def _metadata_from_opf(self, opf):
from calibre.ebooks.metadata.opf2 import OPF
from calibre.ebooks.oeb.transforms.metadata import meta_info_to_oeb_metadata
stream = cStringIO.StringIO(etree.tostring(opf, xml_declaration=True, encoding='utf-8'))
o = OPF(stream)
pwm = o.primary_writing_mode
if pwm:
self.oeb.metadata.primary_writing_mode = pwm
mi = o.to_book_metadata()
if not mi.language:
mi.language = get_lang().replace('_', '-')
self.oeb.metadata.add('language', mi.language)
if not mi.book_producer:
mi.book_producer = '%(a)s (%(v)s) [http://%(a)s-ebook.com]'%\
dict(a=__appname__, v=__version__)
meta_info_to_oeb_metadata(mi, self.oeb.metadata, self.logger)
m = self.oeb.metadata
m.add('identifier', str(uuid.uuid4()), id='uuid_id', scheme='uuid')
self.oeb.uid = self.oeb.metadata.identifier[-1]
if not m.title:
m.add('title', self.oeb.translate(__('Unknown')))
has_aut = False
for x in m.creator:
if getattr(x, 'role', '').lower() in ('', 'aut'):
has_aut = True
break
if not has_aut:
m.add('creator', self.oeb.translate(__('Unknown')), role='aut')
def mark_as_titlepage(container, name, move_to_start=True):
if move_to_start:
for item, q, linear in container.spine_iter:
if name == q:
break
if not linear:
item.set('linear', 'yes')
if item.getparent().index(item) > 0:
container.insert_into_xml(item.getparent(), item, 0)
for ref in container.opf_xpath('//opf:guide/opf:reference[@type="cover"]'):
ref.getparent().remove(ref)
for guide in container.opf_xpath('//opf:guide'):
container.insert_into_xml(
guide,
guide.makeelement(OPF('reference'),
type='cover',
href=container.name_to_href(
name, container.opf_name)))
container.dirty(container.opf_name)
def convert_text(self, oeb_book):
from calibre.ebooks.metadata.opf2 import OPF
from calibre.ebooks.pdf.render.from_html import PDFWriter
self.log.debug('Serializing oeb input to disk for processing...')
self.get_cover_data()
self.process_fonts()
with TemporaryDirectory('_pdf_out') as oeb_dir:
from calibre.customize.ui import plugin_for_output_format
oeb_output = plugin_for_output_format('oeb')
oeb_output.convert(oeb_book, oeb_dir, self.input_plugin, self.opts,
self.log)
opfpath = glob.glob(os.path.join(oeb_dir, '*.opf'))[0]
opf = OPF(opfpath, os.path.dirname(opfpath))
self.write(PDFWriter, [s.path for s in opf.spine],
getattr(opf, 'toc', None))
def set_azw3_cover(container, cover_path, report):
name = None
found = True
for gi in container.opf_xpath(
'//opf:guide/opf:reference[@href and contains(@type, "cover")]'):
href = gi.get('href')
name = container.href_to_name(href, container.opf_name)
container.remove_from_xml(gi)
if name is None or not container.has_name(name):
item = container.generate_item(name='cover.jpeg', id_prefix='cover')
name = container.href_to_name(item.get('href'), container.opf_name)
found = False
href = container.name_to_href(name, container.opf_name)
guide = container.opf_xpath('//opf:guide')[0]
container.insert_into_xml(
guide, guide.makeelement(OPF('reference'), href=href, type='cover'))
with open(cover_path, 'rb') as src, container.open(name, 'wb') as dest:
shutil.copyfileobj(src, dest)
container.dirty(container.opf_name)
report('Cover updated' if found else 'Cover inserted')
def writer(root, prefixes, refines, ival=None):
uid = root.get('unique-identifier')
package_identifier = None
for ident in XPath('./opf:metadata/dc:identifier')(root):
is_package_id = uid is not None and uid == ident.get('id')
if is_package_id:
package_identifier = ident
val = (ident.text or '').strip()
if (val.startswith(name + ':')
or ident.get(OPF('scheme')) == name) and not is_package_id:
remove_element(ident, refines)
metadata = XPath('./opf:metadata')(root)[0]
if ival:
ident = metadata.makeelement(DC('identifier'))
ident.text = '%s:%s' % (name, ival)
if package_identifier is None:
metadata.append(ident)
else:
p = package_identifier.getparent()
p.insert(p.index(package_identifier), ident)
def generate_item(self, name, id_prefix=None, media_type=None):
'''Add an item to the manifest with href derived from the given
name. Ensures uniqueness of href and id automatically. Returns
generated item.'''
id_prefix = id_prefix or 'id'
media_type = media_type or guess_type(name)
href = self.name_to_href(name, self.opf_name)
base, ext = href.rpartition('.')[0::2]
all_ids = {x.get('id') for x in self.opf_xpath('//*[@id]')}
c = 0
item_id = id_prefix
while item_id in all_ids:
c += 1
item_id = id_prefix + '%d'%c
all_names = {x.get('href') for x in self.opf_xpath(
'//opf:manifest/opf:item[@href]')}
def exists(h):
return self.exists(self.href_to_name(h, self.opf_name))
c = 0
while href in all_names or exists(href):
c += 1
href = '%s_%d.%s'%(base, c, ext)
manifest = self.opf_xpath('//opf:manifest')[0]
item = manifest.makeelement(OPF('item'),
id=item_id, href=href)
item.set('media-type', media_type)
self.insert_into_xml(manifest, item)
self.dirty(self.opf_name)
name = self.href_to_name(href, self.opf_name)
self.name_path_map[name] = path = self.name_to_abspath(name)
self.mime_map[name] = media_type
# Ensure that the file corresponding to the newly created item exists
# otherwise cloned containers will fail when they try to get the number
# of links to the file
base = os.path.dirname(path)
if not os.path.exists(base):
os.makedirs(base)
open(path, 'wb').close()
return item
def convert_text(self, oeb_book):
from calibre.ebooks.metadata.opf2 import OPF
from calibre.ebooks.pdf.render.from_html import PDFWriter
self.log.debug('Serializing oeb input to disk for processing...')
self.get_cover_data()
self.process_fonts()
if self.opts.pdf_use_document_margins and self.stored_page_margins:
import json
for href, margins in iteritems(self.stored_page_margins):
item = oeb_book.manifest.hrefs.get(href)
if item is not None:
root = item.data
if hasattr(root, 'xpath') and margins:
root.set('data-calibre-pdf-output-page-margins',
json.dumps(margins))
# Remove javascript
for item in self.oeb.spine:
root = item.data
if hasattr(root, 'xpath'):
for script in root.xpath('//*[local-name()="script"]'):
script.text = None
script.attrib.clear()
for elem in root.iter('*'):
for attr in tuple(elem.attrib):
if attr.startswith('on'):
elem.set(attr, '')
with TemporaryDirectory('_pdf_out') as oeb_dir:
from calibre.customize.ui import plugin_for_output_format
oeb_output = plugin_for_output_format('oeb')
oeb_output.convert(oeb_book, oeb_dir, self.input_plugin, self.opts,
self.log)
opfpath = glob.glob(os.path.join(oeb_dir, '*.opf'))[0]
opf = OPF(opfpath, os.path.dirname(opfpath))
self.write(PDFWriter, [s.path for s in opf.spine],
getattr(opf, 'toc', None))
def mark_as_titlepage(container, name, move_to_start=True):
'''
Mark the specified HTML file as the titlepage of the EPUB.
