Skip to content

LLM-Driven Extraction of Unstructured Data — Built for API Deployments & ETL Pipeline Workflows

feat: Structure tool and prompt-service error handling improvments (for SDK v0.71) (#1277)

Chandrasekharan M35ce5600c272
diff --git a/backend/prompt_studio/prompt_studio_core_v2/prompt_studio_helper.py b/backend/prompt_studio/prompt_studio_core_v2/prompt_studio_helper.py
index 056f3a785..5e45837ed 100644
--- a/backend/prompt_studio/prompt_studio_core_v2/prompt_studio_helper.py
+++ b/backend/prompt_studio/prompt_studio_core_v2/prompt_studio_helper.py
@@ -952,25 +952,24 @@ class PromptStudioHelper:
TSPKeys.EXECUTION_SOURCE: ExecutionSource.IDE.value,
}
- responder = PromptTool(
- tool=util,
- prompt_host=settings.PROMPT_HOST,
- prompt_port=settings.PROMPT_PORT,
- )
- include_metadata = {TSPKeys.INCLUDE_METADATA: True}
- headers = {Common.X_REQUEST_ID: StateStore.get(Common.REQUEST_ID)}
- answer = responder.answer_prompt(
- payload=payload, params=include_metadata, headers=headers
- )
- if answer["status"] == "ERROR":
- error_message = answer.get("error", "")
+ try:
+ responder = PromptTool(
+ tool=util,
+ prompt_host=settings.PROMPT_HOST,
+ prompt_port=settings.PROMPT_PORT,
+ request_id=StateStore.get(Common.REQUEST_ID),
+ )
+ params = {TSPKeys.INCLUDE_METADATA: True}
+ return responder.answer_prompt(payload=payload, params=params)
+ except SdkError as e:
+ msg = str(e)
+ if e.actual_err and hasattr(e.actual_err, "response"):
+ msg = e.actual_err.response.json().get("error", str(e))
raise AnswerFetchError(
"Error while fetching response for "
- f"'{prompt.prompt_key}' with '{doc_name}'. {error_message}",
- status_code=int(answer.get("status_code")),
+ f"'{prompt.prompt_key}' with '{doc_name}'. {msg}",
+ status_code=int(e.status_code or 500),
)
- output_response = json.loads(answer["structure_output"])
- return output_response
@staticmethod
def fetch_table_settings_if_enabled(
@@ -1101,22 +1100,22 @@ class PromptStudioHelper:
util = PromptIdeBaseTool(log_level=LogLevel.INFO, org_id=org_id)
- responder = PromptTool(
- tool=util,
- prompt_host=settings.PROMPT_HOST,
- prompt_port=settings.PROMPT_PORT,
- )
- headers = {Common.X_REQUEST_ID: StateStore.get(Common.REQUEST_ID)}
- response = responder.index(payload=payload, headers=headers)
-
- status_code = response.get("status_code")
- if status_code == 200:
- doc_id = json.loads(response.get("structure_output")).get("doc_id")
- else:
- error_message = f"Failed to index '{filename}'. " + response.get(
- "error", ""
+ try:
+ responder = PromptTool(
+ tool=util,
+ prompt_host=settings.PROMPT_HOST,
+ prompt_port=settings.PROMPT_PORT,
+ request_id=StateStore.get(Common.REQUEST_ID),
+ )
+ doc_id = responder.index(payload=payload)
+ except SdkError as e:
+ msg = str(e)
+ if e.actual_err and hasattr(e.actual_err, "response"):
+ msg = e.actual_err.response.json().get("error", str(e))
+ raise IndexingAPIError(
+ f"Failed to index '{filename}'. {msg}",
+ status_code=int(e.status_code or 500),
)
- raise IndexingAPIError(error_message)
PromptStudioIndexHelper.handle_index_manager(
document_id=document_id,
@@ -1128,15 +1127,19 @@ class PromptStudioHelper:
)
return {"status": IndexingStatus.COMPLETED_STATUS.value, "output": doc_id}
except (IndexingError, IndexingAPIError, SdkError) as e:
- logger.error(f"Indexing failed : {e} ", stack_info=True, exc_info=True)
- doc_name = os.path.split(file_path)[1]
+ msg = str(e)
+ if isinstance(e, SdkError) and hasattr(e.actual_err, "response"):
+ msg = e.actual_err.response.json().get("error", str(e))
+
+ msg = f"Error while indexing '{filename}'. {msg}"
+ logger.error(msg, stack_info=True, exc_info=True)
PromptStudioHelper._publish_log(
- {"tool_id": tool_id, "run_id": run_id, "doc_name": doc_name},
+ {"tool_id": tool_id, "run_id": run_id, "doc_name": filename},
LogLevels.ERROR,
LogLevels.RUN,
- f"Indexing failed : {e}",
+ msg,
)
- raise IndexingAPIError(f"Error while indexing '{doc_name}'. {str(e)}") from e
+ raise IndexingAPIError(msg) from e
@staticmethod
def _fetch_single_pass_response(
@@ -1254,22 +1257,10 @@ class PromptStudioHelper:
tool=util,
prompt_host=settings.PROMPT_HOST,
prompt_port=settings.PROMPT_PORT,
+ request_id=StateStore.get(Common.REQUEST_ID),
)
- include_metadata = {TSPKeys.INCLUDE_METADATA: True}
- headers = {Common.X_REQUEST_ID: StateStore.get(Common.REQUEST_ID)}
- answer = responder.single_pass_extraction(
- payload=payload,
- params=include_metadata,
- headers=headers,
- )
- if answer["status"] == "ERROR":
- error_message = answer.get("error", None)
- logger.info(f"{str(answer)}")
- raise AnswerFetchError(
- f"Error while fetching response for prompt(s). {error_message}"
- )
- output_response = json.loads(answer["structure_output"])
- return output_response
+ params = {TSPKeys.INCLUDE_METADATA: True}
+ return responder.single_pass_extraction(payload=payload, params=params)
