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 2012-04-20 22:40:30 +05:30
commit 80c062c62b
2 changed files with 11 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/calibre/ebooks/conversion/preprocess.py
View File

@ -559,7 +559,7 @@ class HTMLPreProcessor(object):
end_rules.append((re.compile(u'(?<=.{%i}[–—])\s*<p>\s*(?=[[a-z\d])' % length), lambda match: '')) end_rules.append((re.compile(u'(?<=.{%i}[–—])\s*<p>\s*(?=[[a-z\d])' % length), lambda match: ''))
end_rules.append( end_rules.append(
# Un wrap using punctuation # Un wrap using punctuation
(re.compile(u'(?<=.{%i}([a-zäëïöüàèìòùáćéíĺóŕńśúýâêîôûçąężıãõñæøþðßěľščťžňďřů,:“”)\IA\u00DF]|(?<!\&\w{4});))\s*(?P<ital></(i|b|u)>)?\s*(</p>\s*<p>\s*)+\s*(?=(<(i|b|u)>)?\s*[\w\d$(])' % length, re.UNICODE), wrap_lines), (re.compile(u'(?<=.{%i}([a-zäëïöüàèìòùáćéíĺóŕńśúýâêîôûçąężıãõñæøþðßěľščťžňďřů,:)\IA\u00DF]|(?<!\&\w{4});))\s*(?P<ital></(i|b|u)>)?\s*(</p>\s*<p>\s*)+\s*(?=(<(i|b|u)>)?\s*[\w\d$(])' % length, re.UNICODE), wrap_lines),
) )
for rule in self.PREPROCESS + start_rules: for rule in self.PREPROCESS + start_rules:

12
src/calibre/ebooks/conversion/utils.py
View File

@ -316,10 +316,18 @@ class HeuristicProcessor(object):
''' '''
Unwraps lines based on line length and punctuation Unwraps lines based on line length and punctuation
supports a range of html markup and text files supports a range of html markup and text files
the lookahead regex below is meant look for any non-full stop characters - punctuation
characters which can be used as a full stop should *not* be added below - e.g. ?!“”. etc
the reason for this is to prevent false positive wrapping. False positives are more
difficult to detect than false negatives during a manual review of the doc
This function intentionally leaves hyphenated content alone as that is handled by the
dehyphenate routine in a separate step
''' '''
# 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 # 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
em_en_lookahead = "(?<=.{"+str(length)+u"}[\u2013\u2014])" em_en_lookahead = "(?<=.{"+str(length)+u"}[\u2013\u2014])"
soft_hyphen = u"\xad" soft_hyphen = u"\xad"
line_ending = "\s*</(span|[iubp]|div)>\s*(</(span|[iubp]|div)>)?" line_ending = "\s*</(span|[iubp]|div)>\s*(</(span|[iubp]|div)>)?"