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

preserve style changes across line breaks in heuristics, fixes bug 1066507

Browse Source
This commit is contained in:
Lee 2012-12-08 21:26:14 +08:00
parent 92afbd1ab5
commit 67deb3adcd
1 changed files with 24 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

33
src/calibre/ebooks/conversion/utils.py
View File

@ -335,14 +335,26 @@ class HeuristicProcessor(object):
This function intentionally leaves hyphenated content alone as that is handled by the This function intentionally leaves hyphenated content alone as that is handled by the
dehyphenate routine in a separate step dehyphenate routine in a separate step
''' '''
def style_unwrap(match):
style_close = match.group('style_close')
style_open = match.group('style_open')
if style_open and style_close:
return style_close+' '+style_open
elif style_open and not style_close:
return ' '+style_open
elif not style_open and style_close:
return style_close+' '
else:
return ' '
# 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"
line_ending = "\s*</(span|[iubp]|div)>\s*(</(span|[iubp]|div)>)?" line_ending = "\s*(?P<style_close></(span|[iub])>)?\s*(</(p|div)>)?"
blanklines = "\s*(?P<up2threeblanks><(p|span|div)[^>]*>\s*(<(p|span|div)[^>]*>\s*</(span|p|div)>\s*)</(span|p|div)>\s*){0,3}\s*" blanklines = "\s*(?P<up2threeblanks><(p|span|div)[^>]*>\s*(<(p|span|div)[^>]*>\s*</(span|p|div)>\s*)</(span|p|div)>\s*){0,3}\s*"
line_opening = "<(span|[iubp]|div)[^>]*>\s*(<(span|[iubp]|div)[^>]*>)?\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 unwrap_regex = lookahead+line_ending+blanklines+line_opening
@ -353,14 +365,17 @@ class HeuristicProcessor(object):
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.sub(' ', content)
content = em_en_unwrap.sub('', content)
content = shy_unwrap.sub('', content)
else:
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)
content = unwrap.sub(style_unwrap, content)
content = em_en_unwrap.sub(style_unwrap, content)
content = shy_unwrap.sub(style_unwrap, content)
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)
content = unwrap.sub(' ', content)
content = em_en_unwrap.sub('', content)
content = shy_unwrap.sub('', content)
return content return content
def txt_process(self, match): def txt_process(self, match):