def update_cover(self):
# get the calibre cover
cover_path = self.db.cover(self.book_id, as_path=True)
fmt = cover_path.rpartition('.')[-1]
new_cover_name = "00000000_cover." + fmt
self.make_temp_cbz_file()
# search for a previously embeded cover
zf = ZipFile(self.file)
cover_info = ""
for name in zf.namelist():
if name.rsplit(".", 1)[0] == "00000000_cover":
cover_info = name
break
# delete previous cover
if cover_info != "":
with open(self.file, 'r+b') as zf, open(cover_path, 'r+b') as cp:
safe_replace(zf, cover_info, cp)
# save the cover in the file
else:
zf = ZipFile(self.file, "a")
zf.write(cover_path, new_cover_name)
zf.close()
delete_temp_file(cover_path)
def get_metadata(stream):
from calibre.ebooks.metadata.meta import get_metadata
from calibre.ebooks.metadata.archive import is_comic
stream_type = None
zf = ZipFile(stream, 'r')
names = zf.namelist()
if is_comic(names):
# Is probably a comic
return get_metadata(stream, 'cbz')
for f in names:
stream_type = os.path.splitext(f)[1].lower()
if stream_type:
stream_type = stream_type[1:]
if stream_type in ('lit', 'opf', 'prc', 'mobi', 'fb2', 'epub',
'rb', 'imp', 'pdf', 'lrf', 'azw', 'azw1', 'azw3'):
with TemporaryDirectory() as tdir:
with CurrentDir(tdir):
path = zf.extract(f)
mi = get_metadata(open(path,'rb'), stream_type)
if stream_type == 'opf' and mi.application_id is None:
try:
# zip archive opf files without an application_id were assumed not to have a cover
# reparse the opf and if cover exists read its data from zip archive for the metadata
nmi = zip_opf_metadata(path, zf)
nmi.timestamp = None
return nmi
except:
pass
mi.timestamp = None
return mi
raise ValueError('No ebook found in ZIP archive (%s)' % os.path.basename(getattr(stream, 'name', '') or '<stream>'))
def get_picture_size(self):
from calibre.utils.magick import Image
self.make_temp_cbz_file()
zf = ZipFile(self.file)
files = zf.namelist()
size_x, size_y = 0, 0
index = 1
while index < 10 and index < len(files):
fname = files[index]
if fname.lower().rpartition('.')[-1] in IMG_EXTENSIONS:
with zf.open(fname) as ffile:
img = Image()
try:
img.open(ffile)
size_x, size_y = img.size
except:
pass
if size_x < size_y:
break
index += 1
zf.close()
size = round(size_x * size_y / 1000000, 2)
return size
def update_cover(self):
# get the calibre cover
cover_path = self.db.cover(self.book_id, as_path=True)
fmt = cover_path.rpartition('.')[-1]
new_cover_name = "00000000_cover." + fmt
self.make_temp_cbz_file()
# search for a previously embeded cover
zf = ZipFile(self.file)
cover_info = ""
for name in zf.namelist():
if name.rsplit(".", 1)[0] == "00000000_cover":
cover_info = name
break
zf.close()
# delete previous cover
if cover_info != "":
with open(self.file, 'r+b') as zf, open(cover_path, 'r+b') as cp:
safe_replace(zf, cover_info, cp)
# save the cover in the file
else:
zf = ZipFile(self.file, "a")
zf.write(cover_path, new_cover_name)
zf.close()
delete_temp_file(cover_path)
def __enter__(self, *args):
'''
Add this plugin to the python path so that it's contents become directly importable.
