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

HTML Input: Handle absolute paths in resource links on windows correctly. Fixes #3031 (HTML input: Improper handling of local URLs)

Browse Source
This commit is contained in:
Kovid Goyal 2010-06-23 14:54:54 -06:00
parent 344141ff45
commit 7ea679768e
1 changed files with 25 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

32
src/calibre/ebooks/html/input.py
View File

@ -20,7 +20,7 @@ from itertools import izip
from calibre.customize.conversion import InputFormatPlugin from calibre.customize.conversion import InputFormatPlugin
from calibre.ebooks.chardet import xml_to_unicode from calibre.ebooks.chardet import xml_to_unicode
from calibre.customize.conversion import OptionRecommendation from calibre.customize.conversion import OptionRecommendation
from calibre.constants import islinux, isfreebsd from calibre.constants import islinux, isfreebsd, iswindows
from calibre import unicode_path from calibre import unicode_path
from calibre.utils.localization import get_lang from calibre.utils.localization import get_lang
from calibre.utils.filenames import ascii_filename from calibre.utils.filenames import ascii_filename
@ -32,9 +32,14 @@ class Link(object):
@classmethod @classmethod
def url_to_local_path(cls, url, base): def url_to_local_path(cls, url, base):
path = urlunparse(('', '', url.path, url.params, url.query, '')) path = url.path
isabs = False
if iswindows and path.startswith('/'):
path = path[1:]
isabs = True
path = urlunparse(('', '', path, url.params, url.query, ''))
path = unquote(path) path = unquote(path)
if os.path.isabs(path): if isabs or os.path.isabs(path):
return path return path
return os.path.abspath(os.path.join(base, path)) return os.path.abspath(os.path.join(base, path))
@ -411,10 +416,23 @@ class HTMLInput(InputFormatPlugin):
def resource_adder(self, link_, base=None): def resource_adder(self, link_, base=None):
link = self.urlnormalize(link_) if not isinstance(link_, unicode):
link, frag = self.urldefrag(link) try:
link = unquote(link).replace('/', os.sep) link_ = link_.decode('utf-8', 'error')
if not link.strip(): except:
self.log.warn('Failed to decode link %r. Ignoring'%link_)
return link_
try:
l = Link(link_, base if base else os.path.getcwdu())
except:
self.log.exception('Failed to process link: %r'%link_)
return link_
if l.path is None:
# Not a local resource
return link_
link = l.path.replace('/', os.sep).strip()
frag = l.fragment
if not link:
return link_ return link_
try: try:
if base and not os.path.isabs(link): if base and not os.path.isabs(link):