@staticmethod
def get_tool_from_tool_id(tool_id: str) -> CustomTool | None:
@@ -1334,27 +1325,26 @@ class PromptStudioHelper:
util = PromptIdeBaseTool(log_level=LogLevel.INFO, org_id=org_id)
- responder = PromptTool(
- tool=util,
- prompt_host=settings.PROMPT_HOST,
- prompt_port=settings.PROMPT_PORT,
- )
- headers = {Common.X_REQUEST_ID: StateStore.get(Common.REQUEST_ID)}
- response = responder.extract(payload=payload, headers=headers)
- status_code = response.get("status_code")
- if status_code == 200:
- response_data = response.get("structure_output")
- structure_output = json.loads(response_data)
- extracted_text = structure_output.get("extracted_text")
+ try:
+ responder = PromptTool(
+ tool=util,
+ prompt_host=settings.PROMPT_HOST,
+ prompt_port=settings.PROMPT_PORT,
+ request_id=StateStore.get(Common.REQUEST_ID),
+ )
+ extracted_text = responder.extract(payload=payload)
PromptStudioIndexHelper.mark_extraction_status(
document_id=document_id,
profile_manager=profile_manager,
doc_id=doc_id,
)
- else:
- error_message = f"Failed to extract '{filename}'. " + response.get(
- "error", ""
+ except SdkError as e:
+ msg = str(e)
+ if e.actual_err and hasattr(e.actual_err, "response"):
+ msg = e.actual_err.response.json().get("error", str(e))
+ raise ExtractionAPIError(
+ f"Failed to extract '{filename}'. {msg}",
+ status_code=int(e.status_code or 500),
)
- raise ExtractionAPIError(error_message)
return extracted_text
diff --git a/backend/pyproject.toml b/backend/pyproject.toml
index bacfded4e..5971e97f7 100644
--- a/backend/pyproject.toml
+++ b/backend/pyproject.toml
@@ -37,7 +37,7 @@ dependencies = [
"social-auth-core==4.4.2", # For OAuth
# TODO: Temporarily removing the extra dependencies of aws and gcs from unstract-sdk
# to resolve lock file. Will have to be re-looked into
- "unstract-sdk~=0.70.0",
+ "unstract-sdk~=0.71.0",
"gcsfs==2024.10.0",
"s3fs==2024.10.0",
"azure-identity==1.16.0",
@@ -63,7 +63,6 @@ dev = [
"unstract-tool-sandbox",
"unstract-workflow-execution",
"unstract-filesystem",
- "unstract-sdk~=0.70.0rc1",
# For file watching
"inotify>=0.2.10",
"poethepoet>=0.33.1",
diff --git a/backend/sample.env b/backend/sample.env
index c5c6f53c6..e6dd7250c 100644
--- a/backend/sample.env
+++ b/backend/sample.env
@@ -75,9 +75,9 @@ PROMPT_STUDIO_FILE_PATH=/app/prompt-studio-data
# Structure Tool Image (Runs prompt studio exported tools)
# https://hub.docker.com/r/unstract/tool-structure
-STRUCTURE_TOOL_IMAGE_URL="docker:unstract/tool-structure:0.0.77"
+STRUCTURE_TOOL_IMAGE_URL="docker:unstract/tool-structure:0.0.78"
STRUCTURE_TOOL_IMAGE_NAME="unstract/tool-structure"
-STRUCTURE_TOOL_IMAGE_TAG="0.0.77"
+STRUCTURE_TOOL_IMAGE_TAG="0.0.78"
# Feature Flags
EVALUATION_SERVER_IP=unstract-flipt
diff --git a/backend/uv.lock b/backend/uv.lock
index b8b82e46b..ba36766f0 100644
--- a/backend/uv.lock
+++ b/backend/uv.lock
@@ -3890,7 +3890,6 @@ dev = [
{ name = "unstract-core" },
{ name = "unstract-filesystem" },
{ name = "unstract-flags" },
- { name = "unstract-sdk" },
{ name = "unstract-tool-registry" },
{ name = "unstract-tool-sandbox" },
{ name = "unstract-workflow-execution" },
@@ -3933,7 +3932,7 @@ requires-dist = [
{ name = "unstract-core", editable = "../unstract/core" },
{ name = "unstract-filesystem", editable = "../unstract/filesystem" },
{ name = "unstract-flags", editable = "../unstract/flags" },
- { name = "unstract-sdk", specifier = "~=0.70.0" },
+ { name = "unstract-sdk", specifier = "~=0.71.0" },
{ name = "unstract-tool-registry", editable = "../unstract/tool-registry" },
{ name = "unstract-tool-sandbox", editable = "../unstract/tool-sandbox" },
{ name = "unstract-workflow-execution", editable = "../unstract/workflow-execution" },
@@ -3952,7 +3951,6 @@ dev = [
{ name = "unstract-core", editable = "../unstract/core" },
{ name = "unstract-filesystem", editable = "../unstract/filesystem" },
{ name = "unstract-flags", editable = "../unstract/flags" },
- { name = "unstract-sdk", specifier = "~=0.70.0rc1" },
{ name = "unstract-tool-registry", editable = "../unstract/tool-registry" },
{ name = "unstract-tool-sandbox", editable = "../unstract/tool-sandbox" },
{ name = "unstract-workflow-execution", editable = "../unstract/workflow-execution" },
@@ -4037,7 +4035,7 @@ dependencies = [
]
[package.metadata]
-requires-dist = [{ name = "unstract-sdk", specifier = "~=0.70.0" }]
+requires-dist = [{ name = "unstract-sdk", specifier = "~=0.71.0" }]
[[package]]
name = "unstract-flags"
@@ -4056,7 +4054,7 @@ requires-dist = [
[[package]]
name = "unstract-sdk"
-version = "0.70.0"
+version = "0.71.0"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "filetype" },
@@ -4095,9 +4093,9 @@ dependencies = [
{ name = "tiktoken" },
{ name = "transformers" },
]
