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

Fix #213

Browse Source
This commit is contained in:
Kovid Goyal 2007-09-23 06:36:37 +00:00
parent 5a21453397
commit ffd3ac2b21
1 changed files with 10 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
src/libprs500/ebooks/lrf/html/convert_from.py
View File

@ -196,10 +196,18 @@ class HTMLConverter(object):
raw = open(self.file_name, 'rb').read()
if self.pdftohtml:
nmassage.extend(HTMLConverter.PDFTOHTML)
raw = unicode(raw, 'utf8', 'replace')
soup = BeautifulSoup(raw,
#raw = unicode(raw, 'utf8', 'replace')
try:
soup = BeautifulSoup(raw,
convertEntities=BeautifulSoup.HTML_ENTITIES,
markupMassage=nmassage)
except ConversionError, err:
if 'Failed to coerce to unicode' in str(err):
raw = unicode(raw, 'utf8', 'replace')
soup = BeautifulSoup(raw,
convertEntities=BeautifulSoup.HTML_ENTITIES,
markupMassage=nmassage)
if not self.baen and self.is_baen(soup):
self.baen = True
self.logger.info('Baen file detected. Re-parsing...')