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 #6928 (Calibre corrupts EPUB)

Browse Source
This commit is contained in:
Kovid Goyal 2010-09-24 13:22:59 -06:00
parent ee0b2541c8
commit 54c3e009d5
2 changed files with 17 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
src/calibre/ebooks/oeb/iterator.py
View File

@ -15,7 +15,7 @@ from calibre.customize.ui import available_input_formats
from calibre.ebooks.metadata.opf2 import OPF from calibre.ebooks.metadata.opf2 import OPF
from calibre.ptempfile import TemporaryDirectory from calibre.ptempfile import TemporaryDirectory
from calibre.ebooks.chardet import xml_to_unicode from calibre.ebooks.chardet import xml_to_unicode
from calibre.utils.zipfile import safe_replace, ZipFile from calibre.utils.zipfile import safe_replace
from calibre.utils.config import DynamicConfig from calibre.utils.config import DynamicConfig
from calibre.utils.logging import Log from calibre.utils.logging import Log
from calibre import guess_type, prints from calibre import guess_type, prints
@ -294,12 +294,8 @@ class EbookIterator(object):
zf = open(self.pathtoebook, 'r+b') zf = open(self.pathtoebook, 'r+b')
except IOError: except IOError:
return return
zipf = ZipFile(zf, mode='a') safe_replace(zf, 'META-INF/calibre_bookmarks.txt', StringIO(dat),
for name in zipf.namelist(): add_missing=True)
if name == 'META-INF/calibre_bookmarks.txt':
safe_replace(zf, 'META-INF/calibre_bookmarks.txt', StringIO(dat))
return
zipf.writestr('META-INF/calibre_bookmarks.txt', dat)
else: else:
self.config['bookmarks_'+self.pathtoebook] = dat self.config['bookmarks_'+self.pathtoebook] = dat

16
src/calibre/utils/zipfile.py
View File

@ -281,7 +281,7 @@ class ZipInfo (object):
'file_offset', 'file_offset',
) )
def __init__(self, filename="NoName", date_time=(1980,1,1,0,0,0)): def __init__(self, filename=u"NoName", date_time=(1980,1,1,0,0,0)):
self.orig_filename = filename # Original file name in archive self.orig_filename = filename # Original file name in archive
# Terminate the file name at the first null byte. Null bytes in file # Terminate the file name at the first null byte. Null bytes in file
@ -1362,30 +1362,42 @@ class ZipFile:
self.fp.close() self.fp.close()
self.fp = None self.fp = None
def safe_replace(zipstream, name, datastream, extra_replacements={}): def safe_replace(zipstream, name, datastream, extra_replacements={},
add_missing=False):
''' '''
Replace a file in a zip file in a safe manner. This proceeds by extracting Replace a file in a zip file in a safe manner. This proceeds by extracting
and re-creating the zipfile. This is necessary because :method:`ZipFile.replace` and re-creating the zipfile. This is necessary because :method:`ZipFile.replace`
sometimes created corrupted zip files. sometimes created corrupted zip files.
:param zipstream: Stream from a zip file :param zipstream: Stream from a zip file
:param name: The name of the file to replace :param name: The name of the file to replace
:param datastream: The data to replace the file with. :param datastream: The data to replace the file with.
:param extra_replacements: Extra replacements. Mapping of name to file-like :param extra_replacements: Extra replacements. Mapping of name to file-like
objects objects
:param add_missing: If a replacement does not exist in the zip file, it is
added. Use with care as currently parent directories
are not created.
''' '''
z = ZipFile(zipstream, 'r') z = ZipFile(zipstream, 'r')
replacements = {name:datastream} replacements = {name:datastream}
replacements.update(extra_replacements) replacements.update(extra_replacements)
names = frozenset(replacements.keys()) names = frozenset(replacements.keys())
found = set([])
with SpooledTemporaryFile(max_size=100*1024*1024) as temp: with SpooledTemporaryFile(max_size=100*1024*1024) as temp:
ztemp = ZipFile(temp, 'w') ztemp = ZipFile(temp, 'w')
for obj in z.infolist(): for obj in z.infolist():
if isinstance(obj.filename, unicode):
obj.flag_bits |= 0x16 # Set isUTF-8 bit
if obj.filename in names: if obj.filename in names:
ztemp.writestr(obj, replacements[obj.filename].read()) ztemp.writestr(obj, replacements[obj.filename].read())
found.add(obj.filename)
else: else:
ztemp.writestr(obj, z.read_raw(obj), raw_bytes=True) ztemp.writestr(obj, z.read_raw(obj), raw_bytes=True)
if add_missing:
for name in names - found:
ztemp.writestr(name, replacements[name].read())
ztemp.close() ztemp.close()
z.close() z.close()
temp.seek(0) temp.seek(0)