name: Optional[str] = "text",) -> WorkingMemory:stuff = StuffFactory.make_stuff(- concept_code=concept_code,+ concept_str=concept_str,content=TextContent(text=text),name=name,)@@ -30,11 +30,11 @@ class WorkingMemoryFactory(BaseModel):def make_from_image(cls,image_url: str,- concept_code: str = NativeConcept.IMAGE.code,+ concept_str: str = NativeConcept.IMAGE.code,name: Optional[str] = "image",) -> WorkingMemory:stuff = StuffFactory.make_stuff(- concept_code=concept_code,+ concept_str=concept_str,content=ImageContent(url=image_url),name=name,)@@ -44,11 +44,11 @@ class WorkingMemoryFactory(BaseModel):def make_from_pdf(cls,pdf_url: str,- concept_code: str = NativeConcept.PDF.code,+ concept_str: str = NativeConcept.PDF.code,name: Optional[str] = "pdf",) -> WorkingMemory:stuff = StuffFactory.make_stuff(- concept_code=concept_code,+ concept_str=concept_str,content=PDFContent(url=pdf_url),name=name,)diff --git a/pipelex/libraries/library_manager.py b/pipelex/libraries/library_manager.pyindex eaa7fce2d..4226d38e7 100644--- a/pipelex/libraries/library_manager.py+++ b/pipelex/libraries/library_manager.py@@ -10,7 +10,6 @@ from pipelex import logfrom pipelex.cogt.llm.llm_models.llm_deck import LLMDeckfrom pipelex.core.concept_factory import ConceptFactoryfrom pipelex.core.concept_library import ConceptLibrary-from pipelex.core.concept_native import NativeConceptfrom pipelex.core.domain import Domainfrom pipelex.core.domain_library import DomainLibraryfrom pipelex.core.pipe_abstract import PipeAbstract@@ -86,7 +85,7 @@ class LibraryManager:)library_paths += [LibraryConfig.test_pipelines_path]- native_concepts = NativeConcept.all_concepts()+ native_concepts = ConceptFactory.list_native_concepts()self.concept_library.add_concepts(concepts=native_concepts)self._load_combo_libraries(library_paths=library_paths)diff --git a/pipelex/libraries/pipelines/documents.toml b/pipelex/libraries/pipelines/documents.tomlindex ea9b86f9b..c2716ad45 100644--- a/pipelex/libraries/pipelines/documents.toml+++ b/pipelex/libraries/pipelines/documents.toml@@ -13,14 +13,14 @@ TextAndImagesContent = "A content that comprises text and images where the text# but it must be either an image or a pdf or a concept which refines one of them[pipe.extract_page_contents_from_pdf]PipeOcr = "Extract page contents from a PDF document"-inputs = { pdf = "native.PDF" }-output = "PageContent"+inputs = { pdf = "PDF" }+output = "Page"page_images = truepage_views = false[pipe.extract_page_contents_and_views_from_pdf]PipeOcr = "Extract page contents from a PDF document as well aspage views"-inputs = { pdf = "native.PDF" }-output = "PageContent"+inputs = { pdf = "PDF" }+output = "Page"page_images = truepage_views = truediff --git a/pipelex/libraries/pipelines/images.toml b/pipelex/libraries/pipelines/images.tomlindex 9efb44e07..0bb7cd01b 100644--- a/pipelex/libraries/pipelines/images.toml+++ b/pipelex/libraries/pipelines/images.toml@@ -8,12 +8,12 @@ VisualDescription = "Visual description of something"[concept.ImgGenPrompt]Concept = "Prompt to generate an image"-refines = ["native.Text"]+refines = "Text"[concept.Photo]Concept = "Photo"structure = "ImageContent"-refines = ["native.Image"]+refines = "Image"[pipe]diff --git a/pipelex/pipe_controllers/pipe_batch.py b/pipelex/pipe_controllers/pipe_batch.pyindex 99ac6d4ba..190557ccd 100644--- a/pipelex/pipe_controllers/pipe_batch.py+++ b/pipelex/pipe_controllers/pipe_batch.py@@ -83,7 +83,7 @@ class PipeBatch(PipeController):branch_input_item_code = f"{input_stuff_code}-branch-{branch_index}"item_input_stuff = StuffFactory.make_stuff(code=branch_input_item_code,- concept_code=input_item_concept_code,+ concept_str=input_item_concept_code,content=item,name=input_item_stuff_name,)@@ -119,7 +119,7 @@ class PipeBatch(PipeController):list_content: ListContent[StuffContent] = ListContent(items=output_items)output_stuff = StuffFactory.make_stuff(code=output_stuff_code,- concept_code=self.output_concept_code,+ concept_str=self.output_concept_code,content=list_content,name=output_name,)diff --git a/pipelex/pipe_operators/pipe_func.py b/pipelex/pipe_operators/pipe_func.pyindex 69605a689..e539398b3 100644--- a/pipelex/pipe_operators/pipe_func.py+++ b/pipelex/pipe_operators/pipe_func.py@@ -48,7 +48,7 @@ class PipeFunc(PipeOperator):output_stuff = StuffFactory.make_stuff(name=output_name,- concept_code=self.output_concept_code,+ concept_str=self.output_concept_code,content=the_content,)diff --git a/pipelex/pipe_operators/pipe_img_gen.py b/pipelex/pipe_operators/pipe_img_gen.pyindex c701be73b..0c6685252 100644--- a/pipelex/pipe_operators/pipe_img_gen.py+++ b/pipelex/pipe_operators/pipe_img_gen.py@@ -260,7 +260,7 @@ class PipeImgGen(PipeOperator):output_stuff = StuffFactory.make_stuff(name=output_name,- concept_code=self.output_concept_code,+ concept_str=self.output_concept_code,content=the_content,)diff --git a/pipelex/pipe_operators/pipe_llm.py b/pipelex/pipe_operators/pipe_llm.pyindex 62b510e67..756562148 100644--- a/pipelex/pipe_operators/pipe_llm.py+++ b/pipelex/pipe_operators/pipe_llm.py@@ -13,7 +13,7 @@ from pipelex.cogt.llm.llm_models.llm_setting import LLMSettingfrom pipelex.cogt.llm.llm_prompt import LLMPromptfrom pipelex.cogt.llm.llm_prompt_factory_abstract import LLMPromptFactoryAbstractfrom pipelex.config import StaticValidationReaction, get_config-from pipelex.core.concept_factory import ConceptFactory+from pipelex.core.concept_code_factory import ConceptCodeFactoryfrom pipelex.core.concept_native import NativeConcept, NativeConceptClassfrom pipelex.core.domain import Domain, SpecialDomainfrom pipelex.core.pipe_input_spec import PipeInputSpec@@ -122,13 +122,19 @@ class PipeLLM(PipeOperator):input_concept_class = class_registry.get_required_subclass(name=input_concept_class_name, base_class=StuffContent)if issubclass(input_concept_class, StructuredContent):continue+ explanation = "The input provided for LLM Vision must be an image or a concept that refines image"+ if inadequate_concept := get_concept_provider().get_concept(concept_code=concept_code_of_declared_input):+ explanation += f",\nconcept = {inadequate_concept}"+ else:+ explanation += ",\nconcept not found"+inadequate_input_concept_error = StaticValidationError(error_type=StaticValidationErrorType.INADEQUATE_INPUT_CONCEPT,domain_code=self.domain,pipe_code=self.code,variable_names=[required_variable_name],provided_concept_code=concept_code_of_declared_input,- explanation="The input provided for LLM Vision must be an image or a concept that refines image",+ explanation=explanation,)match reactions.get(StaticValidationErrorType.INADEQUATE_INPUT_CONCEPT, default_reaction):case StaticValidationReaction.IGNORE:@@ -212,7 +218,7 @@ class PipeLLM(PipeOperator):content_generator = content_generator or get_content_generator()# interpret / unwrap the argumentslog.debug(f"PipeLLM pipe_code = {self.code}")- if self.output_concept_code == ConceptFactory.make_concept_code(+ if self.output_concept_code == ConceptCodeFactory.make_concept_code(SpecialDomain.NATIVE,NativeConcept.DYNAMIC.code,):diff --git a/pipelex/pipe_operators/pipe_llm_prompt.py b/pipelex/pipe_operators/pipe_llm_prompt.pyindex bcbd228e1..c3066a416 100644--- a/pipelex/pipe_operators/pipe_llm_prompt.py+++ b/pipelex/pipe_operators/pipe_llm_prompt.py@@ -202,7 +202,7 @@ class PipeLLMPrompt(PipeOperator):output_stuff = StuffFactory.make_stuff(name=output_name,- concept_code=self.output_concept_code,+ concept_str=self.output_concept_code,content=llm_prompt,)diff --git a/pipelex/pipe_operators/pipe_ocr.py b/pipelex/pipe_operators/pipe_ocr.pyindex 14ce618b5..bbb98d0cd 100644--- a/pipelex/pipe_operators/pipe_ocr.py+++ b/pipelex/pipe_operators/pipe_ocr.py@@ -206,7 +206,7 @@ class PipeOcr(PipeOperator):output_stuff = StuffFactory.make_stuff(name=output_name,- concept_code=self.output_concept_code,+ concept_str=self.output_concept_code,content=content,)diff --git a/pipelex/reporting/reporting_manager.py b/pipelex/reporting/reporting_manager.pyindex 708a4f416..cf6370856 100644--- a/pipelex/reporting/reporting_manager.py+++ b/pipelex/reporting/reporting_manager.py@@ -84,7 +84,7 @@ class ReportingManager(ReportingProtocol):@overridedef report_inference_job(self, inference_job: InferenceJobAbstract):- log.info(f"Inference job '{inference_job.job_metadata.unit_job_id}' completed in {inference_job.job_metadata.duration:.2f} seconds")+ log.debug(f"Inference job '{inference_job.job_metadata.unit_job_id}' completed in {inference_job.job_metadata.duration:.2f} seconds")if not isinstance(inference_job, LLMJob):# ReportingManager does not support reporting for other types of inference jobs yet# TODO: add support for other types of inference jobsdiff --git a/pyproject.toml b/pyproject.tomlindex 62ba78892..571962dad 100644--- a/pyproject.toml+++ b/pyproject.toml@@ -47,8 +47,8 @@ dependencies = [[project.urls]Homepage = "https://pipelex.com"Repository = "https://github.com/Pipelex/pipelex"-Documentation = "https://github.com/Pipelex/pipelex/blob/main/doc/Documentation.md"-Changelog = "https://github.com/Pipelex/pipelex/blob/main/CHANGELOG.md"+Documentation = "https://docs.pipelex.com/"+Changelog = "https://docs.pipelex.com/changelog/"[project.optional-dependencies]anthropic = ["anthropic>=0.49.0"]@@ -58,6 +58,10 @@ google = ["google-auth-oauthlib>=1.2.1"]mistralai = ["mistralai==1.5.2"]compat = ["backports.strenum>=1.3.0 ; python_version < '3.11'"]dev = [+ "mkdocs==1.6.1",+ "mkdocs-glightbox==0.4.0",+ "mkdocs-material==9.6.14",+ "mkdocs-meta-manager==1.1.0","boto3-stubs>=1.35.24","mypy>=1.11.2","pyright==1.1.398",diff --git a/tests/core/test_concept_pydantic_validation.py b/tests/core/test_concept_pydantic_validation.pynew file mode 100644index 000000000..e170387a3--- /dev/null+++ b/tests/core/test_concept_pydantic_validation.py@@ -0,0 +1,197 @@+import pytest++from pipelex.core.concept import Concept+from pipelex.core.concept_native import NativeConcept+from pipelex.exceptions import ConceptCodeError, ConceptDomainError+++class TestConceptPydanticFieldValidation:+ """Test Concept validation through Pydantic field validation by instantiating the class."""++ def test_concept_creation_with_valid_refines(self):+ """Test successful concept creation with valid refines list."""+ concept = Concept(+ code="test_domain.TestConcept",+ domain="test_domain",+ structure_class_name="TextContent",+ definition="A test concept",+ refines=["domain1.Concept1", "domain2.Concept2"],+ )+ assert concept.refines == ["domain1.Concept1", "domain2.Concept2"]++ def test_concept_creation_with_empty_refines(self):+ """Test successful concept creation with empty refines list."""+ concept = Concept(+ code="test_domain.TestConcept", domain="test_domain", structure_class_name="TextContent", definition="A test concept", refines=[]+ )+ assert concept.refines == []++ def test_concept_creation_with_default_refines(self):+ """Test successful concept creation without specifying refines (should default to empty list)."""+ concept = Concept(code="test_domain.TestConcept", domain="test_domain", structure_class_name="TextContent", definition="A test concept")+ assert concept.refines == []++ def test_concept_creation_with_native_concept_refines(self):+ """Test successful concept creation with NativeConcept string values in refines."""+ concept = Concept(+ code="test_domain.TestConcept",+ domain="test_domain",+ structure_class_name="TextContent",+ definition="A test concept",+ refines=[+ NativeConcept.TEXT.value,+ NativeConcept.IMAGE.value,+ NativeConcept.PDF.value,+ ],+ )+ # NativeConcept strings should be converted to full codes+ assert concept.refines == [+ NativeConcept.TEXT.code,+ NativeConcept.IMAGE.code,+ NativeConcept.PDF.code,+ ]++ def test_concept_creation_with_mixed_refines(self):+ """Test successful concept creation with mix of NativeConcept and domain.concept refines."""+ concept = Concept(+ code="test_domain.TestConcept",+ domain="test_domain",+ structure_class_name="TextContent",+ definition="A test concept",+ refines=[+ "valid_domain.ValidConcept",+ NativeConcept.TEXT.value,+ "another_domain.AnotherConcept",+ NativeConcept.IMAGE.value,+ ],+ )+ expected_refines = [+ "valid_domain.ValidConcept",+ NativeConcept.TEXT.code,+ "another_domain.AnotherConcept",+ NativeConcept.IMAGE.code,+ ]+ assert concept.refines == expected_refines++ def test_concept_creation_with_only_native_concept_refines(self):+ """Test successful concept creation with only NativeConcept values in refines."""+ concept = Concept(+ code="test_domain.TestConcept",+ domain="test_domain",+ structure_class_name="TextContent",+ definition="A test concept",+ refines=[+ NativeConcept.DYNAMIC.value,+ NativeConcept.TEXT.value,+ NativeConcept.NUMBER.value,+ ],+ )+ assert concept.refines == [+ NativeConcept.DYNAMIC.code,+ NativeConcept.TEXT.code,+ NativeConcept.NUMBER.code,+ ]++ def test_concept_creation_with_refines_missing_dot(self):+ """Test concept creation fails when refines contain invalid codes missing dot."""+ with pytest.raises(ConceptCodeError) as exc_info:+ Concept(+ code="test_domain.TestConcept",+ domain="test_domain",+ structure_class_name="TextContent",+ definition="A test concept",+ refines=["invalidConcept"],+ )++ assert "Each refine code must contain a single dot" in str(exc_info.value)++ def test_concept_creation_with_refines_invalid_domain(self):+ """Test concept creation fails when refines contain invalid domain format."""+ with pytest.raises(ConceptDomainError) as exc_info:+ Concept(+ code="test_domain.TestConcept",+ domain="test_domain",+ structure_class_name="TextContent",+ definition="A test concept",+ refines=["InvalidDomain.Concept"],+ )++ assert "Domain must be snake_case" in str(exc_info.value)++ def test_concept_creation_with_refines_invalid_concept_code(self):+ """Test concept creation fails when refines contain invalid concept code format."""+ with pytest.raises(ConceptCodeError) as exc_info:+ Concept(+ code="test_domain.TestConcept",+ domain="test_domain",+ structure_class_name="TextContent",+ definition="A test concept",+ refines=["valid_domain.invalid_concept"],+ )++ assert "Code must be PascalCase" in str(exc_info.value)++ def test_concept_creation_with_multiple_refines_errors(self):+ """Test concept creation fails when multiple refines are invalid."""+ with pytest.raises(ConceptCodeError) as exc_info:+ Concept(+ code="test_domain.TestConcept",+ domain="test_domain",+ structure_class_name="TextContent",+ definition="A test concept",+ refines=[+ "invalidConcept",+ "InvalidDomain.Concept",+ "valid_domain.invalid_concept",+ ],+ )++ # The first error should be about the missing dot (validation stops at first error)+ assert "Each refine code must contain a single dot" in str(exc_info.value)++ def test_concept_creation_with_mixed_valid_invalid_refines(self):+ """Test concept creation fails even when some refines are valid but others are invalid."""+ with pytest.raises(ConceptCodeError) as exc_info:+ Concept(+ code="test_domain.TestConcept",+ domain="test_domain",+ structure_class_name="TextContent",+ definition="A test concept",+ refines=[+ "valid_domain.ValidConcept",+ "invalidConcept",+ ],+ )++ assert "Each refine code must contain a single dot" in str(exc_info.value)++ def test_concept_creation_with_refines_multiple_dots(self):+ """Test concept creation fails when refines contain codes with multiple dots."""+ with pytest.raises(ConceptCodeError) as exc_info:+ Concept(+ code="test_domain.TestConcept",+ domain="test_domain",+ structure_class_name="TextContent",+ definition="A test concept",+ refines=["domain.concept.subconcept"],+ )++ assert "Each refine code must contain a single dot" in str(exc_info.value)++ def test_concept_creation_with_valid_complex_refines(self):+ """Test successful concept creation with more complex valid refines."""