From ec8844c7d4440a8338372f6cff352e1f0c038033 Mon Sep 17 00:00:00 2001
From: Ozzie Isaacs <ozzie.fernandez.isaacs@googlemail.com>
Date: Sat, 27 Aug 2022 15:44:21 +0200
Subject: [PATCH] Make pyPDF2 again to the favorite pdf metadata extractor

---
 cps/uploader.py  | 15 +++++++++------
 requirements.txt |  2 +-
 2 files changed, 10 insertions(+), 7 deletions(-)

diff --git a/cps/uploader.py b/cps/uploader.py
index 92035356..ecf4868c 100644
--- a/cps/uploader.py
+++ b/cps/uploader.py
@@ -37,16 +37,15 @@ except (ImportError, RuntimeError) as e:
     use_generic_pdf_cover = True
 
 try:
-    from PyPDF3 import PdfFileReader
-    from PyPDF3 import __version__ as PyPdfVersion
+    from PyPDF2 import PdfFileReader
     use_pdf_meta = True
 except ImportError as ex:
+    log.debug('PyPDF2 is recommended for best performance in metadata extracting from pdf files: %s', ex)
     try:
-        from PyPDF2 import PdfFileReader
-        from PyPDF2 import __version__ as PyPdfVersion
+        from PyPDF3 import PdfFileReader
         use_pdf_meta = True
     except ImportError as e:
-        log.debug('Cannot import PyPDF3/PyPDF2, extracting pdf metadata will not work: %s / %s', ex, e)
+        log.debug('Cannot import PyPDF3/PyPDF2, extracting pdf metadata will not work: %s / %s', e)
         use_pdf_meta = False
 
 try:
@@ -114,7 +113,7 @@ def parse_xmp(pdf_file):
     try:
         xmp_info = pdf_file.getXmpMetadata()
     except Exception as ex:
-        log.debug('Can not read XMP metadata {}'.format(ex))
+        log.debug('Can not read PDF XMP metadata {}'.format(ex))
         return None
 
     if xmp_info:
@@ -160,6 +159,10 @@ def pdf_meta(tmp_file_path, original_file_name, original_file_extension):
         with open(tmp_file_path, 'rb') as f:
             pdf_file = PdfFileReader(f)
             doc_info = pdf_file.getDocumentInfo()
+            try:
+                doc_info = pdf_file.getDocumentInfo()
+            except Exception as exc:
+                log.debug('Can not read PDF DocumentInfo {}'.format(exc))
             xmp_info = parse_xmp(pdf_file)
 
     if xmp_info:
diff --git a/requirements.txt b/requirements.txt
index acaf4a40..0c20c71d 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -7,7 +7,7 @@ Flask-Principal>=0.3.2,<0.5.1
 backports_abc>=0.4
 Flask>=1.0.2,<2.1.0
 iso-639>=0.4.5,<0.5.0
-PyPDF3>=1.0.0,<1.0.7
+PyPDF2>=1.20,<2.11.0
 pytz>=2016.10
 requests>=2.11.1,<2.28.0
 SQLAlchemy>=1.3.0,<1.5.0