Useful when bundling large python libraries into the plugin. Use it like this::
with plugin:
import something
'''
if self.plugin_path is not None:
from calibre.utils.zipfile import ZipFile
zf = ZipFile(self.plugin_path)
extensions = {x.rpartition('.')[-1].lower() for x in zf.namelist()}
zip_safe = True
for ext in ('pyd', 'so', 'dll', 'dylib'):
if ext in extensions:
zip_safe = False
break
if zip_safe:
sys.path.insert(0, self.plugin_path)
self.sys_insertion_path = self.plugin_path
else:
from calibre.ptempfile import TemporaryDirectory
self._sys_insertion_tdir = TemporaryDirectory('plugin_unzip')
self.sys_insertion_path = self._sys_insertion_tdir.__enter__(
*args)
zf.extractall(self.sys_insertion_path)
sys.path.insert(0, self.sys_insertion_path)
zf.close()
def get_comic_metadata_from_cbz(self):
'''
Reads the comic metadata from the comic cbz file as comictagger metadata
'''
self.make_temp_cbz_file()
# open the zipfile
zf = ZipFile(self.file)
# get cix metadata
for name in zf.namelist():
if name.lower() == "comicinfo.xml":
self.cix_metadata = ComicInfoXml().metadataFromString(
zf.read(name))
self.zipinfo = name
if not prefs['auto_count_pages']:
break
elif prefs['auto_count_pages'] and name.lower().rpartition(
'.')[-1] in IMG_EXTENSIONS:
self.pages += 1
# get the cbi metadata
if ComicBookInfo().validateString(zf.comment):
self.cbi_metadata = ComicBookInfo().metadataFromString(zf.comment)
zf.close()
# get combined metadata
self._get_combined_metadata()
def get_metadata(stream):
from calibre.ebooks.metadata.meta import get_metadata
from calibre.ebooks.metadata.archive import is_comic
stream_type = None
zf = ZipFile(stream, 'r')
names = zf.namelist()
if is_comic(names):
# Is probably a comic
return get_metadata(stream, 'cbz')
for f in names:
stream_type = os.path.splitext(f)[1].lower()
if stream_type:
stream_type = stream_type[1:]
if stream_type in ('lit', 'opf', 'prc', 'mobi', 'fb2', 'epub',
'rb', 'imp', 'pdf', 'lrf', 'azw', 'azw1', 'azw3'):
with TemporaryDirectory() as tdir:
with CurrentDir(tdir):
path = zf.extract(f)
mi = get_metadata(open(path,'rb'), stream_type)
if stream_type == 'opf' and mi.application_id is None:
try:
# zip archive opf files without an application_id were assumed not to have a cover
# reparse the opf and if cover exists read its data from zip archive for the metadata
nmi = zip_opf_metadata(path, zf)
nmi.timestamp = None
return nmi
except:
pass
mi.timestamp = None
return mi
raise ValueError('No ebook found in ZIP archive (%s)' % os.path.basename(getattr(stream, 'name', '') or '<stream>'))
def __enter__(self, *args):
'''
Add this plugin to the python path so that it's contents become directly importable.
Useful when bundling large python libraries into the plugin. Use it like this::
with plugin:
import something
'''
if self.plugin_path is not None:
from calibre.utils.zipfile import ZipFile
zf = ZipFile(self.plugin_path)
extensions = set([x.rpartition('.')[-1].lower() for x in
zf.namelist()])
zip_safe = True
for ext in ('pyd', 'so', 'dll', 'dylib'):
if ext in extensions:
zip_safe = False
break
if zip_safe:
sys.path.insert(0, self.plugin_path)
self.sys_insertion_path = self.plugin_path
else:
from calibre.ptempfile import TemporaryDirectory
self._sys_insertion_tdir = TemporaryDirectory('plugin_unzip')
self.sys_insertion_path = self._sys_insertion_tdir.__enter__(*args)
zf.extractall(self.sys_insertion_path)
sys.path.insert(0, self.sys_insertion_path)
zf.close()
def count_pages(self):
self.make_temp_cbz_file()
zf = ZipFile(self.file)
pages = 0
for name in zf.namelist():
if name.lower().rpartition('.')[-1] in IMG_EXTENSIONS:
pages += 1
return pages
def run(self, archive):
from calibre.utils.zipfile import ZipFile
is_rar = archive.lower().endswith('.rar')
if is_rar:
from calibre.utils.unrar import extract_member, names
else:
zf = ZipFile(archive, 'r')
if is_rar:
with open(archive, 'rb') as rf:
fnames = list(names(rf))
else:
fnames = zf.namelist()
def fname_ok(fname):
bn = os.path.basename(fname).lower()
if bn == 'thumbs.db':
return False
if '.' not in bn:
return False
if bn.rpartition('.')[-1] in {'diz', 'nfo'}:
return False
