tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

rclmpdf.py: p2/3 compat

Browse Source
This commit is contained in:
Jean-Francois Dockes 2015-11-21 12:46:58 +01:00
parent 5776c4bc20
commit d115bcfaa2
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/filters/rclmpdf.py
View File

@ -135,9 +135,9 @@ class PDFExtractor:
b'content="text/html; charset=UTF-8">\n' b'content="text/html; charset=UTF-8">\n'
didcs = True didcs = True
m = re.search(rb'(.*<title>)(.*)(<\/title>.*)', line) m = re.search(b'''(.*<title>)(.*)(<\/title>.*)''', line)
if not m: if not m:
m = re.search(rb'(.*content=")(.*)(".*/>.*)', line) m = re.search(b'''(.*content=")(.*)(".*/>.*)''', line)
if m: if m:
line = m.group(1) + self.em.htmlescape(m.group(2)) + \ line = m.group(1) + self.em.htmlescape(m.group(2)) + \
m.group(3) m.group(3)