def structure_toc(entries):
indent_vals = sorted({x.indent for x in entries})
last_found = [None for i in indent_vals]
newtoc = TOC()
if len(indent_vals) > 6:
for x in entries:
newtoc.add_item('index.html', x.anchor, x.text)
return newtoc
def find_parent(level):
candidates = last_found[:level]
for x in reversed(candidates):
if x is not None:
return x
return newtoc
for item in entries:
level = indent_vals.index(item.indent)
parent = find_parent(level)
last_found[level] = parent.add_item('index.html', item.anchor,
item.text)
for i in xrange(level+1, len(last_found)):
last_found[i] = None
return newtoc
def structure_toc(entries):
indent_vals = sorted({x.indent for x in entries})
last_found = [None for i in indent_vals]
newtoc = TOC()
if len(indent_vals) > 6:
for x in entries:
newtoc.add_item('index.html', x.anchor, x.text)
return newtoc
def find_parent(level):
candidates = last_found[:level]
for x in reversed(candidates):
if x is not None:
return x
return newtoc
for item in entries:
level = indent_vals.index(item.indent)
parent = find_parent(level)
last_found[level] = parent.add_item('index.html', item.anchor,
item.text)
for i in range(level+1, len(last_found)):
last_found[i] = None
return newtoc
def _parse_toc(self, ul, basedir=os.getcwdu()):
toc = TOC(play_order=self._playorder, base_path=basedir, text='')
self._playorder += 1
for li in ul('li', recursive=False):
href = li.object('param', {'name': 'Local'})[0]['value']
if href.count('#'):
href, frag = href.split('#')
else:
frag = None
name = self._deentity(li.object('param', {'name': 'Name'})[0]['value'])
#print "========>", name
toc.add_item(href, frag, name, play_order=self._playorder)
self._playorder += 1
if li.ul:
child = self._parse_toc(li.ul)
child.parent = toc
toc.append(child)
#print toc
return toc
def _parse_toc(self, ul, basedir=getcwd()):
toc = TOC(play_order=self._playorder, base_path=basedir, text='')
self._playorder += 1
for li in ul('li', recursive=False):
href = li.object('param', {'name': 'Local'})[0]['value']
if href.count('#'):
href, frag = href.split('#')
else:
frag = None
name = self._deentity(li.object('param', {'name': 'Name'})[0]['value'])
# print "========>", name
toc.add_item(href, frag, name, play_order=self._playorder)
self._playorder += 1
if li.ul:
child = self._parse_toc(li.ul)
child.parent = toc
toc.append(child)
# print toc
return toc
def convert(self, stream, opts, file_ext, log, accelerators):
from calibre.ebooks.metadata import MetaInformation
from calibre.ebooks.metadata.opf2 import OPFCreator
from calibre.ebooks.metadata.toc import TOC
self.opts, self.log= opts, log
if file_ext == 'cbc':
comics_ = self.get_comics_from_collection(stream)
else:
comics_ = [['Comic', os.path.abspath(stream.name)]]
stream.close()
comics = []
for i, x in enumerate(comics_):
title, fname = x
cdir = u'comic_%d'%(i+1) if len(comics_) > 1 else u'.'
