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

When creating ZIP/EPUB files encode all file names in UTF-8 and set the UTF-8 bit marker. When extracting from a ZIP/EPUB file, if the UTF-8 bit marker is not set, still assume that the file name is encoded in UTF-8. If UTF-8 decoding fails try to detect character encoding using chardet. This is because most ZIP/EPUB files calibre comes across seem to be utf-8 encoded.

Browse Source
This commit is contained in:
Kovid Goyal 2010-03-06 11:06:40 -07:00
parent cd531451bb
commit 65c3a9c2a3
1 changed files with 17 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
src/calibre/utils/zipfile.py
View File

@ -138,12 +138,16 @@ _CD64_OFFSET_START_CENTDIR = 9
def decode_arcname(name): def decode_arcname(name):
if not isinstance(name, unicode): if not isinstance(name, unicode):
encoding = detect(name)['encoding']
try: try:
name = name.decode(encoding) name = name.decode('utf-8')
except: except:
name = name.decode('utf-8', 'replace') res = detect(name)
return name.encode(filesystem_encoding, 'replace') encoding = res['encoding']
try:
name = name.decode(encoding)
except:
name = name.decode('utf-8', 'replace')
return name
def is_zipfile(filename): def is_zipfile(filename):
@ -352,10 +356,7 @@ class ZipInfo (object):
def _encodeFilenameFlags(self): def _encodeFilenameFlags(self):
if isinstance(self.filename, unicode): if isinstance(self.filename, unicode):
try: return self.filename.encode('utf-8'), self.flag_bits | 0x800
return self.filename.encode('ascii'), self.flag_bits
except:
return self.filename.encode('utf-8'), self.flag_bits | 0x800
else: else:
return self.filename, self.flag_bits return self.filename, self.flag_bits
@ -363,7 +364,7 @@ class ZipInfo (object):
if self.flag_bits & 0x800: if self.flag_bits & 0x800:
return self.filename.decode('utf-8') return self.filename.decode('utf-8')
else: else:
return self.filename return decode_arcname(self.filename)
def _decodeExtra(self): def _decodeExtra(self):
# Try to decode the extra field. # Try to decode the extra field.
@ -1059,7 +1060,9 @@ class ZipFile:
targetpath = targetpath[:-1] targetpath = targetpath[:-1]
# don't include leading "/" from file name if present # don't include leading "/" from file name if present
fname = decode_arcname(member.filename) fname = member.filename
if isinstance(fname, unicode):
fname = fname.encode(filesystem_encoding, 'replace')
if fname.startswith('/'): if fname.startswith('/'):
fname = fname[1:] fname = fname[1:]
targetpath = os.path.join(targetpath, fname) targetpath = os.path.join(targetpath, fname)
@ -1111,8 +1114,6 @@ class ZipFile:
def write(self, filename, arcname=None, compress_type=None): def write(self, filename, arcname=None, compress_type=None):
"""Put the bytes from filename into the archive under the name """Put the bytes from filename into the archive under the name
arcname.""" arcname."""
if isinstance(filename, unicode):
filename = filename.encode('utf-8')
if not self.fp: if not self.fp:
raise RuntimeError( raise RuntimeError(
"Attempt to write to ZIP archive that was already closed") "Attempt to write to ZIP archive that was already closed")
@ -1126,6 +1127,8 @@ class ZipFile:
arcname = os.path.normpath(os.path.splitdrive(arcname)[1]) arcname = os.path.normpath(os.path.splitdrive(arcname)[1])
while arcname[0] in (os.sep, os.altsep): while arcname[0] in (os.sep, os.altsep):
arcname = arcname[1:] arcname = arcname[1:]
if not isinstance(arcname, unicode):
arcname = arcname.decode(filesystem_encoding)
zinfo = ZipInfo(arcname, date_time) zinfo = ZipInfo(arcname, date_time)
zinfo.external_attr = (st[0] & 0xFFFF) << 16L # Unix attributes zinfo.external_attr = (st[0] & 0xFFFF) << 16L # Unix attributes
if compress_type is None: if compress_type is None:
@ -1187,8 +1190,8 @@ class ZipFile:
assert not raw_bytes or (raw_bytes and assert not raw_bytes or (raw_bytes and
isinstance(zinfo_or_arcname, ZipInfo)) isinstance(zinfo_or_arcname, ZipInfo))
if not isinstance(zinfo_or_arcname, ZipInfo): if not isinstance(zinfo_or_arcname, ZipInfo):
if isinstance(zinfo_or_arcname, unicode): if not isinstance(zinfo_or_arcname, unicode):
zinfo_or_arcname = zinfo_or_arcname.encode('utf-8') zinfo_or_arcname = zinfo_or_arcname.decode(filesystem_encoding)
zinfo = ZipInfo(filename=zinfo_or_arcname, zinfo = ZipInfo(filename=zinfo_or_arcname,
date_time=time.localtime(time.time())[:6]) date_time=time.localtime(time.time())[:6])
zinfo.compress_type = compression zinfo.compress_type = compression