LLM-Driven Extraction of Unstructured Data — Built for API Deployments & ETL Pipeline Workflows
Python ai-agentsdata-engineeringdocument-aigenerative-aiidpjson-extractionllmmcp-serverocrpdf-extractionprompt-engineeringstructured-output
UN-2175 Avoid extracting when doc is already extracted during indexing (#1605)
Chandrasekharan M13642ba376d6
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.pyindex 4e159c3ec..f6bf7086b 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@@ -427,7 +427,6 @@ class PromptStudioHelper:run_id=run_id,enable_highlight=tool.enable_highlight,doc_id=doc_id,- reindex=True,)if tool.summarize_context:summarize_file_path = PromptStudioHelper.summarize(@@ -1332,7 +1331,6 @@ class PromptStudioHelper:profile_manager: ProfileManager,document_id: str,doc_id: str,- reindex: bool | None = False,) -> str:x2text = str(profile_manager.x2text.id)is_extracted: bool = False@@ -1350,7 +1348,7 @@ class PromptStudioHelper:profile_manager=profile_manager,doc_id=doc_id,)- if is_extracted and not reindex:+ if is_extracted:fs_instance = EnvHelper.get_storage(storage_type=StorageType.PERMANENT,env_name=FileStorageKeys.PERMANENT_REMOTE_STORAGE,