-sdist = { url = "https://files.pythonhosted.org/packages/00/f7/e88481d199847003a173c8355b4664fa8b272cfe0b99ba425703e75c4a70/unstract_sdk-0.70.0.tar.gz", hash = "sha256:98ce28e8b1e88b4c43ebc1fbc947cafc9efae0b7914043fcd969061169947d1d", size = 2356341 }
+sdist = { url = "https://files.pythonhosted.org/packages/d0/12/efecbf0581040b1bfc44fe939a68ba07e45c65c16b3e91666227e94b52be/unstract_sdk-0.71.0.tar.gz", hash = "sha256:c9269cbd1656e40a03445d3758bc51d7921273af729e0d3813f5f9e28967e31d", size = 2360756 }
wheels = [
- { url = "https://files.pythonhosted.org/packages/21/a3/92240dd060544348d0d45b5560d064aa63e8c2daee25159dce533b1a137b/unstract_sdk-0.70.0-py3-none-any.whl", hash = "sha256:583f6ec793594d3113379f175883597a8af7d58e9faab5d4ff02f31bf8bf3510", size = 264025 },
+ { url = "https://files.pythonhosted.org/packages/3a/88/4ba63cd8611d7bdaf36619e9fe1b19effb5bc10a13ec50b79bf714aa410a/unstract_sdk-0.71.0-py3-none-any.whl", hash = "sha256:0d3bdaf79468f03991f457d040b07ab5d159be3a601b4bfae7f772083d70b705", size = 265087 },
]
[[package]]
@@ -4119,7 +4117,7 @@ requires-dist = [
{ name = "jsonschema", specifier = "~=4.18.2" },
{ name = "pyyaml", specifier = "~=6.0.1" },
{ name = "unstract-flags", editable = "../unstract/flags" },
- { name = "unstract-sdk", specifier = "~=0.70.0" },
+ { name = "unstract-sdk", specifier = "~=0.71.0" },
{ name = "unstract-tool-sandbox", editable = "../unstract/tool-sandbox" },
]
diff --git a/prompt-service/pyproject.toml b/prompt-service/pyproject.toml
index cc2757859..d5a41769d 100644
--- a/prompt-service/pyproject.toml
+++ b/prompt-service/pyproject.toml
@@ -16,7 +16,7 @@ dependencies = [
"json-repair~=0.42.0",
# TODO: Temporarily removing the extra dependencies of aws and gcs from unstract-sdk
# to resolve lock file. Will have to be re-looked into
- "unstract-sdk~=0.70.0",
+ "unstract-sdk~=0.71.0",
"gcsfs==2024.10.0",
"s3fs==2024.10.0",
"redis>=5.0.3,<5.3",
diff --git a/prompt-service/uv.lock b/prompt-service/uv.lock
index 5b134d083..22c87f64f 100644
--- a/prompt-service/uv.lock
+++ b/prompt-service/uv.lock
@@ -3053,7 +3053,7 @@ requires-dist = [
{ name = "s3fs", specifier = "==2024.10.0" },
{ name = "unstract-core", editable = "../unstract/core" },
{ name = "unstract-flags", editable = "../unstract/flags" },
- { name = "unstract-sdk", specifier = "~=0.70.0" },
+ { name = "unstract-sdk", specifier = "~=0.71.0" },
]
[package.metadata.requires-dev]
@@ -3073,7 +3073,7 @@ test = [
[[package]]
name = "unstract-sdk"
-version = "0.70.0"
+version = "0.71.0"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "filetype" },
@@ -3112,9 +3112,9 @@ dependencies = [
{ name = "tiktoken" },
{ name = "transformers" },
]
-sdist = { url = "https://files.pythonhosted.org/packages/00/f7/e88481d199847003a173c8355b4664fa8b272cfe0b99ba425703e75c4a70/unstract_sdk-0.70.0.tar.gz", hash = "sha256:98ce28e8b1e88b4c43ebc1fbc947cafc9efae0b7914043fcd969061169947d1d", size = 2356341 }
+sdist = { url = "https://files.pythonhosted.org/packages/d0/12/efecbf0581040b1bfc44fe939a68ba07e45c65c16b3e91666227e94b52be/unstract_sdk-0.71.0.tar.gz", hash = "sha256:c9269cbd1656e40a03445d3758bc51d7921273af729e0d3813f5f9e28967e31d", size = 2360756 }
wheels = [
- { url = "https://files.pythonhosted.org/packages/21/a3/92240dd060544348d0d45b5560d064aa63e8c2daee25159dce533b1a137b/unstract_sdk-0.70.0-py3-none-any.whl", hash = "sha256:583f6ec793594d3113379f175883597a8af7d58e9faab5d4ff02f31bf8bf3510", size = 264025 },
+ { url = "https://files.pythonhosted.org/packages/3a/88/4ba63cd8611d7bdaf36619e9fe1b19effb5bc10a13ec50b79bf714aa410a/unstract_sdk-0.71.0-py3-none-any.whl", hash = "sha256:0d3bdaf79468f03991f457d040b07ab5d159be3a601b4bfae7f772083d70b705", size = 265087 },
]
[[package]]
diff --git a/pyproject.toml b/pyproject.toml
index e02bb6023..f174a4597 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -37,7 +37,7 @@ hook-check-django-migrations = [
"psycopg2-binary==2.9.9",
"python-dotenv==1.0.0",
"python-magic==0.4.27",
- "unstract-sdk~=0.70.0",
+ "unstract-sdk~=0.71.0",
"unstract-connectors",
"unstract-core",
"unstract-flags",
diff --git a/tools/classifier/requirements.txt b/tools/classifier/requirements.txt
index 6b9ef80b6..271fe4332 100644
--- a/tools/classifier/requirements.txt
+++ b/tools/classifier/requirements.txt
@@ -4,4 +4,4 @@
# aws alone is needed here
# because tools use transient temporary storage.
-unstract-sdk[aws]~=0.70.0
+unstract-sdk[aws]~=0.71.0
diff --git a/tools/classifier/src/config/properties.json b/tools/classifier/src/config/properties.json
index 203265bdc..4a8a77143 100644
--- a/tools/classifier/src/config/properties.json
+++ b/tools/classifier/src/config/properties.json
@@ -2,7 +2,7 @@
"schemaVersion": "0.0.1",
"displayName": "File Classifier",
"functionName": "classify",
- "toolVersion": "0.0.58",
+ "toolVersion": "0.0.59",
"description": "Classifies a file into a bin based on its contents",
"input": {
"description": "File to be classified"
