def get_cover(opf, opf_path, stream, reader=None):
raster_cover = opf.raster_cover
stream.seek(0)
try:
zf = ZipFile(stream)
except:
stream.seek(0)
zf = LocalZipFile(stream)
if raster_cover:
base = posixpath.dirname(opf_path)
cpath = posixpath.normpath(posixpath.join(base, raster_cover))
if reader is not None and \
reader.encryption_meta.is_encrypted(cpath):
return
try:
member = zf.getinfo(cpath)
except:
pass
else:
f = zf.open(member)
data = f.read()
f.close()
zf.close()
return data
return render_cover(opf, opf_path, zf, reader=reader)
class OCFZipReader(OCFReader):
def __init__(self, stream, mode='r', root=None):
if isinstance(stream, (LocalZipFile, ZipFile)):
self.archive = stream
else:
try:
self.archive = ZipFile(stream, mode=mode)
except BadZipfile:
raise EPubException("not a ZIP .epub OCF container")
self.root = root
if self.root is None:
name = getattr(stream, 'name', False)
if name:
self.root = os.path.abspath(os.path.dirname(name))
else:
self.root = getcwd()
super().__init__()
def open(self, name):
if isinstance(self.archive, LocalZipFile):
return self.archive.open(name)
return io.BytesIO(self.archive.read(name))
def read_bytes(self, name):
return self.archive.read(name)
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
class OCFZipReader(OCFReader):
def __init__(self, stream, mode='r', root=None):
if isinstance(stream, (LocalZipFile, ZipFile)):
self.archive = stream
else:
try:
self.archive = ZipFile(stream, mode=mode)
except BadZipfile:
raise EPubException("not a ZIP .epub OCF container")
self.root = root
if self.root is None:
name = getattr(stream, 'name', False)
if name:
self.root = os.path.abspath(os.path.dirname(name))
else:
self.root = getcwd()
super(OCFZipReader, self).__init__()
def open(self, name, mode='r'):
if isinstance(self.archive, LocalZipFile):
return self.archive.open(name)
return io.BytesIO(self.archive.read(name))
def read_bytes(self, name):
return self.archive.read(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
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
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