cdir = os.path.abspath(cdir)
if not os.path.exists(cdir):
os.makedirs(cdir)
pages = self.get_pages(fname, cdir)
if not pages: continue
wrappers = self.create_wrappers(pages)
comics.append((title, pages, wrappers))
if not comics:
raise ValueError('No comic pages found in %s'%stream.name)
mi = MetaInformation(os.path.basename(stream.name).rpartition('.')[0],
[_('Unknown')])
opf = OPFCreator(os.getcwdu(), mi)
entries = []
def href(x):
if len(comics) == 1: return os.path.basename(x)
return '/'.join(x.split(os.sep)[-2:])
for comic in comics:
pages, wrappers = comic[1:]
entries += [(w, None) for w in map(href, wrappers)] + \
[(x, None) for x in map(href, pages)]
opf.create_manifest(entries)
spine = []
for comic in comics:
spine.extend(map(href, comic[2]))
self._images = []
for comic in comics:
self._images.extend(comic[1])
opf.create_spine(spine)
toc = TOC()
if len(comics) == 1:
wrappers = comics[0][2]
for i, x in enumerate(wrappers):
toc.add_item(href(x), None, _('Page')+' %d'%(i+1),
play_order=i)
else:
po = 0
for comic in comics:
po += 1
wrappers = comic[2]
stoc = toc.add_item(href(wrappers[0]),
None, comic[0], play_order=po)
if not opts.dont_add_comic_pages_to_toc:
for i, x in enumerate(wrappers):
stoc.add_item(href(x), None,
_('Page')+' %d'%(i+1), play_order=po)
po += 1
opf.set_toc(toc)
m, n = open(u'metadata.opf', 'wb'), open('toc.ncx', 'wb')
opf.render(m, n, u'toc.ncx')
return os.path.abspath(u'metadata.opf')
def convert(self, stream, opts, file_ext, log, accelerators):
from calibre.ebooks.metadata import MetaInformation
from calibre.ebooks.metadata.opf2 import OPFCreator
from calibre.ebooks.metadata.toc import TOC
self.opts, self.log = opts, log
if file_ext == 'cbc':
comics_ = self.get_comics_from_collection(stream)
else:
comics_ = [['Comic', os.path.abspath(stream.name)]]
stream.close()
comics = []
num_pages_per_comic = []
for i, x in enumerate(comics_):
title, fname = x
cdir = 'comic_%d' % (i + 1) if len(comics_) > 1 else '.'
cdir = os.path.abspath(cdir)
if not os.path.exists(cdir):
os.makedirs(cdir)
pages = self.get_pages(fname, cdir)
if not pages:
continue
num_pages_per_comic.append(len(pages))
if self.for_viewer:
comics.append(
(title, pages, [self.create_viewer_wrapper(pages, cdir)]))
else:
wrappers = self.create_wrappers(pages)
comics.append((title, pages, wrappers))
if not comics:
raise ValueError('No comic pages found in %s' % stream.name)
mi = MetaInformation(
os.path.basename(stream.name).rpartition('.')[0], [_('Unknown')])
opf = OPFCreator(os.getcwd(), mi)
entries = []
def href(x):
if len(comics) == 1:
return os.path.basename(x)
return '/'.join(x.split(os.sep)[-2:])
cover_href = None
for comic in comics:
pages, wrappers = comic[1:]
page_entries = [(x, None) for x in map(href, pages)]
entries += [(w, None) for w in map(href, wrappers)] + page_entries
if cover_href is None and page_entries:
cover_href = page_entries[0][0]
opf.create_manifest(entries)
spine = []
for comic in comics:
spine.extend(map(href, comic[2]))
self._images = []
for comic in comics:
self._images.extend(comic[1])
opf.create_spine(spine)
if self.for_viewer and cover_href:
if os.path.isabs(cover_href):
cover_href = os.path.relpath(cover_href).replace(os.sep, '/')
opf.guide.set_cover(cover_href)
toc = TOC()
if len(comics) == 1:
wrappers = comics[0][2]
if self.for_viewer:
wrapper_page_href = href(wrappers[0])
for i in range(num_pages_per_comic[0]):
toc.add_item('{}#page_{}'.format(wrapper_page_href, i + 1),
None,
_('Page') + ' %d' % (i + 1),
play_order=i)
else:
for i, x in enumerate(wrappers):
toc.add_item(href(x),
None,
_('Page') + ' %d' % (i + 1),
play_order=i)
else:
po = 0
for num_pages, comic in zip(num_pages_per_comic, comics):
po += 1
wrappers = comic[2]
stoc = toc.add_item(href(wrappers[0]),
None,
comic[0],
play_order=po)
if not opts.dont_add_comic_pages_to_toc:
if self.for_viewer:
wrapper_page_href = href(wrappers[0])
for i in range(num_pages):
stoc.add_item('{}#page_{}'.format(
wrapper_page_href, i + 1),
None,
_('Page') + ' %d' % (i + 1),
play_order=po)
po += 1
else:
for i, x in enumerate(wrappers):
stoc.add_item(href(x),
None,
_('Page') + ' %d' % (i + 1),
play_order=po)
po += 1
opf.set_toc(toc)
with open('metadata.opf', 'wb') as m, open('toc.ncx', 'wb') as n:
opf.render(m, n, 'toc.ncx')
return os.path.abspath('metadata.opf')
def get_toc(self):
'''
Toc can have up to 5 levels, 0 - 4 inclusive.