diff --git a/tools/structure/requirements.txt b/tools/structure/requirements.txt
index 6df69121a..e3c5821b3 100644
--- a/tools/structure/requirements.txt
+++ b/tools/structure/requirements.txt
@@ -3,4 +3,4 @@
# Required for all unstract tools
# aws alone is needed here
# because tools use transient temporary storage.
-unstract-sdk[aws]~=0.70.0
+unstract-sdk[aws]~=0.71.0
diff --git a/tools/structure/src/config/properties.json b/tools/structure/src/config/properties.json
index 26102158a..b96a5c5c9 100644
--- a/tools/structure/src/config/properties.json
+++ b/tools/structure/src/config/properties.json
@@ -2,7 +2,7 @@
"schemaVersion": "0.0.1",
"displayName": "Structure Tool",
"functionName": "structure_tool",
- "toolVersion": "0.0.77",
+ "toolVersion": "0.0.78",
"description": "This is a template tool which can answer set of input prompts designed in the Prompt Studio",
"input": {
"description": "File that needs to be indexed and parsed for answers"
diff --git a/tools/structure/src/helpers.py b/tools/structure/src/helpers.py
index 7462228cb..ae05be02d 100644
--- a/tools/structure/src/helpers.py
+++ b/tools/structure/src/helpers.py
@@ -1,5 +1,5 @@
import datetime
-import json
+import logging
from typing import Any
from constants import IndexingConstants as IKeys
@@ -8,6 +8,8 @@ from constants import SettingsKeys # type: ignore [attr-defined]
from unstract.sdk.prompt import PromptTool
from unstract.sdk.tool.base import BaseTool
+logger = logging.getLogger(__name__)
+
class StructureToolHelper:
@staticmethod
@@ -22,7 +24,6 @@ class StructureToolHelper:
execution_run_data_folder: str,
) -> str:
x2text = tool_settings[SettingsKeys.X2TEXT_ADAPTER]
- tool.stream_log(f"Extracting text from {file_path} into {extract_file_path}")
payload = {
IKeys.X2TEXT_INSTANCE_ID: x2text,
IKeys.FILE_PATH: file_path,
@@ -36,18 +37,15 @@ class StructureToolHelper:
IKeys.EXECUTION_DATA_DIR: str(execution_run_data_folder),
}
- tool.stream_log(f"Payload constructed : {payload}")
- responder = PromptTool(
+ logger.info(f"Prompt service payload for text extraction:\n{payload}")
+
+ prompt_tool = PromptTool(
tool=tool,
prompt_host=tool.get_env_or_die(SettingsKeys.PROMPT_HOST),
prompt_port=tool.get_env_or_die(SettingsKeys.PROMPT_PORT),
+ request_id=run_id,
)
- tool.stream_log(f"responder : {responder}")
- response = responder.extract(payload=payload)
- response_data = response.get("structure_output")
- structure_output = json.loads(response_data)
- extracted_text = structure_output.get("extracted_text")
- return extracted_text
+ return prompt_tool.extract(payload=payload)
@staticmethod
def dynamic_indexing(
@@ -87,13 +85,16 @@ class StructureToolHelper:
IKeys.EXTRACTED_TEXT: extracted_text,
}
+ sensitive_keys = [IKeys.EXTRACTED_TEXT]
+ payload_to_log = {k: v for k, v in payload.items() if k not in sensitive_keys}
+ logger.info(f"Prompt service payload for indexing:\n{payload_to_log}")
responder = PromptTool(
tool=tool,
prompt_host=tool.get_env_or_die(SettingsKeys.PROMPT_HOST),
prompt_port=tool.get_env_or_die(SettingsKeys.PROMPT_PORT),
+ request_id=run_id,
)
- doc_id = responder.index(payload=payload)
- return doc_id
+ return responder.index(payload=payload)
@staticmethod
def elapsed_time(start_time) -> float:
diff --git a/tools/structure/src/main.py b/tools/structure/src/main.py
index 189b5506c..15e1a2987 100644
--- a/tools/structure/src/main.py
+++ b/tools/structure/src/main.py
@@ -10,7 +10,8 @@ from constants import SettingsKeys # type: ignore [attr-defined]
from helpers import StructureToolHelper as STHelper
from utils import json_to_markdown
-from unstract.sdk.constants import LogLevel, LogState, MetadataKey, ToolEnv, UsageKwargs
+from unstract.sdk.constants import LogState, MetadataKey, ToolEnv, UsageKwargs
+from unstract.sdk.platform import PlatformHelper
from unstract.sdk.prompt import PromptTool
from unstract.sdk.tool.base import BaseTool
from unstract.sdk.tool.entrypoint import ToolEntrypoint
@@ -48,30 +49,45 @@ class StructureTool(BaseTool):
tool=self,
prompt_port=self.get_env_or_die(SettingsKeys.PROMPT_PORT),
prompt_host=self.get_env_or_die(SettingsKeys.PROMPT_HOST),
+ request_id=self.file_execution_id,
)
self.stream_log(
f"Fetching prompt studio exported tool with UUID '{prompt_registry_id}'"
)
try:
- exported_tool = responder.get_exported_tool(
- tool=self, prompt_registry_id=prompt_registry_id
+ platform_helper: PromptTool = PlatformHelper(
+ tool=self,
+ platform_port=self.get_env_or_die(ToolEnv.PLATFORM_PORT),
+ platform_host=self.get_env_or_die(ToolEnv.PLATFORM_HOST),
+ request_id=self.file_execution_id,
+ )
+ exported_tool = platform_helper.get_prompt_studio_tool(
+ prompt_registry_id=prompt_registry_id
)
tool_metadata = exported_tool[SettingsKeys.TOOL_METADATA]
ps_project_name = tool_metadata.get("name", prompt_registry_id)
+ # Count only the active (enabled) prompts
total_prompt_count = len(tool_metadata[SettingsKeys.OUTPUTS])
- tool.stream_log(
+ self.stream_log(
f"Retrieved prompt studio exported tool '{ps_project_name}' having "
f"'{total_prompt_count}' prompts"
)
except Exception as e:
- self.stream_error_and_exit(f"Error loading structure definition: {e}")
+ self.stream_error_and_exit(f"Error fetching prompt studio project: {e}")
+ active_prompt_count = len(
+ [
+ output
+ for output in tool_metadata[SettingsKeys.OUTPUTS]
+ if output.get("active", False)
+ ]
+ )
# Update GUI
input_log = f"## Loaded '{ps_project_name}'\n{json_to_markdown(tool_metadata)}\n"
output_log = (
f"## Processing '{self.source_file_name}'\nThis might take a while and "
"involve...\n- Extracting text\n- Indexing\n- Retrieving answers "
- f"for possibly '{total_prompt_count}' prompts"
+ f"for '{active_prompt_count}' prompts"
)
self.stream_update(input_log, state=LogState.INPUT_UPDATE)
self.stream_update(output_log, state=LogState.OUTPUT_UPDATE)
@@ -87,7 +103,6 @@ class StructureTool(BaseTool):
)
tool_settings[SettingsKeys.SUMMARIZE_AS_SOURCE] = summarize_as_source
tool_settings[SettingsKeys.ENABLE_HIGHLIGHT] = enable_highlight
- prompt_service_resp = None
_, file_name = os.path.split(input_file)
if summarize_as_source:
file_name = SettingsKeys.SUMMARIZE
@@ -124,86 +139,65 @@ class StructureTool(BaseTool):
execution_run_data_folder=str(execution_run_data_folder),
)
+ summarize_file_hash = None
+ if summarize_as_source:
+ summarize_file_path, summarize_file_hash = self._summarize_and_index(
+ tool_settings=tool_settings,
+ tool_data_dir=tool_data_dir,
+ responder=responder,
+ outputs=outputs,
+ usage_kwargs=usage_kwargs,
+ )
+ payload[SettingsKeys.FILE_HASH] = summarize_file_hash
+ payload[SettingsKeys.FILE_PATH] = summarize_file_path
+
if tool_settings[SettingsKeys.ENABLE_SINGLE_PASS_EXTRACTION]:
- if summarize_as_source:
- summarize_file_path, summarize_file_hash = self._summarize_and_index(
- tool_settings=tool_settings,
- tool_data_dir=tool_data_dir,
- responder=responder,
- outputs=outputs,
- usage_kwargs=usage_kwargs,
- )
- payload[SettingsKeys.FILE_HASH] = summarize_file_hash
- payload[SettingsKeys.FILE_PATH] = summarize_file_path
- self.stream_log("Fetching response for single pass extraction")
+ self.stream_log("Fetching response for single pass extraction...")