if '__MACOSX' in fname.split('/'):
return False
return True
fnames = list(filter(fname_ok, fnames))
if is_comic(fnames):
ext = '.cbr' if is_rar else '.cbz'
of = self.temporary_file('_archive_extract' + ext)
with open(archive, 'rb') as f:
of.write(f.read())
of.close()
return of.name
if len(fnames) > 1 or not fnames:
return archive
fname = fnames[0]
ext = os.path.splitext(fname)[1][1:]
if ext.lower() not in {
'lit', 'epub', 'mobi', 'prc', 'rtf', 'pdf', 'mp3', 'pdb',
'azw', 'azw1', 'azw3', 'fb2', 'docx', 'doc', 'odt'
}:
return archive
of = self.temporary_file('_archive_extract.' + ext)
with closing(of):
if is_rar:
with open(archive, 'rb') as f:
data = extract_member(f, match=None, name=fname)[1]
of.write(data)
else:
of.write(zf.read(fname))
return of.name
def run(self, archive):
from calibre.utils.zipfile import ZipFile
is_rar = archive.lower().endswith('.rar')
if is_rar:
from calibre.utils.unrar import extract_member, names
else:
zf = ZipFile(archive, 'r')
if is_rar:
with open(archive, 'rb') as rf:
fnames = list(names(rf))
else:
fnames = zf.namelist()
def fname_ok(fname):
bn = os.path.basename(fname).lower()
if bn == 'thumbs.db':
return False
if '.' not in bn:
return False
if bn.rpartition('.')[-1] in {'diz', 'nfo'}:
return False
if '__MACOSX' in fname.split('/'):
return False
return True
fnames = list(filter(fname_ok, fnames))
if is_comic(fnames):
ext = '.cbr' if is_rar else '.cbz'
of = self.temporary_file('_archive_extract'+ext)
with open(archive, 'rb') as f:
of.write(f.read())
of.close()
return of.name
if len(fnames) > 1 or not fnames:
return archive
fname = fnames[0]
ext = os.path.splitext(fname)[1][1:]
if ext.lower() not in {
'lit', 'epub', 'mobi', 'prc', 'rtf', 'pdf', 'mp3', 'pdb',
'azw', 'azw1', 'azw3', 'fb2', 'docx', 'doc', 'odt'}:
return archive
of = self.temporary_file('_archive_extract.'+ext)
with closing(of):
if is_rar:
with open(archive, 'rb') as f:
data = extract_member(f, match=None, name=fname)[1]
of.write(data)
else:
of.write(zf.read(fname))
return of.name
def get_fb2_data(stream):
from calibre.utils.zipfile import ZipFile, BadZipfile
pos = stream.tell()
try:
zf = ZipFile(stream)
except BadZipfile:
stream.seek(pos)
ans = stream.read()
zip_file_name = None
else:
names = zf.namelist()
names = [x for x in names if x.lower().endswith('.fb2')] or names
zip_file_name = names[0]
ans = zf.open(zip_file_name).read()
return ans, zip_file_name
def get_fb2_data(stream):
from calibre.utils.zipfile import ZipFile, BadZipfile
pos = stream.tell()
try:
zf = ZipFile(stream)
except BadZipfile:
stream.seek(pos)
ans = stream.read()
zip_file_name = None
else:
names = zf.namelist()
names = [x for x in names if x.lower().endswith('.fb2')] or names
zip_file_name = names[0]
ans = zf.open(zip_file_name).read()
return ans, zip_file_name
def count_pages(self):
self.make_temp_cbz_file()
# open the zipfile
zf = ZipFile(self.file)
# count the pages
for name in zf.namelist():
if name.lower().rpartition('.')[-1] in IMG_EXTENSIONS:
self.pages += 1
if self.pages == 0:
return False
update_custom_column(prefs['pages_column'], self.pages,
self.calibre_metadata,
self.db.field_metadata.custom_field_metadata())
self.db.set_metadata(self.book_id, self.calibre_metadata)
return True
def run(self, archive):
from calibre.utils.zipfile import ZipFile
is_rar = archive.lower().endswith('.rar')
if is_rar:
from calibre.utils.unrar import extract_member, names
else:
zf = ZipFile(archive, 'r')
if is_rar:
with open(archive, 'rb') as rf:
fnames = list(names(rf))
else:
fnames = zf.namelist()
fnames = [
x for x in fnames
if '.' in x and x.lower().rpartition('/')[-1] != 'thumbs.db'
]
if is_comic(fnames):
ext = '.cbr' if is_rar else '.cbz'
of = self.temporary_file('_archive_extract' + ext)
with open(archive, 'rb') as f:
of.write(f.read())
of.close()
return of.name
if len(fnames) > 1 or not fnames:
return archive
fname = fnames[0]
ext = os.path.splitext(fname)[1][1:]
if ext.lower() not in ('lit', 'epub', 'mobi', 'prc', 'rtf', 'pdf',
'mp3', 'pdb', 'azw', 'azw1', 'azw3', 'fb2'):
return archive
of = self.temporary_file('_archive_extract.' + ext)
with closing(of):
if is_rar:
with open(archive, 'rb') as f:
