From 9a01a23e4ca8985420c154220c270d49d57d8291 Mon Sep 17 00:00:00 2001 From: Roberto Rosario Date: Sat, 24 Jan 2015 04:27:57 -0400 Subject: [PATCH] Detech empty PDF parsing by detecting unicode /u000c and not ASCII /x0c --- mayan/apps/ocr/parsers/__init__.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mayan/apps/ocr/parsers/__init__.py b/mayan/apps/ocr/parsers/__init__.py index f505b3c531..4438a25c3f 100644 --- a/mayan/apps/ocr/parsers/__init__.py +++ b/mayan/apps/ocr/parsers/__init__.py @@ -88,7 +88,7 @@ class SlateParser(Parser): pdf_pages = slate.PDF(descriptor) descriptor.close() - if pdf_pages[document_page.page_number - 1] == '\x0c': + if unicode(pdf_pages[document_page.page_number - 1]) == '\u000c': raise ParserError document_page.content = pdf_pages[document_page.page_number - 1] @@ -164,7 +164,7 @@ class PopplerParser(Parser): raise ParserError output = proc.stdout.read() - if output == '\x0c': + if unicode(output) == '\u000c': logger.debug('Parser didn\'t any output') raise ParserError('No output')