# Since indexing is not involved for single pass
index_metrics = {"time_taken(s)": 0}
- prompt_service_resp = responder.single_pass_extraction(
+ structured_output = responder.single_pass_extraction(
payload=payload,
)
else:
- try:
- # To reindex even if file is already
- # indexed to get the output in required path
- reindex = True
- for output in outputs:
- if summarize_as_source:
- summarize_file_path, summarize_file_hash = (
- self._summarize_and_index(
- tool_settings=tool_settings,
- tool_data_dir=tool_data_dir,
- responder=responder,
- outputs=outputs,
- usage_kwargs=usage_kwargs,
- )
- )
- payload[SettingsKeys.OUTPUTS] = outputs
- payload[SettingsKeys.FILE_HASH] = summarize_file_hash
- payload[SettingsKeys.FILE_PATH] = summarize_file_path
- # Since indexing is not involved for summary
- index_metrics[output[SettingsKeys.NAME]] = {"time_taken(s)": 0}
- break
+ # To reindex even if file is already
+ # indexed to get the output in required path
+ reindex = True
+ index_metrics = {}
+ for output in outputs:
+ if summarize_as_source:
+ # Since indexing is not involved for summary
+ index_metrics[output[SettingsKeys.NAME]] = {"time_taken(s)": 0}
+ break
+ if (reindex or not summarize_as_source) and output[
+ SettingsKeys.CHUNK_SIZE
+ ] != 0:
+ indexing_start_time = datetime.datetime.now()
self.stream_log(
- f"Chunk size '{output[SettingsKeys.CHUNK_SIZE]}', "
- f"indexing '{tool_data_dir / SettingsKeys.EXTRACT}'.."
+ f"Indexing document with chunk size '{output[SettingsKeys.CHUNK_SIZE]}' and overlap '{output[SettingsKeys.CHUNK_OVERLAP]}'"
)
- if (reindex or not summarize_as_source) and output[
- SettingsKeys.CHUNK_SIZE
- ] != 0:
- indexing_start_time = datetime.datetime.now()
- self.stream_log(
- f"Sucessfully extracted text, "
- f"indexing {tool_data_dir / SettingsKeys.EXTRACT}.."
- )
- STHelper.dynamic_indexing(
- tool_settings=tool_settings,
- run_id=self.file_execution_id,
- file_path=tool_data_dir / SettingsKeys.EXTRACT,
- tool=self,
- execution_run_data_folder=str(execution_run_data_folder),
- chunk_overlap=output[SettingsKeys.CHUNK_OVERLAP],
- reindex=reindex,
- usage_kwargs=usage_kwargs,
- enable_highlight=enable_highlight,
- chunk_size=output[SettingsKeys.CHUNK_SIZE],
- tool_id=tool_metadata[SettingsKeys.TOOL_ID],
- file_hash=file_hash,
- extracted_text=extracted_text,
- )
- index_metrics[output[SettingsKeys.NAME]] = {
- SettingsKeys.INDEXING: {
- "time_taken(s)": STHelper.elapsed_time(
- start_time=indexing_start_time
- )
- }
+ STHelper.dynamic_indexing(
+ tool_settings=tool_settings,
+ run_id=self.file_execution_id,
+ file_path=tool_data_dir / SettingsKeys.EXTRACT,
+ tool=self,
+ execution_run_data_folder=str(execution_run_data_folder),
+ chunk_overlap=output[SettingsKeys.CHUNK_OVERLAP],
+ reindex=reindex,
+ usage_kwargs=usage_kwargs,
+ enable_highlight=enable_highlight,
+ chunk_size=output[SettingsKeys.CHUNK_SIZE],
+ tool_id=tool_metadata[SettingsKeys.TOOL_ID],
+ file_hash=file_hash,
+ extracted_text=extracted_text,
+ )
+ index_metrics[output[SettingsKeys.NAME]] = {
+ SettingsKeys.INDEXING: {
+ "time_taken(s)": STHelper.elapsed_time(
+ start_time=indexing_start_time
+ )
}
-
- reindex = False
- except Exception as e:
- self.stream_log(
- f"Error fetching data and indexing: {e}", level=LogLevel.ERROR
- )
- raise
+ }
+ reindex = False
for output in outputs:
if SettingsKeys.TABLE_SETTINGS in output:
@@ -215,33 +209,23 @@ class StructureTool(BaseTool):
table_settings[SettingsKeys.IS_DIRECTORY_MODE] = is_directory_mode
output.update({SettingsKeys.TABLE_SETTINGS: table_settings})
- self.stream_log(f"Fetching responses for {len(outputs)} prompt(s)...")
- prompt_service_resp = responder.answer_prompt(
+ self.stream_log(f"Fetching responses for '{len(outputs)}' prompt(s)...")
