Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Fix #1091631 (Private bug)

Browse Source
This commit is contained in:
Kovid Goyal 2012-12-18 18:58:21 +05:30
parent 0797811191
commit 925afbf8c1
1 changed files with 13 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
src/calibre/ebooks/conversion/utils.py
View File

@ -357,21 +357,24 @@ class HeuristicProcessor(object):
line_opening = "<(p|div)[^>]*>\s*(?P<style_open><(span|[iub])[^>]*>)?\s*" line_opening = "<(p|div)[^>]*>\s*(?P<style_open><(span|[iub])[^>]*>)?\s*"
txt_line_wrap = u"((\u0020|\u0009)*\n){1,4}" txt_line_wrap = u"((\u0020|\u0009)*\n){1,4}"
unwrap_regex = lookahead+line_ending+blanklines+line_opening
em_en_unwrap_regex = em_en_lookahead+line_ending+blanklines+line_opening
shy_unwrap_regex = soft_hyphen+line_ending+blanklines+line_opening
if format == 'txt': if format == 'txt':
unwrap_regex = lookahead+txt_line_wrap unwrap_regex = lookahead+txt_line_wrap
em_en_unwrap_regex = em_en_lookahead+txt_line_wrap em_en_unwrap_regex = em_en_lookahead+txt_line_wrap
shy_unwrap_regex = soft_hyphen+txt_line_wrap shy_unwrap_regex = soft_hyphen+txt_line_wrap
content = unwrap_regex.sub(' ', content)
content = em_en_unwrap_regex.sub('', content)
content = shy_unwrap_regex.sub('', content)
else: else:
unwrap = re.compile(u"%s" % unwrap_regex, re.UNICODE) unwrap_regex = lookahead+line_ending+blanklines+line_opening
em_en_unwrap = re.compile(u"%s" % em_en_unwrap_regex, re.UNICODE) em_en_unwrap_regex = em_en_lookahead+line_ending+blanklines+line_opening
shy_unwrap = re.compile(u"%s" % shy_unwrap_regex, re.UNICODE) shy_unwrap_regex = soft_hyphen+line_ending+blanklines+line_opening
unwrap = re.compile(u"%s" % unwrap_regex, re.UNICODE)
em_en_unwrap = re.compile(u"%s" % em_en_unwrap_regex, re.UNICODE)
shy_unwrap = re.compile(u"%s" % shy_unwrap_regex, re.UNICODE)
if format == 'txt':
content = unwrap.sub(' ', content)
content = em_en_unwrap.sub('', content)
content = shy_unwrap.sub('', content)
else:
content = unwrap.sub(style_unwrap, content) content = unwrap.sub(style_unwrap, content)
content = em_en_unwrap.sub(style_unwrap, content) content = em_en_unwrap.sub(style_unwrap, content)
content = shy_unwrap.sub(style_unwrap, content) content = shy_unwrap.sub(style_unwrap, content)