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

...

Browse Source
This commit is contained in:
Kovid Goyal 2011-09-28 09:29:25 -06:00
parent c798239fd7
commit 9a366a92d9
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
src/calibre/ebooks/conversion/utils.py
View File

@ -315,7 +315,7 @@ class HeuristicProcessor(object):
supports a range of html markup and text files supports a range of html markup and text files
''' '''
# define the pieces of the regex # define the pieces of the regex
lookahead = "(?<=.{"+str(length)+u"}([a-zäëïöüàèìòùáćéíĺóŕńśúýâêîôûçąężıãõñæøþðßěľščťžňďřů,:“”)\IA\u00DF]|(?<!\&\w{4});))" # (?<!\&\w{4});) is a semicolon not part of an entity lookahead = "(?<=.{"+str(length)+u"}([a-zäëïöüàèìòùáćéíĺóŕńśúýâêîôûçąężıãõñæøþðßěľščťžňďřů,:“”)\IA\u00DF]|(?<!\&\w{4});))" # (?<!\&\w{4});) is a semicolon not part of an entity
em_en_lookahead = "(?<=.{"+str(length)+u"}[\u2013\u2014])" em_en_lookahead = "(?<=.{"+str(length)+u"}[\u2013\u2014])"
soft_hyphen = u"\xad" soft_hyphen = u"\xad"
@ -770,6 +770,7 @@ class HeuristicProcessor(object):
# Multiple sequential blank paragraphs are merged with appropriate margins # Multiple sequential blank paragraphs are merged with appropriate margins
# If non-blank scene breaks exist they are center aligned and styled with appropriate margins. # If non-blank scene breaks exist they are center aligned and styled with appropriate margins.
if getattr(self.extra_opts, 'format_scene_breaks', False): if getattr(self.extra_opts, 'format_scene_breaks', False):
self.log.debug('Formatting scene breaks')
html = re.sub('(?i)<div[^>]*>\s*<br(\s?/)?>\s*</div>', '<p></p>', html) html = re.sub('(?i)<div[^>]*>\s*<br(\s?/)?>\s*</div>', '<p></p>', html)
html = self.detect_whitespace(html) html = self.detect_whitespace(html)
html = self.detect_soft_breaks(html) html = self.detect_soft_breaks(html)