Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-08-07 09:01:38 -04:00
Code Issues Packages Projects Releases Wiki Activity

HTML Input: Fix handling of files with ä characters in their filenames. Fixes #919931 (ebook-convert crash on converting from html-source)

Browse Source
This commit is contained in:
Kovid Goyal 2012-01-22 16:32:20 +05:30
parent f998c8bf4b
commit 7dd371e5e3
2 changed files with 16 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/calibre/ebooks/html/input.py
View File

@ -475,7 +475,9 @@ class HTMLInput(InputFormatPlugin):
# bhref refers to an already existing file. The read() method of
# DirContainer will call unquote on it before trying to read the
# file, therefore we quote it here.
item.html_input_href = quote(bhref)
if isinstance(bhref, unicode):
bhref = bhref.encode('utf-8')
item.html_input_href = quote(bhref).decode('utf-8')
if guessed in self.OEB_STYLES:
item.override_css_fetch = partial(
self.css_import_handler, os.path.dirname(link))

17
src/calibre/ebooks/oeb/base.py
View File

@ -425,15 +425,24 @@ class DirContainer(object):
self.opfname = path
return
def _unquote(self, path):
# urlunquote must run on a bytestring and will return a bytestring
# if it runs on a unicode object, it returns a double encoded unicode
# string: unquote(u'%C3%A4') != unquote(b'%C3%A4').decode('utf-8')
# and the latter is correct
if isinstance(path, unicode):
path = path.encode('utf-8')
return urlunquote(path).decode('utf-8')
def read(self, path):
if path is None:
path = self.opfname
path = os.path.join(self.rootdir, path)
with open(urlunquote(path), 'rb') as f:
path = os.path.join(self.rootdir, self._unquote(path))
with open(path, 'rb') as f:
return f.read()
def write(self, path, data):
path = os.path.join(self.rootdir, urlunquote(path))
path = os.path.join(self.rootdir, self._unquote(path))
dir = os.path.dirname(path)
if not os.path.isdir(dir):
os.makedirs(dir)
@ -442,7 +451,7 @@ class DirContainer(object):
def exists(self, path):
try:
path = os.path.join(self.rootdir, urlunquote(path))
path = os.path.join(self.rootdir, self._unquote(path))
except ValueError: #Happens if path contains quoted special chars
return False
return os.path.isfile(path)