:param move_to_start: If True the HTML file is moved to the start of the spine
'''
if move_to_start:
for item, q, linear in container.spine_iter:
if name == q:
break
if not linear:
item.set('linear', 'yes')
if item.getparent().index(item) > 0:
container.insert_into_xml(item.getparent(), item, 0)
for ref in container.opf_xpath('//opf:guide/opf:reference[@type="cover"]'):
ref.getparent().remove(ref)
for guide in get_guides(container):
container.insert_into_xml(guide, guide.makeelement(
OPF('reference'), type='cover', href=container.name_to_href(name, container.opf_name)))
container.dirty(container.opf_name)
def fix_opf(self, container):
spine_names = {n for n, l in container.spine_names}
spine = container.opf_xpath('//opf:spine')[0]
rmap = {v: k for k, v in container.manifest_id_map.iteritems()}
# Add unreferenced text files to the spine
for name, mt in container.mime_map.iteritems():
if mt in OEB_DOCS and name not in spine_names:
spine_names.add(name)
container.insert_into_xml(
spine, spine.makeelement(OPF('itemref'), idref=rmap[name]))
# Remove duplicate entries from spine
seen = set()
for item, name, linear in container.spine_iter:
if name in seen:
container.remove_from_xml(item)
seen.add(name)
# Ensure that the meta cover tag is correct
cover_id = rmap['_static/' + self.config.epub_cover[0]]
for meta in container.opf_xpath('//opf:meta[@name="cover"]'):
meta.set('content', cover_id)
# Add description metadata
metadata = container.opf_xpath('//opf:metadata')[0]
container.insert_into_xml(metadata,
metadata.makeelement(DC('description')))
metadata[-1].text = 'Comprehensive documentation for calibre'
# Remove search.html since it is useless in EPUB
container.remove_item('search.html')
# Remove unreferenced files
for error in check_links(container):
if error.__class__ is UnreferencedResource:
container.remove_item(error.name)
# Pretty print the OPF
pretty_opf(container.parsed(container.opf_name))
container.dirty(container.opf_name)
def fix_opf(self, container):
spine_names = {n for n, l in container.spine_names}
spine = container.opf_xpath('//opf:spine')[0]
rmap = {v: k for k, v in iteritems(container.manifest_id_map)}
# Add unreferenced text files to the spine
for name, mt in iteritems(container.mime_map):
if mt in OEB_DOCS and name not in spine_names:
spine_names.add(name)
container.insert_into_xml(
spine, spine.makeelement(OPF('itemref'), idref=rmap[name]))
# Remove duplicate entries from spine
seen = set()
for item, name, linear in container.spine_iter:
if name in seen:
container.remove_from_xml(item)
seen.add(name)
# Remove the <guide> which is not needed in EPUB 3
for guide in container.opf_xpath('//*[local-name()="guide"]'):
guide.getparent().remove(guide)
# Ensure that the cover-image property is set
cover_id = rmap['_static/' + self.config.epub_cover[0]]
for item in container.opf_xpath(
'//opf:item[@id="{}"]'.format(cover_id)):
item.set('properties', 'cover-image')
# Remove any <meta cover> tag as it is not needed in epub 3
for meta in container.opf_xpath('//opf:meta[@name="cover"]'):
meta.getparent().remove(meta)
# Remove unreferenced files
for error in check_links(container):
if error.__class__ is UnreferencedResource:
container.remove_item(error.name)
# Pretty print the OPF
pretty_opf(container.parsed(container.opf_name))
container.dirty(container.opf_name)
def add_or_replace_jacket(container):
name = find_existing_jacket(container)
found = True
if name is None:
jacket_item = container.generate_item('jacket.xhtml',
id_prefix='jacket')
name = container.href_to_name(jacket_item.get('href'),
container.opf_name)
found = False
replace_jacket(container, name)
if not found:
# Insert new jacket into spine
index = 0
sp = container.abspath_to_name(container.spine_items.next())
if sp == find_cover_page(container):
index = 1
itemref = container.opf.makeelement(OPF('itemref'),
idref=jacket_item.get('id'))
container.insert_into_xml(container.opf_xpath('//opf:spine')[0],
itemref,
index=index)
return found
def convert_metadata(self, oeb):
E = ElementMaker(
namespace=namespaces['cp'],
nsmap={x: namespaces[x]
for x in 'cp dc dcterms xsi'.split()})
cp = E.coreProperties(E.revision("1"), E.lastModifiedBy('calibre'))
ts = utcnow().isoformat(str('T')).rpartition('.')[0] + 'Z'
for x in 'created modified'.split():
x = cp.makeelement(
'{%s}%s' % (namespaces['dcterms'], x),
**{'{%s}type' % namespaces['xsi']: 'dcterms:W3CDTF'})
x.text = ts
cp.append(x)
package = etree.Element(OPF('package'),
attrib={'version': '2.0'},
nsmap={None: OPF2_NS})
oeb.metadata.to_opf2(package)
self.mi = ReadOPF(BytesIO(xml2str(package)),
populate_spine=False,
try_to_guess_cover=False).to_book_metadata()
update_doc_props(cp, self.mi)
return xml2str(cp)
def parse_identifier(ident, val, refines):
idid = ident.get('id')
refines = refines[idid]
scheme = None
lval = val.lower()
def finalize(scheme, val):
if not scheme or not val:
return None, None
scheme = scheme.lower()
if scheme in ('http', 'https'):
return None, None
if scheme.startswith('isbn'):
scheme = 'isbn'
if scheme == 'isbn':
val = val.split(':')[-1]
val = check_isbn(val)
if val is None:
return None, None
return scheme, val
# Try the OPF 2 style opf:scheme attribute, which will be present, for
# example, in EPUB 3 files that have had their metadata set by an
# application that only understands EPUB 2.
scheme = ident.get(OPF('scheme'))
if scheme and not lval.startswith('urn:'):
return finalize(scheme, val)
# Technically, we should be looking for refines that define the scheme, but
# the IDioticPF created such a bad spec that they got their own
# examples wrong, so I cannot be bothered doing this.
# http://www.idpf.org/epub/301/spec/epub-publications-errata/
# Parse the value for the scheme
if lval.startswith('urn:'):
val = val[4:]
prefix, rest = val.partition(':')[::2]
return finalize(prefix, rest)
def create_cover_page(self, input_fmt):
templ = '''
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">
<head><style>
html, body, img { height: 100vh; display: block; margin: 0; padding: 0; border-width: 0; }
img {
width: auto; height: auto;
margin-left: auto; margin-right: auto;
max-width: 100vw; max-height: 100vh
}
</style></head><body><img src="%s"/></body></html>
'''
if input_fmt == 'epub':
def cover_path(action, data):
if action == 'write_image':
data.write(BLANK_JPEG)
return set_epub_cover(self,
cover_path, (lambda *a: None),
options={'template': templ})
raster_cover_name = find_cover_image(self, strict=True)
if raster_cover_name is None:
item = self.generate_item(name='cover.jpeg', id_prefix='cover')
raster_cover_name = self.href_to_name(item.get('href'),
self.opf_name)
with self.open(raster_cover_name, 'wb') as dest:
dest.write(BLANK_JPEG)
item = self.generate_item(name='titlepage.html', id_prefix='titlepage')
titlepage_name = self.href_to_name(item.get('href'), self.opf_name)
raw = templ % prepare_string_for_xml(
self.name_to_href(raster_cover_name, titlepage_name), True)
with self.open(titlepage_name, 'wb') as f:
f.write(raw.encode('utf-8'))
spine = self.opf_xpath('//opf:spine')[0]
ref = spine.makeelement(OPF('itemref'), idref=item.get('id'))
self.insert_into_xml(spine, ref, index=0)
self.dirty(self.opf_name)
return raster_cover_name, titlepage_name
def create_cover_page(self, input_fmt):
if input_fmt == 'epub':
def cover_path(action, data):
if action == 'write_image':
data.write(BLANK_JPEG)
return set_epub_cover(self, cover_path, (lambda *a: None))
from calibre.ebooks.oeb.transforms.cover import CoverManager
raster_cover_name = find_cover_image(self, strict=True)
if raster_cover_name is None:
item = self.generate_item(name='cover.jpeg', id_prefix='cover')
raster_cover_name = self.href_to_name(item.get('href'), self.opf_name)
with self.open(raster_cover_name, 'wb') as dest:
dest.write(BLANK_JPEG)
item = self.generate_item(name='titlepage.html', id_prefix='titlepage')
titlepage_name = self.href_to_name(item.get('href'), self.opf_name)
templ = CoverManager.SVG_TEMPLATE
raw = templ % self.name_to_href(raster_cover_name, titlepage_name)
with self.open(titlepage_name, 'wb') as f:
f.write(raw.encode('utf-8'))
spine = self.opf_xpath('//opf:spine')[0]
ref = spine.makeelement(OPF('itemref'), idref=item.get('id'))
self.insert_into_xml(spine, ref, index=0)
self.dirty(self.opf_name)
return raster_cover_name, titlepage_name
def convert(self, oeb, output_path, input_plugin, opts, log):
self.log, self.opts, self.oeb = log, opts, oeb
if self.opts.epub_inline_toc:
from calibre.ebooks.mobi.writer8.toc import TOCAdder
opts.mobi_toc_at_start = not opts.epub_toc_at_end
opts.mobi_passthrough = False
opts.no_inline_toc = False
TOCAdder(oeb,
opts,
replace_previous_inline_toc=True,
ignore_existing_toc=True)
if self.opts.epub_flatten:
from calibre.ebooks.oeb.transforms.filenames import FlatFilenames
FlatFilenames()(oeb, opts)
else:
from calibre.ebooks.oeb.transforms.filenames import UniqueFilenames
UniqueFilenames()(oeb, opts)
self.workaround_ade_quirks()
self.workaround_webkit_quirks()
self.upshift_markup()
from calibre.ebooks.oeb.transforms.rescale import RescaleImages
RescaleImages(check_colorspaces=True)(oeb, opts)
from calibre.ebooks.oeb.transforms.split import Split
split = Split(not self.opts.dont_split_on_page_breaks,
max_flow_size=self.opts.flow_size * 1024)
split(self.oeb, self.opts)
from calibre.ebooks.oeb.transforms.cover import CoverManager
cm = CoverManager(
no_default_cover=self.opts.no_default_epub_cover,
no_svg_cover=self.opts.no_svg_cover,
preserve_aspect_ratio=self.opts.preserve_cover_aspect_ratio)
cm(self.oeb, self.opts, self.log)
self.workaround_sony_quirks()
if self.oeb.toc.count() == 0:
self.log.warn('This EPUB file has no Table of Contents. '
'Creating a default TOC')
first = next(iter(self.oeb.spine))
self.oeb.toc.add(_('Start'), first.href)
from calibre.ebooks.oeb.base import OPF
identifiers = oeb.metadata['identifier']
uuid = None
for x in identifiers:
if x.get(OPF('scheme'),
None).lower() == 'uuid' or str(x).startswith('urn:uuid:'):
uuid = str(x).split(':')[-1]
break
encrypted_fonts = getattr(input_plugin, 'encrypted_fonts', [])
if uuid is None:
self.log.warn('No UUID identifier found')
from uuid import uuid4
uuid = str(uuid4())
oeb.metadata.add('identifier', uuid, scheme='uuid', id=uuid)
if encrypted_fonts and not uuid.startswith('urn:uuid:'):
