def extract_embedded_content(self, doc):
from calibre.ebooks.fb2 import base64_decode
self.binary_map = {}
for elem in doc.xpath('./*'):
if elem.text and 'binary' in elem.tag and elem.attrib.has_key(
'id'):
ct = elem.get('content-type', '')
fname = elem.attrib['id']
ext = ct.rpartition('/')[-1].lower()
if ext in ('png', 'jpeg', 'jpg'):
if fname.lower().rpartition('.')[-1] not in {
'jpg', 'jpeg', 'png'
}:
fname += '.' + ext
self.binary_map[elem.get('id')] = fname
raw = elem.text.strip()
try:
data = base64_decode(raw)
except TypeError:
self.log.exception(
'Binary data with id=%s is corrupted, ignoring' %
(elem.get('id')))
else:
with open(fname, 'wb') as f:
f.write(data)
def _parse_cover_data(root, imgid, mi, ctx):
from calibre.ebooks.fb2 import base64_decode
elm_binary = ctx.XPath('//fb:binary[@id="%s"]'%imgid)(root)
if elm_binary:
mimetype = elm_binary[0].get('content-type', 'image/jpeg')
mime_extensions = guess_all_extensions(mimetype)
if not mime_extensions and mimetype.startswith('image/'):
mimetype_fromid = guess_type(imgid)[0]
if mimetype_fromid and mimetype_fromid.startswith('image/'):
mime_extensions = guess_all_extensions(mimetype_fromid)
if mime_extensions:
pic_data = elm_binary[0].text
if pic_data:
mi.cover_data = (mime_extensions[0][1:],
base64_decode(pic_data.strip()))
else:
prints("WARNING: Unsupported coverpage mime-type '%s' (id=#%s)" % (mimetype, imgid))
def _parse_cover_data(root, imgid, mi, ctx):
from calibre.ebooks.fb2 import base64_decode
elm_binary = ctx.XPath('//fb:binary[@id="%s"]'%imgid)(root)
if elm_binary:
mimetype = elm_binary[0].get('content-type', 'image/jpeg')
mime_extensions = guess_all_extensions(mimetype)
if not mime_extensions and mimetype.startswith('image/'):
mimetype_fromid = guess_type(imgid)[0]
if mimetype_fromid and mimetype_fromid.startswith('image/'):
mime_extensions = guess_all_extensions(mimetype_fromid)
if mime_extensions:
pic_data = elm_binary[0].text
if pic_data:
mi.cover_data = (mime_extensions[0][1:],
base64_decode(pic_data.strip()))
else:
prints("WARNING: Unsupported coverpage mime-type '%s' (id=#%s)" % (mimetype, imgid))
def extract_embedded_content(self, doc):
from calibre.ebooks.fb2 import base64_decode
self.binary_map = {}
for elem in doc.xpath('./*'):
if elem.text and 'binary' in elem.tag and 'id' in elem.attrib:
ct = elem.get('content-type', '')
fname = elem.attrib['id']
ext = ct.rpartition('/')[-1].lower()
if ext in ('png', 'jpeg', 'jpg'):
if fname.lower().rpartition('.')[-1] not in {'jpg', 'jpeg',
'png'}:
fname += '.' + ext
self.binary_map[elem.get('id')] = fname
raw = elem.text.strip()
try:
data = base64_decode(raw)
except TypeError:
self.log.exception('Binary data with id=%s is corrupted, ignoring'%(
elem.get('id')))
else:
with open(fname, 'wb') as f:
f.write(data)