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

delete microsoft smart tags during preprocess

Browse Source
This commit is contained in:
ldolse 2011-01-07 03:40:47 +08:00
parent fb893a7c37
commit 439b8c0f21
1 changed files with 3 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

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

@ -234,8 +234,11 @@ class PreProcessor(object):
self.log("replaced "+unicode(self.found_indents)+ " nbsp indents with inline styles") self.log("replaced "+unicode(self.found_indents)+ " nbsp indents with inline styles")
# remove remaining non-breaking spaces # remove remaining non-breaking spaces
html = re.sub(ur'\u00a0', ' ', html) html = re.sub(ur'\u00a0', ' ', html)
# Get rid of various common microsoft specific tags which can cause issues later
# Get rid of empty <o:p> tags to simplify other processing # Get rid of empty <o:p> tags to simplify other processing
html = re.sub(ur'\s*<o:p>\s*</o:p>', ' ', html) html = re.sub(ur'\s*<o:p>\s*</o:p>', ' ', html)
# Delete microsoft 'smart' tags
html = re.sub('(?i)</?st1:\w+>', '', html)
# Get rid of empty span, bold, & italics tags # Get rid of empty span, bold, & italics tags
html = re.sub(r"\s*<span[^>]*>\s*(<span[^>]*>\s*</span>){0,2}\s*</span>\s*", " ", html) html = re.sub(r"\s*<span[^>]*>\s*(<span[^>]*>\s*</span>){0,2}\s*</span>\s*", " ", html)
html = re.sub(r"\s*<[ibu][^>]*>\s*(<[ibu][^>]*>\s*</[ibu]>\s*){0,2}\s*</[ibu]>", " ", html) html = re.sub(r"\s*<[ibu][^>]*>\s*(<[ibu][^>]*>\s*</[ibu]>\s*){0,2}\s*</[ibu]>", " ", html)