+ concept = Concept(+ code="my_domain.MyTestConcept",+ domain="my_domain",+ structure_class_name="TextContent",+ definition="A complex test concept",+ refines=[+ "base_domain.BaseConcept",+ "another_domain.AnotherConcept123",+ "third_domain.ThirdConcept",+ "snake_case_domain.PascalCaseConcept",+ ],+ )+ assert len(concept.refines) == 4+ assert "base_domain.BaseConcept" in concept.refines+ assert "snake_case_domain.PascalCaseConcept" in concept.refinesdiff --git a/tests/core/test_concept_refines.py b/tests/core/test_concept_refines.pynew file mode 100644index 000000000..c5b3cc696--- /dev/null+++ b/tests/core/test_concept_refines.py@@ -0,0 +1,94 @@+import pytest++from pipelex.core.concept import Concept+from pipelex.core.concept_native import NativeConcept+from pipelex.exceptions import ConceptCodeError, ConceptDomainError, ConceptError+++class TestConceptRefinesValidationFunction:+ def test_validate_refines_success(self):+ # Test valid refines list+ valid_refines = ["domain1.Concept1", "domain2.Concept2", NativeConcept.TEXT.value]+ result = Concept.validate_refines(valid_refines)+ # NativeConcept.TEXT.value ("Text") should be converted to its full code "native.Text"+ expected = ["domain1.Concept1", "domain2.Concept2", NativeConcept.TEXT.code]+ assert result == expected++ def test_validate_refines_empty_list(self):+ # Test empty refines list+ result = Concept.validate_refines([])+ assert result == []++ def test_validate_refines_with_native_concept_strings(self):+ # Test refines with NativeConcept string values+ valid_refines = [+ "domain1.Concept1",+ NativeConcept.TEXT.value,+ NativeConcept.IMAGE.value,+ NativeConcept.PDF.value,+ ]+ result = Concept.validate_refines(valid_refines)+ # The NativeConcept strings should be converted to full codes+ expected = [+ "domain1.Concept1",+ NativeConcept.TEXT.code,+ NativeConcept.IMAGE.code,+ NativeConcept.PDF.code,+ ]+ assert result == expected++ def test_validate_refines_with_only_native_concepts(self):+ # Test refines with only NativeConcept values+ valid_refines = [+ NativeConcept.TEXT.value,+ NativeConcept.IMAGE.value,+ NativeConcept.DYNAMIC.value,+ ]+ result = Concept.validate_refines(valid_refines)+ expected = [+ NativeConcept.TEXT.code,+ NativeConcept.IMAGE.code,+ NativeConcept.DYNAMIC.code,+ ]+ assert result == expected++ def test_validate_refines_with_mixed_native_and_domain_concepts(self):+ # Test refines with mix of NativeConcept values and domain.concept codes+ valid_refines = [+ "my_domain.MyClass",+ NativeConcept.TEXT.value,+ "another_domain.AnotherClass",+ NativeConcept.IMAGE.value,+ ]+ result = Concept.validate_refines(valid_refines)+ expected = [+ "my_domain.MyClass",+ NativeConcept.TEXT.code,+ "another_domain.AnotherClass",+ NativeConcept.IMAGE.code,+ ]+ assert result == expected++ def test_validate_refines_missing_dot(self):+ # Test refines with missing dot+ with pytest.raises(ConceptCodeError) as exc_info:+ Concept.validate_refines(["invalidConcept"])+ assert "Each refine code must contain a single dot" in str(exc_info.value)++ def test_validate_refines_invalid_domain(self):+ # Test refines with invalid domain format+ with pytest.raises(ConceptDomainError) as exc_info:+ Concept.validate_refines(["InvalidDomain.Concept"])+ assert "Domain must be snake_case" in str(exc_info.value)++ def test_validate_refines_invalid_concept(self):+ # Test refines with invalid concept format+ with pytest.raises(ConceptCodeError) as exc_info:+ Concept.validate_refines(["valid_domain.invalid_concept"])+ assert "Code must be PascalCase" in str(exc_info.value)++ def test_validate_refines_multiple_dots(self):+ # Test refines with multiple dots+ with pytest.raises(ConceptError) as exc_info:+ Concept.validate_refines(["domain.concept.subconcept"])+ assert "Each refine code must contain a single dot" in str(exc_info.value)diff --git a/tests/pipelex/pipelex_asynch/test_client.py b/tests/pipelex/pipelex_asynch/test_client.pyindex d6a29aa15..76fd4355a 100644--- a/tests/pipelex/pipelex_asynch/test_client.py+++ b/tests/pipelex/pipelex_asynch/test_client.py@@ -30,7 +30,7 @@ class TestPipelexApiClient:pipe_code="retrieve_excerpts",memory=[StuffFactory.make_stuff(- concept_code="native.Text",+ concept_str="Text",name="text",content=TextContent(text="""@@ -75,7 +75,7 @@ class TestPipelexApiClient:),),StuffFactory.make_stuff(- concept_code="answer.Question",+ concept_str="answer.Question",name="question",content=TextContent(text="Aerodynamic features?"),),diff --git a/tests/pipelex/pipelex_asynch/test_pipe_batch.py b/tests/pipelex/pipelex_asynch/test_pipe_batch.pyindex be772c723..a826eec20 100644--- a/tests/pipelex/pipelex_asynch/test_pipe_batch.py+++ b/tests/pipelex/pipelex_asynch/test_pipe_batch.py@@ -21,7 +21,7 @@ class TestPipeBatch:):# Create Stuff objectsinvoice_list_stuff = StuffFactory.make_stuff(- concept_code="test_pipe_batch.TestPipeBatchItem",+ concept_str="test_pipe_batch.TestPipeBatchItem",content=ListContent(items=[TextContent(text="data_1"),diff --git a/tests/pipelex/pipelex_asynch/test_pipe_ocr.py b/tests/pipelex/pipelex_asynch/test_pipe_ocr.pyindex e30351675..7da9bfae7 100644--- a/tests/pipelex/pipelex_asynch/test_pipe_ocr.py+++ b/tests/pipelex/pipelex_asynch/test_pipe_ocr.py@@ -30,7 +30,7 @@ class TestPipeOCR:domain="generic",inputs=PipeInputSpec(root={- "page_scan": "native.Image",+ "page_scan": "Image",}),should_include_images=True,@@ -42,7 +42,7 @@ class TestPipeOCR:pipe_run_params=PipeRunParamsFactory.make_run_params(pipe_run_mode=pipe_run_mode),working_memory=WorkingMemoryFactory.make_from_image(image_url=image_url,- concept_code="ocr.PageScan",+ concept_str="ocr.PageScan",name="page_scan",),)@@ -64,7 +64,7 @@ class TestPipeOCR:domain="generic",inputs=PipeInputSpec(root={- "pdf": "native.PDF",+ "pdf": "PDF",}),should_include_images=True,@@ -76,7 +76,7 @@ class TestPipeOCR:pipe_run_params=PipeRunParamsFactory.make_run_params(pipe_run_mode=pipe_run_mode),working_memory=WorkingMemoryFactory.make_from_pdf(pdf_url=pdf_url,- concept_code=NativeConcept.PDF.code,+ concept_str=NativeConcept.PDF.code,name="pdf",),)diff --git a/tests/pipelex/test_data.py b/tests/pipelex/test_data.pyindex 32f04e5ba..ab53765fe 100644--- a/tests/pipelex/test_data.py+++ b/tests/pipelex/test_data.py@@ -46,25 +46,25 @@ class PipeTestCases:# Create simple Stuff objectsSIMPLE_STUFF_TEXT = StuffFactory.make_stuff(name="text",- concept_code="native.Text",+ concept_str="Text",content=TextContent(text="Describe a t-shirt in 2 sentences"),pipelex_session_id="unit_test",)SIMPLE_STUFF_IMAGE = StuffFactory.make_stuff(name="image",- concept_code="native.Image",+ concept_str="Image",content=ImageContent(url=URL_IMG_FASHION_PHOTO_1),pipelex_session_id="unit_test",)SIMPLE_STUFF_PDF = StuffFactory.make_stuff(name="pdf",- concept_code="native.PDF",+ concept_str="PDF",content=PDFContent(url=PDFTestCases.DOCUMENT_URLS[0]),pipelex_session_id="unit_test",)COMPLEX_STUFF = StuffFactory.make_stuff(name="complex",- concept_code="tests.Complex",+ concept_str="tests.Complex",content=ListContent(items=[TextContent(text="The quick brown fox jumps over the lazy dog"),@@ -76,7 +76,7 @@ class PipeTestCases:STUFF_CONTENT_WITH_IMAGE_ATTRIBUTE_1 = SomeContentWithImageAttribute(image_attribute=ImageContent(url=URL_IMG_FASHION_PHOTO_1))STUFF_WITH_IMAGE_ATTRIBUTE = StuffFactory.make_stuff(- concept_code="native.Image",+ concept_str="Image",content=STUFF_CONTENT_WITH_IMAGE_ATTRIBUTE_1,name="stuff_with_image",pipelex_session_id="unit_test",@@ -86,7 +86,7 @@ class PipeTestCases:sub_object=STUFF_CONTENT_WITH_IMAGE_ATTRIBUTE_1,)STUFF_WITH_IMAGE_ATTRIBUTE_IN_SUB_OBJECT = StuffFactory.make_stuff(- concept_code="native.Image",+ concept_str="Image",content=STUFF_CONTENT_WITH_IMAGE_ATTRIBUTE_IN_SUB_OBJECT,name="stuff_with_image_in_sub_object",pipelex_session_id="unit_test",@@ -143,7 +143,7 @@ class PipeTestCases:("batch_test",StuffFactory.make_stuff(- concept_code="flows.Color",+ concept_str="flows.Color",name="colors",content=ListContent(items=[@@ -170,13 +170,6 @@ class PipeTestCases:"extract_page_contents_from_pdf",),]- SIMPLE_PIPE_RUN_FROM_STR: ClassVar[List[Tuple[str, str, str]]] = [ # pipe_code, input_concept_code, str_value- (- "extract_colors",- "native.Text",- USER_TEXT_COLORS,- ),- ]FAILURE_PIPES: ClassVar[List[Tuple[str, Type[Exception], str]]] = [("infinite_loop_1",@@ -290,7 +283,7 @@ Format filter html:]ANY_OBJECT: ClassVar[List[Tuple[str, Any]]] = [("cherry", PipeTestCases.SIMPLE_STUFF_TEXT),- ("complex", PipeTestCases.COMPLEX_STUFF),+ # ("complex", PipeTestCases.COMPLEX_STUFF),]diff --git a/tests/test_pipelines/__init__.py b/tests/test_pipelines/__init__.pyindex 191fc7df8..e69de29bb 100644--- a/tests/test_pipelines/__init__.py+++ b/tests/test_pipelines/__init__.py@@ -1,7 +0,0 @@-from pipelex.tools.log.log import log-from pipelex.tools.misc.pretty import pretty_print--__all__ = [- "log",- "pretty_print",-]diff --git a/tests/test_pipelines/failure_modes.toml b/tests/test_pipelines/failure_modes.tomlindex a46283c99..3f31bf459 100644--- a/tests/test_pipelines/failure_modes.toml+++ b/tests/test_pipelines/failure_modes.toml@@ -6,7 +6,7 @@ domain = "failure_modes"[pipe.dummy]PipeLLM = "This pipe is a dummy pipe"-output = "native.Text"+output = "Text"llm = { llm_handle = "gpt-4o-mini", temperature = 1, max_tokens = 50 }prompt_template = """This is a dummy prompt, do whatever you want. Something funny but not a joke, ok?@@ -14,7 +14,7 @@ This is a dummy prompt, do whatever you want. Something funny but not a joke, ok[pipe.infinite_loop_1]PipeSequence = "This pipe will cause an infinite loop"-output = "native.Text"+output = "Text"steps = [{ pipe = "dummy", result = "dummy_result" },{ pipe = "infinite_loop_1", result = "disaster" },diff --git a/tests/test_pipelines/misc_tests/__init__.py b/tests/test_pipelines/misc_tests/__init__.pyindex 191fc7df8..e69de29bb 100644--- a/tests/test_pipelines/misc_tests/__init__.py+++ b/tests/test_pipelines/misc_tests/__init__.py@@ -1,7 +0,0 @@-from pipelex.tools.log.log import log-from pipelex.tools.misc.pretty import pretty_print--__all__ = [- "log",- "pretty_print",-]diff --git a/tests/test_pipelines/misc_tests/multiplicity.toml b/tests/test_pipelines/misc_tests/multiplicity.tomlindex 2b4246bc4..dfffeeb0b 100644--- a/tests/test_pipelines/misc_tests/multiplicity.toml+++ b/tests/test_pipelines/misc_tests/multiplicity.toml@@ -49,7 +49,7 @@ Keep it short and concise, just one sentence.[pipe.imagine_nature_product_list]PipeSequence = "Imagine nature products of different colors"-output = "native.Image"+output = "Image"steps = [{ pipe = "choose_colors", result = "color", nb_output = 3 },{ pipe = "imagine_nature_product", result = "product_of_nature" },@@ -59,7 +59,7 @@ steps = [[pipe.imagine_nature_scene_of_original_power_rangers_colors]PipeSequence = "Imagine nature scenes of Power Rangers colors"-output = "native.Image"+output = "Image"steps = [{ pipe = "original_power_ranger_colors", result = "color" },{ pipe = "imagine_nature_product", result = "product_of_nature" },@@ -70,7 +70,7 @@ steps = [[pipe.imagine_nature_scene_of_alltime_power_rangers_colors]PipeSequence = "Imagine nature scenes of Power Rangers colors"-output = "native.Image"+output = "Image"steps = [{ pipe = "alltime_power_ranger_colors", result = "color", multiple_output = true },{ pipe = "imagine_nature_product", result = "product_of_nature" },diff --git a/tests/test_pipelines/misc_tests/test_errors.toml b/tests/test_pipelines/misc_tests/test_errors.tomlindex 1bb013af2..216df442d 100644--- a/tests/test_pipelines/misc_tests/test_errors.toml+++ b/tests/test_pipelines/misc_tests/test_errors.toml@@ -5,32 +5,32 @@ definition = "This library is intended for testing errors"# [pipe.jinja2_syntax_error]# PipeLLM = "Jinja2 syntax error (on purpose, for testing): bad curly braces"-# inputs = { text = "native.Text" }-# output = "native.Text"+# inputs = { text = "Text" }+# output = "Text"# prompt_template = """# {{ a123|tag }# """# [pipe.jinja2_dummy_filter1]# PipeLLM = "Jinja2 undefined filter (on purpose, for testing)"-# inputs = { text = "native.Text" }-# output = "native.Text"+# inputs = { text = "Text" }+# output = "Text"# prompt_template = """# {{ foobar|dummy_filter1 }}# """# [pipe.jinja2_dummy_filter2]# PipeLLM = "Jinja2 undefined filter (on purpose, for testing)"-# inputs = { text = "native.Text" }-# output = "native.Text"+# inputs = { text = "Text" }+# output = "Text"# prompt_template = """# {{ foobar|dummy_filter2() }}# """# [pipe.jinja2_dummy_filter3]# PipeLLM = "Jinja2 undefined filter (on purpose, for testing)"-# inputs = { text = "native.Text" }-# output = "native.Text"+# inputs = { text = "Text" }+# output = "Text"# prompt_template = """# {{ foobar|dummy_filter3("param") }}# """@@ -38,8 +38,8 @@ definition = "This library is intended for testing errors"# [pipe.jinja2_dummy_filter4]# PipeLLM = "Jinja2 undefined filter (on purpose, for testing)"-# inputs = { text = "native.Text" }-# output = "native.Text"+# inputs = { text = "Text" }+# output = "Text"# prompt_template = """# {{ foobar|dummy_filter4(param1, param2) }}# """diff --git a/tests/test_pipelines/misc_tests/test_jinja2.toml b/tests/test_pipelines/misc_tests/test_jinja2.tomlindex 358d348db..0c3bca9c9 100644--- a/tests/test_pipelines/misc_tests/test_jinja2.toml+++ b/tests/test_pipelines/misc_tests/test_jinja2.toml@@ -6,8 +6,8 @@ definition = "This library is intended for testing Jinja2"[pipe.jinja2_test_1]PipeJinja2 = "Jinja2 test 1"-inputs = { text = "native.Text" }-output = "native.Text"+inputs = { text = "Text" }+output = "Text"jinja2 = """This is a simple test prompt:@text@@ -17,6 +17,6 @@ template_category = "llm_prompt"# [pipe.jinja2_test_2]# PipeJinja2 = "Jinja2 test 2"-# input = "native.Text"-# output = "native.Text"+# inputs = { text = "Text" }+# output = "Text"# jinja2_name = "non-existing-name"diff --git a/tests/test_pipelines/misc_tests/tests.toml b/tests/test_pipelines/misc_tests/tests.tomlindex f4096f04e..c7e5083ae 100644--- a/tests/test_pipelines/misc_tests/tests.toml+++ b/tests/test_pipelines/misc_tests/tests.toml@@ -10,16 +10,16 @@ Complex = "A complex object"[pipe.simple_llm_test_from_text]PipeLLM = "Simple LLM test from text"-inputs = { text = "native.Text" }-output = "native.Text"+inputs = { text = "Text" }+output = "Text"prompt_template = """This is a simple test prompt:@text"""[pipe.simple_llm_test_from_image]PipeLLM = "Simple LLM test from image"-inputs = { image = "native.Image" }-output = "native.Text"+inputs = { image = "Image" }+output = "Text"prompt_template = """Describe the using rap lyrics, including puns and references to the image."""@@ -37,8 +37,8 @@ multiple_output = true[pipe.create_synopsis]PipeLLM = "Create Synopsis"-inputs = { text = "native.Text" }-output = "native.Text"+inputs = { text = "Text" }+output = "Text"llm = "llm_for_creative_writing"prompt_template = """Create a short synopsis for a story about:@@ -50,7 +50,7 @@ Be concise: 3 sentences max.[pipe.test_no_input]PipeLLM = "No Input"-output = "native.Text"+output = "Text"llm = "llm_for_testing_gen_text"prompt_template = """Explain that this is a test prompt which took no input from the user.@@ -58,7 +58,7 @@ Explain that this is a test prompt which took no input from the user.[pipe.test_no_input_that_could_be_long]PipeLLM = "No Input but generates a text that could be long"-output = "native.Text"+output = "Text"llm = { llm_handle = "gpt-4o-mini", temperature = 0.5, max_tokens = 100 }prompt_template = """Tell me a short story about a red baloon.@@ -67,5 +67,5 @@ Tell me a short story about a red baloon.[pipe.generate_3_images]PipeImgGen = "Generate 3 images"inputs = { prompt = "images.ImgGenPrompt" }-output = "native.Image"+output = "Image"nb_output = 3diff --git a/tests/test_pipelines/tricky_questions.toml b/tests/test_pipelines/tricky_questions.tomlindex a44246dfc..a59a93dba 100644--- a/tests/test_pipelines/tricky_questions.toml+++ b/tests/test_pipelines/tricky_questions.toml@@ -9,7 +9,7 @@ ThoughtfulAnswerConclusion = "Conclusion of a thoughtful answer"[concept.ThoughtfulAnswer]Concept = "A thoughtful answer to a question"structure = "ThoughtfulAnswer"-refines = ["AnswerToAQuestion"]+refines = "AnswerToAQuestion"[pipe]diff --git a/uv.lock b/uv.lockindex 29e9b6a9c..9d33aacbc 100644--- a/uv.lock+++ b/uv.lock@@ -63,7 +63,7 @@ wheels = [[[package]]name = "aiohttp"-version = "3.12.12"+version = "3.12.13"source = { registry = "https://pypi.org/simple" }dependencies = [{ name = "aiohappyeyeballs" },@@ -75,76 +75,76 @@ dependencies = [{ name = "propcache" },{ name = "yarl" },]-sdist = { url = "https://files.pythonhosted.org/packages/f2/84/ea27e6ad14747d8c51afe201fb88a5c8282b6278256d30a6f71f730add88/aiohttp-3.12.12.tar.gz", hash = "sha256:05875595d2483d96cb61fa9f64e75262d7ac6251a7e3c811d8e26f7d721760bd", size = 7818643, upload-time = "2025-06-10T05:22:00.247Z" }-wheels = [- { url = "https://files.pythonhosted.org/packages/b4/d9/cfde93b9cb75253c716b8b1c773565209e3d4dd0772dd3ce3a2adcaa4639/aiohttp-3.12.12-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:6f25e9d274d6abbb15254f76f100c3984d6b9ad6e66263cc60a465dd5c7e48f5", size = 702071, upload-time = "2025-06-10T05:18:23.986Z" },- { url = "https://files.pythonhosted.org/packages/ee/b0/46e38b8bc0bc645317deec32612af922ad9bafd85a1df255a67c2f2305f6/aiohttp-3.12.12-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:b8ec3c1a1c13d24941b5b913607e57b9364e4c0ea69d5363181467492c4b2ba6", size = 478436, upload-time = "2025-06-10T05:18:28.411Z" },- { url = "https://files.pythonhosted.org/packages/8f/47/9c83db7f02ca71eb99a707ee13657fc24ba703b9babc59000c1f58ac1198/aiohttp-3.12.12-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:81ef2f9253c327c211cb7b06ea2edd90e637cf21c347b894d540466b8d304e08", size = 466213, upload-time = "2025-06-10T05:18:30.706Z" },- { url = "https://files.pythonhosted.org/packages/31/fe/4690c112e269e06c9182c32eeb43f3a95c4f203fdb095502717327993b80/aiohttp-3.12.12-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:28ded835c3663fd41c9ad44685811b11e34e6ac9a7516a30bfce13f6abba4496", size = 1648258, upload-time = "2025-06-10T05:18:32.498Z" },- { url = "https://files.pythonhosted.org/packages/c8/1f/dacca6c7bbe69c77d8535d7a672478803e7078cc20fd9993fe09aa5be880/aiohttp-3.12.12-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:a4b78ccf254fc10605b263996949a94ca3f50e4f9100e05137d6583e266b711e", size = 1622316, upload-time = "2025-06-10T05:18:34.357Z" },- { url = "https://files.pythonhosted.org/packages/ff/65/5ef47708f70524fcdecda735e0aea06e0feb7b8679e976e9bd1e7900f4c0/aiohttp-3.12.12-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4f4a5af90d5232c41bb857568fe7d11ed84408653ec9da1ff999cc30258b9bd1", size = 1694723, upload-time = "2025-06-10T05:18:36.858Z" },- { url = "https://files.pythonhosted.org/packages/18/62/ab32bfa59f61292e4096c383316863e10001eec30e5b4b314856ed7156e2/aiohttp-3.12.12-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ffa5205c2f53f1120e93fdf2eca41b0f6344db131bc421246ee82c1e1038a14a", size = 1737037, upload-time = "2025-06-10T05:18:39.663Z" },- { url = "https://files.pythonhosted.org/packages/c1/b9/8b8f793081311e4f63aea63003a519064048e406c627c0454d6ed09dbc99/aiohttp-3.12.12-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f68301660f0d7a3eddfb84f959f78a8f9db98c76a49b5235508fa16edaad0f7c", size = 1641701, upload-time = "2025-06-10T05:18:41.666Z" },- { url = "https://files.pythonhosted.org/packages/1a/5c/72f510d42d626463b526345dcb8d14b390de89a9ba27a4717b518460bcd4/aiohttp-3.12.12-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:db874d3b0c92fdbb553751af9d2733b378c25cc83cd9dfba87f12fafd2dc9cd5", size = 1581824, upload-time = "2025-06-10T05:18:44.136Z" },- { url = "https://files.pythonhosted.org/packages/61/6f/9378c9e1543d1c800ca040e21cd333b8f923ed051ae82b5a49ad96a6ac71/aiohttp-3.12.12-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:5e53cf9c201b45838a2d07b1f2d5f7fec9666db7979240002ce64f9b8a1e0cf2", size = 1625674, upload-time = "2025-06-10T05:18:46.716Z" },- { url = "https://files.pythonhosted.org/packages/bb/85/4eef9bd52b497a405c88469cc099f4d15d33b149b5746ca4ef8ec6ab6388/aiohttp-3.12.12-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:8687cc5f32b4e328c233acd387d09a1b477007896b2f03c1c823a0fd05f63883", size = 1636460, upload-time = "2025-06-10T05:18:49.305Z" },- { url = "https://files.pythonhosted.org/packages/56/59/d8e954830b375fd658843cf7d88d27ca5e38dd5fcbfe62db3d1ba415d0fe/aiohttp-3.12.12-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:5ee537ad29de716a3d8dc46c609908de0c25ffeebf93cd94a03d64cdc07d66d0", size = 1611912, upload-time = "2025-06-10T05:18:51.694Z" },- { url = "https://files.pythonhosted.org/packages/c3/5d/d0096a02f0515a38dff67db42d966273a12d17fc895e91466bfb4ab3875e/aiohttp-3.12.12-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:411f821be5af6af11dc5bed6c6c1dc6b6b25b91737d968ec2756f9baa75e5f9b", size = 1691498, upload-time = "2025-06-10T05:18:54.36Z" },- { url = "https://files.pythonhosted.org/packages/87/8d/d3a02397a6345c06623ae4648e2aef18fced858510b4a89d7262cfa4c683/aiohttp-3.12.12-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:f90319d94cf5f9786773237f24bd235a7b5959089f1af8ec1154580a3434b503", size = 1714737, upload-time = "2025-06-10T05:18:56.806Z" },- { url = "https://files.pythonhosted.org/packages/a9/40/b81000bf07c96db878703ea3dc561393d82441597729910459a8e06acc9a/aiohttp-3.12.12-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:73b148e606f34e9d513c451fd65efe1091772659ca5703338a396a99f60108ff", size = 1643078, upload-time = "2025-06-10T05:18:59.33Z" },- { url = "https://files.pythonhosted.org/packages/41/e5/31830642ce2c6d3dba74ed8a94933213df5e1651c1e8b4efc81cc88105ab/aiohttp-3.12.12-cp310-cp310-win32.whl", hash = "sha256:d40e7bfd577fdc8a92b72f35dfbdd3ec90f1bc8a72a42037fefe34d4eca2d4a1", size = 427517, upload-time = "2025-06-10T05:19:01.535Z" },- { url = "https://files.pythonhosted.org/packages/55/9d/a4e5379d44679e5f8d7d7ebecb0dae8cafab95176c4e753da6bc4b4aebb5/aiohttp-3.12.12-cp310-cp310-win_amd64.whl", hash = "sha256:65c7804a2343893d6dea9fce69811aea0a9ac47f68312cf2e3ee1668cd9a387f", size = 450725, upload-time = "2025-06-10T05:19:03.874Z" },- { url = "https://files.pythonhosted.org/packages/47/1f/b1b66e05dc3066a9ba7862d50e2e95b3871db82ccf9652568845f353eeba/aiohttp-3.12.12-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:38823fe0d8bc059b3eaedb263fe427d887c7032e72b4ef92c472953285f0e658", size = 709385, upload-time = "2025-06-10T05:19:05.763Z" },- { url = "https://files.pythonhosted.org/packages/43/e6/3230e42af16438b450b1e193c537fd3d2d31771dafda3c2105a8d11af707/aiohttp-3.12.12-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:10237f2c34711215d04ed21da63852ce023608299554080a45c576215d9df81c", size = 481660, upload-time = "2025-06-10T05:19:08.332Z" },- { url = "https://files.pythonhosted.org/packages/06/ba/cfa91fe5cc262535e1175b1522d8fcc09f9d6ad18b85241f4ee3be1d780f/aiohttp-3.12.12-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:563ec477c0dc6d56fc7f943a3475b5acdb399c7686c30f5a98ada24bb7562c7a", size = 469924, upload-time = "2025-06-10T05:19:10.342Z" },- { url = "https://files.pythonhosted.org/packages/9a/f0/5c706cfddd4769b55c0cda466aa6034412d39e416f0b30dda81c4a24616f/aiohttp-3.12.12-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f3d05c46a61aca7c47df74afff818bc06a251ab95d95ff80b53665edfe1e0bdf", size = 1740116, upload-time = "2025-06-10T05:19:12.783Z" },- { url = "https://files.pythonhosted.org/packages/4d/9f/04dba2e1c8bee53c3c623d11a1f947c9e2712500f734dc0dfd06daad32ec/aiohttp-3.12.12-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:277c882916759b4a6b6dc7e2ceb124aad071b3c6456487808d9ab13e1b448d57", size = 1688784, upload-time = "2025-06-10T05:19:15.36Z" },- { url = "https://files.pythonhosted.org/packages/df/24/19d6d4c41fbf8304fe7c111fcc701e0aa5a2232ee3ac16272677a11f9cfe/aiohttp-3.12.12-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:216abf74b324b0f4e67041dd4fb2819613909a825904f8a51701fbcd40c09cd7", size = 1787575, upload-time = "2025-06-10T05:19:18.586Z" },- { url = "https://files.pythonhosted.org/packages/0c/59/01f4c55a1f91ad3b5255b2498b3a22362a3fe6ee9bc9ba1af3cc668244da/aiohttp-3.12.12-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:65d6cefad286459b68e7f867b9586a821fb7f121057b88f02f536ef570992329", size = 1826621, upload-time = "2025-06-10T05:19:21.284Z" },- { url = "https://files.pythonhosted.org/packages/55/85/6357166918ff5025602a7cc41332c1ae7a5b57f2fe3da4d755ae30f24bd0/aiohttp-3.12.12-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:feaaaff61966b5f4b4eae0b79fc79427f49484e4cfa5ab7d138ecd933ab540a8", size = 1729082, upload-time = "2025-06-10T05:19:23.307Z" },- { url = "https://files.pythonhosted.org/packages/e3/ca/de3b5ccd5a2aa9352f6ec6f446565f6e1601ebb54860c94c686a9ff76660/aiohttp-3.12.12-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:a05917780b7cad1755784b16cfaad806bc16029a93d15f063ca60185b7d9ba05", size = 1666159, upload-time = "2025-06-10T05:19:25.929Z" },- { url = "https://files.pythonhosted.org/packages/d1/69/a1006021a1d3244c0872ee75cd8da150e0098b3b2ec6945c225754d11a60/aiohttp-3.12.12-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:082c5ec6d262c1b2ee01c63f4fb9152c17f11692bf16f0f100ad94a7a287d456", size = 1714433, upload-time = "2025-06-10T05:19:28.044Z" },- { url = "https://files.pythonhosted.org/packages/d2/2a/15aa1179e9fbdd0d17cdf117b4296dedad098abb5a93f8e9c8ab4626f6ea/aiohttp-3.12.12-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:b265a3a8b379b38696ac78bdef943bdc4f4a5d6bed1a3fb5c75c6bab1ecea422", size = 1709590, upload-time = "2025-06-10T05:19:30.165Z" },- { url = "https://files.pythonhosted.org/packages/a2/f0/95ed9e21250815f1d1a0cd3e868a3f39400a16010ae59f19ddd4ccc4e787/aiohttp-3.12.12-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:2e0f2e208914ecbc4b2a3b7b4daa759d0c587d9a0b451bb0835ac47fae7fa735", size = 1689776, upload-time = "2025-06-10T05:19:32.965Z" },- { url = "https://files.pythonhosted.org/packages/81/4d/370ecc133c648c98a85445f2d331c1272859c89cd52c29a293015bc352c7/aiohttp-3.12.12-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:9923b025845b72f64d167bca221113377c8ffabd0a351dc18fb839d401ee8e22", size = 1783378, upload-time = "2025-06-10T05:19:35.14Z" },- { url = "https://files.pythonhosted.org/packages/a8/86/414e3dae7e07caf6b02cd75d7148d0d8673d4c5077f407be3627d6e33fac/aiohttp-3.12.12-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:1ebb213445900527831fecc70e185bf142fdfe5f2a691075f22d63c65ee3c35a", size = 1803841, upload-time = "2025-06-10T05:19:37.41Z" },- { url = "https://files.pythonhosted.org/packages/88/df/486f10df681cd1a8c898acc8dc2edbd46ffb088b886757b71ae362bf44d3/aiohttp-3.12.12-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:6fc369fb273a8328077d37798b77c1e65676709af5c182cb74bd169ca9defe81", size = 1716896, upload-time = "2025-06-10T05:19:40.172Z" },- { url = "https://files.pythonhosted.org/packages/07/1e/1cacaf5d838869432e96ece1580d0b51494ebb66351f0e8118b74b38d2f0/aiohttp-3.12.12-cp311-cp311-win32.whl", hash = "sha256:58ecd10fda6a44c311cd3742cfd2aea8c4c600338e9f27cb37434d9f5ca9ddaa", size = 427030, upload-time = "2025-06-10T05:19:42.152Z" },- { url = "https://files.pythonhosted.org/packages/30/dd/e89c1d190da2c84e0ca03c2970b9988a9c56005d18db7f447cf62b3ae6d0/aiohttp-3.12.12-cp311-cp311-win_amd64.whl", hash = "sha256:b0066e88f30be00badffb5ef8f2281532b9a9020863d873ae15f7c147770b6ec", size = 451419, upload-time = "2025-06-10T05:19:44.176Z" },- { url = "https://files.pythonhosted.org/packages/df/e6/df14ec151942818ecc5e685fa8a4b07d3d3d8a9e4a7d2701047c89290551/aiohttp-3.12.12-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:98451ce9ce229d092f278a74a7c2a06b3aa72984673c87796126d7ccade893e9", size = 700494, upload-time = "2025-06-10T05:19:46.18Z" },- { url = "https://files.pythonhosted.org/packages/4f/dc/7bc6e17adcd7a82b0d0317ad3e792ac22c93fb672077f0eade93e8d70182/aiohttp-3.12.12-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:adbac7286d89245e1aff42e948503fdc6edf6d5d65c8e305a67c40f6a8fb95f4", size = 475095, upload-time = "2025-06-10T05:19:48.246Z" },- { url = "https://files.pythonhosted.org/packages/80/fd/c4e8846ad9d9ecdb7d5ba96de65b7bf2c1582f0b2732f2023080c1c05255/aiohttp-3.12.12-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:0728882115bfa85cbd8d0f664c8ccc0cfd5bd3789dd837596785450ae52fac31", size = 467929, upload-time = "2025-06-10T05:19:50.79Z" },- { url = "https://files.pythonhosted.org/packages/70/40/abebcf5c81f5e65b4379c05929773be2731ce12414264d3e0fe09ee241eb/aiohttp-3.12.12-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6bf3b9d9e767f9d0e09fb1a31516410fc741a62cc08754578c40abc497d09540", size = 1714729, upload-time = "2025-06-10T05:19:52.989Z" },- { url = "https://files.pythonhosted.org/packages/8e/67/4c4f96ef6f16405e7c5205ab3c28852c7e904493b6ddc1c744dda1c97a81/aiohttp-3.12.12-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:c944860e86b9f77a462321a440ccf6fa10f5719bb9d026f6b0b11307b1c96c7b", size = 1697380, upload-time = "2025-06-10T05:19:55.832Z" },- { url = "https://files.pythonhosted.org/packages/e9/a2/dae9ebea4caa8030170c0237e55fa0960df44b3596a849ab9ea621964054/aiohttp-3.12.12-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3b1979e1f0c98c06fd0cd940988833b102fa3aa56751f6c40ffe85cabc51f6fd", size = 1752474, upload-time = "2025-06-10T05:19:58.007Z" },- { url = "https://files.pythonhosted.org/packages/31/ef/f3d9073565ac7ad5257aaa1490ebfc2f182dfc817d3ccfd38c8ab35b2247/aiohttp-3.12.12-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:120b7dd084e96cfdad85acea2ce1e7708c70a26db913eabb8d7b417c728f5d84", size = 1798631, upload-time = "2025-06-10T05:20:00.393Z" },- { url = "https://files.pythonhosted.org/packages/8b/0b/8b1978662274c80c8e4a739d9be1ae9ef25e5ce42b55838d6a9d1a4e3497/aiohttp-3.12.12-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0e58f5ae79649ffa247081c2e8c85e31d29623cf2a3137dda985ae05c9478aae", size = 1718071, upload-time = "2025-06-10T05:20:02.812Z" },- { url = "https://files.pythonhosted.org/packages/56/aa/35786137db867901b41cb3d2c19c0f4c56dfe581694dba99dec2683d8f8d/aiohttp-3.12.12-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9aa5f049e3e2745b0141f13e5a64e7c48b1a1427ed18bbb7957b348f282fee56", size = 1633871, upload-time = "2025-06-10T05:20:05.127Z" },- { url = "https://files.pythonhosted.org/packages/63/1d/34d45497dd04d08d662ecda875c44e91d271bbc5d21f4c9e4cbd3ddf7ae2/aiohttp-3.12.12-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:7163cc9cf3722d90f1822f8a38b211e3ae2fc651c63bb55449f03dc1b3ff1d44", size = 1694933, upload-time = "2025-06-10T05:20:07.431Z" },- { url = "https://files.pythonhosted.org/packages/29/c7/41e09a4517449eabbb0a7fe6d60f584fe5b21d4bff761197eb0b81e70034/aiohttp-3.12.12-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:ef97c4d035b721de6607f3980fa3e4ef0ec3aca76474b5789b7fac286a8c4e23", size = 1716386, upload-time = "2025-06-10T05:20:09.787Z" },- { url = "https://files.pythonhosted.org/packages/3a/32/907bd2010b51b70de5314ad707dfc4e898ea0011ff3d678cdf43d6f8980a/aiohttp-3.12.12-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:1c14448d6a86acadc3f7b2f4cc385d1fb390acb6f37dce27f86fe629410d92e3", size = 1657039, upload-time = "2025-06-10T05:20:12.198Z" },- { url = "https://files.pythonhosted.org/packages/60/27/8d87344a33346dcd39273adc33060aeb135e0ef70d1d6e71a3b03894a8e9/aiohttp-3.12.12-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:a1b6df6255cfc493454c79221183d64007dd5080bcda100db29b7ff181b8832c", size = 1736599, upload-time = "2025-06-10T05:20:14.519Z" },- { url = "https://files.pythonhosted.org/packages/ca/45/57c7ef1af694a6d0906abab6edde03787c8c6b0cf5d8359b69d1eb0679df/aiohttp-3.12.12-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:60fc7338dfb0626c2927bfbac4785de3ea2e2bbe3d328ba5f3ece123edda4977", size = 1764575, upload-time = "2025-06-10T05:20:16.993Z" },- { url = "https://files.pythonhosted.org/packages/2a/cc/b1f918cd702efa9ead9d41f89214e9225cda4e5d013d6eed7f1915c17d0a/aiohttp-3.12.12-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:d2afc72207ef4c9d4ca9fcd00689a6a37ef2d625600c3d757b5c2b80c9d0cf9a", size = 1724184, upload-time = "2025-06-10T05:20:19.296Z" },- { url = "https://files.pythonhosted.org/packages/47/55/089762ee32c2a2e0f523d9ab38c9da2a344cac0e0cc8d16ecf206517ef7e/aiohttp-3.12.12-cp312-cp312-win32.whl", hash = "sha256:8098a48f93b2cbcdb5778e7c9a0e0375363e40ad692348e6e65c3b70d593b27c", size = 421762, upload-time = "2025-06-10T05:20:22.063Z" },- { url = "https://files.pythonhosted.org/packages/ab/47/151f657e429972916f61399bd52b410e9072d5a2cae1b794f890930e5797/aiohttp-3.12.12-cp312-cp312-win_amd64.whl", hash = "sha256:d1c1879b2e0fc337d7a1b63fe950553c2b9e93c071cf95928aeea1902d441403", size = 447863, upload-time = "2025-06-10T05:20:24.326Z" },- { url = "https://files.pythonhosted.org/packages/ee/3e/396a7d1c47aa7a74612b186dc716857506c61afac72337a7a96215c2a124/aiohttp-3.12.12-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ea5d604318234427929d486954e3199aded65f41593ac57aa0241ab93dda3d15", size = 694901, upload-time = "2025-06-10T05:20:26.58Z" },- { url = "https://files.pythonhosted.org/packages/cc/97/235e48eadf73a1854b4d4da29b88d00049309d897d55a511e1cbe4412603/aiohttp-3.12.12-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:e03ff38250b8b572dce6fcd7b6fb6ee398bb8a59e6aa199009c5322d721df4fc", size = 472552, upload-time = "2025-06-10T05:20:28.957Z" },- { url = "https://files.pythonhosted.org/packages/6b/73/cd7c9439e8cab4113650541017c6524bd0e675b219dfdbbf945a78305e3f/aiohttp-3.12.12-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:71125b1fc2b6a94bccc63bbece620906a4dead336d2051f8af9cbf04480bc5af", size = 464853, upload-time = "2025-06-10T05:20:31.652Z" },- { url = "https://files.pythonhosted.org/packages/d1/33/eea88ee55ed4b3f74732d9fc773e6fcf134a2971a19c7ecc49a291e7e57f/aiohttp-3.12.12-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:784a66f9f853a22c6b8c2bd0ff157f9b879700f468d6d72cfa99167df08c5c9c", size = 1703671, upload-time = "2025-06-10T05:20:33.969Z" },- { url = "https://files.pythonhosted.org/packages/2a/e3/a67ecf9c154b13bad9e2a86ea3782a4b73e889343ffde8c1aadcf9099c09/aiohttp-3.12.12-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:a5be0b58670b54301404bd1840e4902570a1c3be00358e2700919cb1ea73c438", size = 1684934, upload-time = "2025-06-10T05:20:36.721Z" },- { url = "https://files.pythonhosted.org/packages/89/f0/3aaea866531be2f2fcf3a87607e1f55fa72e6ce5acd6b058941a4fc35e15/aiohttp-3.12.12-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ce8f13566fc7bf5a728275b434bc3bdea87a7ed3ad5f734102b02ca59d9b510f", size = 1737004, upload-time = "2025-06-10T05:20:39.533Z" },- { url = "https://files.pythonhosted.org/packages/a7/7a/15867a4c7d39d8fd9bd02191cf60b1d06415fc407bbd4ff2f9660845f1cb/aiohttp-3.12.12-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:d736e57d1901683bc9be648aa308cb73e646252c74b4c639c35dcd401ed385ea", size = 1786378, upload-time = "2025-06-10T05:20:42.03Z" },- { url = "https://files.pythonhosted.org/packages/bd/61/82b15f87088b35705e01fce55806241b45a1099b3470bbca0bed8ee98662/aiohttp-3.12.12-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e2007eaa7aae9102f211c519d1ec196bd3cecb1944a095db19eeaf132b798738", size = 1708707, upload-time = "2025-06-10T05:20:44.474Z" },- { url = "https://files.pythonhosted.org/packages/28/f2/aed0786d5a1c2ed1f5a13ff2a98baacc27206b81d93812da28fc49d8a5d0/aiohttp-3.12.12-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2a813e61583cab6d5cdbaa34bc28863acdb92f9f46e11de1b3b9251a1e8238f6", size = 1622410, upload-time = "2025-06-10T05:20:46.957Z" },- { url = "https://files.pythonhosted.org/packages/17/54/8305f49a960376136ada977be1370fddb584c63d40bd1b9bef59469f28c7/aiohttp-3.12.12-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:e408293aa910b0aea48b86a28eace41d497a85ba16c20f619f0c604597ef996c", size = 