def create_inline_toc(container, title=None):
'''
Create an inline (HTML) Table of Contents from an existing NCX table of contents.
:param title: The title for this table of contents.
'''
lang = get_book_language(container)
default_title = 'Table of Contents'
if lang:
lang = lang_as_iso639_1(lang) or lang
default_title = translate(lang, default_title)
title = title or default_title
toc = get_toc(container)
if len(toc) == 0:
return None
toc_name = find_inline_toc(container)
name = toc_name
html = toc_to_html(toc, container, name, title, lang)
raw = serialize(html, 'text/html')
if name is None:
name, c = 'toc.xhtml', 0
while container.has_name(name):
c += 1
name = 'toc%d.xhtml' % c
container.add_file(name, raw, spine_index=0)
else:
with container.open(name, 'wb') as f:
f.write(raw)
set_guide_item(container,
'toc',
title,
name,
frag='calibre_generated_inline_toc')
return name
def create_inline_toc(container, title=None):
title = title or _('Table of Contents')
toc = get_toc(container)
if len(toc) == 0:
return None
toc_name = find_inline_toc(container)
def process_node(html_parent, toc, level=1, indent=' '):
li = html_parent.makeelement(XHTML('li'))
li.tail = '\n' + (indent * level)
html_parent.append(li)
name, frag = toc.dest, toc.frag
href = '#'
if name:
href = container.name_to_href(name, toc_name)
if frag:
href += '#' + frag
a = li.makeelement(XHTML('a'), href=href)
a.text = toc.title
li.append(a)
if len(toc) > 0:
parent = li.makeelement(XHTML('ul'))
li.append(parent)
a.tail = '\n\n' + (indent * (level + 2))
parent.text = '\n' + (indent * (level + 3))
parent.tail = '\n\n' + (indent * (level + 1))
for child in toc:
process_node(parent, child, level + 3)
parent[-1].tail = '\n' + (indent * (level + 2))
E = ElementMaker(namespace=XHTML_NS, nsmap={None: XHTML_NS})
html = E.html(
E.head(
E.title(title),
E.style('''
li { list-style-type: none; padding-left: 2em; margin-left: 0}
a { text-decoration: none }
a:hover { color: red }''',
type='text/css'),
), E.body(
E.h2(title),
E.ul(),
id="calibre_generated_inline_toc",
))
name = toc_name
for child in toc:
process_node(html[1][1], child)
pretty_html_tree(container, html)
raw = serialize(html, 'text/html')
if name is None:
name, c = 'toc.xhtml', 0
while container.has_name(name):
c += 1
name = 'toc%d.xhtml' % c
container.add_file(name, raw, spine_index=0)
else:
with container.open(name, 'wb') as f:
f.write(raw)
return name
def pretty_xml(container, name, raw):
' Pretty print the XML represented as a string in raw. If ``name`` is the name of the OPF, extra OPF-specific prettying is performed. '
root = container.parse_xml(raw)
if name == container.opf_name:
pretty_opf(root)
pretty_xml_tree(root)
return serialize(root, 'text/xml')
def pretty_xml(container, name, raw):
' Pretty print the XML represented as a string in raw. If ``name`` is the name of the OPF, extra OPF-specific prettying is performed. '
root = container.parse_xml(raw)
if name == container.opf_name:
pretty_opf(root)
pretty_xml_tree(root)
return serialize(root, 'text/xml')
def beautify_text(raw, syntax):
from lxml import etree
from calibre.ebooks.oeb.polish.parsing import parse
from calibre.ebooks.oeb.polish.pretty import pretty_xml_tree, pretty_html_tree
from calibre.ebooks.chardet import strip_encoding_declarations
if syntax == 'xml':
root = etree.fromstring(strip_encoding_declarations(raw))
pretty_xml_tree(root)
elif syntax == 'css':
import logging
from calibre.ebooks.oeb.base import serialize, _css_logger
from calibre.ebooks.oeb.polish.utils import setup_css_parser_serialization
from css_parser import CSSParser, log
setup_css_parser_serialization(tprefs['editor_tab_stop_width'])
log.setLevel(logging.WARN)
log.raiseExceptions = False
parser = CSSParser(loglevel=logging.WARNING,