# Apparently ADE requires this value to start with urn:uuid:
# for some absurd reason, or it will throw a hissy fit and refuse
# to use the obfuscated fonts.
for x in identifiers:
if str(x) == uuid:
x.content = 'urn:uuid:' + uuid
with TemporaryDirectory('_epub_output') as tdir:
from calibre.customize.ui import plugin_for_output_format
metadata_xml = None
extra_entries = []
if self.is_periodical:
if self.opts.output_profile.epub_periodical_format == 'sony':
from calibre.ebooks.epub.periodical import sony_metadata
metadata_xml, atom_xml = sony_metadata(oeb)
extra_entries = [('atom.xml', 'application/atom+xml',
atom_xml)]
oeb_output = plugin_for_output_format('oeb')
oeb_output.convert(oeb, tdir, input_plugin, opts, log)
opf = [x for x in os.listdir(tdir) if x.endswith('.opf')][0]
self.condense_ncx([
os.path.join(tdir, x) for x in os.listdir(tdir)
if x.endswith('.ncx')
][0])
encryption = None
if encrypted_fonts:
encryption = self.encrypt_fonts(encrypted_fonts, tdir, uuid)
from calibre.ebooks.epub import initialize_container
with initialize_container(output_path,
os.path.basename(opf),
extra_entries=extra_entries) as epub:
epub.add_dir(tdir)
if encryption is not None:
epub.writestr('META-INF/encryption.xml', encryption)
if metadata_xml is not None:
epub.writestr('META-INF/metadata.xml',
metadata_xml.encode('utf-8'))
if opts.extract_to is not None:
from calibre.utils.zipfile import ZipFile
if os.path.exists(opts.extract_to):
if os.path.isdir(opts.extract_to):
shutil.rmtree(opts.extract_to)
else:
os.remove(opts.extract_to)
os.mkdir(opts.extract_to)
with ZipFile(output_path) as zf:
zf.extractall(path=opts.extract_to)
self.log.info('EPUB extracted to', opts.extract_to)
def split(container, name, loc_or_xpath, before=True):
''' Split the file specified by name at the position specified by loc_or_xpath. '''
root = container.parsed(name)
if isinstance(loc_or_xpath, type('')):
split_point = root.xpath(loc_or_xpath)[0]
else:
split_point = node_from_loc(root, loc_or_xpath)
if in_table(split_point):
raise AbortError('Cannot split inside tables')
if split_point.tag.endswith('}body'):
raise AbortError('Cannot split on the <body> tag')
tree1, tree2 = do_split(split_point, container.log, before=before)
root1, root2 = tree1.getroot(), tree2.getroot()
anchors_in_top = frozenset(root1.xpath('//*/@id')) | frozenset(
root1.xpath('//*/@name')) | {''}
anchors_in_bottom = frozenset(root2.xpath('//*/@id')) | frozenset(
root2.xpath('//*/@name'))
manifest_item = container.generate_item(
name, media_type=container.mime_map[name])
bottom_name = container.href_to_name(manifest_item.get('href'),
container.opf_name)
# Fix links in the split trees
for r, rname, anchors in [(root1, bottom_name, anchors_in_bottom),
(root2, name, anchors_in_top)]:
for a in r.xpath('//*[@href]'):
url = a.get('href')
if url.startswith('#'):
fname = name
else:
fname = container.href_to_name(url, name)
if fname == name:
purl = urlparse(url)
if purl.fragment in anchors:
a.set(
'href', '%s#%s' %
(container.name_to_href(rname, name), purl.fragment))
# Fix all links in the container that point to anchors in the bottom tree
for fname, media_type in container.mime_map.iteritems():
if fname not in {name, bottom_name}:
repl = SplitLinkReplacer(fname, anchors_in_bottom, name,
bottom_name, container)
container.replace_links(fname, repl)
container.replace(name, root1)
container.replace(bottom_name, root2)
spine = container.opf_xpath('//opf:spine')[0]
for spine_item, spine_name, linear in container.spine_iter:
if spine_name == name:
break
index = spine.index(spine_item) + 1
si = spine.makeelement(OPF('itemref'), idref=manifest_item.get('id'))
if not linear:
si.set('linear', 'no')
container.insert_into_xml(spine, si, index=index)
container.dirty(container.opf_name)
return bottom_name
def create_cover_page(self, input_fmt):
templ = '''
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">
<head><style>
html, body, img { height: 100vh; display: block; margin: 0; padding: 0; border-width: 0; }
img {
width: 100%%; height: 100%%;
object-fit: contain;
margin-left: auto; margin-right: auto;
max-width: 100vw; max-height: 100vh;
top: 50vh; transform: translateY(-50%%);
position: relative;
}
body.cover-fill img { object-fit: fill; }
</style></head><body><img src="%s"/></body></html>
'''
def generic_cover():
if self.book_metadata is not None:
from calibre.ebooks.covers import create_cover
mi = self.book_metadata
return create_cover(mi.title, mi.authors, mi.series,
mi.series_index)
return BLANK_JPEG
if input_fmt == 'epub':
def image_callback(cover_image, wrapped_image):
if cover_image:
image_callback.cover_data = self.raw_data(cover_image,
decode=False)
if wrapped_image and not getattr(image_callback, 'cover_data',
None):
image_callback.cover_data = self.raw_data(wrapped_image,
decode=False)
def cover_path(action, data):
if action == 'write_image':
cdata = getattr(image_callback, 'cover_data',
None) or generic_cover()
data.write(cdata)
raster_cover_name, titlepage_name = set_epub_cover(
self,
cover_path, (lambda *a: None),
options={'template': templ},
image_callback=image_callback)
else:
raster_cover_name = find_cover_image(self, strict=True)
if raster_cover_name is None:
item = self.generate_item(name='cover.jpeg', id_prefix='cover')
raster_cover_name = self.href_to_name(item.get('href'),
self.opf_name)
with self.open(raster_cover_name, 'wb') as dest:
dest.write(generic_cover())
item = self.generate_item(name='titlepage.html',
id_prefix='titlepage')
titlepage_name = self.href_to_name(item.get('href'), self.opf_name)
raw = templ % prepare_string_for_xml(
self.name_to_href(raster_cover_name, titlepage_name), True)
with self.open(titlepage_name, 'wb') as f:
f.write(raw.encode('utf-8'))
spine = self.opf_xpath('//opf:spine')[0]
ref = spine.makeelement(OPF('itemref'), idref=item.get('id'))
self.insert_into_xml(spine, ref, index=0)
self.dirty(self.opf_name)
return raster_cover_name, titlepage_name
def rationalize_cover2(self, opf, log):
''' Ensure that the cover information in the guide is correct. That
means, at most one entry with type="cover" that points to a raster
cover and at most one entry with type="titlepage" that points to an
HTML titlepage. '''
from calibre.ebooks.oeb.base import OPF
removed = None
from lxml import etree
guide_cover, guide_elem = None, None
for guide_elem in opf.iterguide():
if guide_elem.get('type', '').lower() == 'cover':
guide_cover = guide_elem.get('href', '').partition('#')[0]
break
if not guide_cover:
raster_cover = opf.raster_cover
if raster_cover:
if guide_elem is None:
g = opf.root.makeelement(OPF('guide'))
opf.root.append(g)
else:
g = guide_elem.getparent()
guide_cover = raster_cover
guide_elem = g.makeelement(OPF('reference'),
attrib={
'href': raster_cover,
'type': 'cover'
})
g.append(guide_elem)
return
spine = list(opf.iterspine())
if not spine:
return
# Check if the cover specified in the guide is also
# the first element in spine
idref = spine[0].get('idref', '')
manifest = list(opf.itermanifest())
if not manifest:
return
elem = [x for x in manifest if x.get('id', '') == idref]
if not elem or elem[0].get('href', None) != guide_cover:
return
log('Found HTML cover', guide_cover)
# Remove from spine as covers must be treated
# specially
if not self.for_viewer:
if len(spine) == 1:
log.warn(
'There is only a single spine item and it is marked as the cover. Removing cover marking.'
)
for guide_elem in tuple(opf.iterguide()):
if guide_elem.get('type', '').lower() == 'cover':
guide_elem.getparent().remove(guide_elem)
return
else:
spine[0].getparent().remove(spine[0])
removed = guide_cover
else:
# Ensure the cover is displayed as the first item in the book, some
# epub files have it set with linear='no' which causes the cover to
# display in the end
spine[0].attrib.pop('linear', None)