data = extract_member(f, match=None, name=fname)[1]
of.write(data)
else:
of.write(zf.read(fname))
return of.name
def __enter__(self, *args):
if self.plugin_path is not None:
from calibre.utils.zipfile import ZipFile
zf = ZipFile(self.plugin_path)
extensions = set([x.rpartition('.')[-1].lower() for x in
zf.namelist()])
zip_safe = True
for ext in ('pyd', 'so', 'dll', 'dylib'):
if ext in extensions:
zip_safe = False
if zip_safe:
sys.path.insert(0, self.plugin_path)
self.sys_insertion_path = self.plugin_path
else:
from calibre.ptempfile import TemporaryDirectory
self._sys_insertion_tdir = TemporaryDirectory('plugin_unzip')
self.sys_insertion_path = self._sys_insertion_tdir.__enter__(*args)
zf.extractall(self.sys_insertion_path)
sys.path.insert(0, self.sys_insertion_path)
zf.close()
def __enter__(self, *args):
if self.plugin_path is not None:
from calibre.utils.zipfile import ZipFile
zf = ZipFile(self.plugin_path)
extensions = set([x.rpartition('.')[-1].lower() for x in
zf.namelist()])
zip_safe = True
for ext in ('pyd', 'so', 'dll', 'dylib'):
if ext in extensions:
zip_safe = False
if zip_safe:
sys.path.insert(0, self.plugin_path)
self.sys_insertion_path = self.plugin_path
else:
from calibre.ptempfile import TemporaryDirectory
self._sys_insertion_tdir = TemporaryDirectory('plugin_unzip')
self.sys_insertion_path = self._sys_insertion_tdir.__enter__(*args)
zf.extractall(self.sys_insertion_path)
sys.path.insert(0, self.sys_insertion_path)
zf.close()
def run(self, archive):
from calibre.utils.zipfile import ZipFile
is_rar = archive.lower().endswith('.rar')
if is_rar:
from calibre.utils.unrar import extract_member, names
else:
zf = ZipFile(archive, 'r')
if is_rar:
with open(archive, 'rb') as rf:
fnames = list(names(rf))
else:
fnames = zf.namelist()
fnames = [x for x in fnames if '.' in x and x.lower().rpartition('/')[-1] != 'thumbs.db']
if is_comic(fnames):
ext = '.cbr' if is_rar else '.cbz'
of = self.temporary_file('_archive_extract'+ext)
with open(archive, 'rb') as f:
of.write(f.read())
of.close()
return of.name
if len(fnames) > 1 or not fnames:
return archive
fname = fnames[0]
ext = os.path.splitext(fname)[1][1:]
if ext.lower() not in ('lit', 'epub', 'mobi', 'prc', 'rtf', 'pdf',
'mp3', 'pdb', 'azw', 'azw1', 'azw3', 'fb2'):
return archive
of = self.temporary_file('_archive_extract.'+ext)
with closing(of):
if is_rar:
with open(archive, 'rb') as f:
data = extract_member(f, match=None, name=fname)[1]
of.write(data)
else:
of.write(zf.read(fname))
return of.name
def get_comic_metadata_from_cbz(self):
'''
Reads the comic metadata from the comic cbz file as comictagger metadata
'''
self.make_temp_cbz_file()
# open the zipfile
zf = ZipFile(self.file)
# get cix metadata
for name in zf.namelist():
if name.lower() == "comicinfo.xml":
self.cix_metadata = ComicInfoXml().metadataFromString(zf.read(name))
self.zipinfo = name
break
# get the cbi metadata
if ComicBookInfo().validateString(zf.comment):
self.cbi_metadata = ComicBookInfo().metadataFromString(zf.comment)
zf.close()
# get combined metadata
self._get_combined_metadata()
def get_comic_metadata_from_cbz(self):
'''
Reads the comic metadata from the comic cbz file as comictagger metadata
'''
self.make_temp_cbz_file()
# open the zipfile
zf = ZipFile(self.file)
# get cix metadata
for name in zf.namelist():
if name.lower() == "comicinfo.xml":
self.cix_metadata = ComicInfoXml().metadataFromString(zf.read(name))
self.zipinfo = name
break
# get the cbi metadata
if ComicBookInfo().validateString(zf.comment):
self.cbi_metadata = ComicBookInfo().metadataFromString(zf.comment)
zf.close()
# get combined metadata
self._get_combined_metadata()
class DOCX(object):
def __init__(self, path_or_stream, log=None, extract=True):
stream = path_or_stream if hasattr(path_or_stream, 'read') else open(path_or_stream, 'rb')
self.name = getattr(stream, 'name', None) or '<stream>'
self.log = log or default_log
if extract:
self.extract(stream)
else:
self.init_zipfile(stream)
self.read_content_types()
self.read_package_relationships()
def init_zipfile(self, stream):
self.zipf = ZipFile(stream)
self.names = frozenset(self.zipf.namelist())