This function will add items to their appropriate
depth in the TOC tree. If the specified depth is
invalid (item would not have a valid parent) add
it to the next valid level above the specified
level.
'''
# Base toc object all items will be added to.
n_toc = TOC()
# Used to track nodes in the toc so we can add
# sub items to the appropriate place in tree.
t_l0 = None
t_l1 = None
t_l2 = None
t_l3 = None
for level, (href, id, text) in self.toc:
if level == '0':
t_l0 = n_toc.add_item(href, id, text)
t_l1 = None
t_l2 = None
t_l3 = None
elif level == '1':
if t_l0 is None:
t_l0 = n_toc
t_l1 = t_l0.add_item(href, id, text)
t_l2 = None
t_l3 = None
elif level == '2':
if t_l1 is None:
if t_l0 is None:
t_l1 = n_toc
else:
t_l1 = t_l0
t_l2 = t_l1.add_item(href, id, text)
t_l3 = None
elif level == '3':
if t_l2 is None:
if t_l1 is None:
if t_l0 is None:
t_l2 = n_toc
else:
t_l2 = t_l0
else:
t_l2 = t_l1
t_l3 = t_l2.add_item(href, id, text)
# Level 4.
# Anything above 4 is invalid but we will count
# it as level 4.
else:
if t_l3 is None:
if t_l2 is None:
if t_l1 is None:
if t_l0 is None:
t_l3 = n_toc
else:
t_l3 = t_l0
else:
t_l3 = t_l1
else:
t_l3 = t_l2
t_l3.add_item(href, id, text)
return n_toc
def get_toc(self):
'''
Toc can have up to 5 levels, 0 - 4 inclusive.
This function will add items to their appropriate
depth in the TOC tree. If the specified depth is
invalid (item would not have a valid parent) add
it to the next valid level above the specified
level.
'''
# Base toc object all items will be added to.
n_toc = TOC()
# Used to track nodes in the toc so we can add
# sub items to the appropriate place in tree.
t_l0 = None
t_l1 = None
t_l2 = None
t_l3 = None
for level, (href, id, text) in self.toc:
if level == u'0':
t_l0 = n_toc.add_item(href, id, text)
t_l1 = None
t_l2 = None
t_l3 = None
elif level == u'1':
if t_l0 is None:
t_l0 = n_toc
t_l1 = t_l0.add_item(href, id, text)
t_l2 = None
t_l3 = None
elif level == u'2':
if t_l1 is None:
if t_l0 is None:
t_l1 = n_toc
else:
t_l1 = t_l0
t_l2 = t_l1.add_item(href, id, text)
t_l3 = None
elif level == u'3':
if t_l2 is None:
if t_l1 is None:
if t_l0 is None:
t_l2 = n_toc
else:
t_l2 = t_l0
else:
t_l2 = t_l1
t_l3 = t_l2.add_item(href, id, text)