# We dont care about @import rules
fetcher=lambda x: (None, None), log=_css_logger)
data = parser.parseString(raw, href='<string>', validate=False)
return serialize(data, 'text/css')
else:
root = parse(raw, line_numbers=False)
pretty_html_tree(None, root)
return etree.tostring(root, encoding=unicode)
def create_book(mi, path, fmt='epub', opf_name='metadata.opf', html_name='start.xhtml', toc_name='toc.ncx'):
''' Create an empty book in the specified format at the specified location. '''
path = os.path.abspath(path)
lang = 'und'
opf = metadata_to_opf(mi, as_string=False)
for l in opf.xpath('//*[local-name()="language"]'):
if l.text:
lang = l.text
break
lang = lang_as_iso639_1(lang) or lang
opfns = OPF_NAMESPACES['opf']
m = opf.makeelement('{%s}manifest' % opfns)
opf.insert(1, m)
i = m.makeelement('{%s}item' % opfns, href=html_name, id='start')
i.set('media-type', guess_type('a.xhtml'))
m.append(i)
i = m.makeelement('{%s}item' % opfns, href=toc_name, id='ncx')
i.set('media-type', guess_type(toc_name))
m.append(i)
s = opf.makeelement('{%s}spine' % opfns, toc="ncx")
opf.insert(2, s)
i = s.makeelement('{%s}itemref' % opfns, idref='start')
s.append(i)
CONTAINER = '''\
<?xml version="1.0"?>
<container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">
<rootfiles>
<rootfile full-path="{0}" media-type="application/oebps-package+xml"/>
</rootfiles>
</container>
'''.format(prepare_string_for_xml(opf_name, True)).encode('utf-8')
HTML = P('templates/new_book.html', data=True).decode('utf-8').replace(
'_LANGUAGE_', prepare_string_for_xml(lang, True)
).replace(
'_TITLE_', prepare_string_for_xml(mi.title)
).replace(
'_AUTHORS_', prepare_string_for_xml(authors_to_string(mi.authors))
).encode('utf-8')
h = parse(HTML)
pretty_html_tree(None, h)
HTML = serialize(h, 'text/html')
ncx = etree.tostring(create_toc(mi, opf, html_name, lang), encoding='utf-8', xml_declaration=True, pretty_print=True)
pretty_xml_tree(opf)
opf = etree.tostring(opf, encoding='utf-8', xml_declaration=True, pretty_print=True)
if fmt == 'azw3':
with TemporaryDirectory('create-azw3') as tdir, CurrentDir(tdir):
for name, data in ((opf_name, opf), (html_name, HTML), (toc_name, ncx)):
with open(name, 'wb') as f:
f.write(data)
c = Container(os.path.dirname(os.path.abspath(opf_name)), opf_name, DevNull())
opf_to_azw3(opf_name, path, c)
else:
with ZipFile(path, 'w', compression=ZIP_STORED) as zf:
zf.writestr('mimetype', b'application/epub+zip', compression=ZIP_STORED)
zf.writestr('META-INF/', b'', 0755)
zf.writestr('META-INF/container.xml', CONTAINER)
zf.writestr(opf_name, opf)
zf.writestr(html_name, HTML)
zf.writestr(toc_name, ncx)
def beautify_text(raw, syntax):
from lxml import etree
from calibre.ebooks.oeb.polish.parsing import parse
from calibre.ebooks.oeb.polish.pretty import pretty_xml_tree, pretty_html_tree
from calibre.ebooks.chardet import strip_encoding_declarations
if syntax == 'xml':
root = etree.fromstring(strip_encoding_declarations(raw))
pretty_xml_tree(root)
elif syntax == 'css':
import logging
from calibre.ebooks.oeb.base import serialize, _css_logger
from calibre.ebooks.oeb.polish.utils import setup_cssutils_serialization
from cssutils import CSSParser, log
setup_cssutils_serialization(tprefs['editor_tab_stop_width'])
log.setLevel(logging.WARN)
log.raiseExceptions = False
parser = CSSParser(loglevel=logging.WARNING,
# We dont care about @import rules
fetcher=lambda x: (None, None), log=_css_logger)
data = parser.parseString(raw, href='<string>', validate=False)
return serialize(data, 'text/css')
else:
root = parse(raw, line_numbers=False)
pretty_html_tree(None, root)
return etree.tostring(root, encoding=unicode)
def create_inline_toc(container, title=None):
'''
Create an inline (HTML) Table of Contents from an existing NCX table of contents.
:param title: The title for this table of contents.