+ structured_output = responder.answer_prompt(
payload=payload,
)
- if prompt_service_resp[SettingsKeys.STATUS] != SettingsKeys.OK:
- self.stream_error_and_exit(
- f"Failed to fetch responses for "
- f"prompts: {prompt_service_resp[SettingsKeys.ERROR]}"
- )
-
- structured_output = prompt_service_resp[SettingsKeys.STRUCTURE_OUTPUT]
- structured_output_dict = json.loads(structured_output)
-
# HACK: Replacing actual file's name instead of INFILE
- if SettingsKeys.METADATA in structured_output_dict:
- structured_output_dict[SettingsKeys.METADATA][SettingsKeys.FILE_NAME] = (
+ if SettingsKeys.METADATA in structured_output:
+ structured_output[SettingsKeys.METADATA][SettingsKeys.FILE_NAME] = (
self.source_file_name
)
if not summarize_as_source:
- metadata = structured_output_dict[SettingsKeys.METADATA]
+ metadata = structured_output[SettingsKeys.METADATA]
# Update the dictionary with modified metadata
- structured_output_dict[SettingsKeys.METADATA] = metadata
- structured_output = json.dumps(structured_output_dict)
+ structured_output[SettingsKeys.METADATA] = metadata
- metrics = structured_output_dict.get(SettingsKeys.METRICS, {})
+ metrics = structured_output.get(SettingsKeys.METRICS, {})
new_metrics = {}
if tool_settings[SettingsKeys.ENABLE_SINGLE_PASS_EXTRACTION]:
new_metrics = {
@@ -256,27 +240,26 @@ class StructureTool(BaseTool):
| set(index_metrics) # Union of keys from both dictionaries
}
if new_metrics:
- structured_output_dict[SettingsKeys.METRICS] = new_metrics
+ structured_output[SettingsKeys.METRICS] = new_metrics
# Update GUI
output_log = (
f"## Result\n**NOTE:** In case of a deployed pipeline, the result would "
"be a JSON. This has been rendered for readability here\n"
- f"{json_to_markdown(structured_output_dict)}\n"
+ f"{json_to_markdown(structured_output)}\n"
)
self.stream_update(output_log, state=LogState.OUTPUT_UPDATE)
- # Write the translated text to output file
try:
- self.stream_log("Writing parsed output...")
- output_path = Path(output_dir) / f"{Path(self.source_file_name).stem}.json"
- self.workflow_filestorage.json_dump(
- path=output_path, data=structured_output_dict
+ self.stream_log(
+ "Dumping prompt studio project's output to workflow's storage"
)
+ output_path = Path(output_dir) / f"{Path(self.source_file_name).stem}.json"
+ self.workflow_filestorage.json_dump(path=output_path, data=structured_output)
except OSError as e:
self.stream_error_and_exit(f"Error creating output file: {e}")
except json.JSONDecodeError as e:
self.stream_error_and_exit(f"Error encoding JSON: {e}")
- self.write_tool_result(data=structured_output_dict)
+ self.write_tool_result(data=structured_output)
def _summarize_and_index(
self,
@@ -310,7 +293,9 @@ class StructureTool(BaseTool):
summarize_file_path = tool_data_dir / SettingsKeys.SUMMARIZE
summarized_context = ""
- self.stream_log(f"Checking if summarized context exists at {summarize_file_path}")
+ self.stream_log(
+ f"Checking if summarized context exists at '{summarize_file_path}'..."
+ )
if self.workflow_filestorage.exists(summarize_file_path):
summarized_context = self.workflow_filestorage.read(
path=summarize_file_path, mode="r"
@@ -326,7 +311,7 @@ class StructureTool(BaseTool):
output[SettingsKeys.X2TEXT_ADAPTER] = x2text_instance_id
output[SettingsKeys.CHUNK_SIZE] = 0
output[SettingsKeys.CHUNK_OVERLAP] = 0
- self.stream_log("Summarizing context")
+ self.stream_log("Summarized context not found, summarizing...")
payload = {
SettingsKeys.RUN_ID: run_id,
SettingsKeys.LLM_ADAPTER_INSTANCE_ID: llm_adapter_instance_id,
@@ -334,14 +319,9 @@ class StructureTool(BaseTool):
SettingsKeys.CONTEXT: context,
SettingsKeys.PROMPT_KEYS: prompt_keys,
}
- response = responder.summarize(payload=payload)
- if response[SettingsKeys.STATUS] != SettingsKeys.OK:
- self.stream_error_and_exit(
- f"Error summarizing response: {response[SettingsKeys.ERROR]}"
- )
- structure_output = json.loads(response[SettingsKeys.STRUCTURE_OUTPUT])
+ structure_output = responder.summarize(payload=payload)
summarized_context = structure_output.get(SettingsKeys.DATA, "")
- self.stream_log("Writing summarized context to a file")
+ self.stream_log(f"Writing summarized context to '{summarize_file_path}'")
self.workflow_filestorage.write(
path=summarize_file_path, mode="w", data=summarized_context
)
diff --git a/tools/text_extractor/requirements.txt b/tools/text_extractor/requirements.txt
