def periodicalize_toc(self):
#return #commit by arroz, you have to create a toc hiberarchy youself.
from calibre.ebooks.oeb.base import TOC
toc = self.oeb.toc
if not toc: # or len(self.oeb.spine) < 3:
return
if toc and toc[0].klass != 'periodical':
#spine[0]为index.html
#one, two = self.oeb.spine[0], self.oeb.spine[1]
self.log.info('Converting TOC for MOBI periodical indexing...')
articles = {}
if toc.depth() < 3:
# single section periodical
#self.oeb.manifest.remove(one) #manifest删除的同时也会将spine项删除
#self.oeb.manifest.remove(two)
#新增一个节点，将之前的所有节点归于此节点下
sections = [
TOC(klass='section',
title=_('All articles'),
href=self.oeb.spine[0].href)
]
for x in toc:
sections[0].nodes.append(x)
else:
# multi-section periodical
#self.oeb.manifest.remove(one)
sections = list(toc) # 顶节点和第一层节点的列表
for x in sections:
x.klass = 'section'
articles_ = list(x) #第二层节点
if articles_: #去掉原先节点的链接，改变链接到第一篇文章
#self.oeb.manifest.remove(self.oeb.manifest.hrefs[x.href])
x.href = articles_[0].href
for sec in sections:
articles[id(sec)] = []
for a in list(sec):
a.klass = 'article'
articles[id(sec)].append(a)
sec.nodes.remove(a)
root = TOC(klass='periodical',
href=self.oeb.spine[0].href,
title=unicode(self.oeb.metadata.title[0]))
#建立 root/Sections/artcicles三层结构
for s in sections:
if articles[id(s)]:
for a in articles[id(s)]:
s.nodes.append(a)
root.nodes.append(s)
for x in list(toc.nodes):
toc.nodes.remove(x)
toc.nodes.append(root)
# Fix up the periodical href to point to first section href
toc.nodes[0].href = toc.nodes[0].nodes[0].href
def flatten_toc(self):
from calibre.ebooks.oeb.base import TOC
nroot = TOC()
for x in self.oeb.toc.iterdescendants():
nroot.add(x.title, x.href)
self.oeb.toc = nroot
def periodicalize_toc(self):
from calibre.ebooks.oeb.base import TOC
toc = self.oeb.toc
if not toc or len(self.oeb.spine) < 3:
return
if toc and toc[0].klass != 'periodical':
one, two = self.oeb.spine[0], self.oeb.spine[1]
self.log('Converting TOC for MOBI periodical indexing...')
articles = {}
if toc.depth() < 3:
# single section periodical
self.oeb.manifest.remove(one)
self.oeb.manifest.remove(two)
sections = [
TOC(klass='section',
title=_('All articles'),
href=self.oeb.spine[0].href)
]
for x in toc:
sections[0].nodes.append(x)
else:
# multi-section periodical
self.oeb.manifest.remove(one)
sections = list(toc)
for i, x in enumerate(sections):
x.klass = 'section'
articles_ = list(x)
if articles_:
self.oeb.manifest.remove(
self.oeb.manifest.hrefs[x.href])
x.href = articles_[0].href
for sec in sections:
articles[id(sec)] = []
for a in list(sec):
a.klass = 'article'
articles[id(sec)].append(a)
sec.nodes.remove(a)
root = TOC(klass='periodical',
href=self.oeb.spine[0].href,
title=unicode_type(self.oeb.metadata.title[0]))
for s in sections:
if articles[id(s)]:
for a in articles[id(s)]:
s.nodes.append(a)
root.nodes.append(s)
for x in list(toc.nodes):
toc.nodes.remove(x)
toc.nodes.append(root)
# Fix up the periodical href to point to first section href
toc.nodes[0].href = toc.nodes[0].nodes[0].href
def _process_nodes(self, root):
from calibre.ebooks.oeb.base import TOC
toc = TOC()
ancestor_map = {}
for node in root.xpath('//object'):
self.add_node(node, toc, ancestor_map)
return toc
def __call__(self, oeb, opts):
self.log = oeb.log
self.oeb = oeb
self.opts = opts
self.log('Detecting structure...')
self.detect_chapters()
if self.oeb.auto_generated_toc or opts.use_auto_toc:
orig_toc = self.oeb.toc
self.oeb.toc = TOC()
self.create_level_based_toc()
if self.oeb.toc.count() < 1:
if not opts.no_chapters_in_toc and self.detected_chapters:
self.create_toc_from_chapters()
if self.oeb.toc.count() < opts.toc_threshold:
self.create_toc_from_links()
if self.oeb.toc.count() < 2 and orig_toc.count() > 2:
self.oeb.toc = orig_toc
else:
self.oeb.auto_generated_toc = True
self.log('Auto generated TOC with %d entries.' %
self.oeb.toc.count())
if opts.toc_filter is not None:
regexp = re.compile(opts.toc_filter)
for node in list(self.oeb.toc.iter()):
if not node.title or regexp.search(node.title) is not None:
self.log('Filtering',
node.title if node.title else 'empty node',
'from TOC')
self.oeb.toc.remove(node)
if opts.page_breaks_before is not None:
pb_xpath = XPath(opts.page_breaks_before)
for item in oeb.spine:
for elem in pb_xpath(item.data):
try:
prev = next(
elem.itersiblings(tag=etree.Element,
preceding=True))
if (barename(elem.tag) in {'h1', 'h2'}
and barename(prev.tag) in {'h1', 'h2'}
and (not prev.tail or not prev.tail.split())):
# We have two adjacent headings, do not put a page
# break on the second one
continue
except StopIteration:
pass
style = elem.get('style', '')
if style:
style += '; '
elem.set('style', style + 'page-break-before:always')
for node in self.oeb.toc.iter():
if not node.title or not node.title.strip():
node.title = _('Unnamed')
if self.opts.start_reading_at:
self.detect_start_reading()
def parse_html_toc(self, item):
from calibre.ebooks.oeb.base import TOC, XPath
dx = XPath('./h:div')
ax = XPath('./h:a[1]')
def do_node(parent, div):
for child in dx(div):
a = ax(child)[0]
c = parent.add(a.text, a.attrib['href'])
do_node(c, child)
toc = TOC()
root = XPath('//h:div[1]')(item.data)[0]
do_node(toc, root)
return toc
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 flatten_toc(self):
from calibre.ebooks.oeb.base import TOC
nroot = TOC()
for x in self.oeb.toc.iterdescendants():
nroot.add(x.title, x.href)
self.oeb.toc = nroot