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

modified blank line handling to better handle lrf formats and preserve scene breaks

Browse Source
This commit is contained in:
ldolse 2011-02-04 18:06:13 +08:00
parent 96b6dede27
commit 070df88d20
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/utils.py
View File

@ -28,8 +28,8 @@ class HeuristicProcessor(object):
self.linereg = re.compile('(?<=<p).*?(?=</p>)', re.IGNORECASE|re.DOTALL) self.linereg = re.compile('(?<=<p).*?(?=</p>)', re.IGNORECASE|re.DOTALL)
self.blankreg = re.compile(r'\s*(?P<openline><p(?!\sclass=\"(softbreak|whitespace)\")[^>]*>)\s*(?P<closeline></p>)', re.IGNORECASE) self.blankreg = re.compile(r'\s*(?P<openline><p(?!\sclass=\"(softbreak|whitespace)\")[^>]*>)\s*(?P<closeline></p>)', re.IGNORECASE)
self.anyblank = re.compile(r'\s*(?P<openline><p[^>]*>)\s*(?P<closeline></p>)', re.IGNORECASE) self.anyblank = re.compile(r'\s*(?P<openline><p[^>]*>)\s*(?P<closeline></p>)', re.IGNORECASE)
self.multi_blank = re.compile(r'(\s*<p[^>]*>\s*</p>){2,}(?!\s*<h\d)', re.IGNORECASE) self.multi_blank = re.compile(r'(\s*(<div[^>]*>\s*)?<p[^>]*>\s*</p>(\s*</div>)?(\s*<div[^>]*>\s*</div>\s*)*){2,}(?!\s*<h\d)', re.IGNORECASE)
self.any_multi_blank = re.compile(r'(\s*<p[^>]*>\s*</p>){2,}', re.IGNORECASE) self.any_multi_blank = re.compile(r'(\s*(<div[^>]*>\s*)?<p[^>]*>\s*</p>(\s*</div>)?(\s*<div[^>]*>\s*</div>\s*)*){2,}', re.IGNORECASE)
self.line_open = "<(?P<outer>p|div)[^>]*>\s*(<(?P<inner1>font|span|[ibu])[^>]*>)?\s*(<(?P<inner2>font|span|[ibu])[^>]*>)?\s*(<(?P<inner3>font|span|[ibu])[^>]*>)?\s*" self.line_open = "<(?P<outer>p|div)[^>]*>\s*(<(?P<inner1>font|span|[ibu])[^>]*>)?\s*(<(?P<inner2>font|span|[ibu])[^>]*>)?\s*(<(?P<inner3>font|span|[ibu])[^>]*>)?\s*"
self.line_close = "(</(?P=inner3)>)?\s*(</(?P=inner2)>)?\s*(</(?P=inner1)>)?\s*</(?P=outer)>" self.line_close = "(</(?P=inner3)>)?\s*(</(?P=inner2)>)?\s*(</(?P=inner1)>)?\s*</(?P=outer)>"
self.single_blank = re.compile(r'(\s*<p[^>]*>\s*</p>)', re.IGNORECASE) self.single_blank = re.compile(r'(\s*<p[^>]*>\s*</p>)', re.IGNORECASE)