def extract(self, stream):
self.tdir = PersistentTemporaryDirectory('docx_container')
try:
zf = ZipFile(stream)
zf.extractall(self.tdir)
except:
self.log.exception('DOCX appears to be invalid ZIP file, trying a'
' more forgiving ZIP parser')
from calibre.utils.localunzip import extractall
stream.seek(0)
extractall(stream, self.tdir)
self.names = {}
for f in walk(self.tdir):
name = os.path.relpath(f, self.tdir).replace(os.sep, '/')
self.names[name] = f
def exists(self, name):
return name in self.names
def read(self, name):
if hasattr(self, 'zipf'):
return self.zipf.open(name).read()
path = self.names[name]
with open(path, 'rb') as f:
return f.read()
def read_content_types(self):
try:
raw = self.read('[Content_Types].xml')
except KeyError:
raise InvalidDOCX('The file %s docx file has no [Content_Types].xml' % self.name)
root = fromstring(raw)
self.content_types = {}
self.default_content_types = {}
for item in root.xpath('//*[local-name()="Types"]/*[local-name()="Default" and @Extension and @ContentType]'):
self.default_content_types[item.get('Extension').lower()] = item.get('ContentType')
for item in root.xpath('//*[local-name()="Types"]/*[local-name()="Override" and @PartName and @ContentType]'):
name = item.get('PartName').lstrip('/')
self.content_types[name] = item.get('ContentType')
def content_type(self, name):
if name in self.content_types:
return self.content_types[name]
ext = name.rpartition('.')[-1].lower()
if ext in self.default_content_types:
return self.default_content_types[ext]
return guess_type(name)[0]
def read_package_relationships(self):
try:
raw = self.read('_rels/.rels')
except KeyError:
raise InvalidDOCX('The file %s docx file has no _rels/.rels' % self.name)
root = fromstring(raw)
self.relationships = {}
self.relationships_rmap = {}
for item in root.xpath('//*[local-name()="Relationships"]/*[local-name()="Relationship" and @Type and @Target]'):
target = item.get('Target').lstrip('/')
typ = item.get('Type')
self.relationships[typ] = target
self.relationships_rmap[target] = typ
@property
def document_name(self):
name = self.relationships.get(DOCUMENT, None)
if name is None:
names = tuple(n for n in self.names if n == 'document.xml' or n.endswith('/document.xml'))
if not names:
raise InvalidDOCX('The file %s docx file has no main document' % self.name)
name = names[0]
return name
@property
def document(self):
return fromstring(self.read(self.document_name))
@property
def document_relationships(self):
return self.get_relationships(self.document_name)
def get_relationships(self, name):
base = '/'.join(name.split('/')[:-1])
by_id, by_type = {}, {}
parts = name.split('/')
name = '/'.join(parts[:-1] + ['_rels', parts[-1] + '.rels'])
try:
raw = self.read(name)
except KeyError:
pass
else:
root = fromstring(raw)
for item in root.xpath('//*[local-name()="Relationships"]/*[local-name()="Relationship" and @Type and @Target]'):
target = item.get('Target')
if item.get('TargetMode', None) != 'External' and not target.startswith('#'):
target = '/'.join((base, target.lstrip('/')))
typ = item.get('Type')
Id = item.get('Id')
by_id[Id] = by_type[typ] = target
return by_id, by_type
@property
def metadata(self):
mi = Metadata(_('Unknown'))
name = self.relationships.get(DOCPROPS, None)
if name is None:
names = tuple(n for n in self.names if n.lower() == 'docprops/core.xml')
if names:
name = names[0]
if name:
try:
raw = self.read(name)
except KeyError:
pass
else:
read_doc_props(raw, mi)
if mi.is_null('language'):
try:
raw = self.read('word/styles.xml')
except KeyError:
pass
else:
read_default_style_language(raw, mi)
name = self.relationships.get(APPPROPS, None)
if name is None:
names = tuple(n for n in self.names if n.lower() == 'docprops/app.xml')
if names:
name = names[0]
if name:
try:
raw = self.read(name)
except KeyError:
pass
else:
read_app_props(raw, mi)
return mi
def close(self):
if hasattr(self, 'zipf'):
self.zipf.close()
else:
try:
shutil.rmtree(self.tdir)
except EnvironmentError:
pass
def _get_epub_toc(self, cid=None, path=None, prepend_title=None):
'''
Given a calibre id, return the epub TOC indexed by section