1675435, upload-time = "2025-06-10T05:20:49.379Z" },- { url = "https://files.pythonhosted.org/packages/bb/dc/0a55350025bc297265cfa6c6b1b1f7508f4226ca3238697cbe5e772a7d76/aiohttp-3.12.12-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:f3d31faf290f5a30acba46b388465b67c6dbe8655d183e9efe2f6a1d594e6d9d", size = 1707099, upload-time = "2025-06-10T05:20:51.974Z" },- { url = "https://files.pythonhosted.org/packages/d8/70/d949a1612b996e49d540c10ed77a0a1465c482a590e9a59c1c7897746119/aiohttp-3.12.12-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:0b84731697325b023902aa643bd1726d999f5bc7854bc28b17ff410a81151d4b", size = 1649693, upload-time = "2025-06-10T05:20:54.973Z" },- { url = "https://files.pythonhosted.org/packages/c1/ea/fb87beb7135e25576a1e6fbe98106c037d9fcf1543f19108f9ceb73c192c/aiohttp-3.12.12-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:a324c6852b6e327811748446e56cc9bb6eaa58710557922183175816e82a4234", size = 1725825, upload-time = "2025-06-10T05:20:57.433Z" },- { url = "https://files.pythonhosted.org/packages/f1/1f/adbeb3e440d49b733cef499ace94723ab1fe9fb516425e219379e03b7c9a/aiohttp-3.12.12-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:22fd867fbd72612dcf670c90486dbcbaf702cb807fb0b42bc0b7a142a573574a", size = 1759300, upload-time = "2025-06-10T05:21:00.444Z" },- { url = "https://files.pythonhosted.org/packages/f2/c1/2fe007ad930f409d0d7fd9916cd55ec9b78b6a611a237424266ed71da48b/aiohttp-3.12.12-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:3e092f1a970223794a4bf620a26c0e4e4e8e36bccae9b0b5da35e6d8ee598a03", size = 1708189, upload-time = "2025-06-10T05:21:02.969Z" },- { url = "https://files.pythonhosted.org/packages/85/5e/ed3ed640fafae3972eae6cd26f66240108cf62452ac8128d59970d538cb1/aiohttp-3.12.12-cp313-cp313-win32.whl", hash = "sha256:7f5f5eb8717ef8ba15ab35fcde5a70ad28bbdc34157595d1cddd888a985f5aae", size = 420783, upload-time = "2025-06-10T05:21:06.287Z" },- { url = "https://files.pythonhosted.org/packages/a6/db/57d2bb4af52dd0c6f62c42c7d34b82495b2902e50440134f70bfb7ee0fdd/aiohttp-3.12.12-cp313-cp313-win_amd64.whl", hash = "sha256:ace2499bdd03c329c054dc4b47361f2b19d5aa470f7db5c7e0e989336761b33c", size = 446721, upload-time = "2025-06-10T05:21:08.738Z" },+sdist = { url = "https://files.pythonhosted.org/packages/42/6e/ab88e7cb2a4058bed2f7870276454f85a7c56cd6da79349eb314fc7bbcaa/aiohttp-3.12.13.tar.gz", hash = "sha256:47e2da578528264a12e4e3dd8dd72a7289e5f812758fe086473fab037a10fcce", size = 7819160, upload-time = "2025-06-14T15:15:41.354Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/8b/2d/27e4347660723738b01daa3f5769d56170f232bf4695dd4613340da135bb/aiohttp-3.12.13-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:5421af8f22a98f640261ee48aae3a37f0c41371e99412d55eaf2f8a46d5dad29", size = 702090, upload-time = "2025-06-14T15:12:58.938Z" },+ { url = "https://files.pythonhosted.org/packages/10/0b/4a8e0468ee8f2b9aff3c05f2c3a6be1dfc40b03f68a91b31041d798a9510/aiohttp-3.12.13-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0fcda86f6cb318ba36ed8f1396a6a4a3fd8f856f84d426584392083d10da4de0", size = 478440, upload-time = "2025-06-14T15:13:02.981Z" },+ { url = "https://files.pythonhosted.org/packages/b9/c8/2086df2f9a842b13feb92d071edf756be89250f404f10966b7bc28317f17/aiohttp-3.12.13-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:4cd71c9fb92aceb5a23c4c39d8ecc80389c178eba9feab77f19274843eb9412d", size = 466215, upload-time = "2025-06-14T15:13:04.817Z" },+ { url = "https://files.pythonhosted.org/packages/a7/3d/d23e5bd978bc8012a65853959b13bd3b55c6e5afc172d89c26ad6624c52b/aiohttp-3.12.13-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:34ebf1aca12845066c963016655dac897651e1544f22a34c9b461ac3b4b1d3aa", size = 1648271, upload-time = "2025-06-14T15:13:06.532Z" },+ { url = "https://files.pythonhosted.org/packages/31/31/e00122447bb137591c202786062f26dd383574c9f5157144127077d5733e/aiohttp-3.12.13-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:893a4639694c5b7edd4bdd8141be296042b6806e27cc1d794e585c43010cc294", size = 1622329, upload-time = "2025-06-14T15:13:08.394Z" },+ { url = "https://files.pythonhosted.org/packages/04/01/caef70be3ac38986969045f21f5fb802ce517b3f371f0615206bf8aa6423/aiohttp-3.12.13-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:663d8ee3ffb3494502ebcccb49078faddbb84c1d870f9c1dd5a29e85d1f747ce", size = 1694734, upload-time = "2025-06-14T15:13:09.979Z" },+ { url = "https://files.pythonhosted.org/packages/3f/15/328b71fedecf69a9fd2306549b11c8966e420648a3938d75d3ed5bcb47f6/aiohttp-3.12.13-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f0f8f6a85a0006ae2709aa4ce05749ba2cdcb4b43d6c21a16c8517c16593aabe", size = 1737049, upload-time = "2025-06-14T15:13:11.672Z" },+ { url = "https://files.pythonhosted.org/packages/e6/7a/d85866a642158e1147c7da5f93ad66b07e5452a84ec4258e5f06b9071e92/aiohttp-3.12.13-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1582745eb63df267c92d8b61ca655a0ce62105ef62542c00a74590f306be8cb5", size = 1641715, upload-time = "2025-06-14T15:13:13.548Z" },+ { url = "https://files.pythonhosted.org/packages/14/57/3588800d5d2f5f3e1cb6e7a72747d1abc1e67ba5048e8b845183259c2e9b/aiohttp-3.12.13-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d59227776ee2aa64226f7e086638baa645f4b044f2947dbf85c76ab11dcba073", size = 1581836, upload-time = "2025-06-14T15:13:15.086Z" },+ { url = "https://files.pythonhosted.org/packages/2f/55/c913332899a916d85781aa74572f60fd98127449b156ad9c19e23135b0e4/aiohttp-3.12.13-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:06b07c418bde1c8e737d8fa67741072bd3f5b0fb66cf8c0655172188c17e5fa6", size = 1625685, upload-time = "2025-06-14T15:13:17.163Z" },+ { url = "https://files.pythonhosted.org/packages/4c/34/26cded195f3bff128d6a6d58d7a0be2ae7d001ea029e0fe9008dcdc6a009/aiohttp-3.12.13-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:9445c1842680efac0f81d272fd8db7163acfcc2b1436e3f420f4c9a9c5a50795", size = 1636471, upload-time = "2025-06-14T15:13:19.086Z" },+ { url = "https://files.pythonhosted.org/packages/19/21/70629ca006820fccbcec07f3cd5966cbd966e2d853d6da55339af85555b9/aiohttp-3.12.13-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:09c4767af0b0b98c724f5d47f2bf33395c8986995b0a9dab0575ca81a554a8c0", size = 1611923, upload-time = "2025-06-14T15:13:20.997Z" },+ { url = "https://files.pythonhosted.org/packages/31/80/7fa3f3bebf533aa6ae6508b51ac0de9965e88f9654fa679cc1a29d335a79/aiohttp-3.12.13-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:f3854fbde7a465318ad8d3fc5bef8f059e6d0a87e71a0d3360bb56c0bf87b18a", size = 1691511, upload-time = "2025-06-14T15:13:22.54Z" },+ { url = "https://files.pythonhosted.org/packages/0f/7a/359974653a3cdd3e9cee8ca10072a662c3c0eb46a359c6a1f667b0296e2f/aiohttp-3.12.13-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:2332b4c361c05ecd381edb99e2a33733f3db906739a83a483974b3df70a51b40", size = 1714751, upload-time = "2025-06-14T15:13:24.366Z" },+ { url = "https://files.pythonhosted.org/packages/2d/24/0aa03d522171ce19064347afeefadb008be31ace0bbb7d44ceb055700a14/aiohttp-3.12.13-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:1561db63fa1b658cd94325d303933553ea7d89ae09ff21cc3bcd41b8521fbbb6", size = 1643090, upload-time = "2025-06-14T15:13:26.231Z" },+ { url = "https://files.pythonhosted.org/packages/86/2e/7d4b0026a41e4b467e143221c51b279083b7044a4b104054f5c6464082ff/aiohttp-3.12.13-cp310-cp310-win32.whl", hash = "sha256:a0be857f0b35177ba09d7c472825d1b711d11c6d0e8a2052804e3b93166de1ad", size = 427526, upload-time = "2025-06-14T15:13:27.988Z" },+ { url = "https://files.pythonhosted.org/packages/17/de/34d998da1e7f0de86382160d039131e9b0af1962eebfe53dda2b61d250e7/aiohttp-3.12.13-cp310-cp310-win_amd64.whl", hash = "sha256:fcc30ad4fb5cb41a33953292d45f54ef4066746d625992aeac33b8c681173178", size = 450734, upload-time = "2025-06-14T15:13:29.394Z" },+ { url = "https://files.pythonhosted.org/packages/6a/65/5566b49553bf20ffed6041c665a5504fb047cefdef1b701407b8ce1a47c4/aiohttp-3.12.13-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:7c229b1437aa2576b99384e4be668af1db84b31a45305d02f61f5497cfa6f60c", size = 709401, upload-time = "2025-06-14T15:13:30.774Z" },+ { url = "https://files.pythonhosted.org/packages/14/b5/48e4cc61b54850bdfafa8fe0b641ab35ad53d8e5a65ab22b310e0902fa42/aiohttp-3.12.13-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:04076d8c63471e51e3689c93940775dc3d12d855c0c80d18ac5a1c68f0904358", size = 481669, upload-time = "2025-06-14T15:13:32.316Z" },+ { url = "https://files.pythonhosted.org/packages/04/4f/e3f95c8b2a20a0437d51d41d5ccc4a02970d8ad59352efb43ea2841bd08e/aiohttp-3.12.13-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:55683615813ce3601640cfaa1041174dc956d28ba0511c8cbd75273eb0587014", size = 469933, upload-time = "2025-06-14T15:13:34.104Z" },+ { url = "https://files.pythonhosted.org/packages/41/c9/c5269f3b6453b1cfbd2cfbb6a777d718c5f086a3727f576c51a468b03ae2/aiohttp-3.12.13-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:921bc91e602d7506d37643e77819cb0b840d4ebb5f8d6408423af3d3bf79a7b7", size = 1740128, upload-time = "2025-06-14T15:13:35.604Z" },+ { url = "https://files.pythonhosted.org/packages/6f/49/a3f76caa62773d33d0cfaa842bdf5789a78749dbfe697df38ab1badff369/aiohttp-3.12.13-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:e72d17fe0974ddeae8ed86db297e23dba39c7ac36d84acdbb53df2e18505a013", size = 1688796, upload-time = "2025-06-14T15:13:37.125Z" },+ { url = "https://files.pythonhosted.org/packages/ad/e4/556fccc4576dc22bf18554b64cc873b1a3e5429a5bdb7bbef7f5d0bc7664/aiohttp-3.12.13-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0653d15587909a52e024a261943cf1c5bdc69acb71f411b0dd5966d065a51a47", size = 1787589, upload-time = "2025-06-14T15:13:38.745Z" },+ { url = "https://files.pythonhosted.org/packages/b9/3d/d81b13ed48e1a46734f848e26d55a7391708421a80336e341d2aef3b6db2/aiohttp-3.12.13-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a77b48997c66722c65e157c06c74332cdf9c7ad00494b85ec43f324e5c5a9b9a", size = 1826635, upload-time = "2025-06-14T15:13:40.733Z" },+ { url = "https://files.pythonhosted.org/packages/75/a5/472e25f347da88459188cdaadd1f108f6292f8a25e62d226e63f860486d1/aiohttp-3.12.13-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d6946bae55fd36cfb8e4092c921075cde029c71c7cb571d72f1079d1e4e013bc", size = 1729095, upload-time = "2025-06-14T15:13:42.312Z" },+ { url = "https://files.pythonhosted.org/packages/b9/fe/322a78b9ac1725bfc59dfc301a5342e73d817592828e4445bd8f4ff83489/aiohttp-3.12.13-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:4f95db8c8b219bcf294a53742c7bda49b80ceb9d577c8e7aa075612b7f39ffb7", size = 1666170, upload-time = "2025-06-14T15:13:44.884Z" },+ { url = "https://files.pythonhosted.org/packages/7a/77/ec80912270e231d5e3839dbd6c065472b9920a159ec8a1895cf868c2708e/aiohttp-3.12.13-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:03d5eb3cfb4949ab4c74822fb3326cd9655c2b9fe22e4257e2100d44215b2e2b", size = 1714444, upload-time = "2025-06-14T15:13:46.401Z" },+ { url = "https://files.pythonhosted.org/packages/21/b2/fb5aedbcb2b58d4180e58500e7c23ff8593258c27c089abfbcc7db65bd40/aiohttp-3.12.13-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:6383dd0ffa15515283c26cbf41ac8e6705aab54b4cbb77bdb8935a713a89bee9", size = 1709604, upload-time = "2025-06-14T15:13:48.377Z" },+ { url = "https://files.pythonhosted.org/packages/e3/15/a94c05f7c4dc8904f80b6001ad6e07e035c58a8ebfcc15e6b5d58500c858/aiohttp-3.12.13-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:6548a411bc8219b45ba2577716493aa63b12803d1e5dc70508c539d0db8dbf5a", size = 1689786, upload-time = "2025-06-14T15:13:50.401Z" },+ { url = "https://files.pythonhosted.org/packages/1d/fd/0d2e618388f7a7a4441eed578b626bda9ec6b5361cd2954cfc5ab39aa170/aiohttp-3.12.13-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:81b0fcbfe59a4ca41dc8f635c2a4a71e63f75168cc91026c61be665945739e2d", size = 1783389, upload-time = "2025-06-14T15:13:51.945Z" },+ { url = "https://files.pythonhosted.org/packages/a6/6b/6986d0c75996ef7e64ff7619b9b7449b1d1cbbe05c6755e65d92f1784fe9/aiohttp-3.12.13-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:6a83797a0174e7995e5edce9dcecc517c642eb43bc3cba296d4512edf346eee2", size = 1803853, upload-time = "2025-06-14T15:13:53.533Z" },+ { url = "https://files.pythonhosted.org/packages/21/65/cd37b38f6655d95dd07d496b6d2f3924f579c43fd64b0e32b547b9c24df5/aiohttp-3.12.13-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:a5734d8469a5633a4e9ffdf9983ff7cdb512524645c7a3d4bc8a3de45b935ac3", size = 1716909, upload-time = "2025-06-14T15:13:55.148Z" },+ { url = "https://files.pythonhosted.org/packages/fd/20/2de7012427dc116714c38ca564467f6143aec3d5eca3768848d62aa43e62/aiohttp-3.12.13-cp311-cp311-win32.whl", hash = "sha256:fef8d50dfa482925bb6b4c208b40d8e9fa54cecba923dc65b825a72eed9a5dbd", size = 427036, upload-time = "2025-06-14T15:13:57.076Z" },+ { url = "https://files.pythonhosted.org/packages/f8/b6/98518bcc615ef998a64bef371178b9afc98ee25895b4f476c428fade2220/aiohttp-3.12.13-cp311-cp311-win_amd64.whl", hash = "sha256:9a27da9c3b5ed9d04c36ad2df65b38a96a37e9cfba6f1381b842d05d98e6afe9", size = 451427, upload-time = "2025-06-14T15:13:58.505Z" },+ { url = "https://files.pythonhosted.org/packages/b4/6a/ce40e329788013cd190b1d62bbabb2b6a9673ecb6d836298635b939562ef/aiohttp-3.12.13-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:0aa580cf80558557285b49452151b9c69f2fa3ad94c5c9e76e684719a8791b73", size = 700491, upload-time = "2025-06-14T15:14:00.048Z" },+ { url = "https://files.pythonhosted.org/packages/28/d9/7150d5cf9163e05081f1c5c64a0cdf3c32d2f56e2ac95db2a28fe90eca69/aiohttp-3.12.13-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:b103a7e414b57e6939cc4dece8e282cfb22043efd0c7298044f6594cf83ab347", size = 475104, upload-time = "2025-06-14T15:14:01.691Z" },+ { url = "https://files.pythonhosted.org/packages/f8/91/d42ba4aed039ce6e449b3e2db694328756c152a79804e64e3da5bc19dffc/aiohttp-3.12.13-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:78f64e748e9e741d2eccff9597d09fb3cd962210e5b5716047cbb646dc8fe06f", size = 467948, upload-time = "2025-06-14T15:14:03.561Z" },+ { url = "https://files.pythonhosted.org/packages/99/3b/06f0a632775946981d7c4e5a865cddb6e8dfdbaed2f56f9ade7bb4a1039b/aiohttp-3.12.13-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:29c955989bf4c696d2ededc6b0ccb85a73623ae6e112439398935362bacfaaf6", size = 1714742, upload-time = "2025-06-14T15:14:05.558Z" },+ { url = "https://files.pythonhosted.org/packages/92/a6/2552eebad9ec5e3581a89256276009e6a974dc0793632796af144df8b740/aiohttp-3.12.13-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:d640191016763fab76072c87d8854a19e8e65d7a6fcfcbf017926bdbbb30a7e5", size = 1697393, upload-time = "2025-06-14T15:14:07.194Z" },+ { url = "https://files.pythonhosted.org/packages/d8/9f/bd08fdde114b3fec7a021381b537b21920cdd2aa29ad48c5dffd8ee314f1/aiohttp-3.12.13-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4dc507481266b410dede95dd9f26c8d6f5a14315372cc48a6e43eac652237d9b", size = 1752486, upload-time = "2025-06-14T15:14:08.808Z" },+ { url = "https://files.pythonhosted.org/packages/f7/e1/affdea8723aec5bd0959171b5490dccd9a91fcc505c8c26c9f1dca73474d/aiohttp-3.12.13-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8a94daa873465d518db073bd95d75f14302e0208a08e8c942b2f3f1c07288a75", size = 1798643, upload-time = "2025-06-14T15:14:10.767Z" },+ { url = "https://files.pythonhosted.org/packages/f3/9d/666d856cc3af3a62ae86393baa3074cc1d591a47d89dc3bf16f6eb2c8d32/aiohttp-3.12.13-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:177f52420cde4ce0bb9425a375d95577fe082cb5721ecb61da3049b55189e4e6", size = 1718082, upload-time = "2025-06-14T15:14:12.38Z" },+ { url = "https://files.pythonhosted.org/packages/f3/ce/3c185293843d17be063dada45efd2712bb6bf6370b37104b4eda908ffdbd/aiohttp-3.12.13-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0f7df1f620ec40f1a7fbcb99ea17d7326ea6996715e78f71a1c9a021e31b96b8", size = 1633884, upload-time = "2025-06-14T15:14:14.415Z" },+ { url = "https://files.pythonhosted.org/packages/3a/5b/f3413f4b238113be35dfd6794e65029250d4b93caa0974ca572217745bdb/aiohttp-3.12.13-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:3062d4ad53b36e17796dce1c0d6da0ad27a015c321e663657ba1cc7659cfc710", size = 1694943, upload-time = "2025-06-14T15:14:16.48Z" },+ { url = "https://files.pythonhosted.org/packages/82/c8/0e56e8bf12081faca85d14a6929ad5c1263c146149cd66caa7bc12255b6d/aiohttp-3.12.13-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:8605e22d2a86b8e51ffb5253d9045ea73683d92d47c0b1438e11a359bdb94462", size = 1716398, upload-time = "2025-06-14T15:14:18.589Z" },+ { url = "https://files.pythonhosted.org/packages/ea/f3/33192b4761f7f9b2f7f4281365d925d663629cfaea093a64b658b94fc8e1/aiohttp-3.12.13-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:54fbbe6beafc2820de71ece2198458a711e224e116efefa01b7969f3e2b3ddae", size = 1657051, upload-time = "2025-06-14T15:14:20.223Z" },+ { url = "https://files.pythonhosted.org/packages/5e/0b/26ddd91ca8f84c48452431cb4c5dd9523b13bc0c9766bda468e072ac9e29/aiohttp-3.12.13-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:050bd277dfc3768b606fd4eae79dd58ceda67d8b0b3c565656a89ae34525d15e", size = 1736611, upload-time = "2025-06-14T15:14:21.988Z" },+ { url = "https://files.pythonhosted.org/packages/c3/8d/e04569aae853302648e2c138a680a6a2f02e374c5b6711732b29f1e129cc/aiohttp-3.12.13-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:2637a60910b58f50f22379b6797466c3aa6ae28a6ab6404e09175ce4955b4e6a", size = 1764586, upload-time = "2025-06-14T15:14:23.979Z" },+ { url = "https://files.pythonhosted.org/packages/ac/98/c193c1d1198571d988454e4ed75adc21c55af247a9fda08236602921c8c8/aiohttp-3.12.13-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:e986067357550d1aaa21cfe9897fa19e680110551518a5a7cf44e6c5638cb8b5", size = 1724197, upload-time = "2025-06-14T15:14:25.692Z" },+ { url = "https://files.pythonhosted.org/packages/e7/9e/07bb8aa11eec762c6b1ff61575eeeb2657df11ab3d3abfa528d95f3e9337/aiohttp-3.12.13-cp312-cp312-win32.whl", hash = "sha256:ac941a80aeea2aaae2875c9500861a3ba356f9ff17b9cb2dbfb5cbf91baaf5bf", size = 421771, upload-time = "2025-06-14T15:14:27.364Z" },+ { url = "https://files.pythonhosted.org/packages/52/66/3ce877e56ec0813069cdc9607cd979575859c597b6fb9b4182c6d5f31886/aiohttp-3.12.13-cp312-cp312-win_amd64.whl", hash = "sha256:671f41e6146a749b6c81cb7fd07f5a8356d46febdaaaf07b0e774ff04830461e", size = 447869, upload-time = "2025-06-14T15:14:29.05Z" },+ { url = "https://files.pythonhosted.org/packages/11/0f/db19abdf2d86aa1deec3c1e0e5ea46a587b97c07a16516b6438428b3a3f8/aiohttp-3.12.13-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:d4a18e61f271127465bdb0e8ff36e8f02ac4a32a80d8927aa52371e93cd87938", size = 694910, upload-time = "2025-06-14T15:14:30.604Z" },+ { url = "https://files.pythonhosted.org/packages/d5/81/0ab551e1b5d7f1339e2d6eb482456ccbe9025605b28eed2b1c0203aaaade/aiohttp-3.12.13-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:532542cb48691179455fab429cdb0d558b5e5290b033b87478f2aa6af5d20ace", size = 472566, upload-time = "2025-06-14T15:14:32.275Z" },+ { url = "https://files.pythonhosted.org/packages/34/3f/6b7d336663337672d29b1f82d1f252ec1a040fe2d548f709d3f90fa2218a/aiohttp-3.12.13-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:d7eea18b52f23c050ae9db5d01f3d264ab08f09e7356d6f68e3f3ac2de9dfabb", size = 464856, upload-time = "2025-06-14T15:14:34.132Z" },+ { url = "https://files.pythonhosted.org/packages/26/7f/32ca0f170496aa2ab9b812630fac0c2372c531b797e1deb3deb4cea904bd/aiohttp-3.12.13-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ad7c8e5c25f2a26842a7c239de3f7b6bfb92304593ef997c04ac49fb703ff4d7", size = 1703683, upload-time = "2025-06-14T15:14:36.034Z" },+ { url = "https://files.pythonhosted.org/packages/ec/53/d5513624b33a811c0abea8461e30a732294112318276ce3dbf047dbd9d8b/aiohttp-3.12.13-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:6af355b483e3fe9d7336d84539fef460120c2f6e50e06c658fe2907c69262d6b", size = 1684946, upload-time = "2025-06-14T15:14:38Z" },+ { url = "https://files.pythonhosted.org/packages/37/72/4c237dd127827b0247dc138d3ebd49c2ded6114c6991bbe969058575f25f/aiohttp-3.12.13-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a95cf9f097498f35c88e3609f55bb47b28a5ef67f6888f4390b3d73e2bac6177", size = 1737017, upload-time = "2025-06-14T15:14:39.951Z" },+ { url = "https://files.pythonhosted.org/packages/0d/67/8a7eb3afa01e9d0acc26e1ef847c1a9111f8b42b82955fcd9faeb84edeb4/aiohttp-3.12.13-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b8ed8c38a1c584fe99a475a8f60eefc0b682ea413a84c6ce769bb19a7ff1c5ef", size = 1786390, upload-time = "2025-06-14T15:14:42.151Z" },+ { url = "https://files.pythonhosted.org/packages/48/19/0377df97dd0176ad23cd8cad4fd4232cfeadcec6c1b7f036315305c98e3f/aiohttp-3.12.13-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7a0b9170d5d800126b5bc89d3053a2363406d6e327afb6afaeda2d19ee8bb103", size = 1708719, upload-time = "2025-06-14T15:14:44.039Z" },+ { url = "https://files.pythonhosted.org/packages/61/97/ade1982a5c642b45f3622255173e40c3eed289c169f89d00eeac29a89906/aiohttp-3.12.13-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:372feeace612ef8eb41f05ae014a92121a512bd5067db8f25101dd88a8db11da", size = 1622424, upload-time = "2025-06-14T15:14:45.945Z" },+ { url = "https://files.pythonhosted.org/packages/99/ab/00ad3eea004e1d07ccc406e44cfe2b8da5acb72f8c66aeeb11a096798868/aiohttp-3.12.13-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:a946d3702f7965d81f7af7ea8fb03bb33fe53d311df48a46eeca17e9e0beed2d", size = 1675447, upload-time = "2025-06-14T15:14:47.911Z" },+ { url = "https://files.pythonhosted.org/packages/3f/fe/74e5ce8b2ccaba445fe0087abc201bfd7259431d92ae608f684fcac5d143/aiohttp-3.12.13-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:a0c4725fae86555bbb1d4082129e21de7264f4ab14baf735278c974785cd2041", size = 1707110, upload-time = "2025-06-14T15:14:50.334Z" },+ { url = "https://files.pythonhosted.org/packages/ef/c4/39af17807f694f7a267bd8ab1fbacf16ad66740862192a6c8abac2bff813/aiohttp-3.12.13-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:9b28ea2f708234f0a5c44eb6c7d9eb63a148ce3252ba0140d050b091b6e842d1", size = 1649706, upload-time = "2025-06-14T15:14:52.378Z" },+ { url = "https://files.pythonhosted.org/packages/38/e8/f5a0a5f44f19f171d8477059aa5f28a158d7d57fe1a46c553e231f698435/aiohttp-3.12.13-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:d4f5becd2a5791829f79608c6f3dc745388162376f310eb9c142c985f9441cc1", size = 1725839, upload-time = "2025-06-14T15:14:54.617Z" },+ { url = "https://files.pythonhosted.org/packages/fd/ac/81acc594c7f529ef4419d3866913f628cd4fa9cab17f7bf410a5c3c04c53/aiohttp-3.12.13-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:60f2ce6b944e97649051d5f5cc0f439360690b73909230e107fd45a359d3e911", size = 1759311, upload-time = "2025-06-14T15:14:56.597Z" },+ { url = "https://files.pythonhosted.org/packages/38/0d/aabe636bd25c6ab7b18825e5a97d40024da75152bec39aa6ac8b7a677630/aiohttp-3.12.13-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:69fc1909857401b67bf599c793f2183fbc4804717388b0b888f27f9929aa41f3", size = 1708202, upload-time = "2025-06-14T15:14:58.598Z" },+ { url = "https://files.pythonhosted.org/packages/1f/ab/561ef2d8a223261683fb95a6283ad0d36cb66c87503f3a7dde7afe208bb2/aiohttp-3.12.13-cp313-cp313-win32.whl", hash = "sha256:7d7e68787a2046b0e44ba5587aa723ce05d711e3a3665b6b7545328ac8e3c0dd", size = 420794, upload-time = "2025-06-14T15:15:00.939Z" },+ { url = "https://files.pythonhosted.org/packages/9d/47/b11d0089875a23bff0abd3edb5516bcd454db3fefab8604f5e4b07bd6210/aiohttp-3.12.13-cp313-cp313-win_amd64.whl", hash = "sha256:5a178390ca90419bfd41419a809688c368e63c86bd725e1186dd97f6b89c2706", size = 446735, upload-time = "2025-06-14T15:15:02.858Z" },][[package]]@@ -228,6 +228,15 @@ wheels = [{ url = "https://files.pythonhosted.org/packages/77/06/bb80f5f86020c4551da315d78b3ab75e8228f89f0162f2c3a819e407941a/attrs-25.3.0-py3-none-any.whl", hash = "sha256:427318ce031701fea540783410126f03899a97ffc6f61596ad581ac2e40e3bc3", size = 63815, upload-time = "2025-03-13T11:10:21.14Z" },]+[[package]]+name = "babel"+version = "2.17.0"+source = { registry = "https://pypi.org/simple" }+sdist = { url = "https://files.pythonhosted.org/packages/7d/6b/d52e42361e1aa00709585ecc30b3f9684b3ab62530771402248b1b1d6240/babel-2.17.0.tar.gz", hash = "sha256:0c54cffb19f690cdcc52a3b50bcbf71e07a808d1c80d549f2459b9d2cf0afb9d", size = 9951852, upload-time = "2025-02-01T15:17:41.026Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/b7/b8/3fe70c75fe32afc4bb507f75563d39bc5642255d1d94f1f23604725780bf/babel-2.17.0-py3-none-any.whl", hash = "sha256:4d0b53093fdfb4b21c92b5213dba5a1b23885afa8383709427046b21c366e5f2", size = 10182537, upload-time = "2025-02-01T15:17:37.39Z" },+]+[[package]]name = "backports-strenum"version = "1.3.1"@@ -237,6 +246,19 @@ wheels = [{ url = "https://files.pythonhosted.org/packages/d6/50/56cf20e2ee5127b603b81d5a69580a1a325083e2b921aa8f067da83927c0/backports_strenum-1.3.1-py3-none-any.whl", hash = "sha256:cdcfe36dc897e2615dc793b7d3097f54d359918fc448754a517e6f23044ccf83", size = 8304, upload-time = "2023-12-09T14:36:39.905Z" },]+[[package]]+name = "backrefs"+version = "5.8"+source = { registry = "https://pypi.org/simple" }+sdist = { url = "https://files.pythonhosted.org/packages/6c/46/caba1eb32fa5784428ab401a5487f73db4104590ecd939ed9daaf18b47e0/backrefs-5.8.tar.gz", hash = "sha256:2cab642a205ce966af3dd4b38ee36009b31fa9502a35fd61d59ccc116e40a6bd", size = 6773994, upload-time = "2025-02-25T18:15:32.003Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/bf/cb/d019ab87fe70e0fe3946196d50d6a4428623dc0c38a6669c8cae0320fbf3/backrefs-5.8-py310-none-any.whl", hash = "sha256:c67f6638a34a5b8730812f5101376f9d41dc38c43f1fdc35cb54700f6ed4465d", size = 380337, upload-time = "2025-02-25T16:53:14.607Z" },+ { url = "https://files.pythonhosted.org/packages/a9/86/abd17f50ee21b2248075cb6924c6e7f9d23b4925ca64ec660e869c2633f1/backrefs-5.8-py311-none-any.whl", hash = "sha256:2e1c15e4af0e12e45c8701bd5da0902d326b2e200cafcd25e49d9f06d44bb61b", size = 392142, upload-time = "2025-02-25T16:53:17.266Z" },+ { url = "https://files.pythonhosted.org/packages/b3/04/7b415bd75c8ab3268cc138c76fa648c19495fcc7d155508a0e62f3f82308/backrefs-5.8-py312-none-any.whl", hash = "sha256:bbef7169a33811080d67cdf1538c8289f76f0942ff971222a16034da88a73486", size = 398021, upload-time = "2025-02-25T16:53:26.378Z" },+ { url = "https://files.pythonhosted.org/packages/04/b8/60dcfb90eb03a06e883a92abbc2ab95c71f0d8c9dd0af76ab1d5ce0b1402/backrefs-5.8-py313-none-any.whl", hash = "sha256:e3a63b073867dbefd0536425f43db618578528e3896fb77be7141328642a1585", size = 399915, upload-time = "2025-02-25T16:53:28.167Z" },+ { url = "https://files.pythonhosted.org/packages/0c/37/fb6973edeb700f6e3d6ff222400602ab1830446c25c7b4676d8de93e65b8/backrefs-5.8-py39-none-any.whl", hash = "sha256:a66851e4533fb5b371aa0628e1fee1af05135616b86140c9d787a2ffdf4b8fdc", size = 380336, upload-time = "2025-02-25T16:53:29.858Z" },+]+[[package]]name = "boto3"version = "1.37.3"@@ -302,11 +324,11 @@ wheels = [[[package]]name = "certifi"-version = "2025.4.26"+version = "2025.6.15"source = { registry = "https://pypi.org/simple" }-sdist = { url = "https://files.pythonhosted.org/packages/e8/9e/c05b3920a3b7d20d3d3310465f50348e5b3694f4f88c6daf736eef3024c4/certifi-2025.4.26.tar.gz", hash = "sha256:0a816057ea3cdefcef70270d2c515e4506bbc954f417fa5ade2021213bb8f0c6", size = 160705, upload-time = "2025-04-26T02:12:29.51Z" }+sdist = { url = "https://files.pythonhosted.org/packages/73/f7/f14b46d4bcd21092d7d3ccef689615220d8a08fb25e564b65d20738e672e/certifi-2025.6.15.tar.gz", hash = "sha256:d747aa5a8b9bbbb1bb8c22bb13e22bd1f18e9796defa16bab421f7f7a317323b", size = 158753, upload-time = "2025-06-15T02:45:51.329Z" }wheels = [- { url = "https://files.pythonhosted.org/packages/4a/7e/3db2bd1b1f9e95f7cddca6d6e75e2f2bd9f51b1246e546d88addca0106bd/certifi-2025.4.26-py3-none-any.whl", hash = "sha256:30350364dfe371162649852c63336a15c70c6510c2ad5015b21c2345311805f3", size = 159618, upload-time = "2025-04-26T02:12:27.662Z" },+ { url = "https://files.pythonhosted.org/packages/84/ae/320161bd181fc06471eed047ecce67b693fd7515b16d495d8932db763426/certifi-2025.6.15-py3-none-any.whl", hash = "sha256:2e0c7ce7cb5d8f8634ca55d2ba7e6ec2689a2fd6537d8dec1296a477a4910057", size = 157650, upload-time = "2025-06-15T02:45:49.977Z" },][[package]]@@ -393,66 +415,66 @@ wheels = [[[package]]name = "coverage"-version = "7.9.0"-source = { registry = "https://pypi.org/simple" }-sdist = { url = "https://files.pythonhosted.org/packages/06/d1/7b18a2e0d2994e4e108dadf16580ec192e0a9c65f7456ccb82ced059f9bf/coverage-7.9.0.tar.gz", hash = "sha256:1a93b43de2233a7670a8bf2520fed8ebd5eea6a65b47417500a9d882b0533fa2", size = 813385, upload-time = "2025-06-11T23:23:34.004Z" }-wheels = [- { url = "https://files.pythonhosted.org/packages/37/25/c83935ed228bd0ce277a9a92b505a4f67b0b15ba0344680974a77452c5dd/coverage-7.9.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:3d494fa4256e3cb161ca1df14a91d2d703c27d60452eb0d4a58bb05f52f676e4", size = 211940, upload-time = "2025-06-11T23:21:47.353Z" },- { url = "https://files.pythonhosted.org/packages/36/42/c58ca1fec2a346ad12356fac955a9b6d848ab37f632a7cb1bc7476efcf90/coverage-7.9.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:b613efceeabf242978d14e1a65626ec3be67c5261918a82a985f56c2a05475ee", size = 212329, upload-time = "2025-06-11T23:21:50.216Z" },- { url = "https://files.pythonhosted.org/packages/64/0a/6b61e4348cf7b0a70f7995247cde5cc4b5ef0b61d9718109896c77d9ed0e/coverage-7.9.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:673a4d2cb7ec78e1f2f6f41039f6785f27bca0f6bc0e722b53a58286d12754e1", size = 241447, upload-time = "2025-06-11T23:21:51.757Z" },- { url = "https://files.pythonhosted.org/packages/a9/1e/5f7060b909352cba70d34be0e34619659c0ddbef426665e036d5d3046b3c/coverage-7.9.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1edc2244932e9fed92ad14428b9480a97ecd37c970333688bd35048f6472f260", size = 239322, upload-time = "2025-06-11T23:21:53.826Z" },- { url = "https://files.pythonhosted.org/packages/f5/78/f4ba669c9bf15b537136b663ccb846032cfb73e28b59458ef6899f18fe07/coverage-7.9.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ec8b92a7617faa2017bd44c94583830bab8be175722d420501680abc4f5bc794", size = 240467, upload-time = "2025-06-11T23:21:55.415Z" },- { url = "https://files.pythonhosted.org/packages/79/38/3246ea3ac68dc6f85afac0cb0362d3703647378b9882d55796c71fe83a1a/coverage-7.9.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:d8f3ca1f128f11812d3baf0a482e7f36ffb856ac1ae14de3b5d1adcfb7af955d", size = 240376, upload-time = "2025-06-11T23:21:57.108Z" },- { url = "https://files.pythonhosted.org/packages/c0/58/ef1f20afbaf9affe2941e7b077a8cf08075c6e3fe5e1dfc3160908b6a1de/coverage-7.9.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:c30eed34eb8206d9b8c2d0d9fa342fa98e10f34b1e9e1eb05f79ccbf4499c8ff", size = 239046, upload-time = "2025-06-11T23:21:58.709Z" },- { url = "https://files.pythonhosted.org/packages/09/ba/d510b05b3ca0da8fe746acf8ac815b2d560d6c4d5c4e0f6eafb2ec27dc33/coverage-7.9.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:24e6f8e5f125cd8bff33593a484a079305c9f0be911f76c6432f580ade5c1a17", size = 239318, upload-time = "2025-06-11T23:21:59.987Z" },- { url = "https://files.pythonhosted.org/packages/82/c7/328a412e3bd78c049180df3f4374bb13a332ed8731ff66f49578d5ebf98c/coverage-7.9.0-cp310-cp310-win32.whl", hash = "sha256:a1b0317b4a8ff4d3703cd7aa642b4f963a71255abe4e878659f768238fab6602", size = 214430, upload-time = "2025-06-11T23:22:01.663Z" },- { url = "https://files.pythonhosted.org/packages/db/a5/0e788cc4796989d77bfb6b1c58819edc2c65522926f0c08cfe42d1529f2b/coverage-7.9.0-cp310-cp310-win_amd64.whl", hash = "sha256:512b1ea57a11dfa23b7f3d8fe8690fcf8cd983a70ae4c2c262cf5c972618fa15", size = 215350, upload-time = "2025-06-11T23:22:02.957Z" },- { url = "https://files.pythonhosted.org/packages/9d/91/721a7df15263babfe89caf535a08bacbadebdef87338cf37d40f7400161b/coverage-7.9.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:55b7b9df45174956e0f719a56cf60c0cb4a7f155668881d00de6384e2a3402f4", size = 212055, upload-time = "2025-06-11T23:22:04.389Z" },- { url = "https://files.pythonhosted.org/packages/8d/d6/1f4c1eae67e698a8535ede02a6958a7587d06869d33a9b134ecc0e17ee07/coverage-7.9.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:87bceebbc91a58c9264c43638729fcb45910805b9f86444f93654d988305b3a2", size = 212445, upload-time = "2025-06-11T23:22:06.044Z" },- { url = "https://files.pythonhosted.org/packages/bd/48/c375a6e6a266efa2d5fbf9b04eac88c87430d1a337b4f383ea8beeeedd44/coverage-7.9.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:81da3b6e289bf9fc7dc159ab6d5222f5330ac6e94a6d06f147ba46e53fa6ec82", size = 245010, upload-time = "2025-06-11T23:22:07.367Z" },- { url = "https://files.pythonhosted.org/packages/7a/43/ec070ad02a1ee10837555a852b6fa256f8c71a953c209488e027673fc5b6/coverage-7.9.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b361684a91224d4362879c1b1802168d2435ff76666f1b7ba52fc300ad832dbc", size = 242725, upload-time = "2025-06-11T23:22:08.64Z" },- { url = "https://files.pythonhosted.org/packages/fa/ff/8b8efbd058dd59b489d9c5e27ba5766e895c396dd3bd1b78bebef9808c5f/coverage-7.9.0-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f9a384ea4f77ac0a7e36c9a805ed95ef10f423bdb68b4e9487646cdf548a6a05", size = 244527, upload-time = "2025-06-11T23:22:10.416Z" },- { url = "https://files.pythonhosted.org/packages/9d/e7/3863f458a3af009a4817656f5b56fa90c7e363d73fef338601b275e979c4/coverage-7.9.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:38a5642aa82ea6de0e4331e346f5ba188a9fdb7d727e00199f55031b85135d0a", size = 244174, upload-time = "2025-06-11T23:22:12.046Z" },- { url = "https://files.pythonhosted.org/packages/4b/f0/2ff1fa06ccd3c3d653e352b10ddeec511b018890b28dbd3c29b6ea3f742e/coverage-7.9.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:8c5ff4ca4890c0b57d3e80850534609493280c0f9e6ea2bd314b10cb8cbd76e0", size = 242227, upload-time = "2025-06-11T23:22:13.438Z" },- { url = "https://files.pythonhosted.org/packages/32/e2/bae13555436f1d0278e70cfe22a0980eab9809e89361e859c96ffa788cb9/coverage-7.9.