github/privateGPT
1
0
Fork 1
mirror of https://github.com/imartinez/privateGPT.git synced 2025-06-30 17:22:43 +00:00
Code Issues Packages Projects Releases Wiki Activity

Try catch added on ingest

Browse Source
This commit is contained in:
Saurab-Shrestha 2024-02-27 10:38:44 +05:45
parent a8a10e412e
commit 484b6b7223
2 changed files with 34 additions and 26 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
private_gpt/components/ocr_components/table_ocr_api.py
View File

@ -44,7 +44,7 @@ async def process_images_and_generate_doc(pdf_path: str, upload_dir: str):
ocr = GetOCRText() ocr = GetOCRText()
img_tab = ImageToTable() img_tab = ImageToTable()
pdf_doc = fitz.open(pdf_path) pdf_doc = fitz.open(pdf_path)
for page_index in range(len(pdf_doc)): for page_index in range(len(pdf_doc)):
page = pdf_doc[page_index] page = pdf_doc[page_index]
image_list = page.get_images() image_list = page.get_images()

58
private_gpt/server/ingest/ingest_router.py
View File

@ -294,34 +294,42 @@ async def common_ingest_logic(
) )
# Handling Original File # Handling Original File
if original_file: if original_file:
print("ORIGINAL PDF FILE PATH IS :: ", original_file) try:
file_name = Path(original_file).name print("ORIGINAL PDF FILE PATH IS :: ", original_file)
upload_path = Path(f"{UPLOAD_DIR}/{file_name}") file_name = Path(original_file).name
upload_path = Path(f"{UPLOAD_DIR}/{file_name}")
file_ingested = crud.documents.get_by_filename( file_ingested = crud.documents.get_by_filename(
db, file_name=file_name) db, file_name=file_name)
if file_ingested: if file_ingested:
raise HTTPException(
status_code=status.HTTP_409_CONFLICT,
detail="File already exists. Choose a different file.",
)
if file_name is None:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail="No file name provided",
)
docs_in = schemas.DocumentCreate(
filename=file_name, uploaded_by=current_user.id, department_id=current_user.department_id)
crud.documents.create(db=db, obj_in=docs_in)
with open(upload_path, "wb") as f:
with open(original_file, "rb") as original_file_reader:
f.write(original_file_reader.read())
with open(upload_path, "rb") as f:
ingested_documents = service.ingest_bin_data(file_name, f)
except Exception as e:
print(traceback.print_exc())
raise HTTPException( raise HTTPException(
status_code=status.HTTP_409_CONFLICT, status_code=500,
detail="File already exists. Choose a different file.", detail="Internal Server Error: Unable to ingest file.",
) )
if file_name is None:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail="No file name provided",
)
docs_in = schemas.DocumentCreate(
filename=file_name, uploaded_by=current_user.id, department_id=current_user.department_id)
crud.documents.create(db=db, obj_in=docs_in)
with open(upload_path, "wb") as f:
with open(original_file, "rb") as original_file_reader:
f.write(original_file_reader.read())
with open(upload_path, "rb") as f:
ingested_documents = service.ingest_bin_data(file_name, f)
logger.info( logger.info(
f"{file_name} is uploaded by the {current_user.fullname}.") f"{file_name} is uploaded by the {current_user.fullname}.")