index 6b9ef80b6..271fe4332 100644
--- a/tools/text_extractor/requirements.txt
+++ b/tools/text_extractor/requirements.txt
@@ -4,4 +4,4 @@
# aws alone is needed here
# because tools use transient temporary storage.
-unstract-sdk[aws]~=0.70.0
+unstract-sdk[aws]~=0.71.0
diff --git a/tools/text_extractor/src/config/properties.json b/tools/text_extractor/src/config/properties.json
index cd9adfd00..8c0b73dde 100644
--- a/tools/text_extractor/src/config/properties.json
+++ b/tools/text_extractor/src/config/properties.json
@@ -2,7 +2,7 @@
"schemaVersion": "0.0.1",
"displayName": "Text Extractor",
"functionName": "text_extractor",
- "toolVersion": "0.0.55",
+ "toolVersion": "0.0.56",
"description": "The Text Extractor is a powerful tool designed to convert documents to its text form or Extract texts from documents",
"input": {
"description": "Document"
diff --git a/unstract/filesystem/pyproject.toml b/unstract/filesystem/pyproject.toml
index 27edf53f9..0e8f16e5e 100644
--- a/unstract/filesystem/pyproject.toml
+++ b/unstract/filesystem/pyproject.toml
@@ -6,7 +6,7 @@ authors = [{ name = "Zipstack Inc.", email = "[email protected]" }]
requires-python = ">=3.12,<3.13"
readme = "README.md"
-dependencies = ["unstract-sdk~=0.70.0"]
+dependencies = ["unstract-sdk~=0.71.0"]
# [tool.uv.sources]
# unstract-sdk = { git = "https://github.com/Zipstack/unstract-sdk", branch = "python-upgrade" }
diff --git a/unstract/filesystem/uv.lock b/unstract/filesystem/uv.lock
index 5a7394b13..fd66b00ae 100644
--- a/unstract/filesystem/uv.lock
+++ b/unstract/filesystem/uv.lock
@@ -2574,11 +2574,11 @@ dependencies = [
]
[package.metadata]
-requires-dist = [{ name = "unstract-sdk", specifier = "~=0.70.0" }]
+requires-dist = [{ name = "unstract-sdk", specifier = "~=0.71.0" }]
[[package]]
name = "unstract-sdk"
-version = "0.70.0"
+version = "0.71.0"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "filetype" },
@@ -2617,9 +2617,9 @@ dependencies = [
{ name = "tiktoken" },
{ name = "transformers" },
]
-sdist = { url = "https://files.pythonhosted.org/packages/00/f7/e88481d199847003a173c8355b4664fa8b272cfe0b99ba425703e75c4a70/unstract_sdk-0.70.0.tar.gz", hash = "sha256:98ce28e8b1e88b4c43ebc1fbc947cafc9efae0b7914043fcd969061169947d1d", size = 2356341 }
+sdist = { url = "https://files.pythonhosted.org/packages/d0/12/efecbf0581040b1bfc44fe939a68ba07e45c65c16b3e91666227e94b52be/unstract_sdk-0.71.0.tar.gz", hash = "sha256:c9269cbd1656e40a03445d3758bc51d7921273af729e0d3813f5f9e28967e31d", size = 2360756 }
wheels = [
- { url = "https://files.pythonhosted.org/packages/21/a3/92240dd060544348d0d45b5560d064aa63e8c2daee25159dce533b1a137b/unstract_sdk-0.70.0-py3-none-any.whl", hash = "sha256:583f6ec793594d3113379f175883597a8af7d58e9faab5d4ff02f31bf8bf3510", size = 264025 },
+ { url = "https://files.pythonhosted.org/packages/3a/88/4ba63cd8611d7bdaf36619e9fe1b19effb5bc10a13ec50b79bf714aa410a/unstract_sdk-0.71.0-py3-none-any.whl", hash = "sha256:0d3bdaf79468f03991f457d040b07ab5d159be3a601b4bfae7f772083d70b705", size = 265087 },
]
[[package]]
diff --git a/unstract/tool-registry/pyproject.toml b/unstract/tool-registry/pyproject.toml
index 265d7fa35..eb9749662 100644
--- a/unstract/tool-registry/pyproject.toml
+++ b/unstract/tool-registry/pyproject.toml
@@ -11,7 +11,7 @@ dependencies = [
"docker~=6.1.3",
"jsonschema~=4.18.2",
"PyYAML~=6.0.1",
- "unstract-sdk~=0.70.0",
+ "unstract-sdk~=0.71.0",
"unstract-tool-sandbox",
"unstract-flags",
]
diff --git a/unstract/tool-registry/tool_registry_config/public_tools.json b/unstract/tool-registry/tool_registry_config/public_tools.json
index 26f2f4c28..f7e55fdc9 100644
--- a/unstract/tool-registry/tool_registry_config/public_tools.json
+++ b/unstract/tool-registry/tool_registry_config/public_tools.json
@@ -5,7 +5,7 @@
"schemaVersion": "0.0.1",
"displayName": "File Classifier",
"functionName": "classify",
- "toolVersion": "0.0.58",
+ "toolVersion": "0.0.59",
"description": "Classifies a file into a bin based on its contents",
"input": {
"description": "File to be classified"
@@ -106,9 +106,9 @@
"properties": {}
},
"icon": "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n<svg\n enable-background=\"new 0 0 20 20\"\n height=\"48\"\n viewBox=\"0 0 20 20\"\n width=\"48\"\n fill=\"#000000\"\n version=\"1.1\"\n id=\"svg8109\"\n sodipodi:docname=\"folder_copy_black_48dp.svg\"\n xmlns:inkscape=\"http://www.inkscape.org/namespaces/inkscape\"\n xmlns:sodipodi=\"http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd\"\n xmlns=\"http://www.w3.org/2000/svg\"\n xmlns:svg=\"http://www.w3.org/2000/svg\">\n <defs\n id=\"defs8113\" />\n <sodipodi:namedview\n id=\"namedview8111\"\n pagecolor=\"#ffffff\"\n bordercolor=\"#000000\"\n borderopacity=\"0.25\"\n inkscape:showpageshadow=\"2\"\n inkscape:pageopacity=\"0.0\"\n inkscape:pagecheckerboard=\"0\"\n inkscape:deskcolor=\"#d1d1d1\"\n showgrid=\"false\" />\n <g\n id=\"g8099\">\n <rect\n fill=\"none\"\n height=\"20\"\n width=\"20\"\n x=\"0\"\n id=\"rect8097\"\n y=\"0\" />\n </g>\n <g\n id=\"g8107\"\n style=\"fill:#ff4d6d;fill-opacity:1\">\n <g\n id=\"g8105\"\n style=\"fill:#ff4d6d;fill-opacity:1\">\n <path\n d=\"M 2.5,5 H 1 V 15.5 C 1,16.33 1.67,17 2.5,17 H 15.68 V 15.5 H 2.5 Z\"\n id=\"path8101\"\n style=\"fill:#ff4d6d;fill-opacity:1\" />\n <path\n d=\"M 16.5,4 H 11 L 9,2 H 5.5 C 4.67,2 4,2.67 4,3.5 v 9 C 4,13.33 4.67,14 5.5,14 h 11 c 0.83,0 1.5,-0.67 1.5,-1.5 v -7 C 18,4.67 17.33,4 16.5,4 Z m 0,8.5 h -11 v -9 h 2.88 l 2,2 h 6.12 z\"\n id=\"path8103\"\n style=\"fill:#ff4d6d;fill-opacity:1\" />\n </g>\n </g>\n</svg>\n",