opf.spine[0].is_linear = True
# Ensure that the guide has a cover entry pointing to a raster cover
# and a titlepage entry pointing to the html titlepage. The titlepage
# entry will be used by the epub output plugin, the raster cover entry
# by other output plugins.
# Search for a raster cover identified in the OPF
raster_cover = opf.raster_cover
# Set the cover guide entry
if raster_cover is not None:
guide_elem.set('href', raster_cover)
else:
# Render the titlepage to create a raster cover
from calibre.ebooks import render_html_svg_workaround
guide_elem.set('href', 'calibre_raster_cover.jpg')
t = etree.SubElement(elem[0].getparent(),
OPF('item'),
href=guide_elem.get('href'),
id='calibre_raster_cover')
t.set('media-type', 'image/jpeg')
if os.path.exists(guide_cover):
renderer = render_html_svg_workaround(guide_cover, log)
if renderer is not None:
open('calibre_raster_cover.jpg', 'wb').write(renderer)
# Set the titlepage guide entry
self.set_guide_type(opf, 'titlepage', guide_cover, 'Title Page')
return removed
def convert(self, stream, options, file_ext, log, accelerators):
from calibre.utils.zipfile import ZipFile
from calibre import walk
from calibre.ebooks import DRMError
from calibre.ebooks.metadata.opf2 import OPF
try:
zf = ZipFile(stream)
zf.extractall(os.getcwdu())
except:
log.exception('EPUB appears to be invalid ZIP file, trying a'
' more forgiving ZIP parser')
from calibre.utils.localunzip import extractall
stream.seek(0)
extractall(stream)
encfile = os.path.abspath(os.path.join('META-INF', 'encryption.xml'))
opf = self.find_opf()
if opf is None:
for f in walk(u'.'):
if f.lower().endswith('.opf') and '__MACOSX' not in f and \
not os.path.basename(f).startswith('.'):
opf = os.path.abspath(f)
break
path = getattr(stream, 'name', 'stream')
if opf is None:
raise ValueError(
'%s is not a valid EPUB file (could not find opf)' % path)
opf = os.path.relpath(opf, os.getcwdu())
parts = os.path.split(opf)
opf = OPF(opf, os.path.dirname(os.path.abspath(opf)))
self._encrypted_font_uris = []
if os.path.exists(encfile):
if not self.process_encryption(encfile, opf, log):
raise DRMError(os.path.basename(path))
self.encrypted_fonts = self._encrypted_font_uris
if len(parts) > 1 and parts[0]:
delta = '/'.join(parts[:-1]) + '/'
def normpath(x):
return posixpath.normpath(delta + elem.get('href'))
for elem in opf.itermanifest():
elem.set('href', normpath(elem.get('href')))
for elem in opf.iterguide():
elem.set('href', normpath(elem.get('href')))
f = self.rationalize_cover3 if opf.package_version >= 3.0 else self.rationalize_cover2
self.removed_cover = f(opf, log)
if self.removed_cover:
self.removed_items_to_ignore = (self.removed_cover, )
epub3_nav = opf.epub3_nav
if epub3_nav is not None:
self.convert_epub3_nav(epub3_nav, opf, log, options)
for x in opf.itermanifest():
if x.get('media-type', '') == 'application/x-dtbook+xml':
raise ValueError(
'EPUB files with DTBook markup are not supported')
not_for_spine = set()
for y in opf.itermanifest():
id_ = y.get('id', None)
if id_:
mt = y.get('media-type', None)
if mt in {
'application/vnd.adobe-page-template+xml',
'application/vnd.adobe.page-template+xml',
'application/adobe-page-template+xml',
'application/adobe.page-template+xml',
'application/text'
}:
not_for_spine.add(id_)
ext = y.get('href', '').rpartition('.')[-1].lower()
if mt == 'text/plain' and ext in {'otf', 'ttf'}:
# some epub authoring software sets font mime types to
# text/plain
not_for_spine.add(id_)
y.set('media-type', 'application/font')
seen = set()
for x in list(opf.iterspine()):
ref = x.get('idref', None)
if not ref or ref in not_for_spine or ref in seen:
x.getparent().remove(x)
continue
seen.add(ref)
if len(list(opf.iterspine())) == 0:
raise ValueError('No valid entries in the spine of this EPUB')
with lopen('content.opf', 'wb') as nopf:
nopf.write(opf.render())
return os.path.abspath(u'content.opf')
def build_exth(metadata,
prefer_author_sort=False,
is_periodical=False,
share_not_sync=True,
cover_offset=None,
thumbnail_offset=None,
start_offset=None,
mobi_doctype=2,
num_of_resources=None,
kf8_unknown_count=0,
be_kindlegen2=False,
kf8_header_index=None,
opts=None):
exth = BytesIO()
nrecs = 0
for term in metadata:
if term not in EXTH_CODES: continue
code = EXTH_CODES[term]
items = metadata[term]
if term == 'creator':
if prefer_author_sort:
creators = [
authors_to_sort_string([unicode(c)]) for c in items
]
else:
creators = [unicode(c) for c in items]
items = creators
elif term == 'rights':
try:
rights = utf8_text(unicode(metadata.rights[0]))
except:
rights = b'Unknown'
exth.write(pack(b'>II', EXTH_CODES['rights'], len(rights) + 8))
exth.write(rights)
nrecs += 1
continue
for item in items:
data = unicode(item)
if term != 'description':
data = COLLAPSE_RE.sub(' ', data)
if term == 'identifier':
if data.lower().startswith('urn:isbn:'):
data = data[9:]
elif item.scheme.lower() == 'isbn':
pass
else:
continue
if term == 'language':
d2 = usr_lang_as_iso639_1(data)
if d2:
data = d2
data = utf8_text(data)
exth.write(pack(b'>II', code, len(data) + 8))
exth.write(data)
nrecs += 1
# Write UUID as ASIN
uuid = None
from calibre.ebooks.oeb.base import OPF
for x in metadata['identifier']:
if (x.get(OPF('scheme'), None).lower() == 'uuid'
or unicode(x).startswith('urn:uuid:')):
uuid = unicode(x).split(':')[-1]
break
if uuid is None:
from uuid import uuid4
uuid = str(uuid4())
if isinstance(uuid, unicode):
uuid = uuid.encode('utf-8')
if not share_not_sync:
exth.write(pack(b'>II', 113, len(uuid) + 8))
exth.write(uuid)
nrecs += 1
# Write UUID as SOURCE
c_uuid = b'calibre:%s' % uuid
exth.write(pack(b'>II', 112, len(c_uuid) + 8))
exth.write(c_uuid)
nrecs += 1
# Write cdetype
if not is_periodical:
if not share_not_sync:
exth.write(pack(b'>II', 501, 12))
exth.write(b'EBOK')
nrecs += 1
else:
ids = {0x101: b'NWPR', 0x103: b'MAGZ'}.get(mobi_doctype, None)
if ids:
exth.write(pack(b'>II', 501, 12))
exth.write(ids)
nrecs += 1
# Add a publication date entry
datestr = None
if metadata['date']:
datestr = str(metadata['date'][0])
elif metadata['timestamp']:
datestr = str(metadata['timestamp'][0])
if not datestr:
raise ValueError("missing date or timestamp")
datestr = bytes(datestr)
exth.write(pack(b'>II', EXTH_CODES['pubdate'], len(datestr) + 8))
exth.write(datestr)
nrecs += 1
if is_periodical:
exth.write(pack(b'>II', EXTH_CODES['lastupdatetime'],
len(datestr) + 8))
exth.write(datestr)
nrecs += 1
if be_kindlegen2:
vals = {204: 201, 205: 2, 206: 5, 207: 0}
elif is_periodical:
# Pretend to be amazon's super secret periodical generator
vals = {204: 201, 205: 2, 206: 0, 207: 101}
else:
# Pretend to be kindlegen 1.2
vals = {204: 201, 205: 1, 206: 2, 207: 33307}
for code, val in vals.iteritems():
exth.write(pack(b'>III', code, 12, val))
nrecs += 1
if cover_offset is not None:
exth.write(pack(b'>III', EXTH_CODES['coveroffset'], 12, cover_offset))
exth.write(pack(b'>III', EXTH_CODES['hasfakecover'], 12, 0))
nrecs += 2
if thumbnail_offset is not None:
exth.write(
pack(b'>III', EXTH_CODES['thumboffset'], 12, thumbnail_offset))
thumbnail_uri_str = bytes(
'kindle:embed:%s' %
(to_base(thumbnail_offset, base=32, min_num_digits=4)))
exth.write(
pack(b'>II', EXTH_CODES['kf8_thumbnail_uri'],
len(thumbnail_uri_str) + 8))
exth.write(thumbnail_uri_str)
nrecs += 2
if start_offset is not None:
try:
len(start_offset)
except TypeError:
start_offset = [start_offset]
for so in start_offset:
if so is not None:
exth.write(pack(b'>III', EXTH_CODES['startreading'], 12, so))
nrecs += 1
if kf8_header_index is not None:
exth.write(
pack(b'>III', EXTH_CODES['kf8_header_index'], 12,
kf8_header_index))
nrecs += 1
if num_of_resources is not None:
exth.write(
pack(b'>III', EXTH_CODES['num_of_resources'], 12,
num_of_resources))
nrecs += 1
if kf8_unknown_count is not None:
exth.write(
pack(b'>III', EXTH_CODES['kf8_unknown_count'], 12,
kf8_unknown_count))
nrecs += 1
#Extra metadata for fullscrenn
if opts and opts.book_mode == 'comic': #added for kindleear [insert0003 2017-09-03]
exth.write(pack(b'>II', EXTH_CODES['RegionMagnification'], 13))
exth.write(b'false')
exth.write(pack(b'>II', EXTH_CODES['book-type'], 13))
exth.write(b'comic')
exth.write(pack(b'>II', EXTH_CODES['zero-gutter'], 12))
exth.write(b'true')
exth.write(pack(b'>II', EXTH_CODES['zero-margin'], 12))
exth.write(b'true')
exth.write(pack(b'>II', EXTH_CODES['primary-writing-mode'], 21))
exth.write(b'horizontal-lr')
exth.write(pack(b'>II', EXTH_CODES['fixed-layout'], 12))
exth.write(b'true')
exth.write(pack(b'>II', EXTH_CODES['orientation-lock'], 16))
exth.write(b'portrait')
original_resolution = b'%dx%d' % opts.dest.comic_screen_size #sth like comic_screen_size = (1072, 1430)
exth.write(
pack(b'>II', EXTH_CODES['original-resolution'],
