class NonLinearItems(BaseError):
level = WARN
has_multiple_locations = True
HELP = xml(
_('There are items marked as non-linear in the <spine>.'
' These will be displayed in random order by different ebook readers.'
' Some will ignore the non-linear attribute, some will display'
' them at the end or the beginning of the book and some will'
' fail to display them at all. Instead of using non-linear items'
' simply place the items in the order you want them to be displayed.'
))
INDIVIDUAL_FIX = _('Mark all non-linear items as linear')
def __init__(self, name, locs):
BaseError.__init__(self, _('Non-linear items in the spine'), name)
self.all_locations = [(name, x, None) for x in locs]
def __call__(self, container):
[
elem.attrib.pop('linear')
for elem in container.opf_xpath('//opf:spine/opf:itemref[@linear]')
]
container.dirty(container.opf_name)
return True
def __init__(self, name, section_name):
BaseError.__init__(
self,
_('The <%s> section is missing from the OPF') % section_name, name)
self.HELP = xml(
_('The <%s> section is required in the OPF file. You have to create one.'
) % section_name)
class NoUID(BaseError):
HELP = xml(
_('The OPF must have a unique identifier, i.e. a <dc:identifier> element whose id is referenced'
' by the <package> element'))
INDIVIDUAL_FIX = _('Auto-generate a unique identifier')
def __init__(self, name):
BaseError.__init__(self, _('The OPF has no unique identifier'), name)
def __call__(self, container):
import uuid
opf = container.opf
uid = str(uuid.uuid4())
opf.set('unique-identifier', uid)
m = container.opf_xpath('/opf:package/opf:metadata')
if not m:
m = [
container.opf.makeelement(OPF('metadata'),
nsmap={'dc': DC11_NS})
]
container.insert_into_xml(container.opf, m[0], 0)
m = m[0]
dc = m.makeelement(DC('identifier'), id=uid, nsmap={'opf': OPF2_NS})
dc.set(OPF('scheme'), 'uuid')
dc.text = uid
container.insert_into_xml(m, dc)
container.dirty(container.opf_name)
return True
def __init__(self, name, idref, lnum):
BaseError.__init__(self,
_('idref="%s" points to unknown id') % idref, name,
lnum)
self.HELP = xml(
_('The idref="%s" points to an id that does not exist in the OPF')
% idref)
def __init__(self, name, lnum, cover):
BaseError.__init__(
self, _('The meta cover tag points to an non-existent item'), name,
lnum)
self.HELP = xml(
_('The meta cover tag points to an item with id="%s" which does not exist in the manifest'
) % cover)
def ACQUISITION_ENTRY(book_id, updated, request_context):
field_metadata = request_context.db.field_metadata
mi = request_context.db.get_metadata(book_id)
extra = []
if (mi.rating or 0) > 0:
rating = rating_to_stars(mi.rating)
extra.append(_('RATING: %s<br />')%rating)
if mi.tags:
extra.append(_('TAGS: %s<br />')%xml(format_tag_string(mi.tags, None)))
if mi.series:
extra.append(_('SERIES: %(series)s [%(sidx)s]<br />')%
dict(series=xml(mi.series),
sidx=fmt_sidx(float(mi.series_index))))
for key in filter(request_context.ctx.is_field_displayable, field_metadata.ignorable_field_keys()):
name, val = mi.format_field(key)
if val:
fm = field_metadata[key]
datatype = fm['datatype']
if datatype == 'text' and fm['is_multiple']:
extra.append('%s: %s<br />'%
(xml(name),
xml(format_tag_string(val,
fm['is_multiple']['ui_to_list'],
joinval=fm['is_multiple']['list_to_ui']))))
elif datatype == 'comments' or (fm['datatype'] == 'composite' and fm['display'].get('contains_html', False)):
extra.append('%s: %s<br />'%(xml(name), comments_to_html(str(val))))
else:
extra.append('%s: %s<br />'%(xml(name), xml(str(val))))
if mi.comments:
comments = comments_to_html(mi.comments)
extra.append(comments)
if extra:
extra = html_to_lxml('\n'.join(extra))
ans = E.entry(TITLE(mi.title), E.author(E.name(authors_to_string(mi.authors))), ID('urn:uuid:' + mi.uuid), UPDATED(mi.last_modified),
E.published(mi.timestamp.isoformat()))
if mi.pubdate and not is_date_undefined(mi.pubdate):
ans.append(ans.makeelement('{%s}date' % DC_NS))
ans[-1].text = mi.pubdate.isoformat()
if len(extra):
ans.append(E.content(extra, type='xhtml'))
get = partial(request_context.ctx.url_for, '/get', book_id=book_id, library_id=request_context.library_id)
if mi.formats:
fm = mi.format_metadata
for fmt in mi.formats:
fmt = fmt.lower()
mt = guess_type('a.'+fmt)[0]
if mt:
link = E.link(type=mt, href=get(what=fmt), rel="http://opds-spec.org/acquisition")
ffm = fm.get(fmt.upper())
if ffm:
link.set('length', str(ffm['size']))
link.set('mtime', ffm['mtime'].isoformat())
ans.append(link)
ans.append(E.link(type='image/jpeg', href=get(what='cover'), rel="http://opds-spec.org/cover"))
ans.append(E.link(type='image/jpeg', href=get(what='thumb'), rel="http://opds-spec.org/thumbnail"))
ans.append(E.link(type='image/jpeg', href=get(what='cover'), rel="http://opds-spec.org/image"))
ans.append(E.link(type='image/jpeg', href=get(what='thumb'), rel="http://opds-spec.org/image/thumbnail"))
return ans
def ACQUISITION_ENTRY(book_id, updated, request_context):
field_metadata = request_context.db.field_metadata
mi = request_context.db.get_metadata(book_id)
extra = []
if mi.rating > 0:
rating = u"".join(repeat(u"\u2605", int(mi.rating / 2.0)))
extra.append(_("RATING: %s<br />") % rating)
if mi.tags:
extra.append(_("TAGS: %s<br />") % xml(format_tag_string(mi.tags, None)))
if mi.series:
extra.append(
_("SERIES: %(series)s [%(sidx)s]<br />")
% dict(series=xml(mi.series), sidx=fmt_sidx(float(mi.series_index)))
)
for key in filter(request_context.ctx.is_field_displayable, field_metadata.ignorable_field_keys()):
name, val = mi.format_field(key)
if val:
fm = field_metadata[key]
datatype = fm["datatype"]
if datatype == "text" and fm["is_multiple"]:
extra.append(
"%s: %s<br />"
% (
xml(name),
xml(
format_tag_string(
val, fm["is_multiple"]["ui_to_list"], joinval=fm["is_multiple"]["list_to_ui"]
)
),
)
)
elif datatype == "comments" or (
fm["datatype"] == "composite" and fm["display"].get("contains_html", False)
):
extra.append("%s: %s<br />" % (xml(name), comments_to_html(unicode(val))))
else:
extra.append("%s: %s<br />" % (xml(name), xml(unicode(val))))
if mi.comments:
comments = comments_to_html(mi.comments)
extra.append(comments)
if extra:
extra = html_to_lxml("\n".join(extra))
ans = E.entry(
TITLE(mi.title), E.author(E.name(authors_to_string(mi.authors))), ID("urn:uuid:" + mi.uuid), UPDATED(updated)
)
if len(extra):
ans.append(E.content(extra, type="xhtml"))
get = partial(request_context.ctx.url_for, "/get", book_id=book_id, library_id=request_context.library_id)
if mi.formats:
for fmt in mi.formats:
fmt = fmt.lower()
mt = guess_type("a." + fmt)[0]
if mt:
ans.append(E.link(type=mt, href=get(what=fmt), rel="http://opds-spec.org/acquisition"))
ans.append(E.link(type="image/jpeg", href=get(what="cover"), rel="http://opds-spec.org/cover"))
ans.append(E.link(type="image/jpeg", href=get(what="thumb"), rel="http://opds-spec.org/thumbnail"))
return ans
def ACQUISITION_ENTRY(book_id, updated, request_context):
field_metadata = request_context.db.field_metadata