If cid, use copy in library, else use path to copy on device
'''
toc = None
# if cid is not None:
# mi = self.opts.gui.current_db.get_metadata(cid, index_is_id=True)
# toc = None
# if 'EPUB' in mi.formats:
# fpath = self.opts.gui.current_db.format(cid, 'EPUB', index_is_id=True, as_path=True)
# else:
# return toc
# elif path is not None:
# fpath = path
# else:
# return toc
fpath = path
# iBooks stores books unzipped
# Marvin stores books zipped
# Need spine, ncx_tree to construct toc
if self.ios.stat(fpath) and self.ios.stat(
fpath)['st_ifmt'] == 'S_IFDIR':
# Find the OPF in the unzipped ePub
fp = '/'.join([fpath, 'META-INF', 'container.xml'])
cf = io.BytesIO(self.ios.read(fp))
container = etree.parse(cf)
opf_file = container.xpath('.//*[local-name()="rootfile"]')[0].get(
'full-path')
oebps = opf_file.rpartition('/')[0]
fp = '/'.join([fpath, opf_file])
opf = io.BytesIO(self.ios.read(fp))
opf_tree = etree.parse(opf)
spine = opf_tree.xpath('.//*[local-name()="spine"]')[0]
ncx_fs = spine.get('toc')
manifest = opf_tree.xpath('.//*[local-name()="manifest"]')[0]
ncx_file = manifest.find('.//*[@id="%s"]' % ncx_fs).get('href')
fp = '/'.join([fpath, oebps, ncx_file])
ncxf = io.BytesIO(self.ios.read(fp))
ncx_tree = etree.parse(ncxf)
#self._log(etree.tostring(ncx_tree, pretty_print=True))
else:
# Find the OPF file in the zipped ePub
zfo = io.BytesIO(self.ios.read(fpath, mode='rb'))
try:
zf = ZipFile(zfo, 'r')
container = etree.fromstring(zf.read('META-INF/container.xml'))
opf_tree = etree.fromstring(
zf.read(
container.xpath('.//*[local-name()="rootfile"]')
[0].get('full-path')))
spine = opf_tree.xpath('.//*[local-name()="spine"]')[0]
ncx_fs = spine.get('toc')
manifest = opf_tree.xpath('.//*[local-name()="manifest"]')[0]
ncx = manifest.find('.//*[@id="%s"]' % ncx_fs).get('href')
# Find the ncx file
fnames = zf.namelist()
_ncx = [x for x in fnames if ncx in x][0]
ncx_tree = etree.fromstring(zf.read(_ncx))
except:
import traceback
self._log_location()
self._log(" unable to unzip '%s'" % fpath)
self._log(traceback.format_exc())
return toc
# fpath points to epub (zipped or unzipped dir)
# spine, ncx_tree populated
try:
toc = OrderedDict()
# 1. capture idrefs from spine
for i, el in enumerate(spine):
toc[str(i)] = el.get('idref')
# 2. Resolve <spine> idrefs to <manifest> hrefs
for el in toc:
toc[el] = manifest.find('.//*[@id="%s"]' % toc[el]).get('href')
# 3. Build a dict of src:toc_entry
src_map = OrderedDict()
navMap = ncx_tree.xpath('.//*[local-name()="navMap"]')[0]
for navPoint in navMap:
# Get the first-level entry
src = re.sub(
r'#.*$', '',
navPoint.xpath('.//*[local-name()="content"]')[0].get(
'src'))
toc_entry = navPoint.xpath('.//*[local-name()="text"]')[0].text
src_map[src] = toc_entry
# Get any nested navPoints
nested_navPts = navPoint.xpath('.//*[local-name()="navPoint"]')
for nnp in nested_navPts:
src = re.sub(
r'#.*$', '',
nnp.xpath('.//*[local-name()="content"]')[0].get(
'src'))
toc_entry = nnp.xpath('.//*[local-name()="text"]')[0].text
src_map[src] = toc_entry
# Resolve src paths to toc_entry
for section in toc:
if toc[section] in src_map:
if prepend_title:
toc[section] = "%s · %s" % (
prepend_title, src_map[toc[section]])
else:
toc[section] = src_map[toc[section]]
else:
toc[section] = None
# 5. Fill in the gaps
current_toc_entry = None
for section in toc:
if toc[section] is None:
toc[section] = current_toc_entry
else:
current_toc_entry = toc[section]
except:
import traceback
self._log_location()
self._log("{:~^80}".format(" error parsing '%s' " % fpath))
self._log(traceback.format_exc())
self._log("{:~^80}".format(" end traceback "))
return toc
class DOCX(object):
def __init__(self, path_or_stream, log=None, extract=True):
self.docx_is_transitional = True
stream = path_or_stream if hasattr(path_or_stream, 'read') else open(path_or_stream, 'rb')
self.name = getattr(stream, 'name', None) or '<stream>'
self.log = log or default_log
if extract:
self.extract(stream)
else:
self.init_zipfile(stream)
self.read_content_types()
self.read_package_relationships()
self.namespace = DOCXNamespace(self.docx_is_transitional)
def init_zipfile(self, stream):
self.zipf = ZipFile(stream)