len(original_resolution) + 8))
exth.write(original_resolution)
nrecs += 8
exth = exth.getvalue()
trail = len(exth) % 4
pad = b'\0' * (4 - trail) # Always pad w/ at least 1 byte
exth = [b'EXTH', pack(b'>II', len(exth) + 12, nrecs), exth, pad]
return b''.join(exth)
def create_rating(root, prefixes, val):
ensure_prefix(root, prefixes, 'calibre', CALIBRE_PREFIX)
m = XPath('./opf:metadata')(root)[0]
d = m.makeelement(OPF('meta'), attrib={'property': 'calibre:rating'})
d.text = val
m.append(d)
def fb2_header(self):
from calibre.ebooks.oeb.base import OPF
metadata = {}
metadata['title'] = self.oeb_book.metadata.title[0].value
metadata['appname'] = __appname__
metadata['version'] = __version__
metadata['date'] = '%i.%i.%i' % (
datetime.now().day, datetime.now().month, datetime.now().year)
if self.oeb_book.metadata.language:
lc = lang_as_iso639_1(self.oeb_book.metadata.language[0].value)
if not lc:
lc = self.oeb_book.metadata.language[0].value
metadata['lang'] = lc or 'en'
else:
metadata['lang'] = u'en'
metadata['id'] = None
metadata['cover'] = self.get_cover()
metadata['genre'] = self.opts.fb2_genre
metadata['author'] = u''
for auth in self.oeb_book.metadata.creator:
author_first = u''
author_middle = u''
author_last = u''
author_parts = auth.value.split(' ')
if len(author_parts) == 1:
author_last = author_parts[0]
elif len(author_parts) == 2:
author_first = author_parts[0]
author_last = author_parts[1]
else:
author_first = author_parts[0]
author_middle = ' '.join(author_parts[1:-1])
author_last = author_parts[-1]
metadata['author'] += '<author>'
metadata[
'author'] += '<first-name>%s</first-name>' % prepare_string_for_xml(
author_first)
if author_middle:
metadata[
'author'] += '<middle-name>%s</middle-name>' % prepare_string_for_xml(
author_middle)
metadata[
'author'] += '<last-name>%s</last-name>' % prepare_string_for_xml(
author_last)
metadata['author'] += '</author>'
if not metadata['author']:
metadata[
'author'] = u'<author><first-name></first-name><last-name><last-name></author>'
metadata['sequence'] = u''
if self.oeb_book.metadata.series:
index = '1'
if self.oeb_book.metadata.series_index:
index = self.oeb_book.metadata.series_index[0]
metadata['sequence'] = u'<sequence name="%s" number="%s" />' % (
prepare_string_for_xml(
u'%s' % self.oeb_book.metadata.series[0]), index)
identifiers = self.oeb_book.metadata['identifier']
for x in identifiers:
if x.get(OPF('scheme'), None).lower() == 'uuid' or unicode(
x).startswith('urn:uuid:'):
metadata['id'] = unicode(x).split(':')[-1]
break
if metadata['id'] is None:
self.log.warn('No UUID identifier found')
metadata['id'] = str(uuid.uuid4())
for key, value in metadata.items():
if key not in ('author', 'cover', 'sequence'):
metadata[key] = prepare_string_for_xml(value)
return u'<FictionBook xmlns="http://www.gribuser.ru/xml/fictionbook/2.0" xmlns:xlink="http://www.w3.org/1999/xlink">' \
'<description>' \
'<title-info>' \
'<genre>%(genre)s</genre>' \
'%(author)s' \
'<book-title>%(title)s</book-title>' \
'%(cover)s' \
'<lang>%(lang)s</lang>' \
'%(sequence)s' \
'</title-info>' \
'<document-info>' \
'%(author)s' \
'<program-used>%(appname)s %(version)s</program-used>' \
'<date>%(date)s</date>' \
'<id>%(id)s</id>' \
'<version>1.0</version>' \
'</document-info>' \
'</description>' % metadata
def check_opf(container):
errors = []
opf_version = container.opf_version_parsed
if container.opf.tag != OPF('package'):
err = BaseError(_('The OPF does not have the correct root element'), container.opf_name, container.opf.sourceline)
err.HELP = xml(_(
'The opf must have the root element <package> in namespace {0}, like this: <package xmlns="{0}">')).format(OPF2_NS)
errors.append(err)
elif container.opf.get('version') is None and container.book_type == 'epub':
err = BaseError(_('The OPF does not have a version'), container.opf_name, container.opf.sourceline)
err.HELP = xml(_(
'The <package> tag in the OPF must have a version attribute. This is usually version="2.0" for EPUB2 and AZW3 and version="3.0" for EPUB3'))
errors.append(err)
for tag in ('metadata', 'manifest', 'spine'):
if not container.opf_xpath('/opf:package/opf:' + tag):
errors.append(MissingSection(container.opf_name, tag))
all_ids = set(container.opf_xpath('//*/@id'))
if '' in all_ids:
for empty_id_tag in container.opf_xpath('//*[@id=""]'):
errors.append(EmptyID(container.opf_name, empty_id_tag.sourceline))
all_ids.discard('')
for elem in container.opf_xpath('//*[@idref]'):
if elem.get('idref') not in all_ids:
errors.append(IncorrectIdref(container.opf_name, elem.get('idref'), elem.sourceline))
nl_items = [elem.sourceline for elem in container.opf_xpath('//opf:spine/opf:itemref[@linear="no"]')]
if nl_items:
errors.append(NonLinearItems(container.opf_name, nl_items))
seen, dups = {}, {}
for item in container.opf_xpath('/opf:package/opf:manifest/opf:item'):
href = item.get('href', None)
if href is None:
errors.append(NoHref(container.opf_name, item.get('id', None), item.sourceline))
else:
hname = container.href_to_name(href, container.opf_name)
if not hname or not container.exists(hname):
errors.append(MissingHref(container.opf_name, href, item.sourceline))
if href in seen:
if href not in dups:
dups[href] = [seen[href]]
dups[href].append(item.sourceline)
else:
seen[href] = item.sourceline
errors.extend(DuplicateHref(container.opf_name, eid, locs) for eid, locs in iteritems(dups))
seen, dups = {}, {}
for item in container.opf_xpath('/opf:package/opf:spine/opf:itemref[@idref]'):
ref = item.get('idref')
if ref in seen:
if ref not in dups:
dups[ref] = [seen[ref]]
dups[ref].append(item.sourceline)
else:
seen[ref] = item.sourceline
errors.extend(DuplicateHref(container.opf_name, eid, locs, for_spine=True) for eid, locs in iteritems(dups))
spine = container.opf_xpath('/opf:package/opf:spine[@toc]')
if spine:
spine = spine[0]
mitems = [x for x in container.opf_xpath('/opf:package/opf:manifest/opf:item[@id]') if x.get('id') == spine.get('toc')]
if mitems:
mitem = mitems[0]
if mitem.get('media-type', '') != guess_type('a.ncx'):
errors.append(IncorrectToc(container.opf_name, mitem.sourceline, bad_mimetype=mitem.get('media-type')))
else:
errors.append(IncorrectToc(container.opf_name, spine.sourceline, bad_idref=spine.get('toc')))
else:
spine = container.opf_xpath('/opf:package/opf:spine')
if spine:
spine = spine[0]
ncx = container.manifest_type_map.get(guess_type('a.ncx'))
if ncx:
ncx_name = ncx[0]
rmap = {v:k for k, v in iteritems(container.manifest_id_map)}
ncx_id = rmap.get(ncx_name)
if ncx_id:
errors.append(MissingNCXRef(container.opf_name, spine.sourceline, ncx_id))
if opf_version.major > 2:
existing_nav = find_existing_nav_toc(container)
if existing_nav is None:
errors.append(MissingNav(container.opf_name, 0))
else:
toc = parse_nav(container, existing_nav)
if len(toc) == 0:
errors.append(EmptyNav(existing_nav, 0))
covers = container.opf_xpath('/opf:package/opf:metadata/opf:meta[@name="cover"]')
if len(covers) > 0:
if len(covers) > 1:
errors.append(MultipleCovers(container.opf_name, [c.sourceline for c in covers]))
manifest_ids = set(container.opf_xpath('/opf:package/opf:manifest/opf:item/@id'))
for cover in covers:
if cover.get('content', None) not in manifest_ids:
errors.append(IncorrectCover(container.opf_name, cover.sourceline, cover.get('content', '')))
raw = etree.tostring(cover)
try:
n, c = raw.index(b'name="'), raw.index(b'content="')
except ValueError:
n = c = -1
if n > -1 and c > -1 and n > c:
errors.append(NookCover(container.opf_name, cover.sourceline))
uid = container.opf.get('unique-identifier', None)
if uid is None or not container.opf_xpath('/opf:package/opf:metadata/dc:identifier[@id=%r]' % uid):
errors.append(NoUID(container.opf_name))
for elem in container.opf_xpath('/opf:package/opf:metadata/dc:identifier'):
if not elem.text or not elem.text.strip():
errors.append(EmptyIdentifier(container.opf_name, elem.sourceline))
for item, name, linear in container.spine_iter:
mt = container.mime_map[name]
if mt != XHTML_MIME:
iid = item.get('idref', None)
lnum = None
if iid:
mitem = container.opf_xpath('/opf:package/opf:manifest/opf:item[@id=%r]' % iid)
if mitem:
lnum = mitem[0].sourceline
else:
iid = None
errors.append(BadSpineMime(name, iid, mt, lnum, container.opf_name))
return errors
def create_series(root, refines, series, series_index):
m = XPath('./opf:metadata')(root)[0]
d = m.makeelement(OPF('meta'), attrib={'property':'belongs-to-collection'})
d.text = series
m.append(d)
set_refines(d, refines, refdef('collection-type', 'series'), refdef('group-position', series_index))
def fb2_header(self):
from calibre.ebooks.oeb.base import OPF
metadata = {}
metadata['title'] = self.oeb_book.metadata.title[0].value