mi = request_context.db.get_metadata(book_id)
extra = []
if mi.rating > 0:
rating = rating_to_stars(mi.rating)
extra.append(_('RATING: %s<br />')%rating)
if mi.tags:
extra.append(_('TAGS: %s<br />')%xml(format_tag_string(mi.tags, None)))
if mi.series:
extra.append(_('SERIES: %(series)s [%(sidx)s]<br />')%
dict(series=xml(mi.series),
sidx=fmt_sidx(float(mi.series_index))))
for key in filter(request_context.ctx.is_field_displayable, field_metadata.ignorable_field_keys()):
name, val = mi.format_field(key)
if val:
fm = field_metadata[key]
datatype = fm['datatype']
if datatype == 'text' and fm['is_multiple']:
extra.append('%s: %s<br />'%
(xml(name),
xml(format_tag_string(val,
fm['is_multiple']['ui_to_list'],
joinval=fm['is_multiple']['list_to_ui']))))
elif datatype == 'comments' or (fm['datatype'] == 'composite' and
fm['display'].get('contains_html', False)):
extra.append('%s: %s<br />'%(xml(name), comments_to_html(unicode(val))))
else:
extra.append('%s: %s<br />'%(xml(name), xml(unicode(val))))
if mi.comments:
comments = comments_to_html(mi.comments)
extra.append(comments)
if extra:
extra = html_to_lxml('\n'.join(extra))
ans = E.entry(TITLE(mi.title), E.author(E.name(authors_to_string(mi.authors))), ID('urn:uuid:' + mi.uuid), UPDATED(mi.last_modified),
E.published(mi.timestamp.isoformat()))
if mi.pubdate and not is_date_undefined(mi.pubdate):
ans.append(ans.makeelement('{%s}date' % DC_NS))
ans[-1].text = mi.pubdate.isoformat()
if len(extra):
ans.append(E.content(extra, type='xhtml'))
get = partial(request_context.ctx.url_for, '/get', book_id=book_id, library_id=request_context.library_id)
if mi.formats:
fm = mi.format_metadata
for fmt in mi.formats:
fmt = fmt.lower()
mt = guess_type('a.'+fmt)[0]
if mt:
link = E.link(type=mt, href=get(what=fmt), rel="http://opds-spec.org/acquisition")
ffm = fm.get(fmt.upper())
if ffm:
link.set('length', str(ffm['size']))
link.set('mtime', ffm['mtime'].isoformat())
ans.append(link)
ans.append(E.link(type='image/jpeg', href=get(what='cover'), rel="http://opds-spec.org/cover"))
ans.append(E.link(type='image/jpeg', href=get(what='thumb'), rel="http://opds-spec.org/thumbnail"))
return ans
class EmptyIdentifier(BaseError):
HELP = xml(_('The <dc:identifier> element must not be empty.'))
INDIVIDUAL_FIX = _('Remove empty identifiers')
def __init__(self, name, lnum):
BaseError.__init__(self, _('Empty identifier element'), name, lnum)
def __call__(self, container):
for dcid in container.opf_xpath(
'/opf:package/opf:metadata/dc:identifier'):
if not dcid.text or not dcid.text.strip():
container.remove_from_xml(dcid)
container.dirty(container.opf_name)
return True
class MultipleCovers(BaseError):
has_multiple_locations = True
HELP = xml(_(
'There is more than one <meta name="cover"> tag defined. There should be only one.'))
INDIVIDUAL_FIX = _('Remove all but the first meta cover tag')
def __init__(self, name, locs):
BaseError.__init__(self, _('There is more than one cover defined'), name)
self.all_locations = [(name, lnum, None) for lnum in sorted(locs)]
def __call__(self, container):
items = [e for e in container.opf_xpath('/opf:package/opf:metadata/opf:meta[@name="cover"]')]
[container.remove_from_xml(e) for e in items[1:]]
container.dirty(self.name)
return True
def ACQUISITION_ENTRY(book_id, updated, request_context):
field_metadata = request_context.db.field_metadata
mi = request_context.db.get_metadata(book_id)
extra = []
if mi.rating > 0:
rating = u''.join(repeat(u'\u2605', int(mi.rating/2.)))