# Level 4.
# Anything above 4 is invalid but we will count
# it as level 4.
else:
if t_l3 is None:
if t_l2 is None:
if t_l1 is None:
if t_l0 is None:
t_l3 = n_toc
else:
t_l3 = t_l0
else:
t_l3 = t_l1
else:
t_l3 = t_l2
t_l3.add_item(href, id, text)
return n_toc
def create_opf(self, htmlfile, guide=None, root=None):
mi = getattr(self.book_header.exth, 'mi', self.embedded_mi)
if mi is None:
mi = MetaInformation(self.book_header.title, [_('Unknown')])
opf = OPFCreator(os.path.dirname(htmlfile), mi)
if hasattr(self.book_header.exth, 'cover_offset'):
opf.cover = u'images/%05d.jpg' % (self.book_header.exth.cover_offset + 1)
elif mi.cover is not None:
opf.cover = mi.cover
else:
opf.cover = u'images/%05d.jpg' % 1
if not os.path.exists(os.path.join(os.path.dirname(htmlfile),
* opf.cover.split('/'))):
opf.cover = None
cover = opf.cover
cover_copied = None
if cover is not None:
cover = cover.replace('/', os.sep)
if os.path.exists(cover):
ncover = u'images'+os.sep+u'calibre_cover.jpg'
if os.path.exists(ncover):
os.remove(ncover)
shutil.copyfile(cover, ncover)
cover_copied = os.path.abspath(ncover)
opf.cover = ncover.replace(os.sep, '/')
manifest = [(htmlfile, 'application/xhtml+xml'),
(os.path.abspath(u'styles.css'), 'text/css')]
bp = os.path.dirname(htmlfile)
added = set([])
for i in getattr(self, 'image_names', []):
path = os.path.join(bp, 'images', i)
added.add(path)
manifest.append((path, 'image/jpeg'))
if cover_copied is not None:
manifest.append((cover_copied, 'image/jpeg'))
opf.create_manifest(manifest)
opf.create_spine([os.path.basename(htmlfile)])
toc = None
if guide is not None:
opf.create_guide(guide)
for ref in opf.guide:
if ref.type.lower() == 'toc':
toc = ref.href()
ncx_manifest_entry = None
if toc:
ncx_manifest_entry = 'toc.ncx'
elems = root.xpath('//*[@id="%s"]' % toc.partition('#')[-1])
tocobj = None
ent_pat = re.compile(r'&(\S+?);')
if elems:
tocobj = TOC()
found = False
reached = False
for x in root.iter():
if x == elems[-1]:
reached = True
continue
if reached and x.tag == 'a':
href = x.get('href', '')
if href and re.match('\\w+://', href) is None:
try:
text = u' '.join([t.strip() for t in
x.xpath('descendant::text()')])
except:
text = ''
text = ent_pat.sub(entity_to_unicode, text)
item = tocobj.add_item(toc.partition('#')[0], href[1:],
text)
item.left_space = int(self.get_left_whitespace(x))
found = True
if reached and found and x.get('class', None) == 'mbp_pagebreak':
break
if tocobj is not None:
tocobj = self.structure_toc(tocobj)
opf.set_toc(tocobj)
return opf, ncx_manifest_entry
def create_opf(self, htmlfile, guide=None, root=None):
mi = getattr(self.book_header.exth, 'mi', self.embedded_mi)
if mi is None:
mi = MetaInformation(self.book_header.title, [_('Unknown')])
opf = OPFCreator(os.path.dirname(htmlfile), mi)
if hasattr(self.book_header.exth, 'cover_offset'):
opf.cover = u'images/%05d.jpg' % (self.book_header.exth.cover_offset + 1)
elif mi.cover is not None:
opf.cover = mi.cover
else:
opf.cover = u'images/%05d.jpg' % 1
if not os.path.exists(os.path.join(os.path.dirname(htmlfile),
* opf.cover.split('/'))):
opf.cover = None
cover = opf.cover
cover_copied = None
if cover is not None:
cover = cover.replace('/', os.sep)
if os.path.exists(cover):
ncover = u'images'+os.sep+u'calibre_cover.jpg'
if os.path.exists(ncover):
os.remove(ncover)
shutil.copyfile(cover, ncover)
cover_copied = os.path.abspath(ncover)
