feat: enable OCR for docling when parsing attachments and update file extractor (#2246)

This commit is contained in:
Alex
2025-12-31 00:08:49 +00:00
committed by GitHub
parent 9e7f1ad1c0
commit 05c835ed02
3 changed files with 15 additions and 5 deletions

View File

@@ -43,6 +43,7 @@ class Settings(BaseSettings):
PARSE_PDF_AS_IMAGE: bool = False
PARSE_IMAGE_REMOTE: bool = False
DOCLING_OCR_ENABLED: bool = True # Enable OCR for docling parsers (PDF, images)
DOCLING_OCR_ATTACHMENTS_ENABLED: bool = False # Enable OCR for docling when parsing attachments
VECTOR_STORE: str = (
"faiss" # "faiss" or "elasticsearch" or "qdrant" or "milvus" or "lancedb" or "pgvector"
)