extra.append(_('RATING: %s<br />')%rating)
if mi.tags:
extra.append(_('TAGS: %s<br />')%xml(format_tag_string(mi.tags, None, no_tag_count=True)))
if mi.series:
extra.append(_('SERIES: %(series)s [%(sidx)s]<br />')%
dict(series=xml(mi.series),
sidx=fmt_sidx(float(mi.series_index))))
for key in field_metadata.ignorable_field_keys():
name, val = mi.format_field(key)
if val:
fm = field_metadata[key]
datatype = fm['datatype']
if datatype == 'text' and fm['is_multiple']:
extra.append('%s: %s<br />'%
(xml(name),
xml(format_tag_string(val,
fm['is_multiple']['ui_to_list'],
no_tag_count=True,
joinval=fm['is_multiple']['list_to_ui']))))
elif datatype == 'comments' or (fm['datatype'] == 'composite' and
fm['display'].get('contains_html', False)):
extra.append('%s: %s<br />'%(xml(name), comments_to_html(unicode(val))))
else:
extra.append('%s: %s<br />'%(xml(name), xml(unicode(val))))
if mi.comments:
comments = comments_to_html(mi.comments)
extra.append(comments)
if extra:
extra = html_to_lxml('\n'.join(extra))
ans = E.entry(TITLE(mi.title), E.author(E.name(authors_to_string(mi.authors))), ID('urn:uuid:' + mi.uuid), UPDATED(updated))
if len(extra):
ans.append(E.content(extra, type='xhtml'))
get = partial(request_context.ctx.url_for, '/get', book_id=book_id, library_id=request_context.library_id)
if mi.formats:
for fmt in mi.formats:
fmt = fmt.lower()
mt = guess_type('a.'+fmt)[0]
if mt:
ans.append(E.link(type=mt, href=get(what=fmt), rel="http://opds-spec.org/acquisition"))
ans.append(E.link(type='image/jpeg', href=get(what='cover'), rel="http://opds-spec.org/cover"))
ans.append(E.link(type='image/jpeg', href=get(what='thumb'), rel="http://opds-spec.org/thumbnail"))
return ans
def ACQUISITION_ENTRY(item, version, db, updated, CFM, CKEYS, prefix):
FM = db.FIELD_MAP
title = item[FM['title']]
if not title:
title = _('Unknown')
authors = item[FM['authors']]
if not authors:
authors = _('Unknown')
authors = ' & '.join([i.replace('|', ',') for i in
authors.split(',')])
extra = []
rating = item[FM['rating']]
if rating > 0:
rating = u''.join(repeat(u'\u2605', int(rating/2.)))
extra.append(_('RATING: %s<br />')%rating)
tags = item[FM['tags']]
if tags:
extra.append(_('TAGS: %s<br />')%xml(format_tag_string(tags, ',',
ignore_max=True,
no_tag_count=True)))
series = item[FM['series']]
if series:
extra.append(_('SERIES: %(series)s [%(sidx)s]<br />')%\
dict(series=xml(series),
sidx=fmt_sidx(float(item[FM['series_index']]))))
for key in CKEYS:
mi = db.get_metadata(item[CFM['id']['rec_index']], index_is_id=True)
name, val = mi.format_field(key)
if val:
datatype = CFM[key]['datatype']
if datatype == 'text' and CFM[key]['is_multiple']:
extra.append('%s: %s<br />'%
(xml(name),
xml(format_tag_string(val,
CFM[key]['is_multiple']['ui_to_list'],
ignore_max=True, no_tag_count=True,
joinval=CFM[key]['is_multiple']['list_to_ui']))))
elif datatype == 'comments' or (CFM[key]['datatype'] == 'composite' and
CFM[key]['display'].get('contains_html', False)):
extra.append('%s: %s<br />'%(xml(name), comments_to_html(unicode(val))))
else:
extra.append('%s: %s<br />'%(xml(name), xml(unicode(val))))
comments = item[FM['comments']]
if comments:
comments = comments_to_html(comments)
extra.append(comments)
if extra:
extra = html_to_lxml('\n'.join(extra))
idm = 'calibre' if version == 0 else 'uuid'
id_ = 'urn:%s:%s'%(idm, item[FM['uuid']])
ans = E.entry(TITLE(title), E.author(E.name(authors)), ID(id_),
UPDATED(updated))
if len(extra):
ans.append(E.content(extra, type='xhtml'))
formats = item[FM['formats']]
if formats:
for fmt in formats.split(','):
fmt = fmt.lower()
mt = guess_type('a.'+fmt)[0]
href = prefix + '/get/%s/%s'%(fmt, item[FM['id']])
if mt:
link = E.link(type=mt, href=href)
if version > 0:
link.set('rel', "http://opds-spec.org/acquisition")
ans.append(link)
ans.append(E.link(type='image/jpeg', href=prefix+'/get/cover/%s'%item[FM['id']],
rel="x-stanza-cover-image" if version == 0 else
"http://opds-spec.org/cover"))
ans.append(E.link(type='image/jpeg', href=prefix+'/get/thumb/%s'%item[FM['id']],
rel="x-stanza-cover-image-thumbnail" if version == 0 else
"http://opds-spec.org/thumbnail"))
return ans
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))
return errors
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 __init__(self, name, lnum):
BaseError.__init__(self, _('Empty id attributes are invalid'), name, lnum)
self.HELP = xml(_(
'Empty ID attributes are invalid in OPF files.'))
