Google Drive handle invalid PDFs (#838)

backend/danswer/connectors/cross_connector_utils/file_utils.py

@@ -9,6 +9,7 @@ from typing import IO
 
 import chardet
 from pypdf import PdfReader
+from pypdf.errors import PdfStreamError
 
 from danswer.utils.logger import setup_logger
 
@@ -37,9 +38,10 @@ def extract_metadata(line: str) -> dict | None:
 
 
 def read_pdf_file(file: IO[Any], file_name: str, pdf_pass: str | None = None) -> str:
+    try:
         pdf_reader = PdfReader(file)
 
-    # if marked as encrypted and a password is provided, try to decrypt
+        # If marked as encrypted and a password is provided, try to decrypt
         if pdf_reader.is_encrypted and pdf_pass is not None:
             decrypt_success = False
             if pdf_pass is not None:
@@ -55,10 +57,14 @@ def read_pdf_file(file: IO[Any], file_name: str, pdf_pass: str | None = None) ->
                 # can be discoverable by title.
                 return ""
 
-    try:
         return "\n".join(page.extract_text() for page in pdf_reader.pages)
+    except PdfStreamError:
+        logger.exception(f"PDF file {file_name} is not a valid PDF")
     except Exception:
         logger.exception(f"Failed to read PDF {file_name}")
+
+    # File is still discoverable by title
+    # but the contents are not included as they cannot be parsed
     return ""