LLM-Driven Extraction of Unstructured Data — Built for API Deployments & ETL Pipeline Workflows
Python ai-agentsdata-engineeringdocument-aigenerative-aiidpjson-extractionllmmcp-serverocrpdf-extractionprompt-engineeringstructured-output
fix: Prompt-service issue for 0 chunk size (#50)
Chandrasekharan M09f1732810bd
diff --git a/prompt-service/src/unstract/prompt_service/main.py b/prompt-service/src/unstract/prompt_service/main.pyindex 995287160..b83e8a5be 100644--- a/prompt-service/src/unstract/prompt_service/main.py+++ b/prompt-service/src/unstract/prompt_service/main.py@@ -486,10 +486,9 @@ def prompt_processor() -> Any:if output[PSKeys.CHUNK_SIZE] == 0:# We can do this only for chunkless indexescontext = tool_index.get_text_from_index(- project_id=tool_id,+ embedding_type=output[PSKeys.EMBEDDING],vector_db=output[PSKeys.VECTOR_DB],doc_id=doc_id,- embedding_dimension=embedding_dimension,)assertion_failed = False