metadata['appname'] = __appname__
metadata['version'] = __version__
metadata['date'] = '%i.%i.%i' % (
datetime.now().day, datetime.now().month, datetime.now().year)
if self.oeb_book.metadata.language:
lc = lang_as_iso639_1(self.oeb_book.metadata.language[0].value)
if not lc:
lc = self.oeb_book.metadata.language[0].value
metadata['lang'] = lc or 'en'
else:
metadata['lang'] = u'en'
metadata['id'] = None
metadata['cover'] = self.get_cover()
metadata['genre'] = self.opts.fb2_genre
metadata['author'] = ''
for auth in self.oeb_book.metadata.creator:
author_first = ''
author_middle = ''
author_last = ''
author_parts = auth.value.split(' ')
if len(author_parts) == 1:
author_last = author_parts[0]
elif len(author_parts) == 2:
author_first = author_parts[0]
author_last = author_parts[1]
else:
author_first = author_parts[0]
author_middle = ' '.join(author_parts[1:-1])
author_last = author_parts[-1]
metadata['author'] += '<author>'
metadata[
'author'] += '<first-name>%s</first-name>' % prepare_string_for_xml(
author_first)
if author_middle:
metadata[
'author'] += '<middle-name>%s</middle-name>' % prepare_string_for_xml(
author_middle)
metadata[
'author'] += '<last-name>%s</last-name>' % prepare_string_for_xml(
author_last)
metadata['author'] += '</author>'
if not metadata['author']:
metadata[
'author'] = '<author><first-name></first-name><last-name></last-name></author>'
metadata['keywords'] = ''
tags = list(map(unicode_type, self.oeb_book.metadata.subject))
if tags:
tags = ', '.join(prepare_string_for_xml(x) for x in tags)
metadata['keywords'] = '<keywords>%s</keywords>' % tags
metadata['sequence'] = ''
if self.oeb_book.metadata.series:
index = '1'
if self.oeb_book.metadata.series_index:
index = self.oeb_book.metadata.series_index[0]
metadata['sequence'] = '<sequence name="%s" number="%s"/>' % (
prepare_string_for_xml(
'%s' % self.oeb_book.metadata.series[0]), index)
year = publisher = isbn = ''
identifiers = self.oeb_book.metadata['identifier']
for x in identifiers:
if x.get(OPF('scheme'), None).lower() == 'uuid' or unicode_type(
x).startswith('urn:uuid:'):
metadata['id'] = unicode_type(x).split(':')[-1]
break
if metadata['id'] is None:
self.log.warn('No UUID identifier found')
metadata['id'] = unicode_type(uuid.uuid4())
try:
date = self.oeb_book.metadata['date'][0]
except IndexError:
pass
else:
year = '<year>%s</year>' % prepare_string_for_xml(
date.value.partition('-')[0])
try:
publisher = self.oeb_book.metadata['publisher'][0]
except IndexError:
pass
else:
publisher = '<publisher>%s</publisher>' % prepare_string_for_xml(
publisher.value)
for x in identifiers:
if x.get(OPF('scheme'), None).lower() == 'isbn':
isbn = '<isbn>%s</isbn>' % prepare_string_for_xml(x.value)
metadata['year'], metadata['isbn'], metadata[
'publisher'] = year, isbn, publisher
for key, value in metadata.items():
if key not in ('author', 'cover', 'sequence', 'keywords', 'year',
'publisher', 'isbn'):
metadata[key] = prepare_string_for_xml(value)
try:
comments = self.oeb_book.metadata['description'][0]
except Exception:
metadata['comments'] = ''
else:
from calibre.utils.html2text import html2text
metadata['comments'] = '<annotation><p>{}</p></annotation>'.format(
prepare_string_for_xml(html2text(comments.value).strip()))
# Keep the indentation level of the description the same as the body.
header = textwrap.dedent('''\
<FictionBook xmlns="http://www.gribuser.ru/xml/fictionbook/2.0" xmlns:l="http://www.w3.org/1999/xlink">
<description>
<title-info>
<genre>%(genre)s</genre>
%(author)s
<book-title>%(title)s</book-title>
%(cover)s
<lang>%(lang)s</lang>
%(keywords)s
%(sequence)s
%(comments)s
</title-info>
<document-info>
%(author)s
<program-used>%(appname)s %(version)s</program-used>
<date>%(date)s</date>
<id>%(id)s</id>
<version>1.0</version>
</document-info>
<publish-info>
%(publisher)s
%(year)s
%(isbn)s
</publish-info>
</description>''') % metadata
# Remove empty lines.
return '\n'.join(filter(unicode_type.strip, header.splitlines()))
def create_epub_cover(container, cover_path, existing_image, options=None):
from calibre.ebooks.conversion.config import load_defaults
from calibre.ebooks.oeb.transforms.cover import CoverManager
ext = cover_path.rpartition('.')[-1].lower()
cname, tname = 'cover.' + ext, 'titlepage.xhtml'
recommended_folders = get_recommended_folders(container, (cname, tname))
if existing_image:
raster_cover = existing_image
manifest_id = {v: k
for k, v in container.manifest_id_map.iteritems()
}[existing_image]
raster_cover_item = container.opf_xpath('//opf:manifest/*[@id="%s"]' %
manifest_id)[0]
else:
folder = recommended_folders[cname]
if folder:
cname = folder + '/' + cname
raster_cover_item = container.generate_item(cname, id_prefix='cover')
raster_cover = container.href_to_name(raster_cover_item.get('href'),
container.opf_name)
with open(cover_path, 'rb') as src, container.open(raster_cover,
'wb') as dest:
shutil.copyfileobj(src, dest)
if options is None:
opts = load_defaults('epub_output')
keep_aspect = opts.get('preserve_cover_aspect_ratio', False)
no_svg = opts.get('no_svg_cover', False)
else:
keep_aspect = options.get('keep_aspect', False)
no_svg = options.get('no_svg', False)
if no_svg:
style = 'style="height: 100%%"'
templ = CoverManager.NONSVG_TEMPLATE.replace('__style__', style)
else:
width, height = 600, 800
try:
if existing_image:
width, height = identify_data(
container.raw_data(existing_image, decode=False))[:2]
else:
width, height = identify(cover_path)[:2]
except:
container.log.exception("Failed to get width and height of cover")
ar = 'xMidYMid meet' if keep_aspect else 'none'
templ = CoverManager.SVG_TEMPLATE.replace('__ar__', ar)
templ = templ.replace('__viewbox__', '0 0 %d %d' % (width, height))
templ = templ.replace('__width__', str(width))
templ = templ.replace('__height__', str(height))
folder = recommended_folders[tname]
if folder:
tname = folder + '/' + tname
titlepage_item = container.generate_item(tname, id_prefix='titlepage')
titlepage = container.href_to_name(titlepage_item.get('href'),
container.opf_name)
raw = templ % container.name_to_href(raster_cover,
titlepage).encode('utf-8')
with container.open(titlepage, 'wb') as f:
f.write(raw)
# We have to make sure the raster cover item has id="cover" for the moron
# that wrote the Nook firmware
if raster_cover_item.get('id') != 'cover':
from calibre.ebooks.oeb.base import uuid_id
newid = uuid_id()
for item in container.opf_xpath('//*[@id="cover"]'):
item.set('id', newid)
for item in container.opf_xpath('//*[@idref="cover"]'):
item.set('idref', newid)
raster_cover_item.set('id', 'cover')
spine = container.opf_xpath('//opf:spine')[0]
ref = spine.makeelement(OPF('itemref'), idref=titlepage_item.get('id'))
container.insert_into_xml(spine, ref, index=0)
guide = container.opf_get_or_create('guide')
container.insert_into_xml(
guide,
guide.makeelement(OPF('reference'),
type='cover',
title=_('Cover'),
href=container.name_to_href(
titlepage, base=container.opf_name)))
metadata = container.opf_get_or_create('metadata')
meta = metadata.makeelement(OPF('meta'), name='cover')
meta.set('content', raster_cover_item.get('id'))
container.insert_into_xml(metadata, meta)
return raster_cover, titlepage
def check_opf(container):
errors = []
if container.opf.tag != OPF('package'):
err = BaseError(_('The OPF does not have the correct root element'),
container.opf_name)
err.HELP = xml(
_('The opf must have the root element <package> in namespace {0}, like this: <package xmlns="{0}">'
)).format(OPF2_NS)
errors.append(err)
for tag in ('metadata', 'manifest', 'spine'):
if not container.opf_xpath('/opf:package/opf:' + tag):
errors.append(MissingSection(container.opf_name, tag))
all_ids = set(container.opf_xpath('//*/@id'))
for elem in container.opf_xpath('//*[@idref]'):
if elem.get('idref') not in all_ids:
errors.append(
IncorrectIdref(container.opf_name, elem.get('idref'),
elem.sourceline))
nl_items = [
elem.sourceline for elem in container.opf_xpath(
'//opf:spine/opf:itemref[@linear="no"]')
]
if nl_items:
errors.append(NonLinearItems(container.opf_name, nl_items))
seen, dups = {}, {}
for item in container.opf_xpath(
'/opf:package/opf:manifest/opf:item[@href]'):
href = item.get('href')
if not container.exists(
container.href_to_name(href, container.opf_name)):
errors.append(
MissingHref(container.opf_name, href, item.sourceline))
if href in seen:
if href not in dups:
dups[href] = [seen[href]]
dups[href].append(item.sourceline)
else:
seen[href] = item.sourceline
errors.extend(
DuplicateHref(container.opf_name, eid, locs)
for eid, locs in dups.iteritems())
seen, dups = {}, {}
for item in container.opf_xpath(
'/opf:package/opf:spine/opf:itemref[@idref]'):
ref = item.get('idref')
if ref in seen:
if ref not in dups:
dups[ref] = [seen[ref]]
dups[ref].append(item.sourceline)
else:
seen[ref] = item.sourceline
errors.extend(
DuplicateHref(container.opf_name, eid, locs, for_spine=True)
for eid, locs in dups.iteritems())
spine = container.opf_xpath('/opf:package/opf:spine[@toc]')
if spine:
spine = spine[0]
mitems = [
x for x in container.opf_xpath(
'/opf:package/opf:manifest/opf:item[@id]')
if x.get('id') == spine.get('toc')
]
if mitems:
mitem = mitems[0]
if mitem.get('media-type', '') != guess_type('a.ncx'):
errors.append(
IncorrectToc(container.opf_name,
mitem.sourceline,
bad_mimetype=mitem.get('media-type')))
else:
errors.append(
IncorrectToc(container.opf_name,
spine.sourceline,
bad_idref=spine.get('toc')))
covers = container.opf_xpath(
'/opf:package/opf:metadata/opf:meta[@name="cover"]')
if len(covers) > 0:
if len(covers) > 1:
errors.append(
MultipleCovers(container.opf_name,
[c.sourceline for c in covers]))
manifest_ids = set(
container.opf_xpath('/opf:package/opf:manifest/opf:item/@id'))
for cover in covers:
if cover.get('content', None) not in manifest_ids:
errors.append(
IncorrectCover(container.opf_name, cover.sourceline,
cover.get('content', '')))
raw = etree.tostring(cover)
try:
n, c = raw.index('name="'), raw.index('content="')
except ValueError:
n = c = -1
if n > -1 and c > -1 and n > c:
errors.append(NookCover(container.opf_name, cover.sourceline))
uid = container.opf.get('unique-identifier', None)
if uid is None or not container.opf_xpath(
'/opf:package/opf:metadata/dc:identifier[@id=%r]' % uid):
errors.append(NoUID(container.opf_name))
for item, name, linear in container.spine_iter:
mt = container.mime_map[name]
if mt != XHTML_MIME:
iid = item.get('idref', None)
lnum = None
if iid:
mitem = container.opf_xpath(
'/opf:package/opf:manifest/opf:item[@id=%r]' % iid)
if mitem:
lnum = mitem[0].sourceline
else:
iid = None
errors.append(BadSpineMime(name, iid, mt, lnum,
container.opf_name))
return errors
def convert(self, stream, options, file_ext, log, accelerators):
from calibre.utils.zipfile import ZipFile
from calibre import walk
from calibre.ebooks import DRMError
from calibre.ebooks.metadata.opf2 import OPF
try:
zf = ZipFile(stream)
zf.extractall(os.getcwdu())
except:
log.exception("EPUB appears to be invalid ZIP file, trying a" " more forgiving ZIP parser")
from calibre.utils.localunzip import extractall
stream.seek(0)
extractall(stream)
encfile = os.path.abspath(os.path.join("META-INF", "encryption.xml"))
opf = self.find_opf()
if opf is None:
for f in walk(u"."):
if f.lower().endswith(".opf") and "__MACOSX" not in f and not os.path.basename(f).startswith("."):
opf = os.path.abspath(f)
break
path = getattr(stream, "name", "stream")
if opf is None:
raise ValueError("%s is not a valid EPUB file (could not find opf)" % path)
opf = os.path.relpath(opf, os.getcwdu())
parts = os.path.split(opf)
opf = OPF(opf, os.path.dirname(os.path.abspath(opf)))
self._encrypted_font_uris = []
if os.path.exists(encfile):
if not self.process_encryption(encfile, opf, log):
raise DRMError(os.path.basename(path))
self.encrypted_fonts = self._encrypted_font_uris
if len(parts) > 1 and parts[0]:
delta = "/".join(parts[:-1]) + "/"
for elem in opf.itermanifest():
elem.set("href", delta + elem.get("href"))
for elem in opf.iterguide():
elem.set("href", delta + elem.get("href"))
self.removed_cover = self.rationalize_cover(opf, log)
self.optimize_opf_parsing = opf
for x in opf.itermanifest():
if x.get("media-type", "") == "application/x-dtbook+xml":
raise ValueError("EPUB files with DTBook markup are not supported")
not_for_spine = set()
for y in opf.itermanifest():
id_ = y.get("id", None)
if id_ and y.get("media-type", None) in ("application/vnd.adobe-page-template+xml", "application/text"):
not_for_spine.add(id_)
seen = set()
for x in list(opf.iterspine()):
ref = x.get("idref", None)
if not ref or ref in not_for_spine or ref in seen:
x.getparent().remove(x)
continue
seen.add(ref)
if len(list(opf.iterspine())) == 0:
raise ValueError("No valid entries in the spine of this EPUB")
with open("content.opf", "wb") as nopf:
nopf.write(opf.render())
return os.path.abspath(u"content.opf")
def meta_info_to_oeb_metadata(mi, m, log, override_input_metadata=False):
from calibre.ebooks.oeb.base import OPF
if not mi.is_null('title'):
m.clear('title')
m.add('title', mi.title)
if mi.title_sort:
if not m.title:
m.add('title', mi.title_sort)
m.clear('title_sort')
m.add('title_sort', mi.title_sort)
if not mi.is_null('authors'):
m.filter('creator', lambda x : x.role.lower() in ['aut', ''])
for a in mi.authors:
attrib = {'role':'aut'}
if mi.author_sort:
attrib[OPF('file-as')] = mi.author_sort
m.add('creator', a, attrib=attrib)
if not mi.is_null('book_producer'):
m.filter('contributor', lambda x : x.role.lower() == 'bkp')
m.add('contributor', mi.book_producer, role='bkp')
elif override_input_metadata:
m.filter('contributor', lambda x : x.role.lower() == 'bkp')
if not mi.is_null('comments'):
m.clear('description')
m.add('description', mi.comments)
elif override_input_metadata:
m.clear('description')
if not mi.is_null('publisher'):
m.clear('publisher')
m.add('publisher', mi.publisher)
elif override_input_metadata:
m.clear('publisher')
if not mi.is_null('series'):
m.clear('series')
m.add('series', mi.series)
elif override_input_metadata:
m.clear('series')
identifiers = mi.get_identifiers()
set_isbn = False
for typ, val in identifiers.iteritems():
has = False
if typ.lower() == 'isbn':
set_isbn = True
for x in m.identifier:
if x.scheme.lower() == typ.lower():
x.content = val
has = True
if not has:
m.add('identifier', val, scheme=typ.upper())
if override_input_metadata and not set_isbn:
m.filter('identifier', lambda x: x.scheme.lower() == 'isbn')
if not mi.is_null('languages'):
m.clear('language')
for lang in mi.languages:
if lang and lang.lower() not in ('und', ''):
m.add('language', lang)
if not mi.is_null('series_index'):
m.clear('series_index')
m.add('series_index', mi.format_series_index())
elif override_input_metadata:
m.clear('series_index')
if not mi.is_null('rating'):
m.clear('rating')
m.add('rating', '%.2f'%mi.rating)
elif override_input_metadata:
m.clear('rating')
if not mi.is_null('tags'):
m.clear('subject')
for t in mi.tags:
m.add('subject', t)
elif override_input_metadata:
m.clear('subject')
if not mi.is_null('pubdate'):
m.clear('date')
m.add('date', isoformat(mi.pubdate))
if not mi.is_null('timestamp'):
m.clear('timestamp')
m.add('timestamp', isoformat(mi.timestamp))
if not mi.is_null('rights'):
m.clear('rights')
m.add('rights', mi.rights)
if not mi.is_null('publication_type'):
m.clear('publication_type')
m.add('publication_type', mi.publication_type)
if not m.timestamp:
m.add('timestamp', isoformat(now()))
def build_exth(metadata, prefer_author_sort=False, is_periodical=False,
share_not_sync=True, cover_offset=None, thumbnail_offset=None,
start_offset=None, mobi_doctype=2, num_of_resources=None,
kf8_unknown_count=0, be_kindlegen2=False, kf8_header_index=None,
page_progression_direction=None, primary_writing_mode=None):
exth = BytesIO()
nrecs = 0
for term in metadata:
if term not in EXTH_CODES:
continue
code = EXTH_CODES[term]
items = metadata[term]
if term == 'creator':
if prefer_author_sort:
creators = [authors_to_sort_string([unicode_type(c)]) for c in
items]
else:
creators = [unicode_type(c) for c in items]
items = creators
elif term == 'rights':
try:
rights = utf8_text(unicode_type(metadata.rights[0]))
except:
rights = b'Unknown'
exth.write(pack(b'>II', EXTH_CODES['rights'], len(rights) + 8))
exth.write(rights)
nrecs += 1
continue
for item in items:
data = unicode_type(item)
if term != 'description':
data = COLLAPSE_RE.sub(' ', data)
if term == 'identifier':
if data.lower().startswith('urn:isbn:'):
data = data[9:]
elif item.scheme.lower() == 'isbn':
pass
else:
continue
if term == 'language':
d2 = lang_as_iso639_1(data)
if d2:
data = d2
data = utf8_text(data)
exth.write(pack(b'>II', code, len(data) + 8))
exth.write(data)
nrecs += 1
# Write UUID as ASIN
uuid = None
from calibre.ebooks.oeb.base import OPF
for x in metadata['identifier']:
if (x.get(OPF('scheme'), None).lower() == 'uuid' or
unicode_type(x).startswith('urn:uuid:')):
uuid = unicode_type(x).split(':')[-1]
break
if uuid is None:
from uuid import uuid4
uuid = str(uuid4())
if isinstance(uuid, unicode_type):
uuid = uuid.encode('utf-8')
if not share_not_sync:
exth.write(pack(b'>II', 113, len(uuid) + 8))
exth.write(uuid)
nrecs += 1
# Write UUID as SOURCE
c_uuid = b'calibre:%s' % uuid
exth.write(pack(b'>II', 112, len(c_uuid) + 8))
exth.write(c_uuid)
nrecs += 1
# Write cdetype
if not is_periodical:
if not share_not_sync:
exth.write(pack(b'>II', 501, 12))
exth.write(b'EBOK')
nrecs += 1
else:
