Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-08 02:34:06 -04:00
Code Issues Packages Projects Releases Wiki Activity

pdftohtml processing: fix spaces rule

Browse Source
This commit is contained in:
John Schember 2009-04-09 19:16:49 -04:00
parent f7ec532d57
commit ac0af1b844
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/calibre/ebooks/conversion/preprocess.py
View File

@ -72,11 +72,11 @@ class HTMLPreProcessor(object):
# Have paragraphs show better # Have paragraphs show better
(re.compile(r'<br.*?>'), lambda match : '<p>'), (re.compile(r'<br.*?>'), lambda match : '<p>'),
# Re wrap lines # Un wrap lines
(re.compile(r'(?<=\w)\s*</i>\s*<p.*?>\s*<i>\s*(?=\w)'), lambda match: ' '), (re.compile(r'(?<=\w)\s*</i>\s*<p.*?>\s*<i>\s*(?=\w)'), lambda match: ' '),
(re.compile(r'(?<=\w)\s*<p.*?>\s*(?=\w)', re.UNICODE), lambda match: ' '), (re.compile(r'(?<=\w)\s*<p.*?>\s*(?=\w)', re.UNICODE), lambda match: ' '),
# Clean up spaces # Clean up spaces
(re.compile(ru'(?<=\.|,|:|;|\?|!|”|"|\')[\s^ ]*(?=<)'), lambda match: ' '), (re.compile(u'(?<=\.|,|:|;|\?|!|”|"|\')[\s^ ]*(?=<)'), lambda match: ' '),
] ]
# Fix Book Designer markup # Fix Book Designer markup