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

MOBI Input: Fix regression affecting detection of table of contents and metadata embedded in content when converting MOBI files. Also fix #2407 (error converting to lrf in 0.5.10)

Browse Source
This commit is contained in:
Kovid Goyal 2009-05-09 01:12:23 -07:00
parent 1c0f55e30c
commit b839cc76e1
1 changed files with 35 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

49
src/calibre/ebooks/mobi/reader.py
View File

@ -276,19 +276,6 @@ class MobiReader(object):
self.replace_page_breaks() self.replace_page_breaks()
self.cleanup_html() self.cleanup_html()
if self.processed_html.startswith('<body'):
self.processed_html = '<html><head></head>'+self.processed_html+'</html>'
self.processed_html = \
re.compile('<head>', re.IGNORECASE).sub(
'\n<head>\n'
'<style type="text/css">\n'
'blockquote { margin: 0em 0em 0em 1.25em; text-align: justify; }\n'
'p { margin: 0em; text-align: justify; }\n'
'.bold { font-weight: bold; }\n'
'.italic { font-style: italic; }\n'
'</style>\n',
self.processed_html)
if self.verbose: if self.verbose:
print 'Parsing HTML...' print 'Parsing HTML...'
root = html.fromstring(self.processed_html) root = html.fromstring(self.processed_html)
@ -296,7 +283,7 @@ class MobiReader(object):
from lxml.html import soupparser from lxml.html import soupparser
print 'Markup contains unclosed <p> tags, parsing using BeatifulSoup' print 'Markup contains unclosed <p> tags, parsing using BeatifulSoup'
root = soupparser.fromstring(self.processed_html) root = soupparser.fromstring(self.processed_html)
if root[0].tag != 'html': if root.tag != 'html':
nroot = html.fromstring('<html><head></head><body></body></html>') nroot = html.fromstring('<html><head></head><body></body></html>')
bod = nroot.find('body') bod = nroot.find('body')
for child in list(root): for child in list(root):
@ -304,6 +291,40 @@ class MobiReader(object):
bod.append(child) bod.append(child)
root = nroot root = nroot
htmls = list(root.xpath('//html'))
if len(htmls) > 1:
print 'Markup contains multiple <html> tags'
# Keep only the largest head and body
bodies, heads = root.xpath('//body'), root.xpath('//head')
def sz(x): return len(list(x.iter()))
def scmp(x, y): return cmp(sz(x), sz(y))
body = list(sorted(bodies, cmp=scmp))
head = list(sorted(heads, cmp=scmp))
for x in root: root.remove(x)
if head:
root.append(head[-1])
if body:
root.append(body[-1])
for x in root.xpath('//script'):
x.getparent().remove(x)
head = root.xpath('//head')
if head:
head = head[0]
else:
head = root.makeelement('head', {})
root.insert(0, head)
head.text = '\n\t'
style = head.makeelement('style', {'type':'text/css'})
head.insert(0, style)
style.tail = '\n\t'
style.text = '''
blockquote { margin: 0em 0em 0em 1.25em; text-align: justify }
p { margin: 0em; text-align: justify }
.bold { font-weight: bold }
.italic { font-style: italic }
'''
self.upshift_markup(root) self.upshift_markup(root)
guides = root.xpath('//guide') guides = root.xpath('//guide')
guide = guides[0] if guides else None guide = guides[0] if guides else None