def sony_metadata(oeb):
m = oeb.metadata
title = short_title = str(m.title[0])
publisher = __appname__ + ' ' + __version__
try:
pt = str(oeb.metadata.publication_type[0])
short_title = ':'.join(pt.split(':')[2:])
except:
pass
try:
date = parse_date(str(m.date[0]),
as_utc=False).strftime('%Y-%m-%d')
except:
date = strftime('%Y-%m-%d')
try:
language = str(m.language[0]).replace('_', '-')
except:
language = 'en'
short_title = xml(short_title, True)
metadata = SONY_METADATA.format(title=xml(title),
short_title=short_title,
publisher=xml(publisher), issue_date=xml(date),
language=xml(language))
updated = strftime('%Y-%m-%dT%H:%M:%SZ', time.gmtime())
def cal_id(x):
for k, v in x.attrib.items():
if k.endswith('scheme') and v == 'uuid':
return True
try:
base_id = str(list(filter(cal_id, m.identifier))[0])
except:
base_id = str(uuid4())
toc = oeb.toc
if False and toc.depth() < 3:
# Single section periodical
# Disabled since I prefer the current behavior
from calibre.ebooks.oeb.base import TOC
section = TOC(klass='section', title=_('All articles'),
href=oeb.spine[2].href)
for x in toc:
section.nodes.append(x)
toc = TOC(klass='periodical', href=oeb.spine[2].href,
title=str(oeb.metadata.title[0]))
toc.nodes.append(section)
entries = []
seen_titles = set()
for i, section in enumerate(toc):
if not section.href:
continue
secid = 'section%d'%i
sectitle = section.title
if not sectitle:
sectitle = _('Unknown')
d = 1
bsectitle = sectitle
while sectitle in seen_titles:
sectitle = bsectitle + ' ' + str(d)
d += 1
seen_titles.add(sectitle)
sectitle = xml(sectitle, True)
secdesc = section.description
if not secdesc:
secdesc = ''
secdesc = xml(secdesc)
entries.append(SONY_ATOM_SECTION.format(title=sectitle,
href=section.href, id=xml(base_id)+'/'+secid,
short_title=short_title, desc=secdesc, updated=updated))
for j, article in enumerate(section):
if not article.href:
continue
atitle = article.title
btitle = atitle
d = 1
while atitle in seen_titles:
atitle = btitle + ' ' + str(d)
d += 1
auth = article.author if article.author else ''
desc = section.description
if not desc:
desc = ''
aid = 'article%d'%j
entries.append(SONY_ATOM_ENTRY.format(
title=xml(atitle),
author=xml(auth),
updated=updated,
desc=desc,
short_title=short_title,
section_title=sectitle,
href=article.href,
word_count=str(1),
id=xml(base_id)+'/'+secid+'/'+aid
))
atom = SONY_ATOM.format(short_title=short_title,
entries='\n\n'.join(entries), updated=updated,
id=xml(base_id)).encode('utf-8')
return metadata, atom
def ACQUISITION_ENTRY(item, version, db, updated, CFM, CKEYS, prefix):
FM = db.FIELD_MAP
title = item[FM["title"]]