ids = {0x101:b'NWPR', 0x103:b'MAGZ'}.get(mobi_doctype, None)
if ids:
exth.write(pack(b'>II', 501, 12))
exth.write(ids)
nrecs += 1
# Add a publication date entry
if metadata['date']:
datestr = str(metadata['date'][0])
elif metadata['timestamp']:
datestr = str(metadata['timestamp'][0])
if datestr is None:
raise ValueError("missing date or timestamp")
datestr = bytes(datestr)
exth.write(pack(b'>II', EXTH_CODES['pubdate'], len(datestr) + 8))
exth.write(datestr)
nrecs += 1
if is_periodical:
exth.write(pack(b'>II', EXTH_CODES['lastupdatetime'], len(datestr) + 8))
exth.write(datestr)
nrecs += 1
if be_kindlegen2:
mv = 200 if iswindows else 202 if isosx else 201
vals = {204:mv, 205:2, 206:9, 207:0}
elif is_periodical:
# Pretend to be amazon's super secret periodical generator
vals = {204:201, 205:2, 206:0, 207:101}
else:
# Pretend to be kindlegen 1.2
vals = {204:201, 205:1, 206:2, 207:33307}
for code, val in vals.iteritems():
exth.write(pack(b'>III', code, 12, val))
nrecs += 1
if be_kindlegen2:
revnum = b'0730-890adc2'
exth.write(pack(b'>II', 535, 8 + len(revnum)) + revnum)
nrecs += 1
if cover_offset is not None:
exth.write(pack(b'>III', EXTH_CODES['coveroffset'], 12,
cover_offset))
exth.write(pack(b'>III', EXTH_CODES['hasfakecover'], 12, 0))
nrecs += 2
if thumbnail_offset is not None:
exth.write(pack(b'>III', EXTH_CODES['thumboffset'], 12,
thumbnail_offset))
thumbnail_uri_str = bytes('kindle:embed:%s' %(to_base(thumbnail_offset, base=32, min_num_digits=4)))
exth.write(pack(b'>II', EXTH_CODES['kf8_thumbnail_uri'], len(thumbnail_uri_str) + 8))
exth.write(thumbnail_uri_str)
nrecs += 2
if start_offset is not None:
try:
len(start_offset)
except TypeError:
start_offset = [start_offset]
for so in start_offset:
if so is not None:
exth.write(pack(b'>III', EXTH_CODES['startreading'], 12,
so))
nrecs += 1
if kf8_header_index is not None:
exth.write(pack(b'>III', EXTH_CODES['kf8_header_index'], 12,
kf8_header_index))
nrecs += 1
if num_of_resources is not None:
exth.write(pack(b'>III', EXTH_CODES['num_of_resources'], 12,
num_of_resources))
nrecs += 1
if kf8_unknown_count is not None:
exth.write(pack(b'>III', EXTH_CODES['kf8_unknown_count'], 12,
kf8_unknown_count))
nrecs += 1
if primary_writing_mode:
pwm = primary_writing_mode.encode('utf-8')
exth.write(pack(b'>II', EXTH_CODES['primary_writing_mode'], len(pwm) + 8))
exth.write(pwm)
nrecs += 1
if page_progression_direction in {'rtl', 'ltr', 'default'}:
ppd = bytes(page_progression_direction)
exth.write(pack(b'>II', EXTH_CODES['page_progression_direction'], len(ppd) + 8))
exth.write(ppd)
nrecs += 1
exth = exth.getvalue()
trail = len(exth) % 4
pad = b'\0' * (4 - trail) # Always pad w/ at least 1 byte
exth = [b'EXTH', pack(b'>II', len(exth) + 12, nrecs), exth, pad]
return b''.join(exth)
def split(container, name, loc_or_xpath, before=True, totals=None):
'''
Split the file specified by name at the position specified by loc_or_xpath.
Splitting automatically migrates all links and references to the affected
files.
:param loc_or_xpath: Should be an XPath expression such as
//h:div[@id="split_here"]. Can also be a *loc* which is used internally to
implement splitting in the preview panel.
:param before: If True the split occurs before the identified element otherwise after it.
:param totals: Used internally
'''
root = container.parsed(name)
if isinstance(loc_or_xpath, type('')):
split_point = root.xpath(loc_or_xpath)[0]
else:
try:
split_point = node_from_loc(root, loc_or_xpath, totals=totals)
except MalformedMarkup:
# The webkit HTML parser and the container parser have yielded
# different node counts, this can happen if the file is valid XML
# but contains constructs like nested <p> tags. So force parse it
# with the HTML 5 parser and try again.
raw = container.raw_data(name)
root = container.parse_xhtml(raw,
fname=name,
force_html5_parse=True)
try:
split_point = node_from_loc(root, loc_or_xpath, totals=totals)
except MalformedMarkup:
raise MalformedMarkup(
_('The file %s has malformed markup. Try running the Fix HTML tool'
' before splitting') % name)
container.replace(name, root)
if in_table(split_point):
raise AbortError('Cannot split inside tables')
if split_point.tag.endswith('}body'):
raise AbortError('Cannot split on the <body> tag')
tree1, tree2 = do_split(split_point, container.log, before=before)
root1, root2 = tree1.getroot(), tree2.getroot()
anchors_in_top = frozenset(root1.xpath('//*/@id')) | frozenset(
root1.xpath('//*/@name')) | {''}
anchors_in_bottom = frozenset(root2.xpath('//*/@id')) | frozenset(
root2.xpath('//*/@name'))
base, ext = name.rpartition('.')[0::2]
base = re.sub(r'_split\d+$', '', base)
nname, s = None, 0
while not nname or container.exists(nname):
s += 1
nname = '%s_split%d.%s' % (base, s, ext)
manifest_item = container.generate_item(
nname, media_type=container.mime_map[name])
bottom_name = container.href_to_name(manifest_item.get('href'),
container.opf_name)
# Fix links in the split trees
for r in (root1, root2):
for a in r.xpath('//*[@href]'):
url = a.get('href')
if url.startswith('#'):
fname = name
else:
fname = container.href_to_name(url, name)
if fname == name:
purl = urlparse(url)
if purl.fragment in anchors_in_top:
if r is root2:
a.set(
'href', '%s#%s' % (container.name_to_href(
name, bottom_name), purl.fragment))
else:
a.set('href', '#' + purl.fragment)
elif purl.fragment in anchors_in_bottom:
if r is root1:
a.set(
'href', '%s#%s' % (container.name_to_href(
bottom_name, name), purl.fragment))
else:
a.set('href', '#' + purl.fragment)
# Fix all links in the container that point to anchors in the bottom tree
for fname, media_type in iteritems(container.mime_map):
if fname not in {name, bottom_name}:
repl = SplitLinkReplacer(fname, anchors_in_bottom, name,
bottom_name, container)
container.replace_links(fname, repl)
container.replace(name, root1)
container.replace(bottom_name, root2)
spine = container.opf_xpath('//opf:spine')[0]
for spine_item, spine_name, linear in container.spine_iter:
if spine_name == name:
break
index = spine.index(spine_item) + 1
si = spine.makeelement(OPF('itemref'), idref=manifest_item.get('id'))
if not linear:
si.set('linear', 'no')
container.insert_into_xml(spine, si, index=index)
container.dirty(container.opf_name)
return bottom_name
def convert(self, stream, options, file_ext, log, accelerators):
from calibre.utils.zipfile import ZipFile
from calibre import walk
from calibre.ebooks import DRMError
from calibre.ebooks.metadata.opf2 import OPF
try:
zf = ZipFile(stream)
zf.extractall(os.getcwdu())
except:
log.exception('EPUB appears to be invalid ZIP file, trying a'
' more forgiving ZIP parser')
from calibre.utils.localunzip import extractall
stream.seek(0)
extractall(stream)
encfile = os.path.abspath(os.path.join('META-INF', 'encryption.xml'))
opf = self.find_opf()
if opf is None:
for f in walk(u'.'):
if f.lower().endswith('.opf') and '__MACOSX' not in f and \
not os.path.basename(f).startswith('.'):
opf = os.path.abspath(f)
break
path = getattr(stream, 'name', 'stream')
if opf is None:
raise ValueError('%s is not a valid EPUB file (could not find opf)'%path)
opf = os.path.relpath(opf, os.getcwdu())
parts = os.path.split(opf)
opf = OPF(opf, os.path.dirname(os.path.abspath(opf)))
self._encrypted_font_uris = []
if os.path.exists(encfile):
if not self.process_encryption(encfile, opf, log):
raise DRMError(os.path.basename(path))
self.encrypted_fonts = self._encrypted_font_uris
epub3_nav = opf.epub3_nav
if epub3_nav is not None:
self.convert_epub3_nav(epub3_nav, opf, log)
if len(parts) > 1 and parts[0]:
delta = '/'.join(parts[:-1])+'/'
for elem in opf.itermanifest():
elem.set('href', delta+elem.get('href'))
for elem in opf.iterguide():
elem.set('href', delta+elem.get('href'))
f = self.rationalize_cover3 if opf.package_version >= 3.0 else self.rationalize_cover2
self.removed_cover = f(opf, log)
for x in opf.itermanifest():
if x.get('media-type', '') == 'application/x-dtbook+xml':
raise ValueError(
'EPUB files with DTBook markup are not supported')
not_for_spine = set()
for y in opf.itermanifest():
id_ = y.get('id', None)
if id_ and y.get('media-type', None) in {
'application/vnd.adobe-page-template+xml', 'application/vnd.adobe.page-template+xml',
'application/adobe-page-template+xml', 'application/adobe.page-template+xml',
'application/text'}:
not_for_spine.add(id_)
seen = set()
for x in list(opf.iterspine()):
ref = x.get('idref', None)
if not ref or ref in not_for_spine or ref in seen:
x.getparent().remove(x)
continue
seen.add(ref)
if len(list(opf.iterspine())) == 0:
raise ValueError('No valid entries in the spine of this EPUB')
with lopen('content.opf', 'wb') as nopf:
nopf.write(opf.render())
return os.path.abspath(u'content.opf')
