def _read_opf(self):
data = self.oeb.container.read(None)
data = self.oeb.decode(data)
data = XMLDECL_RE.sub('', data)
data = re.sub(r'http://openebook.org/namespaces/oeb-package/1.0(/*)',
OPF1_NS, data)
try:
opf = etree.fromstring(data)
except etree.XMLSyntaxError:
data = xml_replace_entities(clean_xml_chars(data), encoding=None)
try:
opf = etree.fromstring(data)
self.logger.warn('OPF contains invalid HTML named entities')
except etree.XMLSyntaxError:
data = re.sub(r'(?is)<tours>.+</tours>', '', data)
data = data.replace('<dc-metadata>',
'<dc-metadata xmlns:dc="http://purl.org/metadata/dublin_core">')
try:
opf = etree.fromstring(data)
self.logger.warn('OPF contains invalid tours section')
except etree.XMLSyntaxError:
from calibre.ebooks.oeb.parse_utils import RECOVER_PARSER
opf = etree.fromstring(data, parser=RECOVER_PARSER)
self.logger.warn('OPF contains invalid markup, trying to parse it anyway')
ns = namespace(opf.tag)
if ns not in ('', OPF1_NS, OPF2_NS):
raise OEBError('Invalid namespace %r for OPF document' % ns)
opf = self._clean_opf(opf)
return opf
def __call__(self, oeb, path):
"""
Write the book in the :class:`OEBBook` object :param:`oeb` to a folder
at :param:`path`.
"""
version = int(self.version[0])
opfname = None
if os.path.splitext(path)[1].lower() == '.opf':
opfname = os.path.basename(path)
path = os.path.dirname(path)
if not os.path.isdir(path):
os.mkdir(path)
output = DirContainer(path, oeb.log)
for item in oeb.manifest.values():
output.write(item.href, item.bytes_representation)
if version == 1:
metadata = oeb.to_opf1()
elif version == 2:
metadata = oeb.to_opf2(page_map=self.page_map)
else:
raise OEBError("Unrecognized OPF version %r" % self.version)
pretty_print = self.pretty_print
for mime, (href, data) in metadata.items():
if opfname and mime == OPF_MIME:
href = opfname
output.write(href, xml2str(data, pretty_print=pretty_print))
return
def _spine_from_opf(self, opf):
spine = self.oeb.spine
manifest = self.oeb.manifest
for elem in xpath(opf, '/o2:package/o2:spine/o2:itemref'):
idref = elem.get('idref')
if idref not in manifest.ids:
self.logger.warn('Spine item %r not found' % idref)
continue
item = manifest.ids[idref]
if item.media_type.lower() in OEB_DOCS and hasattr(
item.data, 'xpath'):
spine.add(item, elem.get('linear'))
else:
if hasattr(item.data,
'tag') and item.data.tag and item.data.tag.endswith(
'}html'):
item.media_type = XHTML_MIME
spine.add(item, elem.get('linear'))
else:
self.oeb.log.warn('The item %s is not a XML document.'
' Removing it from spine.' % item.href)
if len(spine) == 0:
raise OEBError("Spine is empty")
self._spine_add_extra()
for val in xpath(opf,
'/o2:package/o2:spine/@page-progression-direction'):
if val in {'ltr', 'rtl'}:
spine.page_progression_direction = val
def _spine_from_opf(self, opf):
spine = self.oeb.spine
manifest = self.oeb.manifest
for elem in xpath(opf, '/o2:package/o2:spine/o2:itemref'):
idref = elem.get('idref')
if idref not in manifest.ids:
self.logger.warn(u'Spine item %r not found' % idref)
continue
item = manifest.ids[idref]
if item.media_type.lower() in OEB_DOCS and hasattr(item.data, 'xpath'):
spine.add(item, elem.get('linear'))
else:
self.oeb.log.warn('The item %s is not a XML document.'
' Removing it from spine.'%item.href)
if len(spine) == 0:
raise OEBError("Spine is empty")
self._spine_add_extra()