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:cd052a0c4727ede06393da3c1df1ae6ef6c079e6bdfefb39079877404b3edc22", size = 242815, upload-time = "2025-06-11T23:22:14.723Z" },- { url = "https://files.pythonhosted.org/packages/20/7c/e1b5b3313c1e3a5e8f8ced567fee67f18c8f18cebee8af0d69052f445a55/coverage-7.9.0-cp311-cp311-win32.whl", hash = "sha256:f73fd1128165e1d665cb7f863a91d00f073044a672c7dfa04ab400af4d1a9226", size = 214469, upload-time = "2025-06-11T23:22:16.187Z" },- { url = "https://files.pythonhosted.org/packages/6a/c9/0034d3ccbb7b8f80b1ce8a927ea06e2ba265bd0ba4a9a95a83026ac78dfd/coverage-7.9.0-cp311-cp311-win_amd64.whl", hash = "sha256:dd62d62e782d3add529c8e7943f5600efd0d07dadf3819e5f9917edb4acf85d8", size = 215407, upload-time = "2025-06-11T23:22:17.611Z" },- { url = "https://files.pythonhosted.org/packages/f1/e1/7473bf679a43638c5ccba6228f45f68d33c3b7414ffae757dbb0bb2f1127/coverage-7.9.0-cp311-cp311-win_arm64.whl", hash = "sha256:f75288785cc9a67aff3b04dafd8d0f0be67306018b224d319d23867a161578d6", size = 213778, upload-time = "2025-06-11T23:22:19.217Z" },- { url = "https://files.pythonhosted.org/packages/dd/6b/7bdef79e79076c7e3303ce2453072528ed13988210fb7a8702bb3d98ea8c/coverage-7.9.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:969ed1ed0ab0325b50af3204f9024782180e64fb281f5a2952f479ec60a02aba", size = 212252, upload-time = "2025-06-11T23:22:20.662Z" },- { url = "https://files.pythonhosted.org/packages/08/fe/7e08dd50c3c3cfdbe822ee11e24da9f418983faefb4f5e52fbffae5beeb2/coverage-7.9.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:1abd41781c874e716aaeecb8b27db5f4f2bc568f2ed8d41228aa087d567674f0", size = 212491, upload-time = "2025-06-11T23:22:22.002Z" },- { url = "https://files.pythonhosted.org/packages/d4/65/9793cf61b3e4c5647e70aabd5b9470958ffd341c42f90730beeb4d21af9c/coverage-7.9.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0eb6e99487dffd28c88a4fc2ea4286beaf0207a43388775900c93e56cc5a8ae3", size = 246294, upload-time = "2025-06-11T23:22:23.297Z" },- { url = "https://files.pythonhosted.org/packages/2a/c9/fc61695132da06a34b27a49e853010a80d66a5534a1dfa770cb38aca71c0/coverage-7.9.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c425c85ddb62b32d44f83fb20044fe32edceceee1db1f978c062eec020a73ea5", size = 243311, upload-time = "2025-06-11T23:22:24.966Z" },- { url = "https://files.pythonhosted.org/packages/62/0e/559a86887580d0de390e018bddfa632ae0762eeeb065bb5557f319071527/coverage-7.9.0-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d0a1f7676bc90ceba67caa66850d689947d586f204ccf6478400c2bf39da5790", size = 245503, upload-time = "2025-06-11T23:22:26.316Z" },- { url = "https://files.pythonhosted.org/packages/45/09/344d012dc91e60b8c7afee11ffae18338780c703a5b5fb32d8d82987e7cb/coverage-7.9.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:f17055c50768d710d6abc789c9469d0353574780935e1381b83e63edc49ff530", size = 245313, upload-time = "2025-06-11T23:22:27.936Z" },- { url = "https://files.pythonhosted.org/packages/d2/2d/151b23e82aaea28aa7e3c0390d893bd1aef685866132aad36034f7d462b8/coverage-7.9.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:298d2917a6bfadbb272e08545ed026af3965e4d2fe71e3f38bf0a816818b226e", size = 243495, upload-time = "2025-06-11T23:22:29.72Z" },- { url = "https://files.pythonhosted.org/packages/74/5c/0da7fd4ad44259b4b61bd429dc642c6511314a356ffa782b924bd1ea9e5c/coverage-7.9.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:d9be5d26e5f817d478506e4d3c4ff7b92f17d980670b4791bf05baaa37ce2f88", size = 244727, upload-time = "2025-06-11T23:22:31.112Z" },- { url = "https://files.pythonhosted.org/packages/de/08/6ccf2847c5c0d8fcc153bd8f4341d89ab50c85e01a15cabe4a546d3e943e/coverage-7.9.0-cp312-cp312-win32.whl", hash = "sha256:dc2784edd9ac9fe8692fc5505667deb0b05d895c016aaaf641031ed4a5f93d53", size = 214636, upload-time = "2025-06-11T23:22:33.257Z" },- { url = "https://files.pythonhosted.org/packages/79/fa/ae2c14d49475215372772f7638c333deaaacda8f3c5717a75377d1992c82/coverage-7.9.0-cp312-cp312-win_amd64.whl", hash = "sha256:18223198464a6d5549db1934cf77a15deb24bb88652c4f5f7cb21cd3ad853704", size = 215448, upload-time = "2025-06-11T23:22:35.125Z" },- { url = "https://files.pythonhosted.org/packages/62/a9/45309219ba08b89cae84b2cb4ccfed8f941850aa7721c4914282fb3c1081/coverage-7.9.0-cp312-cp312-win_arm64.whl", hash = "sha256:3b00194ff3c84d4b821822ff6c041f245fc55d0d5c7833fc4311d082e97595e8", size = 213817, upload-time = "2025-06-11T23:22:36.557Z" },- { url = "https://files.pythonhosted.org/packages/0b/59/449eb05f795d0050007b57a4efee79b540fa6fcccad813a191351964a001/coverage-7.9.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:122c60e92ab66c9c88e17565f67a91b3b3be5617cb50f73cfd34a4c60ed4aab0", size = 212271, upload-time = "2025-06-11T23:22:38.305Z" },- { url = "https://files.pythonhosted.org/packages/e0/3b/26852a4fb719a6007b0169c1b52116ed14b61267f0bf3ba1e23db516f352/coverage-7.9.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:813c11b367a6b3cf37212ec36b230f8d086c22b69dbf62877b40939fb2c79e74", size = 212538, upload-time = "2025-06-11T23:22:39.665Z" },- { url = "https://files.pythonhosted.org/packages/f6/80/99f82896119f36984a5b9189e71c7310fc036613276560b5884b5ee890d7/coverage-7.9.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3f05e0f5e87f23d43fefe49e86655c6209dd4f9f034786b983e6803cf4554183", size = 245705, upload-time = "2025-06-11T23:22:41.103Z" },- { url = "https://files.pythonhosted.org/packages/a9/29/0b007deb096dd527c42e933129a8e4d5f9f1026f4953979c3a1e60e7ea9f/coverage-7.9.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:62f465886fa4f86d5515da525aead97c5dff13a5cf997fc4c5097a1a59e063b2", size = 242918, upload-time = "2025-06-11T23:22:42.88Z" },- { url = "https://files.pythonhosted.org/packages/6f/eb/273855b57c7fb387dd9787f250b8b333ba8c1c100877c21e32eb1b24ff29/coverage-7.9.0-cp313-cp313-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:549ea4ca901595bbe3270e1afdef98bf5d4d5791596efbdc90b00449a2bb1f91", size = 244902, upload-time = "2025-06-11T23:22:44.563Z" },- { url = "https://files.pythonhosted.org/packages/20/57/4e411b47dbfd831538ecf9e5f407e42888b0c56aedbfe0ea7b102a787559/coverage-7.9.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:8cae1d4450945c74a6a65a09864ed3eaa917055cf70aa65f83ac1b9b0d8d5f9a", size = 245069, upload-time = "2025-06-11T23:22:46.352Z" },- { url = "https://files.pythonhosted.org/packages/91/75/b24cf5703fb325fc4b1899d89984dac117b99e757b9fadd525cad7ecc020/coverage-7.9.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:d7b263910234c0d5ec913ec79ca921152fe874b805a7bcaf67118ef71708e5d2", size = 243040, upload-time = "2025-06-11T23:22:48.147Z" },- { url = "https://files.pythonhosted.org/packages/c7/e1/9495751d5315c3d76ee2c7b5dbc1935ab891d45ad585e1910a333dbdef43/coverage-7.9.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:7d7b7425215963da8f5968096a20c5b5c9af4a86a950fcc25dcc2177ab33e9e5", size = 244424, upload-time = "2025-06-11T23:22:49.574Z" },- { url = "https://files.pythonhosted.org/packages/94/2a/ee504188a586da2379939f37fdc69047d9c46d35c34d1196f2605974a17d/coverage-7.9.0-cp313-cp313-win32.whl", hash = "sha256:e7dcfa92867b0c53d2e22e985c66af946dc09e8bb13c556709e396e90a0adf5c", size = 214677, upload-time = "2025-06-11T23:22:51.394Z" },- { url = "https://files.pythonhosted.org/packages/80/2b/5eab6518643c7560fe180ba5e0f35a0be3d4fc0a88aa6601120407b1fd03/coverage-7.9.0-cp313-cp313-win_amd64.whl", hash = "sha256:aa34ca040785a2b768da489df0c036364d47a6c1c00bdd8f662b98fd3277d3d4", size = 215482, upload-time = "2025-06-11T23:22:53.151Z" },- { url = "https://files.pythonhosted.org/packages/fd/7f/9c9c8b736c4f40d7247bea8339afac40d8f6465491440608b3d73c10ffce/coverage-7.9.0-cp313-cp313-win_arm64.whl", hash = "sha256:9c5dcb5cd3c52d84c5f52045e1c87c16bf189c2fbfa57cc0d811a3b4059939df", size = 213852, upload-time = "2025-06-11T23:22:54.568Z" },- { url = "https://files.pythonhosted.org/packages/e5/83/056464aec8b360dee6f4d7a517dc5ae5a9f462ff895ff536588b42f95b2d/coverage-7.9.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:b52d2fdc1940f90c4572bd48211475a7b102f75a7f9a5e6cfc6e3da7dc380c44", size = 212994, upload-time = "2025-06-11T23:22:56.173Z" },- { url = "https://files.pythonhosted.org/packages/a3/87/f0291ecaa6baaaedbd428cf8b7e1d16b5dc010718fe7739cce955149ef83/coverage-7.9.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:4cc555a3e6ceb8841df01a4634374f5f9635e661f5c307da00bce19819e8bcdf", size = 213212, upload-time = "2025-06-11T23:22:58.051Z" },- { url = "https://files.pythonhosted.org/packages/16/a0/9eb39541774a5beb662dc4ae98fee23afb947414b6aa1443b53d2ad3ea05/coverage-7.9.0-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:244f613617876b7cd32a097788d49c952a8f1698afb25275b2a825a4e895854e", size = 256453, upload-time = "2025-06-11T23:22:59.485Z" },- { url = "https://files.pythonhosted.org/packages/93/33/d0e99f4c809334dfed20f17234080a9003a713ddb80e33ad22697a8aa8e5/coverage-7.9.0-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5c335d77539e66bc6f83e8f1ef207d038129d9b9acd9dc9f0ca42fa9eedf564a", size = 252674, upload-time = "2025-06-11T23:23:00.984Z" },- { url = "https://files.pythonhosted.org/packages/0b/3a/d2a64e7ee5eb783e44e6ca404f8fc2a45afef052ed6593afb4ce9663dae6/coverage-7.9.0-cp313-cp313t-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6b335c7077c8da7bb8173d4f9ebd90ff1a97af6a6bec4fc4e6db4856ae80b31e", size = 254830, upload-time = "2025-06-11T23:23:02.445Z" },- { url = "https://files.pythonhosted.org/packages/e2/6a/9de640f8e2b097d155532d1bc16eb9c5186fccc7c4b8148fe1dd2520875a/coverage-7.9.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:01cbc2c36895b7ab906514042c92b3fc9dd0526bf1c3251cb6aefd9c71ae6dda", size = 256060, upload-time = "2025-06-11T23:23:03.89Z" },- { url = "https://files.pythonhosted.org/packages/07/72/928fa3583b9783fc32e3dfafb6cc0cf73bdd73d1dc41e3a973f203c6aeff/coverage-7.9.0-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:1ac62880a9dff0726a193ce77a1bcdd4e8491009cb3a0510d31381e8b2c46d7a", size = 254174, upload-time = "2025-06-11T23:23:05.366Z" },- { url = "https://files.pythonhosted.org/packages/ad/01/2fd0785f8768693b748e36b442352bc26edf3391246eedcc80d480d06da1/coverage-7.9.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:95314eb306cf54af3d1147e27ba008cf78eed6f1309a1310772f4f05b12c9c65", size = 255011, upload-time = "2025-06-11T23:23:07.212Z" },- { url = "https://files.pythonhosted.org/packages/b7/49/1d0120cfa24e001e0d38795388914183c48cd86fc8640ca3b01337831917/coverage-7.9.0-cp313-cp313t-win32.whl", hash = "sha256:c5cbf3ddfb68de8dc8ce33caa9321df27297a032aeaf2e99b278f183fb4ebc37", size = 215349, upload-time = "2025-06-11T23:23:09.037Z" },- { url = "https://files.pythonhosted.org/packages/9f/48/7625c09621a206fff0b51fcbcf5d6c1162ab10a5ffa546fc132f01c9132b/coverage-7.9.0-cp313-cp313t-win_amd64.whl", hash = "sha256:e3ec9e1525eb7a0f89d31083539b398d921415d884e9f55400002a1e9fe0cf63", size = 216516, upload-time = "2025-06-11T23:23:11.083Z" },- { url = "https://files.pythonhosted.org/packages/bb/50/048b55c34985c3aafcecb32cced3abc4291969bfd967dbcaed95cfc26b2a/coverage-7.9.0-cp313-cp313t-win_arm64.whl", hash = "sha256:a02efe6769f74245ce476e89db3d4e110db07b4c0c3d3f81728e2464bbbbcb8e", size = 214308, upload-time = "2025-06-11T23:23:12.522Z" },- { url = "https://files.pythonhosted.org/packages/e8/b6/d16966f9439ccc3007e1740960d241420d6ba81502642a4be1da1672a103/coverage-7.9.0-pp39.pp310.pp311-none-any.whl", hash = "sha256:ccf1540a0e82ff525844880f988f6caaa2d037005e57bfe203b71cac7626145d", size = 203927, upload-time = "2025-06-11T23:23:30.913Z" },- { url = "https://files.pythonhosted.org/packages/70/0d/534c1e35cb7688b5c40de93fcca07e3ddc0287659ff85cd376b1dd3f770f/coverage-7.9.0-py3-none-any.whl", hash = "sha256:79ea9a26b27c963cdf541e1eb9ac05311b012bc367d0e31816f1833b06c81c02", size = 203917, upload-time = "2025-06-11T23:23:32.413Z" },+version = "7.9.1"+source = { registry = "https://pypi.org/simple" }+sdist = { url = "https://files.pythonhosted.org/packages/e7/e0/98670a80884f64578f0c22cd70c5e81a6e07b08167721c7487b4d70a7ca0/coverage-7.9.1.tar.gz", hash = "sha256:6cf43c78c4282708a28e466316935ec7489a9c487518a77fa68f716c67909cec", size = 813650, upload-time = "2025-06-13T13:02:28.627Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/c1/78/1c1c5ec58f16817c09cbacb39783c3655d54a221b6552f47ff5ac9297603/coverage-7.9.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:cc94d7c5e8423920787c33d811c0be67b7be83c705f001f7180c7b186dcf10ca", size = 212028, upload-time = "2025-06-13T13:00:29.293Z" },+ { url = "https://files.pythonhosted.org/packages/98/db/e91b9076f3a888e3b4ad7972ea3842297a52cc52e73fd1e529856e473510/coverage-7.9.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:16aa0830d0c08a2c40c264cef801db8bc4fc0e1892782e45bcacbd5889270509", size = 212420, upload-time = "2025-06-13T13:00:34.027Z" },+ { url = "https://files.pythonhosted.org/packages/0e/d0/2b3733412954576b0aea0a16c3b6b8fbe95eb975d8bfa10b07359ead4252/coverage-7.9.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cf95981b126f23db63e9dbe4cf65bd71f9a6305696fa5e2262693bc4e2183f5b", size = 241529, upload-time = "2025-06-13T13:00:35.786Z" },+ { url = "https://files.pythonhosted.org/packages/b3/00/5e2e5ae2e750a872226a68e984d4d3f3563cb01d1afb449a17aa819bc2c4/coverage-7.9.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f05031cf21699785cd47cb7485f67df619e7bcdae38e0fde40d23d3d0210d3c3", size = 239403, upload-time = "2025-06-13T13:00:37.399Z" },+ { url = "https://files.pythonhosted.org/packages/37/3b/a2c27736035156b0a7c20683afe7df498480c0dfdf503b8c878a21b6d7fb/coverage-7.9.1-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bb4fbcab8764dc072cb651a4bcda4d11fb5658a1d8d68842a862a6610bd8cfa3", size = 240548, upload-time = "2025-06-13T13:00:39.647Z" },+ { url = "https://files.pythonhosted.org/packages/98/f5/13d5fc074c3c0e0dc80422d9535814abf190f1254d7c3451590dc4f8b18c/coverage-7.9.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:0f16649a7330ec307942ed27d06ee7e7a38417144620bb3d6e9a18ded8a2d3e5", size = 240459, upload-time = "2025-06-13T13:00:40.934Z" },+ { url = "https://files.pythonhosted.org/packages/36/24/24b9676ea06102df824c4a56ffd13dc9da7904478db519efa877d16527d5/coverage-7.9.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:cea0a27a89e6432705fffc178064503508e3c0184b4f061700e771a09de58187", size = 239128, upload-time = "2025-06-13T13:00:42.343Z" },+ { url = "https://files.pythonhosted.org/packages/be/05/242b7a7d491b369ac5fee7908a6e5ba42b3030450f3ad62c645b40c23e0e/coverage-7.9.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:e980b53a959fa53b6f05343afbd1e6f44a23ed6c23c4b4c56c6662bbb40c82ce", size = 239402, upload-time = "2025-06-13T13:00:43.634Z" },+ { url = "https://files.pythonhosted.org/packages/73/e0/4de7f87192fa65c9c8fbaeb75507e124f82396b71de1797da5602898be32/coverage-7.9.1-cp310-cp310-win32.whl", hash = "sha256:70760b4c5560be6ca70d11f8988ee6542b003f982b32f83d5ac0b72476607b70", size = 214518, upload-time = "2025-06-13T13:00:45.622Z" },+ { url = "https://files.pythonhosted.org/packages/d5/ab/5e4e2fe458907d2a65fab62c773671cfc5ac704f1e7a9ddd91996f66e3c2/coverage-7.9.1-cp310-cp310-win_amd64.whl", hash = "sha256:a66e8f628b71f78c0e0342003d53b53101ba4e00ea8dabb799d9dba0abbbcebe", size = 215436, upload-time = "2025-06-13T13:00:47.245Z" },+ { url = "https://files.pythonhosted.org/packages/60/34/fa69372a07d0903a78ac103422ad34db72281c9fc625eba94ac1185da66f/coverage-7.9.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:95c765060e65c692da2d2f51a9499c5e9f5cf5453aeaf1420e3fc847cc060582", size = 212146, upload-time = "2025-06-13T13:00:48.496Z" },+ { url = "https://files.pythonhosted.org/packages/27/f0/da1894915d2767f093f081c42afeba18e760f12fdd7a2f4acbe00564d767/coverage-7.9.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:ba383dc6afd5ec5b7a0d0c23d38895db0e15bcba7fb0fa8901f245267ac30d86", size = 212536, upload-time = "2025-06-13T13:00:51.535Z" },+ { url = "https://files.pythonhosted.org/packages/10/d5/3fc33b06e41e390f88eef111226a24e4504d216ab8e5d1a7089aa5a3c87a/coverage-7.9.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:37ae0383f13cbdcf1e5e7014489b0d71cc0106458878ccde52e8a12ced4298ed", size = 245092, upload-time = "2025-06-13T13:00:52.883Z" },+ { url = "https://files.pythonhosted.org/packages/0a/39/7aa901c14977aba637b78e95800edf77f29f5a380d29768c5b66f258305b/coverage-7.9.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:69aa417a030bf11ec46149636314c24c8d60fadb12fc0ee8f10fda0d918c879d", size = 242806, upload-time = "2025-06-13T13:00:54.571Z" },+ { url = "https://files.pythonhosted.org/packages/43/fc/30e5cfeaf560b1fc1989227adedc11019ce4bb7cce59d65db34fe0c2d963/coverage-7.9.1-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0a4be2a28656afe279b34d4f91c3e26eccf2f85500d4a4ff0b1f8b54bf807338", size = 244610, upload-time = "2025-06-13T13:00:56.932Z" },+ { url = "https://files.pythonhosted.org/packages/bf/15/cca62b13f39650bc87b2b92bb03bce7f0e79dd0bf2c7529e9fc7393e4d60/coverage-7.9.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:382e7ddd5289f140259b610e5f5c58f713d025cb2f66d0eb17e68d0a94278875", size = 244257, upload-time = "2025-06-13T13:00:58.545Z" },+ { url = "https://files.pythonhosted.org/packages/cd/1a/c0f2abe92c29e1464dbd0ff9d56cb6c88ae2b9e21becdb38bea31fcb2f6c/coverage-7.9.