- "image_url": "docker:unstract/tool-classifier:0.0.58",
+ "image_url": "docker:unstract/tool-classifier:0.0.59",
"image_name": "unstract/tool-classifier",
- "image_tag": "0.0.58"
+ "image_tag": "0.0.59"
},
"text_extractor": {
"tool_uid": "text_extractor",
@@ -116,7 +116,7 @@
"schemaVersion": "0.0.1",
"displayName": "Text Extractor",
"functionName": "text_extractor",
- "toolVersion": "0.0.55",
+ "toolVersion": "0.0.56",
"description": "The Text Extractor is a powerful tool designed to convert documents to its text form or Extract texts from documents",
"input": {
"description": "Document"
@@ -191,8 +191,8 @@
}
},
"icon": "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n<svg\n enable-background=\"new 0 0 20 20\"\n height=\"48\"\n viewBox=\"0 0 20 20\"\n width=\"48\"\n fill=\"#000000\"\n version=\"1.1\"\n id=\"svg8109\"\n sodipodi:docname=\"folder_copy_black_48dp.svg\"\n xmlns:inkscape=\"http://www.inkscape.org/namespaces/inkscape\"\n xmlns:sodipodi=\"http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd\"\n xmlns=\"http://www.w3.org/2000/svg\"\n xmlns:svg=\"http://www.w3.org/2000/svg\">\n <defs\n id=\"defs8113\" />\n <sodipodi:namedview\n id=\"namedview8111\"\n pagecolor=\"#ffffff\"\n bordercolor=\"#000000\"\n borderopacity=\"0.25\"\n inkscape:showpageshadow=\"2\"\n inkscape:pageopacity=\"0.0\"\n inkscape:pagecheckerboard=\"0\"\n inkscape:deskcolor=\"#d1d1d1\"\n showgrid=\"false\" />\n <g\n id=\"g8099\">\n <rect\n fill=\"none\"\n height=\"20\"\n width=\"20\"\n x=\"0\"\n id=\"rect8097\"\n y=\"0\" />\n </g>\n <g\n id=\"g8107\"\n style=\"fill:#ff4d6d;fill-opacity:1\">\n <g\n id=\"g8105\"\n style=\"fill:#ff4d6d;fill-opacity:1\">\n <path\n d=\"M 2.5,5 H 1 V 15.5 C 1,16.33 1.67,17 2.5,17 H 15.68 V 15.5 H 2.5 Z\"\n id=\"path8101\"\n style=\"fill:#ff4d6d;fill-opacity:1\" />\n <path\n d=\"M 16.5,4 H 11 L 9,2 H 5.5 C 4.67,2 4,2.67 4,3.5 v 9 C 4,13.33 4.67,14 5.5,14 h 11 c 0.83,0 1.5,-0.67 1.5,-1.5 v -7 C 18,4.67 17.33,4 16.5,4 Z m 0,8.5 h -11 v -9 h 2.88 l 2,2 h 6.12 z\"\n id=\"path8103\"\n style=\"fill:#ff4d6d;fill-opacity:1\" />\n </g>\n </g>\n</svg>\n",
- "image_url": "docker:unstract/tool-text-extractor:0.0.55",
+ "image_url": "docker:unstract/tool-text-extractor:0.0.56",
"image_name": "unstract/tool-text-extractor",
- "image_tag": "0.0.55"
+ "image_tag": "0.0.56"
}
}
diff --git a/uv.lock b/uv.lock
index 98b1771f2..2971c47d0 100644
--- a/uv.lock
+++ b/uv.lock
@@ -3828,7 +3828,7 @@ hook-check-django-migrations = [
{ name = "unstract-core", editable = "unstract/core" },
{ name = "unstract-filesystem", editable = "unstract/filesystem" },
{ name = "unstract-flags", editable = "unstract/flags" },
- { name = "unstract-sdk", specifier = "~=0.70.0" },
+ { name = "unstract-sdk", specifier = "~=0.71.0" },
{ name = "unstract-tool-registry", editable = "unstract/tool-registry" },
{ name = "unstract-tool-sandbox", editable = "unstract/tool-sandbox" },
{ name = "unstract-workflow-execution", editable = "unstract/workflow-execution" },
@@ -3909,7 +3909,7 @@ dependencies = [
]
[package.metadata]
-requires-dist = [{ name = "unstract-sdk", specifier = "~=0.70.0" }]
+requires-dist = [{ name = "unstract-sdk", specifier = "~=0.71.0" }]
[[package]]
name = "unstract-flags"
@@ -3928,7 +3928,7 @@ requires-dist = [
[[package]]
name = "unstract-sdk"
-version = "0.70.0"
+version = "0.71.0"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "filetype" },
@@ -3967,9 +3967,9 @@ dependencies = [
{ name = "tiktoken" },
{ name = "transformers" },
]
-sdist = { url = "https://files.pythonhosted.org/packages/00/f7/e88481d199847003a173c8355b4664fa8b272cfe0b99ba425703e75c4a70/unstract_sdk-0.70.0.tar.gz", hash = "sha256:98ce28e8b1e88b4c43ebc1fbc947cafc9efae0b7914043fcd969061169947d1d", size = 2356341 }
+sdist = { url = "https://files.pythonhosted.org/packages/d0/12/efecbf0581040b1bfc44fe939a68ba07e45c65c16b3e91666227e94b52be/unstract_sdk-0.71.0.tar.gz", hash = "sha256:c9269cbd1656e40a03445d3758bc51d7921273af729e0d3813f5f9e28967e31d", size = 2360756 }
wheels = [
- { url = "https://files.pythonhosted.org/packages/21/a3/92240dd060544348d0d45b5560d064aa63e8c2daee25159dce533b1a137b/unstract_sdk-0.70.0-py3-none-any.whl", hash = "sha256:583f6ec793594d3113379f175883597a8af7d58e9faab5d4ff02f31bf8bf3510", size = 264025 },
+ { url = "https://files.pythonhosted.org/packages/3a/88/4ba63cd8611d7bdaf36619e9fe1b19effb5bc10a13ec50b79bf714aa410a/unstract_sdk-0.71.0-py3-none-any.whl", hash = "sha256:0d3bdaf79468f03991f457d040b07ab5d159be3a601b4bfae7f772083d70b705", size = 265087 },
]
[[package]]
@@ -3991,7 +3991,7 @@ requires-dist = [
{ name = "jsonschema", specifier = "~=4.18.2" },
{ name = "pyyaml", specifier = "~=6.0.1" },
{ name = "unstract-flags", editable = "unstract/flags" },
- { name = "unstract-sdk", specifier = "~=0.70.0" },
+ { name = "unstract-sdk", specifier = "~=0.71.0" },
{ name = "unstract-tool-sandbox", editable = "unstract/tool-sandbox" },
]