self.names = frozenset(self.zipf.namelist())
def extract(self, stream):
self.tdir = PersistentTemporaryDirectory('docx_container')
try:
zf = ZipFile(stream)
zf.extractall(self.tdir)
except:
self.log.exception('DOCX appears to be invalid ZIP file, trying a'
' more forgiving ZIP parser')
from calibre.utils.localunzip import extractall
stream.seek(0)
extractall(stream, self.tdir)
self.names = {}
for f in walk(self.tdir):
name = os.path.relpath(f, self.tdir).replace(os.sep, '/')
self.names[name] = f
def exists(self, name):
return name in self.names
def read(self, name):
if hasattr(self, 'zipf'):
return self.zipf.open(name).read()
path = self.names[name]
with open(path, 'rb') as f:
return f.read()
def read_content_types(self):
try:
raw = self.read('[Content_Types].xml')
except KeyError:
raise InvalidDOCX('The file %s docx file has no [Content_Types].xml' % self.name)
root = fromstring(raw)
self.content_types = {}
self.default_content_types = {}
for item in root.xpath('//*[local-name()="Types"]/*[local-name()="Default" and @Extension and @ContentType]'):
self.default_content_types[item.get('Extension').lower()] = item.get('ContentType')
for item in root.xpath('//*[local-name()="Types"]/*[local-name()="Override" and @PartName and @ContentType]'):
name = item.get('PartName').lstrip('/')
self.content_types[name] = item.get('ContentType')
def content_type(self, name):
if name in self.content_types:
return self.content_types[name]
ext = name.rpartition('.')[-1].lower()
if ext in self.default_content_types:
return self.default_content_types[ext]
return guess_type(name)[0]
def read_package_relationships(self):
try:
raw = self.read('_rels/.rels')
except KeyError:
raise InvalidDOCX('The file %s docx file has no _rels/.rels' % self.name)
root = fromstring(raw)
self.relationships = {}
self.relationships_rmap = {}
for item in root.xpath('//*[local-name()="Relationships"]/*[local-name()="Relationship" and @Type and @Target]'):
target = item.get('Target').lstrip('/')
typ = item.get('Type')
if target == 'word/document.xml':
self.docx_is_transitional = typ != 'http://purl.oclc.org/ooxml/officeDocument/relationships/officeDocument'
self.relationships[typ] = target
self.relationships_rmap[target] = typ
@property
def document_name(self):
name = self.relationships.get(self.namespace.names['DOCUMENT'], None)
if name is None:
names = tuple(n for n in self.names if n == 'document.xml' or n.endswith('/document.xml'))
if not names:
raise InvalidDOCX('The file %s docx file has no main document' % self.name)
name = names[0]
return name
@property
def document(self):
return fromstring(self.read(self.document_name))
@property
def document_relationships(self):
return self.get_relationships(self.document_name)
def get_relationships(self, name):
base = '/'.join(name.split('/')[:-1])
by_id, by_type = {}, {}
parts = name.split('/')
name = '/'.join(parts[:-1] + ['_rels', parts[-1] + '.rels'])
try:
raw = self.read(name)
except KeyError:
pass
else:
root = fromstring(raw)
for item in root.xpath('//*[local-name()="Relationships"]/*[local-name()="Relationship" and @Type and @Target]'):
target = item.get('Target')
if item.get('TargetMode', None) != 'External' and not target.startswith('#'):
target = '/'.join((base, target.lstrip('/')))
typ = item.get('Type')
Id = item.get('Id')
by_id[Id] = by_type[typ] = target
return by_id, by_type
def get_document_properties_names(self):
name = self.relationships.get(self.namespace.names['DOCPROPS'], None)
if name is None:
names = tuple(n for n in self.names if n.lower() == 'docprops/core.xml')
if names:
name = names[0]
yield name
name = self.relationships.get(self.namespace.names['APPPROPS'], None)
if name is None:
names = tuple(n for n in self.names if n.lower() == 'docprops/app.xml')
if names:
name = names[0]
yield name
@property
def metadata(self):
mi = Metadata(_('Unknown'))
dp_name, ap_name = self.get_document_properties_names()
if dp_name:
try:
raw = self.read(dp_name)
except KeyError:
pass
else:
read_doc_props(raw, mi, self.namespace.XPath)
if mi.is_null('language'):
try:
raw = self.read('word/styles.xml')
except KeyError:
pass
else:
read_default_style_language(raw, mi, self.namespace.XPath)
ap_name = self.relationships.get(self.namespace.names['APPPROPS'], None)
if ap_name:
try:
raw = self.read(ap_name)
except KeyError:
pass
else:
read_app_props(raw, mi)
return mi
def close(self):
if hasattr(self, 'zipf'):
self.zipf.close()
else:
try:
shutil.rmtree(self.tdir)
except EnvironmentError:
pass
def _get_epub_toc(self, cid=None, path=None, prepend_title=None):
'''
Given a calibre id, return the epub TOC indexed by section
If cid, use copy in library, else use path to copy on device
'''
toc = None
# if cid is not None:
# mi = self.opts.gui.current_db.get_metadata(cid, index_is_id=True)
# toc = None
# if 'EPUB' in mi.formats:
# fpath = self.opts.gui.current_db.format(cid, 'EPUB', index_is_id=True, as_path=True)
# else:
# return toc
# elif path is not None:
# fpath = path
# else:
# return toc
fpath = path
# iBooks stores books unzipped
# Marvin stores books zipped
# Need spine, ncx_tree to construct toc
if self.ios.stat(fpath) and self.ios.stat(fpath)['st_ifmt'] == 'S_IFDIR':
# Find the OPF in the unzipped ePub
fp = '/'.join([fpath, 'META-INF', 'container.xml'])
cf = cStringIO.StringIO(self.ios.read(fp))
container = etree.parse(cf)
opf_file = container.xpath('.//*[local-name()="rootfile"]')[0].get('full-path')
oebps = opf_file.rpartition('/')[0]
fp = '/'.join([fpath, opf_file])
opf = cStringIO.StringIO(self.ios.read(fp))
opf_tree = etree.parse(opf)
spine = opf_tree.xpath('.//*[local-name()="spine"]')[0]
ncx_fs = spine.get('toc')
manifest = opf_tree.xpath('.//*[local-name()="manifest"]')[0]
ncx_file = manifest.find('.//*[@id="%s"]' % ncx_fs).get('href')
fp = '/'.join([fpath, oebps, ncx_file])
ncxf = cStringIO.StringIO(self.ios.read(fp))
ncx_tree = etree.parse(ncxf)
#self._log(etree.tostring(ncx_tree, pretty_print=True))
else:
# Find the OPF file in the zipped ePub
zfo = cStringIO.StringIO(self.ios.read(fpath, mode='rb'))
try:
zf = ZipFile(zfo, 'r')
container = etree.fromstring(zf.read('META-INF/container.xml'))
opf_tree = etree.fromstring(zf.read(container.xpath('.//*[local-name()="rootfile"]')[0].get('full-path')))
spine = opf_tree.xpath('.//*[local-name()="spine"]')[0]
ncx_fs = spine.get('toc')
manifest = opf_tree.xpath('.//*[local-name()="manifest"]')[0]
ncx = manifest.find('.//*[@id="%s"]' % ncx_fs).get('href')
# Find the ncx file
fnames = zf.namelist()
_ncx = [x for x in fnames if ncx in x][0]
ncx_tree = etree.fromstring(zf.read(_ncx))
except:
import traceback
self._log_location()
self._log(" unable to unzip '%s'" % fpath)
self._log(traceback.format_exc())
return toc
# fpath points to epub (zipped or unzipped dir)
# spine, ncx_tree populated
try:
toc = OrderedDict()
# 1. capture idrefs from spine
for i, el in enumerate(spine):
toc[str(i)] = el.get('idref')
# 2. Resolve <spine> idrefs to <manifest> hrefs
for el in toc:
toc[el] = manifest.find('.//*[@id="%s"]' % toc[el]).get('href')
# 3. Build a dict of src:toc_entry
src_map = OrderedDict()
navMap = ncx_tree.xpath('.//*[local-name()="navMap"]')[0]
for navPoint in navMap:
# Get the first-level entry
src = re.sub(r'#.*$', '', navPoint.xpath('.//*[local-name()="content"]')[0].get('src'))
toc_entry = navPoint.xpath('.//*[local-name()="text"]')[0].text
src_map[src] = toc_entry
# Get any nested navPoints
nested_navPts = navPoint.xpath('.//*[local-name()="navPoint"]')
for nnp in nested_navPts:
src = re.sub(r'#.*$', '', nnp.xpath('.//*[local-name()="content"]')[0].get('src'))
toc_entry = nnp.xpath('.//*[local-name()="text"]')[0].text
src_map[src] = toc_entry
# Resolve src paths to toc_entry
for section in toc:
if toc[section] in src_map:
if prepend_title:
toc[section] = "%s · %s" % (prepend_title, src_map[toc[section]])
else:
toc[section] = src_map[toc[section]]
else:
toc[section] = None
# 5. Fill in the gaps
current_toc_entry = None
for section in toc:
if toc[section] is None:
toc[section] = current_toc_entry
else:
current_toc_entry = toc[section]
except:
import traceback
self._log_location()
self._log("{:~^80}".format(" error parsing '%s' " % fpath))
self._log(traceback.format_exc())
self._log("{:~^80}".format(" end traceback "))
return toc
