fix: skip unreadable PDFs during metadata extraction

backend/src/chitai/services/metadata_extractor.py

@@ -53,7 +53,7 @@ class Extractor:
         # Sort based on file priority
         # EPUB tends to give better metadata results over pdf
         sorted_files = sorted(files, key=lambda f: Extractor._get_file_priority(f))
-        
+
         for file in sorted_files:
             match get_file_extension(file):
                 case "epub":
@@ -155,7 +155,11 @@ class PdfExtractor(FileExtractor):
         if isinstance(data, UploadFile):
             data = data.file
 
-        doc = pypdfium2.PdfDocument(data)
+        try:
+            doc = pypdfium2.PdfDocument(data)
+        except Exception as e:
+            logger.error(f"Error extracting metadata from pdf: {e}")
+            return {}
 
         basic_metadata = doc.get_metadata_dict(skip_empty=False)
         metadata["title"] = basic_metadata["Title"]