opf.cover = ncover.replace(os.sep, '/')
manifest = [(htmlfile, 'application/xhtml+xml'),
(os.path.abspath(u'styles.css'), 'text/css')]
bp = os.path.dirname(htmlfile)
added = set([])
for i in getattr(self, 'image_names', []):
path = os.path.join(bp, 'images', i)
added.add(path)
manifest.append((path, 'image/jpeg'))
if cover_copied is not None:
manifest.append((cover_copied, 'image/jpeg'))
opf.create_manifest(manifest)
opf.create_spine([os.path.basename(htmlfile)])
toc = None
if guide is not None:
opf.create_guide(guide)
for ref in opf.guide:
if ref.type.lower() == 'toc':
toc = ref.href()
ncx_manifest_entry = None
if toc:
ncx_manifest_entry = 'toc.ncx'
elems = root.xpath('//*[@id="%s"]' % toc.partition('#')[-1])
tocobj = None
ent_pat = re.compile(r'&(\S+?);')
if elems:
tocobj = TOC()
found = False
reached = False
for x in root.iter():
if x == elems[-1]:
reached = True
continue
if reached and x.tag == 'a':
href = x.get('href', '')
if href and re.match('\w+://', href) is None:
try:
text = u' '.join([t.strip() for t in
x.xpath('descendant::text()')])
except:
text = ''
text = ent_pat.sub(entity_to_unicode, text)
item = tocobj.add_item(toc.partition('#')[0], href[1:],
text)
item.left_space = int(self.get_left_whitespace(x))
found = True
if reached and found and x.get('class', None) == 'mbp_pagebreak':
break
if tocobj is not None:
tocobj = self.structure_toc(tocobj)
opf.set_toc(tocobj)
return opf, ncx_manifest_entry
def convert(self, stream, opts, file_ext, log, accelerators):
from calibre.ebooks.metadata import MetaInformation
from calibre.ebooks.metadata.opf2 import OPFCreator
from calibre.ebooks.metadata.toc import TOC
self.opts, self.log = opts, log
if file_ext == 'cbc':
comics_ = self.get_comics_from_collection(stream)
else:
comics_ = [['Comic', os.path.abspath(stream.name)]]
stream.close()
comics = []
for i, x in enumerate(comics_):
title, fname = x
cdir = u'comic_%d' % (i + 1) if len(comics_) > 1 else u'.'
cdir = os.path.abspath(cdir)
if not os.path.exists(cdir):
os.makedirs(cdir)
pages = self.get_pages(fname, cdir)
if not pages:
continue
wrappers = self.create_wrappers(pages)
comics.append((title, pages, wrappers))
if not comics:
raise ValueError('No comic pages found in %s' % stream.name)
mi = MetaInformation(
os.path.basename(stream.name).rpartition('.')[0], [_('Unknown')])
opf = OPFCreator(os.getcwdu(), mi)
entries = []
def href(x):
if len(comics) == 1:
return os.path.basename(x)
return '/'.join(x.split(os.sep)[-2:])
for comic in comics:
pages, wrappers = comic[1:]
entries += [(w, None) for w in map(href, wrappers)] + \
[(x, None) for x in map(href, pages)]
opf.create_manifest(entries)
spine = []
for comic in comics:
spine.extend(map(href, comic[2]))
self._images = []
for comic in comics:
self._images.extend(comic[1])
opf.create_spine(spine)
toc = TOC()
if len(comics) == 1:
wrappers = comics[0][2]
for i, x in enumerate(wrappers):
toc.add_item(href(x),
None,
_('Page') + ' %d' % (i + 1),
play_order=i)
else:
po = 0
for comic in comics:
po += 1
wrappers = comic[2]
stoc = toc.add_item(href(wrappers[0]),
None,
comic[0],
play_order=po)
if not opts.dont_add_comic_pages_to_toc:
for i, x in enumerate(wrappers):
stoc.add_item(href(x),
None,
_('Page') + ' %d' % (i + 1),
play_order=po)
po += 1
opf.set_toc(toc)
m, n = open(u'metadata.opf', 'wb'), open('toc.ncx', 'wb')
opf.render(m, n, u'toc.ncx')
return os.path.abspath(u'metadata.opf')