if not title:
title = _("Unknown")
authors = item[FM["authors"]]
if not authors:
authors = _("Unknown")
authors = " & ".join([i.replace("|", ",") for i in authors.split(",")])
extra = []
rating = item[FM["rating"]]
if rating > 0:
rating = u"".join(repeat(u"\u2605", int(rating / 2.0)))
extra.append(_("RATING: %s<br />") % rating)
tags = item[FM["tags"]]
if tags:
extra.append(_("TAGS: %s<br />") % xml(format_tag_string(tags, ",", ignore_max=True, no_tag_count=True)))
series = item[FM["series"]]
if series:
extra.append(
_("SERIES: %(series)s [%(sidx)s]<br />")
% dict(series=xml(series), sidx=fmt_sidx(float(item[FM["series_index"]])))
)
for key in CKEYS:
mi = db.get_metadata(item[CFM["id"]["rec_index"]], index_is_id=True)
name, val = mi.format_field(key)
if val:
datatype = CFM[key]["datatype"]
if datatype == "text" and CFM[key]["is_multiple"]:
extra.append(
"%s: %s<br />"
% (
xml(name),
xml(
format_tag_string(
val,
CFM[key]["is_multiple"]["ui_to_list"],
ignore_max=True,
no_tag_count=True,
joinval=CFM[key]["is_multiple"]["list_to_ui"],
)
),
)
)
elif datatype == "comments" or (
CFM[key]["datatype"] == "composite" and CFM[key]["display"].get("contains_html", False)
):
extra.append("%s: %s<br />" % (xml(name), comments_to_html(unicode(val))))
else:
extra.append("%s: %s<br />" % (xml(name), xml(unicode(val))))
comments = item[FM["comments"]]
if comments:
comments = comments_to_html(comments)
extra.append(comments)
if extra:
extra = html_to_lxml("\n".join(extra))
idm = "calibre" if version == 0 else "uuid"
id_ = "urn:%s:%s" % (idm, item[FM["uuid"]])
ans = E.entry(TITLE(title), E.author(E.name(authors)), ID(id_), UPDATED(updated))
if len(extra):
ans.append(E.content(extra, type="xhtml"))
formats = item[FM["formats"]]
if formats:
for fmt in formats.split(","):
fmt = fmt.lower()
mt = guess_type("a." + fmt)[0]
href = prefix + "/get/%s/%s" % (fmt, item[FM["id"]])
if mt:
link = E.link(type=mt, href=href)
if version > 0:
link.set("rel", "http://opds-spec.org/acquisition")
ans.append(link)
ans.append(
E.link(
type="image/jpeg",
href=prefix + "/get/cover/%s" % item[FM["id"]],
rel="x-stanza-cover-image" if version == 0 else "http://opds-spec.org/cover",
)
)
ans.append(
E.link(
type="image/jpeg",
href=prefix + "/get/thumb/%s" % item[FM["id"]],
rel="x-stanza-cover-image-thumbnail" if version == 0 else "http://opds-spec.org/thumbnail",
)
)
return ans
def ACQUISITION_ENTRY(book_id, updated, request_context):
field_metadata = request_context.db.field_metadata
mi = request_context.db.get_metadata(book_id)
extra = []
if mi.rating > 0:
rating = u''.join(repeat(u'\u2605', int(mi.rating / 2.)))