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:e5532482344186c543c37bfad0ee6069e8ae4fc38d073b8bc836fc8f03c9e250", size = 242309, upload-time = "2025-06-13T13:00:59.836Z" },+ { url = "https://files.pythonhosted.org/packages/57/8d/c6fd70848bd9bf88fa90df2af5636589a8126d2170f3aade21ed53f2b67a/coverage-7.9.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:a39d18b3f50cc121d0ce3838d32d58bd1d15dab89c910358ebefc3665712256c", size = 242898, upload-time = "2025-06-13T13:01:02.506Z" },+ { url = "https://files.pythonhosted.org/packages/c2/9e/6ca46c7bff4675f09a66fe2797cd1ad6a24f14c9c7c3b3ebe0470a6e30b8/coverage-7.9.1-cp311-cp311-win32.whl", hash = "sha256:dd24bd8d77c98557880def750782df77ab2b6885a18483dc8588792247174b32", size = 214561, upload-time = "2025-06-13T13:01:04.012Z" },+ { url = "https://files.pythonhosted.org/packages/a1/30/166978c6302010742dabcdc425fa0f938fa5a800908e39aff37a7a876a13/coverage-7.9.1-cp311-cp311-win_amd64.whl", hash = "sha256:6b55ad10a35a21b8015eabddc9ba31eb590f54adc9cd39bcf09ff5349fd52125", size = 215493, upload-time = "2025-06-13T13:01:05.702Z" },+ { url = "https://files.pythonhosted.org/packages/60/07/a6d2342cd80a5be9f0eeab115bc5ebb3917b4a64c2953534273cf9bc7ae6/coverage-7.9.1-cp311-cp311-win_arm64.whl", hash = "sha256:6ad935f0016be24c0e97fc8c40c465f9c4b85cbbe6eac48934c0dc4d2568321e", size = 213869, upload-time = "2025-06-13T13:01:09.345Z" },+ { url = "https://files.pythonhosted.org/packages/68/d9/7f66eb0a8f2fce222de7bdc2046ec41cb31fe33fb55a330037833fb88afc/coverage-7.9.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:a8de12b4b87c20de895f10567639c0797b621b22897b0af3ce4b4e204a743626", size = 212336, upload-time = "2025-06-13T13:01:10.909Z" },+ { url = "https://files.pythonhosted.org/packages/20/20/e07cb920ef3addf20f052ee3d54906e57407b6aeee3227a9c91eea38a665/coverage-7.9.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:5add197315a054e92cee1b5f686a2bcba60c4c3e66ee3de77ace6c867bdee7cb", size = 212571, upload-time = "2025-06-13T13:01:12.518Z" },+ { url = "https://files.pythonhosted.org/packages/78/f8/96f155de7e9e248ca9c8ff1a40a521d944ba48bec65352da9be2463745bf/coverage-7.9.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:600a1d4106fe66f41e5d0136dfbc68fe7200a5cbe85610ddf094f8f22e1b0300", size = 246377, upload-time = "2025-06-13T13:01:14.87Z" },+ { url = "https://files.pythonhosted.org/packages/3e/cf/1d783bd05b7bca5c10ded5f946068909372e94615a4416afadfe3f63492d/coverage-7.9.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2a876e4c3e5a2a1715a6608906aa5a2e0475b9c0f68343c2ada98110512ab1d8", size = 243394, upload-time = "2025-06-13T13:01:16.23Z" },+ { url = "https://files.pythonhosted.org/packages/02/dd/e7b20afd35b0a1abea09fb3998e1abc9f9bd953bee548f235aebd2b11401/coverage-7.9.1-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:81f34346dd63010453922c8e628a52ea2d2ccd73cb2487f7700ac531b247c8a5", size = 245586, upload-time = "2025-06-13T13:01:17.532Z" },+ { url = "https://files.pythonhosted.org/packages/4e/38/b30b0006fea9d617d1cb8e43b1bc9a96af11eff42b87eb8c716cf4d37469/coverage-7.9.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:888f8eee13f2377ce86d44f338968eedec3291876b0b8a7289247ba52cb984cd", size = 245396, upload-time = "2025-06-13T13:01:19.164Z" },+ { url = "https://files.pythonhosted.org/packages/31/e4/4d8ec1dc826e16791f3daf1b50943e8e7e1eb70e8efa7abb03936ff48418/coverage-7.9.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:9969ef1e69b8c8e1e70d591f91bbc37fc9a3621e447525d1602801a24ceda898", size = 243577, upload-time = "2025-06-13T13:01:22.433Z" },+ { url = "https://files.pythonhosted.org/packages/25/f4/b0e96c5c38e6e40ef465c4bc7f138863e2909c00e54a331da335faf0d81a/coverage-7.9.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:60c458224331ee3f1a5b472773e4a085cc27a86a0b48205409d364272d67140d", size = 244809, upload-time = "2025-06-13T13:01:24.143Z" },+ { url = "https://files.pythonhosted.org/packages/8a/65/27e0a1fa5e2e5079bdca4521be2f5dabf516f94e29a0defed35ac2382eb2/coverage-7.9.1-cp312-cp312-win32.whl", hash = "sha256:5f646a99a8c2b3ff4c6a6e081f78fad0dde275cd59f8f49dc4eab2e394332e74", size = 214724, upload-time = "2025-06-13T13:01:25.435Z" },+ { url = "https://files.pythonhosted.org/packages/9b/a8/d5b128633fd1a5e0401a4160d02fa15986209a9e47717174f99dc2f7166d/coverage-7.9.1-cp312-cp312-win_amd64.whl", hash = "sha256:30f445f85c353090b83e552dcbbdad3ec84c7967e108c3ae54556ca69955563e", size = 215535, upload-time = "2025-06-13T13:01:27.861Z" },+ { url = "https://files.pythonhosted.org/packages/a3/37/84bba9d2afabc3611f3e4325ee2c6a47cd449b580d4a606b240ce5a6f9bf/coverage-7.9.1-cp312-cp312-win_arm64.whl", hash = "sha256:af41da5dca398d3474129c58cb2b106a5d93bbb196be0d307ac82311ca234342", size = 213904, upload-time = "2025-06-13T13:01:29.202Z" },+ { url = "https://files.pythonhosted.org/packages/d0/a7/a027970c991ca90f24e968999f7d509332daf6b8c3533d68633930aaebac/coverage-7.9.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:31324f18d5969feef7344a932c32428a2d1a3e50b15a6404e97cba1cc9b2c631", size = 212358, upload-time = "2025-06-13T13:01:30.909Z" },+ { url = "https://files.pythonhosted.org/packages/f2/48/6aaed3651ae83b231556750280682528fea8ac7f1232834573472d83e459/coverage-7.9.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:0c804506d624e8a20fb3108764c52e0eef664e29d21692afa375e0dd98dc384f", size = 212620, upload-time = "2025-06-13T13:01:32.256Z" },+ { url = "https://files.pythonhosted.org/packages/6c/2a/f4b613f3b44d8b9f144847c89151992b2b6b79cbc506dee89ad0c35f209d/coverage-7.9.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ef64c27bc40189f36fcc50c3fb8f16ccda73b6a0b80d9bd6e6ce4cffcd810bbd", size = 245788, upload-time = "2025-06-13T13:01:33.948Z" },+ { url = "https://files.pythonhosted.org/packages/04/d2/de4fdc03af5e4e035ef420ed26a703c6ad3d7a07aff2e959eb84e3b19ca8/coverage-7.9.1-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d4fe2348cc6ec372e25adec0219ee2334a68d2f5222e0cba9c0d613394e12d86", size = 243001, upload-time = "2025-06-13T13:01:35.285Z" },+ { url = "https://files.pythonhosted.org/packages/f5/e8/eed18aa5583b0423ab7f04e34659e51101135c41cd1dcb33ac1d7013a6d6/coverage-7.9.1-cp313-cp313-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:34ed2186fe52fcc24d4561041979a0dec69adae7bce2ae8d1c49eace13e55c43", size = 244985, upload-time = "2025-06-13T13:01:36.712Z" },+ { url = "https://files.pythonhosted.org/packages/17/f8/ae9e5cce8885728c934eaa58ebfa8281d488ef2afa81c3dbc8ee9e6d80db/coverage-7.9.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:25308bd3d00d5eedd5ae7d4357161f4df743e3c0240fa773ee1b0f75e6c7c0f1", size = 245152, upload-time = "2025-06-13T13:01:39.303Z" },+ { url = "https://files.pythonhosted.org/packages/5a/c8/272c01ae792bb3af9b30fac14d71d63371db227980682836ec388e2c57c0/coverage-7.9.1-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:73e9439310f65d55a5a1e0564b48e34f5369bee943d72c88378f2d576f5a5751", size = 243123, upload-time = "2025-06-13T13:01:40.727Z" },+ { url = "https://files.pythonhosted.org/packages/8c/d0/2819a1e3086143c094ab446e3bdf07138527a7b88cb235c488e78150ba7a/coverage-7.9.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:37ab6be0859141b53aa89412a82454b482c81cf750de4f29223d52268a86de67", size = 244506, upload-time = "2025-06-13T13:01:42.184Z" },+ { url = "https://files.pythonhosted.org/packages/8b/4e/9f6117b89152df7b6112f65c7a4ed1f2f5ec8e60c4be8f351d91e7acc848/coverage-7.9.1-cp313-cp313-win32.whl", hash = "sha256:64bdd969456e2d02a8b08aa047a92d269c7ac1f47e0c977675d550c9a0863643", size = 214766, upload-time = "2025-06-13T13:01:44.482Z" },+ { url = "https://files.pythonhosted.org/packages/27/0f/4b59f7c93b52c2c4ce7387c5a4e135e49891bb3b7408dcc98fe44033bbe0/coverage-7.9.1-cp313-cp313-win_amd64.whl", hash = "sha256:be9e3f68ca9edb897c2184ad0eee815c635565dbe7a0e7e814dc1f7cbab92c0a", size = 215568, upload-time = "2025-06-13T13:01:45.772Z" },+ { url = "https://files.pythonhosted.org/packages/09/1e/9679826336f8c67b9c39a359352882b24a8a7aee48d4c9cad08d38d7510f/coverage-7.9.1-cp313-cp313-win_arm64.whl", hash = "sha256:1c503289ffef1d5105d91bbb4d62cbe4b14bec4d13ca225f9c73cde9bb46207d", size = 213939, upload-time = "2025-06-13T13:01:47.087Z" },+ { url = "https://files.pythonhosted.org/packages/bb/5b/5c6b4e7a407359a2e3b27bf9c8a7b658127975def62077d441b93a30dbe8/coverage-7.9.1-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:0b3496922cb5f4215bf5caaef4cf12364a26b0be82e9ed6d050f3352cf2d7ef0", size = 213079, upload-time = "2025-06-13T13:01:48.554Z" },+ { url = "https://files.pythonhosted.org/packages/a2/22/1e2e07279fd2fd97ae26c01cc2186e2258850e9ec125ae87184225662e89/coverage-7.9.1-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:9565c3ab1c93310569ec0d86b017f128f027cab0b622b7af288696d7ed43a16d", size = 213299, upload-time = "2025-06-13T13:01:49.997Z" },+ { url = "https://files.pythonhosted.org/packages/14/c0/4c5125a4b69d66b8c85986d3321520f628756cf524af810baab0790c7647/coverage-7.9.1-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2241ad5dbf79ae1d9c08fe52b36d03ca122fb9ac6bca0f34439e99f8327ac89f", size = 256535, upload-time = "2025-06-13T13:01:51.314Z" },+ { url = "https://files.pythonhosted.org/packages/81/8b/e36a04889dda9960be4263e95e777e7b46f1bb4fc32202612c130a20c4da/coverage-7.9.1-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3bb5838701ca68b10ebc0937dbd0eb81974bac54447c55cd58dea5bca8451029", size = 252756, upload-time = "2025-06-13T13:01:54.403Z" },+ { url = "https://files.pythonhosted.org/packages/98/82/be04eff8083a09a4622ecd0e1f31a2c563dbea3ed848069e7b0445043a70/coverage-7.9.1-cp313-cp313t-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b30a25f814591a8c0c5372c11ac8967f669b97444c47fd794926e175c4047ece", size = 254912, upload-time = "2025-06-13T13:01:56.769Z" },+ { url = "https://files.pythonhosted.org/packages/0f/25/c26610a2c7f018508a5ab958e5b3202d900422cf7cdca7670b6b8ca4e8df/coverage-7.9.1-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:2d04b16a6062516df97969f1ae7efd0de9c31eb6ebdceaa0d213b21c0ca1a683", size = 256144, upload-time = "2025-06-13T13:01:58.19Z" },+ { url = "https://files.pythonhosted.org/packages/c5/8b/fb9425c4684066c79e863f1e6e7ecebb49e3a64d9f7f7860ef1688c56f4a/coverage-7.9.1-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:7931b9e249edefb07cd6ae10c702788546341d5fe44db5b6108a25da4dca513f", size = 254257, upload-time = "2025-06-13T13:01:59.645Z" },+ { url = "https://files.pythonhosted.org/packages/93/df/27b882f54157fc1131e0e215b0da3b8d608d9b8ef79a045280118a8f98fe/coverage-7.9.1-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:52e92b01041151bf607ee858e5a56c62d4b70f4dac85b8c8cb7fb8a351ab2c10", size = 255094, upload-time = "2025-06-13T13:02:01.37Z" },+ { url = "https://files.pythonhosted.org/packages/41/5f/cad1c3dbed8b3ee9e16fa832afe365b4e3eeab1fb6edb65ebbf745eabc92/coverage-7.9.1-cp313-cp313t-win32.whl", hash = "sha256:684e2110ed84fd1ca5f40e89aa44adf1729dc85444004111aa01866507adf363", size = 215437, upload-time = "2025-06-13T13:02:02.905Z" },+ { url = "https://files.pythonhosted.org/packages/99/4d/fad293bf081c0e43331ca745ff63673badc20afea2104b431cdd8c278b4c/coverage-7.9.1-cp313-cp313t-win_amd64.whl", hash = "sha256:437c576979e4db840539674e68c84b3cda82bc824dd138d56bead1435f1cb5d7", size = 216605, upload-time = "2025-06-13T13:02:05.638Z" },+ { url = "https://files.pythonhosted.org/packages/1f/56/4ee027d5965fc7fc126d7ec1187529cc30cc7d740846e1ecb5e92d31b224/coverage-7.9.1-cp313-cp313t-win_arm64.whl", hash = "sha256:18a0912944d70aaf5f399e350445738a1a20b50fbea788f640751c2ed9208b6c", size = 214392, upload-time = "2025-06-13T13:02:07.642Z" },+ { url = "https://files.pythonhosted.org/packages/3e/e5/c723545c3fd3204ebde3b4cc4b927dce709d3b6dc577754bb57f63ca4a4a/coverage-7.9.1-pp39.pp310.pp311-none-any.whl", hash = "sha256:db0f04118d1db74db6c9e1cb1898532c7dcc220f1d2718f058601f7c3f499514", size = 204009, upload-time = "2025-06-13T13:02:25.787Z" },+ { url = "https://files.pythonhosted.org/packages/08/b8/7ddd1e8ba9701dea08ce22029917140e6f66a859427406579fd8d0ca7274/coverage-7.9.1-py3-none-any.whl", hash = "sha256:66b974b145aa189516b6bf2d8423e888b742517d37872f6ee4c5be0073bd9a3c", size = 204000, upload-time = "2025-06-13T13:02:27.173Z" },][package.optional-dependencies]@@ -645,6 +667,18 @@ wheels = [{ url = "https://files.pythonhosted.org/packages/ee/45/b82e3c16be2182bff01179db177fe144d58b5dc787a7d4492c6ed8b9317f/frozenlist-1.7.0-py3-none-any.whl", hash = "sha256:9a5af342e34f7e97caf8c995864c7a396418ae2859cc6fdf1b1073020d516a7e", size = 13106, upload-time = "2025-06-09T23:02:34.204Z" },]+[[package]]+name = "ghp-import"+version = "2.1.0"+source = { registry = "https://pypi.org/simple" }+dependencies = [+ { name = "python-dateutil" },+]+sdist = { url = "https://files.pythonhosted.org/packages/d9/29/d40217cbe2f6b1359e00c6c307bb3fc876ba74068cbab3dde77f03ca0dc4/ghp-import-2.1.0.tar.gz", hash = "sha256:9c535c4c61193c2df8871222567d7fd7e5014d835f97dc7b7439069e2413d343", size = 10943, upload-time = "2022-05-02T15:47:16.11Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/f7/ec/67fbef5d497f86283db54c22eec6f6140243aae73265799baaaa19cd17fb/ghp_import-2.1.0-py3-none-any.whl", hash = "sha256:8337dd7b50877f163d4c0289bc1f1c7f127550241988d568c1db512c4324a619", size = 11034, upload-time = "2022-05-02T15:47:14.552Z" },+]+[[package]]name = "google-auth"version = "2.40.3"@@ -953,6 +987,15 @@ wheels = [{ url = "https://files.pythonhosted.org/packages/b3/38/89ba8ad64ae25be8de66a6d463314cf1eb366222074cfda9ee839c56a4b4/mdurl-0.1.2-py3-none-any.whl", hash = "sha256:84008a41e51615a49fc9966191ff91509e3c40b939176e643fd50a5c2196b8f8", size = 9979, upload-time = "2022-08-14T12:40:09.779Z" },]+[[package]]+name = "mergedeep"+version = "1.3.4"+source = { registry = "https://pypi.org/simple" }+sdist = { url = "https://files.pythonhosted.org/packages/3a/41/580bb4006e3ed0361b8151a01d324fb03f420815446c7def45d02f74c270/mergedeep-1.3.4.tar.gz", hash = "sha256:0096d52e9dad9939c3d975a774666af186eda617e6ca84df4c94dec30004f2a8", size = 4661, upload-time = "2021-02-05T18:55:30.623Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/2c/19/04f9b178c2d8a15b076c8b5140708fa6ffc5601fb6f1e975537072df5b2a/mergedeep-1.3.4-py3-none-any.whl", hash = "sha256:70775750742b25c0d8f36c55aed03d24c3384d17c951b3175d898bd778ef0307", size = 6354, upload-time = "2021-02-05T18:55:29.583Z" },+]+[[package]]name = "mistralai"version = "1.5.2"@@ -970,6 +1013,97 @@ wheels = [{ url = "https://files.pythonhosted.org/packages/7c/97/5b428225ca4524b9722c8e1b2812c35f958ec5bb6a58c274c6c07a136da8/mistralai-1.5.2-py3-none-any.whl", hash = "sha256:5b1112acebbcad1afd7732ce0bd60614975b64999801c555c54768ac41f506ae", size = 278149, upload-time = "2025-03-19T18:40:28.232Z" },]+[[package]]+name = "mkdocs"+version = "1.6.1"+source = { registry = "https://pypi.org/simple" }+dependencies = [+ { name = "click" },+ { name = "colorama", marker = "sys_platform == 'win32'" },+ { name = "ghp-import" },+ { name = "jinja2" },+ { name = "markdown" },+ { name = "markupsafe" },+ { name = "mergedeep" },+ { name = "mkdocs-get-deps" },+ { name = "packaging" },+ { name = "pathspec" },+ { name = "pyyaml" },+ { name = "pyyaml-env-tag" },+ { name = "watchdog" },+]+sdist = { url = "https://files.pythonhosted.org/packages/bc/c6/bbd4f061bd16b378247f12953ffcb04786a618ce5e904b8c5a01a0309061/mkdocs-1.6.1.tar.gz", hash = "sha256:7b432f01d928c084353ab39c57282f29f92136665bdd6abf7c1ec8d822ef86f2", size = 3889159, upload-time = "2024-08-30T12:24:06.899Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/22/5b/dbc6a8cddc9cfa9c4971d59fb12bb8d42e161b7e7f8cc89e49137c5b279c/mkdocs-1.6.1-py3-none-any.whl", hash = "sha256:db91759624d1647f3f34aa0c3f327dd2601beae39a366d6e064c03468d35c20e", size = 3864451, upload-time = "2024-08-30T12:24:05.054Z" },+]++[[package]]+name = "mkdocs-get-deps"+version = "0.2.0"+source = { registry = "https://pypi.org/simple" }+dependencies = [+ { name = "mergedeep" },+ { name = "platformdirs" },+ { name = "pyyaml" },+]+sdist = { url = "https://files.pythonhosted.org/packages/98/f5/ed29cd50067784976f25ed0ed6fcd3c2ce9eb90650aa3b2796ddf7b6870b/mkdocs_get_deps-0.2.0.tar.gz", hash = "sha256:162b3d129c7fad9b19abfdcb9c1458a651628e4b1dea628ac68790fb3061c60c", size = 10239, upload-time = "2023-11-20T17:51:09.981Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/9f/d4/029f984e8d3f3b6b726bd33cafc473b75e9e44c0f7e80a5b29abc466bdea/mkdocs_get_deps-0.2.0-py3-none-any.whl", hash = "sha256:2bf11d0b133e77a0dd036abeeb06dec8775e46efa526dc70667d8863eefc6134", size = 9521, upload-time = "2023-11-20T17:51:08.587Z" },+]++[[package]]+name = "mkdocs-glightbox"+version = "0.4.0"+source = { registry = "https://pypi.org/simple" }+sdist = { url = "https://files.pythonhosted.org/packages/86/5a/0bc456397ba0acc684b5b1daa4ca232ed717938fd37198251d8bcc4053bf/mkdocs-glightbox-0.4.0.tar.gz", hash = "sha256:392b34207bf95991071a16d5f8916d1d2f2cd5d5bb59ae2997485ccd778c70d9", size = 32010, upload-time = "2024-05-06T14:31:43.063Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/c1/72/b0c2128bb569c732c11ae8e49a777089e77d83c05946062caa19b841e6fb/mkdocs_glightbox-0.4.0-py3-none-any.whl", hash = "sha256:e0107beee75d3eb7380ac06ea2d6eac94c999eaa49f8c3cbab0e7be2ac006ccf", size = 31154, upload-time = "2024-05-06T14:31:41.011Z" },+]++[[package]]+name = "mkdocs-material"+version = "9.6.14"+source = { registry = "https://pypi.org/simple" }+dependencies = [+ { name = "babel" },+ { name = "backrefs" },+ { name = "colorama" },+ { name = "jinja2" },+ { name = "markdown" },+ { name = "mkdocs" },+ { name = "mkdocs-material-extensions" },+ { name = "paginate" },+ { name = "pygments" },+ { name = "pymdown-extensions" },+ { name = "requests" },+]+sdist = { url = "https://files.pythonhosted.org/packages/b3/fa/0101de32af88f87cf5cc23ad5f2e2030d00995f74e616306513431b8ab4b/mkdocs_material-9.6.14.tar.gz", hash = "sha256:39d795e90dce6b531387c255bd07e866e027828b7346d3eba5ac3de265053754", size = 3951707, upload-time = "2025-05-13T13:27:57.173Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/3a/a1/7fdb959ad592e013c01558822fd3c22931a95a0f08cf0a7c36da13a5b2b5/mkdocs_material-9.6.14-py3-none-any.whl", hash = "sha256:3b9cee6d3688551bf7a8e8f41afda97a3c39a12f0325436d76c86706114b721b", size = 8703767, upload-time = "2025-05-13T13:27:54.089Z" },+]++[[package]]+name = "mkdocs-material-extensions"+version = "1.3.1"+source = { registry = "https://pypi.org/simple" }+sdist = { url = "https://files.pythonhosted.org/packages/79/9b/9b4c96d6593b2a541e1cb8b34899a6d021d208bb357042823d4d2cabdbe7/mkdocs_material_extensions-1.3.1.tar.gz", hash = "sha256:10c9511cea88f568257f960358a467d12b970e1f7b2c0e5fb2bb48cab1928443", size = 11847, upload-time = "2023-11-22T19:09:45.208Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/5b/54/662a4743aa81d9582ee9339d4ffa3c8fd40a4965e033d77b9da9774d3960/mkdocs_material_extensions-1.3.1-py3-none-any.whl", hash = "sha256:adff8b62700b25cb77b53358dad940f3ef973dd6db797907c49e3c2ef3ab4e31", size = 8728, upload-time = "2023-11-22T19:09:43.465Z" },+]++[[package]]+name = "mkdocs-meta-manager"+version = "1.1.0"+source = { registry = "https://pypi.org/simple" }+dependencies = [+ { name = "jinja2" },+ { name = "mkdocs" },+]+sdist = { url = "https://files.pythonhosted.org/packages/18/9b/8884e1b8d9eb31bb012bb4309393ddb75162247905cc11826597a705f654/mkdocs_meta_manager-1.1.0.tar.gz", hash = "sha256:545d5976e291b684538421b6ec69b77e1027d6ff6d43c5ade71bcc0a226c9faa", size = 2708, upload-time = "2025-01-20T21:58:03.888Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/fc/5b/0a90ae6576cd7fa2593e7bdde4fa52fe4e79f5812b8f1ada001414aaaa54/mkdocs_meta_manager-1.1.0-py3-none-any.whl", hash = "sha256:bdd1b5e9403bcfe222720dcd93102e0f1df5befef9b40a6ff91ad67926528658", size = 3186, upload-time = "2025-01-20T21:58:02.314Z" },+]+[[package]]name = "multidict"version = "6.4.4"@@ -1325,6 +1459,15 @@ wheels = [{ url = "https://files.pythonhosted.org/packages/20/12/38679034af332785aac8774540895e234f4d07f7545804097de4b666afd8/packaging-25.0-py3-none-any.whl", hash = "sha256:29572ef2b1f17581046b3a2227d5c611fb25ec70ca1ba8554b24b0e69331a484", size = 66469, upload-time = "2025-04-19T11:48:57.875Z" },]+[[package]]+name = "paginate"+version = "0.5.7"+source = { registry = "https://pypi.org/simple" }+sdist = { url = "https://files.pythonhosted.org/packages/ec/46/68dde5b6bc00c1296ec6466ab27dddede6aec9af1b99090e1107091b3b84/paginate-0.5.7.tar.gz", hash = "sha256:22bd083ab41e1a8b4f3690544afb2c60c25e5c9a63a30fa2f483f6c60c8e5945", size = 19252, upload-time = "2024-08-25T14:17:24.139Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/90/96/04b8e52da071d28f5e21a805b19cb9390aa17a47462ac87f5e2696b9566d/paginate-0.5.7-py2.py3-none-any.whl", hash = "sha256:b885e2af73abcf01d9559fd5216b57ef722f8c42affbb63942377668e35c7591", size = 13746, upload-time = "2024-08-25T14:17:22.55Z" },+]+[[package]]name = "pandas"version = "2.3.0"@@ -1519,6 +1662,10 @@ compat = []dev = [{ name = "boto3-stubs" },+ { name = "mkdocs" },+ { name = "mkdocs-glightbox" },+ { name = "mkdocs-material" },+ { name = "mkdocs-meta-manager" },{ name = "mypy" },{ name = "pandas-stubs" },{ name = "pyright" },@@ -1566,6 +1713,10 @@ requires-dist = [{ name = "kajson", specifier = ">=0.1.5" },{ name = "markdown", specifier = ">=3.6" },{ name = "mistralai", marker = "extra == 'mistralai'", specifier = "==1.5.2" },+ { name = "mkdocs", marker = "extra == 'dev'", specifier = "==1.6.1" },+ { name = "mkdocs-glightbox", marker = "extra == 'dev'", specifier = "==0.4.0" },+ { name = "mkdocs-material", marker = "extra == 'dev'", specifier = "==9.6.14" },+ { name = "mkdocs-meta-manager", marker = "extra == 'dev'", specifier = "==1.1.0" },{ name = "mypy", marker = "extra == 'dev'", specifier = ">=1.11.2" },{ name = "networkx", specifier = ">=3.4.2" },{ name = "openai", specifier = ">=1.60.1" },@@ -1603,6 +1754,15 @@ requires-dist = []provides-extras = ["anthropic", "bedrock", "fal", "google", "mistralai", "compat", "dev"]+[[package]]+name = "platformdirs"+version = "4.3.8"+source = { registry = "https://pypi.org/simple" }+sdist = { url = "https://files.pythonhosted.org/packages/fe/8b/3c73abc9c759ecd3f1f7ceff6685840859e8070c4d947c93fae71f6a0bf2/platformdirs-4.3.8.tar.gz", hash = "sha256:3d512d96e16bcb959a814c9f348431070822a6496326a4be0911c40b5a74c2bc", size = 21362, upload-time = "2025-05-07T22:47:42.121Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/fe/39/979e8e21520d4e47a0bbe349e2713c0aac6f3d853d0e5b34d76206c439aa/platformdirs-4.3.8-py3-none-any.whl", hash = "sha256:ff7059bb7eb1179e2685604f4aaf157cfd9535242bd23742eadc3c13542139b4", size = 18567, upload-time = "2025-05-07T22:47:40.376Z" },+]+[[package]]name = "pluggy"version = "1.6.0"@@ -1833,6 +1993,19 @@ wheels = [{ url = "https://files.pythonhosted.org/packages/8a/0b/9fcc47d19c48b59121088dd6da2488a49d5f72dacf8262e2790a1d2c7d15/pygments-2.19.1-py3-none-any.whl", hash = "sha256:9ea1544ad55cecf4b8242fab6dd35a93bbce657034b0611ee383099054ab6d8c", size = 1225293, upload-time = "2025-01-06T17:26:25.553Z" },]+[[package]]+name = "pymdown-extensions"+version = "10.15"+source = { registry = "https://pypi.org/simple" }+dependencies = [+ { name = "markdown" },+ { name = "pyyaml" },+]+sdist = { url = "https://files.pythonhosted.org/packages/08/92/a7296491dbf5585b3a987f3f3fc87af0e632121ff3e490c14b5f2d2b4eb5/pymdown_extensions-10.15.tar.gz", hash = "sha256:0e5994e32155f4b03504f939e501b981d306daf7ec2aa1cd2eb6bd300784f8f7", size = 852320, upload-time = "2025-04-27T23:48:29.183Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/a7/d1/c54e608505776ce4e7966d03358ae635cfd51dff1da6ee421c090dbc797b/pymdown_extensions-10.15-py3-none-any.whl", hash = "sha256:46e99bb272612b0de3b7e7caf6da8dd5f4ca5212c0b273feb9304e236c484e5f", size = 265845, upload-time = "2025-04-27T23:48:27.359Z" },+]+[[package]]name = "pypdfium2"version = "4.30.1"@@ -2011,6 +2184,18 @@ wheels = [{ url = "https://files.pythonhosted.org/packages/fa/de/02b54f42487e3d3c6efb3f89428677074ca7bf43aae402517bc7cca949f3/PyYAML-6.0.2-cp313-cp313-win_amd64.whl", hash = "sha256:8388ee1976c416731879ac16da0aff3f63b286ffdd57cdeb95f3f2e085687563", size = 156446, upload-time = "2024-08-06T20:33:04.33Z" },]+[[package]]+name = "pyyaml-env-tag"+version = "1.1"+source = { registry = "https://pypi.org/simple" }+dependencies = [+ { name = "pyyaml" },+]+sdist = { url = "https://files.pythonhosted.org/packages/eb/2e/79c822141bfd05a853236b504869ebc6b70159afc570e1d5a20641782eaa/pyyaml_env_tag-1.1.tar.gz", hash = "sha256:2eb38b75a2d21ee0475d6d97ec19c63287a7e140231e4214969d0eac923cd7ff", size = 5737, upload-time = "2025-05-13T15:24:01.64Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/04/11/432f32f8097b03e3cd5fe57e88efb685d964e2e5178a48ed61e841f7fdce/pyyaml_env_tag-1.1-py3-none-any.whl", hash = "sha256:17109e1a528561e32f026364712fee1264bc2ea6715120891174ed1b980d2e04", size = 4722, upload-time = "2025-05-13T15:23:59.629Z" },+]+[[package]]name = "requests"version = "2.32.4"@@ -2256,15 +2441,15 @@ bedrock-runtime = [[[package]]name = "types-aiobotocore"-version = "2.22.0"+version = "2.23.0"source = { registry = "https://pypi.org/simple" }dependencies = [{ name = "botocore-stubs" },{ name = "typing-extensions", marker = "python_full_version < '3.12'" },]-sdist = { url = "https://files.pythonhosted.org/packages/ad/e1/2ed95566257867d8916a6abec506ab7fc8ed4985a768e5578eb163ef3329/types_aiobotocore-2.22.0.tar.gz", hash = "sha256:04c6acd984f9a37d22db2ebf670ff84e55f6f227d3cc71e9b297720d4a71b718", size = 84173, upload-time = "2025-05-02T01:46:52.502Z" }+sdist = { url = "https://files.pythonhosted.org/packages/d6/2b/1e039d61079d9f4fd9011eff2aab0ecd1ec1fe16eb9a1c08720b4686248f/types_aiobotocore-2.23.0.tar.gz", hash = "sha256:d363a584726b9582e18cadbed9f88053ac4b488373e0e643d292b57b2f182ac8", size = 84697, upload-time = "2025-06-13T01:59:52.009Z" }wheels = [- { url = "https://files.pythonhosted.org/packages/63/fe/a4f672a70b964e902af18f7eafbad5e793710fe5726040218b953f4141e4/types_aiobotocore-2.22.0-py3-none-any.whl", hash = "sha256:54ffcf8d0144d71fa8b540f3997ae1a555ff3a61200d0c13b3db1c29dec730e5", size = 51827, upload-time = "2025-05-02T01:46:49.449Z" },+ { url = "https://files.pythonhosted.org/packages/4c/4b/821a589e7d02f2ceef9e627139cf0c6c8b554d8e7c44117006f1f2a46c5d/types_aiobotocore-2.23.0-py3-none-any.whl", hash = "sha256:f62d9a58d730bbde825355feedb045713b6de2feab4c1b68b071afd1da60dea8", size = 52122, upload-time = "2025-06-13T01:59:46.918Z" },][[package]]@@ -2449,6 +2634,38 @@ wheels = [{ url = "https://files.pythonhosted.org/packages/6b/11/cc635220681e93a0183390e26485430ca2c7b5f9d33b15c74c2861cb8091/urllib3-2.4.0-py3-none-any.whl", hash = "sha256:4e16665048960a0900c702d4a66415956a584919c03361cac9f1df5c5dd7e813", size = 128680, upload-time = "2025-04-10T15:23:37.377Z" },]+[[package]]+name = "watchdog"+version = "6.0.0"+source = { registry = "https://pypi.org/simple" }+sdist = { url = "https://files.pythonhosted.org/packages/db/7d/7f3d619e951c88ed75c6037b246ddcf2d322812ee8ea189be89511721d54/watchdog-6.0.0.tar.gz", hash = "sha256:9ddf7c82fda3ae8e24decda1338ede66e1c99883db93711d8fb941eaa2d8c282", size = 131220, upload-time = "2024-11-01T14:07:13.037Z" }+wheels = [+ { url = "https://files.pythonhosted.org/packages/0c/56/90994d789c61df619bfc5ce2ecdabd5eeff564e1eb47512bd01b5e019569/watchdog-6.0.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:d1cdb490583ebd691c012b3d6dae011000fe42edb7a82ece80965b42abd61f26", size = 96390, upload-time = "2024-11-01T14:06:24.793Z" },+ { url = "https://files.pythonhosted.org/packages/55/46/9a67ee697342ddf3c6daa97e3a587a56d6c4052f881ed926a849fcf7371c/watchdog-6.0.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:bc64ab3bdb6a04d69d4023b29422170b74681784ffb9463ed4870cf2f3e66112", size = 88389, upload-time = "2024-11-01T14:06:27.112Z" },+ { url = "https://files.pythonhosted.org/packages/44/65/91b0985747c52064d8701e1075eb96f8c40a79df889e59a399453adfb882/watchdog-6.0.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:c897ac1b55c5a1461e16dae288d22bb2e412ba9807df8397a635d88f671d36c3", size = 89020, upload-time = "2024-11-01T14:06:29.876Z" },+ { url = "https://files.pythonhosted.org/packages/e0/24/d9be5cd6642a6aa68352ded4b4b10fb0d7889cb7f45814fb92cecd35f101/watchdog-6.0.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:6eb11feb5a0d452ee41f824e271ca311a09e250441c262ca2fd7ebcf2461a06c", size = 96393, upload-time = "2024-11-01T14:06:31.756Z" },+ { url = "https://files.pythonhosted.org/packages/63/7a/6013b0d8dbc56adca7fdd4f0beed381c59f6752341b12fa0886fa7afc78b/watchdog-6.0.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:ef810fbf7b781a5a593894e4f439773830bdecb885e6880d957d5b9382a960d2", size = 88392, upload-time = "2024-11-01T14:06:32.99Z" },+ { url = "https://files.pythonhosted.org/packages/d1/40/b75381494851556de56281e053700e46bff5b37bf4c7267e858640af5a7f/watchdog-6.0.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:afd0fe1b2270917c5e23c2a65ce50c2a4abb63daafb0d419fde368e272a76b7c", size = 89019, upload-time = "2024-11-01T14:06:34.963Z" },+ { url = "https://files.pythonhosted.org/packages/39/ea/3930d07dafc9e286ed356a679aa02d777c06e9bfd1164fa7c19c288a5483/watchdog-6.0.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:bdd4e6f14b8b18c334febb9c4425a878a2ac20efd1e0b231978e7b150f92a948", size = 96471, upload-time = "2024-11-01T14:06:37.745Z" },+ { url = "https://files.pythonhosted.org/packages/12/87/48361531f70b1f87928b045df868a9fd4e253d9ae087fa4cf3f7113be363/watchdog-6.0.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:c7c15dda13c4eb00d6fb6fc508b3c0ed88b9d5d374056b239c4ad1611125c860", size = 88449, upload-time = "2024-11-01T14:06:39.748Z" },+ { url = "https://files.pythonhosted.org/packages/5b/7e/8f322f5e600812e6f9a31b75d242631068ca8f4ef0582dd3ae6e72daecc8/watchdog-6.0.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:6f10cb2d5902447c7d0da897e2c6768bca89174d0c6e1e30abec5421af97a5b0", size = 89054, upload-time = "2024-11-01T14:06:41.009Z" },+ { url = "https://files.pythonhosted.org/packages/68/98/b0345cabdce2041a01293ba483333582891a3bd5769b08eceb0d406056ef/watchdog-6.0.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:490ab2ef84f11129844c23fb14ecf30ef3d8a6abafd3754a6f75ca1e6654136c", size = 96480, upload-time = "2024-11-01T14:06:42.952Z" },+ { url = "https://files.pythonhosted.org/packages/85/83/cdf13902c626b28eedef7ec4f10745c52aad8a8fe7eb04ed7b1f111ca20e/watchdog-6.0.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:76aae96b00ae814b181bb25b1b98076d5fc84e8a53cd8885a318b42b6d3a5134", size = 88451, upload-time = "2024-11-01T14:06:45.084Z" },+ { url = "https://files.pythonhosted.org/packages/fe/c4/225c87bae08c8b9ec99030cd48ae9c4eca050a59bf5c2255853e18c87b50/watchdog-6.0.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:a175f755fc2279e0b7312c0035d52e27211a5bc39719dd529625b1930917345b", size = 89057, upload-time = "2024-11-01T14:06:47.324Z" },+ { url = "https://files.pythonhosted.org/packages/30/ad/d17b5d42e28a8b91f8ed01cb949da092827afb9995d4559fd448d0472763/watchdog-6.0.0-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:c7ac31a19f4545dd92fc25d200694098f42c9a8e391bc00bdd362c5736dbf881", size = 87902, upload-time = "2024-11-01T14:06:53.119Z" },+ { url = "https://files.pythonhosted.org/packages/5c/ca/c3649991d140ff6ab67bfc85ab42b165ead119c9e12211e08089d763ece5/watchdog-6.0.0-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:9513f27a1a582d9808cf21a07dae516f0fab1cf2d7683a742c498b93eedabb11", size = 88380, upload-time = "2024-11-01T14:06:55.19Z" },+ { url = "https://files.pythonhosted.org/packages/a9/c7/ca4bf3e518cb57a686b2feb4f55a1892fd9a3dd13f470fca14e00f80ea36/watchdog-6.0.0-py3-none-manylinux2014_aarch64.whl", hash = "sha256:7607498efa04a3542ae3e05e64da8202e58159aa1fa4acddf7678d34a35d4f13", size = 79079, upload-time = "2024-11-01T14:06:59.472Z" },+ { url = "https://files.pythonhosted.org/packages/5c/51/d46dc9332f9a647593c947b4b88e2381c8dfc0942d15b8edc0310fa4abb1/watchdog-6.0.0-py3-none-manylinux2014_armv7l.whl", hash = "sha256:9041567ee8953024c83343288ccc458fd0a2d811d6a0fd68c4c22609e3490379", size = 79078, upload-time = "2024-11-01T14:07:01.431Z" },+ { url = "https://files.pythonhosted.org/packages/d4/57/04edbf5e169cd318d5f07b4766fee38e825d64b6913ca157ca32d1a42267/watchdog-6.0.0-py3-none-manylinux2014_i686.whl", hash = "sha256:82dc3e3143c7e38ec49d61af98d6558288c415eac98486a5c581726e0737c00e", size = 79076, upload-time = "2024-11-01T14:07:02.568Z" },+ { url = "https://files.pythonhosted.org/packages/ab/cc/da8422b300e13cb187d2203f20b9253e91058aaf7db65b74142013478e66/watchdog-6.0.0-py3-none-manylinux2014_ppc64.whl", hash = "sha256:212ac9b8bf1161dc91bd09c048048a95ca3a4c4f5e5d4a7d1b1a7d5752a7f96f", size = 79077, upload-time = "2024-11-01T14:07:03.893Z" },+ { url = "https://files.pythonhosted.org/packages/2c/3b/b8964e04ae1a025c44ba8e4291f86e97fac443bca31de8bd98d3263d2fcf/watchdog-6.0.0-py3-none-manylinux2014_ppc64le.whl", hash = "sha256:e3df4cbb9a450c6d49318f6d14f4bbc80d763fa587ba46ec86f99f9e6876bb26", size = 79078, upload-time = "2024-11-01T14:07:05.189Z" },+ { url = "https://files.pythonhosted.org/packages/62/ae/a696eb424bedff7407801c257d4b1afda455fe40821a2be430e173660e81/watchdog-6.0.0-py3-none-manylinux2014_s390x.whl", hash = "sha256:2cce7cfc2008eb51feb6aab51251fd79b85d9894e98ba847408f662b3395ca3c", size = 79077, upload-time = "2024-11-01T14:07:06.376Z" },+ { url = "https://files.pythonhosted.org/packages/b5/e8/dbf020b4d98251a9860752a094d09a65e1b436ad181faf929983f697048f/watchdog-6.0.0-py3-none-manylinux2014_x86_64.whl", hash = "sha256:20ffe5b202af80ab4266dcd3e91aae72bf2da48c0d33bdb15c66658e685e94e2", size = 79078, upload-time = "2024-11-01T14:07:07.547Z" },+ { url = "https://files.pythonhosted.org/packages/07/f6/d0e5b343768e8bcb4cda79f0f2f55051bf26177ecd5651f84c07567461cf/watchdog-6.0.0-py3-none-win32.whl", hash = "sha256:07df1fdd701c5d4c8e55ef6cf55b8f0120fe1aef7ef39a1c6fc6bc2e606d517a", size = 79065, upload-time = "2024-11-01T14:07:09.525Z" },+ { url = "https://files.pythonhosted.org/packages/db/d9/c495884c6e548fce18a8f40568ff120bc3a4b7b99813081c8ac0c936fa64/watchdog-6.0.0-py3-none-win_amd64.whl", hash = "sha256:cbafb470cf848d93b5d013e2ecb245d4aa1c8fd0504e863ccefa32445359d680", size = 79070, upload-time = "2024-11-01T14:07:10.686Z" },+ { url = "https://files.pythonhosted.org/packages/33/e8/e40370e6d74ddba47f002a32919d91310d6074130fe4e17dabcafc15cbf1/watchdog-6.0.0-py3-none-win_ia64.whl", hash = "sha256:a1914259fa9e1454315171103c6a30961236f508b9b623eae470268bbcc6a22f", size = 79067, upload-time = "2024-11-01T14:07:11.845Z" },+]+[[package]]name = "wrapt"version = "1.17.2"