extra.append(_('RATING: %s<br />') % rating)
if mi.tags:
extra.append(
_('TAGS: %s<br />') %
xml(format_tag_string(mi.tags, None, no_tag_count=True)))
if mi.series:
extra.append(
_('SERIES: %(series)s [%(sidx)s]<br />') %
dict(series=xml(mi.series), sidx=fmt_sidx(float(mi.series_index))))
for key in field_metadata.ignorable_field_keys():
name, val = mi.format_field(key)
if val:
fm = field_metadata[key]
datatype = fm['datatype']
if datatype == 'text' and fm['is_multiple']:
extra.append(
'%s: %s<br />' %
(xml(name),
xml(
format_tag_string(
val,
fm['is_multiple']['ui_to_list'],
no_tag_count=True,
joinval=fm['is_multiple']['list_to_ui']))))
elif datatype == 'comments' or (fm['datatype'] == 'composite'
and fm['display'].get(
'contains_html', False)):
extra.append('%s: %s<br />' %
(xml(name), comments_to_html(unicode(val))))
else:
extra.append('%s: %s<br />' % (xml(name), xml(unicode(val))))
if mi.comments:
comments = comments_to_html(mi.comments)
extra.append(comments)
if extra:
extra = html_to_lxml('\n'.join(extra))
ans = E.entry(TITLE(mi.title),
E.author(E.name(authors_to_string(mi.authors))),
ID('urn:uuid:' + mi.uuid), UPDATED(updated))
if len(extra):
ans.append(E.content(extra, type='xhtml'))
get = partial(request_context.ctx.url_for,
'/get',
book_id=book_id,
library_id=request_context.library_id)
if mi.formats:
for fmt in mi.formats:
fmt = fmt.lower()
mt = guess_type('a.' + fmt)[0]
if mt:
ans.append(
E.link(type=mt,
href=get(what=fmt),
rel="http://opds-spec.org/acquisition"))
ans.append(
E.link(type='image/jpeg',
href=get(what='cover'),
rel="http://opds-spec.org/cover"))
ans.append(
E.link(type='image/jpeg',
href=get(what='thumb'),
rel="http://opds-spec.org/thumbnail"))
return ans
def __init__(self, name, idref, lnum):
BaseError.__init__(self, _('idref="%s" points to unknown id') % idref, name, lnum)
self.HELP = xml(_(
'The idref="%s" points to an id that does not exist in the OPF') % idref)
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'))
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 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')))
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 container.manifest_id_map.iteritems()}
ncx_id = rmap.get(ncx_name)
if ncx_id:
errors.append(MissingNCXRef(container.opf_name, spine.sourceline, ncx_id))
if opf_version.major > 2:
if find_existing_nav_toc(container) is None:
errors.append(MissingNav(container.opf_name, 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('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 __init__(self, name, section_name):
BaseError.__init__(self, _('The <%s> section is missing from the OPF') % section_name, name)
self.HELP = xml(_(
'The <%s> section is required in the OPF file. You have to create one.') % section_name)
def ACQUISITION_ENTRY(item, version, db, updated, CFM, CKEYS, prefix):
FM = db.FIELD_MAP
title = item[FM['title']]
if not title:
title = _('Unknown')
authors = item[FM['authors']]
if not authors:
authors = _('Unknown')
authors = ' & '.join([i.replace('|', ',') for i in authors.split(',')])
extra = []
rating = item[FM['rating']]
if rating > 0:
rating = u''.join(repeat(u'\u2605', int(rating / 2.)))
extra.append(_('RATING: %s<br />') % rating)
tags = item[FM['tags']]
if tags:
extra.append(
_('TAGS: %s<br />') % xml(
format_tag_string(
tags, ',', ignore_max=True, no_tag_count=True)))
series = item[FM['series']]
if series:
extra.append(
_('SERIES: %(series)s [%(sidx)s]<br />') %
dict(series=xml(series),
sidx=fmt_sidx(float(item[FM['series_index']]))))
for key in CKEYS:
mi = db.get_metadata(item[CFM['id']['rec_index']], index_is_id=True)
name, val = mi.format_field(key)
if val:
datatype = CFM[key]['datatype']
if datatype == 'text' and CFM[key]['is_multiple']:
extra.append(
'%s: %s<br />' %
(xml(name),
xml(
format_tag_string(
val,
CFM[key]['is_multiple']['ui_to_list'],
ignore_max=True,
no_tag_count=True,
joinval=CFM[key]['is_multiple']['list_to_ui']))))
elif datatype == 'comments' or (CFM[key]['datatype'] == 'composite'
and CFM[key]['display'].get(
'contains_html', False)):
extra.append('%s: %s<br />' %
(xml(name), comments_to_html(unicode(val))))
else:
extra.append('%s: %s<br />' % (xml(name), xml(unicode(val))))
comments = item[FM['comments']]
if comments:
comments = comments_to_html(comments)
extra.append(comments)
if extra:
extra = html_to_lxml('\n'.join(extra))
idm = 'calibre' if version == 0 else 'uuid'
id_ = 'urn:%s:%s' % (idm, item[FM['uuid']])
ans = E.entry(TITLE(title), E.author(E.name(authors)), ID(id_),
UPDATED(updated))
if len(extra):
ans.append(E.content(extra, type='xhtml'))
formats = item[FM['formats']]
if formats:
for fmt in formats.split(','):
fmt = fmt.lower()
mt = guess_type('a.' + fmt)[0]
href = prefix + '/get/%s/%s' % (fmt, item[FM['id']])
if mt:
link = E.link(type=mt, href=href)
if version > 0:
link.set('rel', "http://opds-spec.org/acquisition")
ans.append(link)
ans.append(
E.link(type='image/jpeg',
href=prefix + '/get/cover/%s' % item[FM['id']],
rel="x-stanza-cover-image"
if version == 0 else "http://opds-spec.org/cover"))
ans.append(
E.link(type='image/jpeg',
href=prefix + '/get/thumb/%s' % item[FM['id']],
rel="x-stanza-cover-image-thumbnail"
if version == 0 else "http://opds-spec.org/thumbnail"))
return ans
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
class EmptyIdentifier(BaseError):
HELP = xml(_('The <dc:identifier> element must not be empty.'))
def __init__(self, name, lnum):
BaseError.__init__(self, _('Empty identifier element'), name, lnum)
def __init__(self, name, lnum, cover):
BaseError.__init__(self, _('The meta cover tag points to an non-existent item'), name, lnum)
self.HELP = xml(_(
'The meta cover tag points to an item with id="%s" which does not exist in the manifest') % cover)
def sony_metadata(oeb):
m = oeb.metadata
title = short_title = unicode(m.title[0])
publisher = __appname__ + ' ' + __version__
try:
pt = unicode(oeb.metadata.publication_type[0])
short_title = u':'.join(pt.split(':')[2:])
except:
pass
try:
date = parse_date(unicode(m.date[0]),
as_utc=False).strftime('%Y-%m-%d')
except:
date = strftime('%Y-%m-%d')
try:
language = unicode(m.language[0]).replace('_', '-')
except:
language = 'en'
short_title = xml(short_title, True)
metadata = SONY_METADATA.format(title=xml(title),
short_title=short_title,
publisher=xml(publisher), issue_date=xml(date),
language=xml(language))
updated = strftime('%Y-%m-%dT%H:%M:%SZ', time.gmtime())
def cal_id(x):
for k, v in x.attrib.items():
if k.endswith('scheme') and v == 'uuid':
return True
try:
base_id = unicode(list(filter(cal_id, m.identifier))[0])
except:
base_id = str(uuid4())
toc = oeb.toc
if False and toc.depth() < 3:
# Single section periodical
# Disabled since I prefer the current behavior
from calibre.ebooks.oeb.base import TOC
section = TOC(klass='section', title=_('All articles'),
href=oeb.spine[2].href)
for x in toc:
section.nodes.append(x)
toc = TOC(klass='periodical', href=oeb.spine[2].href,
title=unicode(oeb.metadata.title[0]))
toc.nodes.append(section)
entries = []
seen_titles = set([])
for i, section in enumerate(toc):
if not section.href:
continue
secid = 'section%d'%i
sectitle = section.title
if not sectitle:
sectitle = _('Unknown')
d = 1
bsectitle = sectitle
while sectitle in seen_titles:
sectitle = bsectitle + ' ' + str(d)
d += 1
seen_titles.add(sectitle)
sectitle = xml(sectitle, True)
secdesc = section.description
if not secdesc:
secdesc = ''
secdesc = xml(secdesc)
entries.append(SONY_ATOM_SECTION.format(title=sectitle,
href=section.href, id=xml(base_id)+'/'+secid,
short_title=short_title, desc=secdesc, updated=updated))
for j, article in enumerate(section):
if not article.href:
continue
atitle = article.title
btitle = atitle
d = 1
while atitle in seen_titles:
atitle = btitle + ' ' + str(d)
d += 1
auth = article.author if article.author else ''
desc = section.description
if not desc:
desc = ''
aid = 'article%d'%j
entries.append(SONY_ATOM_ENTRY.format(
title=xml(atitle),
author=xml(auth),
updated=updated,
desc=desc,
short_title=short_title,
section_title=sectitle,
href=article.href,
word_count=str(1),
id=xml(base_id)+'/'+secid+'/'+aid
))
atom = SONY_ATOM.format(short_title=short_title,
entries='\n\n'.join(entries), updated=updated,
id=xml(base_id)).encode('utf-8')
return metadata, atom