'''
lang = get_book_language(container)
default_title = 'Table of Contents'
if lang:
lang = lang_as_iso639_1(lang) or lang
default_title = translate(lang, default_title)
title = title or default_title
toc = get_toc(container)
if len(toc) == 0:
return None
toc_name = find_inline_toc(container)
name = toc_name
html = toc_to_html(toc, container, name, title, lang)
raw = serialize(html, 'text/html')
if name is None:
name, c = 'toc.xhtml', 0
while container.has_name(name):
c += 1
name = 'toc%d.xhtml' % c
container.add_file(name, raw, spine_index=0)
else:
with container.open(name, 'wb') as f:
f.write(raw)
set_guide_item(container, 'toc', title, name, frag='calibre_generated_inline_toc')
return name
def parse_html(raw):
root = parse(raw,
decoder=lambda x: x.decode('utf-8'),
line_numbers=True,
linenumber_attribute='data-lnum')
ans = serialize(root, 'text/html')
if not isinstance(ans, bytes):
ans = ans.encode('utf-8')
return ans
def serialize_item(self, name):
data = self.parsed(name)
if name == self.opf_name:
self.format_opf()
data = serialize(data, self.mime_map[name], pretty_print=name in self.pretty_print)
if name == self.opf_name:
# Needed as I can't get lxml to output opf:role and
# not output <opf:metadata> as well
data = re.sub(br"(<[/]{0,1})opf:", r"\1", data)
return data
def serialize_item(self, name):
data = self.parsed(name)
if name == self.opf_name:
self.format_opf()
data = serialize(data, self.mime_map[name], pretty_print=name in
self.pretty_print)
if name == self.opf_name:
# Needed as I can't get lxml to output opf:role and
# not output <opf:metadata> as well
data = re.sub(br'(<[/]{0,1})opf:', r'\1', data)
return data
def parse_html(raw):
root = parse(raw, decoder=lambda x:x.decode('utf-8'), line_numbers=True, linenumber_attribute='data-lnum')
ans = serialize(root, 'text/html')
if not isinstance(ans, bytes):
ans = ans.encode('utf-8')
return ans
def convert_epub3_nav(self, nav_path, opf, log, opts):
from lxml import etree
from calibre.ebooks.chardet import xml_to_unicode
from calibre.ebooks.oeb.polish.parsing import parse
from calibre.ebooks.oeb.base import EPUB_NS, XHTML, NCX_MIME, NCX, urlnormalize, urlunquote, serialize
from calibre.ebooks.oeb.polish.toc import first_child
from tempfile import NamedTemporaryFile
with lopen(nav_path, 'rb') as f:
raw = f.read()
raw = xml_to_unicode(raw, strip_encoding_pats=True,
assume_utf8=True)[0]
root = parse(raw, log=log)
ncx = etree.fromstring(
'<ncx xmlns="http://www.daisy.org/z3986/2005/ncx/" version="2005-1" xml:lang="eng"><navMap/></ncx>'
)
navmap = ncx[0]
et = '{%s}type' % EPUB_NS
bn = os.path.basename(nav_path)
def add_from_li(li, parent):
href = text = None
for x in li.iterchildren(XHTML('a'), XHTML('span')):
text = etree.tostring(
x, method='text', encoding=unicode,
with_tail=False).strip() or ' '.join(
x.xpath('descendant-or-self::*/@title')).strip()
href = x.get('href')
if href:
if href.startswith('#'):
href = bn + href
break
np = parent.makeelement(NCX('navPoint'))
parent.append(np)
np.append(np.makeelement(NCX('navLabel')))
np[0].append(np.makeelement(NCX('text')))
np[0][0].text = text
if href:
np.append(np.makeelement(NCX('content'), attrib={'src': href}))
return np
def process_nav_node(node, toc_parent):
for li in node.iterchildren(XHTML('li')):
child = add_from_li(li, toc_parent)
ol = first_child(li, XHTML('ol'))
if child is not None and ol is not None:
process_nav_node(ol, child)
for nav in root.iterdescendants(XHTML('nav')):
if nav.get(et) == 'toc':
ol = first_child(nav, XHTML('ol'))
if ol is not None:
process_nav_node(ol, navmap)
break
else:
return
with NamedTemporaryFile(suffix='.ncx',
dir=os.path.dirname(nav_path),
delete=False) as f:
f.write(etree.tostring(ncx, encoding='utf-8'))
ncx_href = os.path.relpath(f.name, os.getcwdu()).replace(os.sep, '/')
ncx_id = opf.create_manifest_item(ncx_href, NCX_MIME,
append=True).get('id')
for spine in opf.root.xpath('//*[local-name()="spine"]'):
spine.set('toc', ncx_id)
opts.epub3_nav_href = urlnormalize(
os.path.relpath(nav_path).replace(os.sep, '/'))
opts.epub3_nav_parsed = root
if getattr(self, 'removed_cover', None):
changed = False
base_path = os.path.dirname(nav_path)
for elem in root.xpath('//*[@href]'):
href, frag = elem.get('href').partition('#')[::2]
link_path = os.path.relpath(
os.path.join(base_path, urlunquote(href)), base_path)
abs_href = urlnormalize(link_path)
if abs_href == self.removed_cover:
changed = True
elem.set('data-calibre-removed-titlepage', '1')
if changed:
with open(nav_path, 'wb') as f:
f.write(serialize(root, 'application/xhtml+xml'))
def create_inline_toc(container, title=None):
lang = get_book_language(container)
default_title = 'Table of Contents'
if lang:
lang = lang_as_iso639_1(lang) or lang
default_title = translate(lang, default_title)
title = title or default_title
toc = get_toc(container)
if len(toc) == 0:
return None
toc_name = find_inline_toc(container)
def process_node(html_parent, toc, level=1, indent=' ', style_level=2):
li = html_parent.makeelement(XHTML('li'))
li.tail = '\n'+ (indent*level)
html_parent.append(li)
name, frag = toc.dest, toc.frag
href = '#'
if name:
href = container.name_to_href(name, toc_name)
if frag:
href += '#' + frag
a = li.makeelement(XHTML('a'), href=href)
a.text = toc.title
li.append(a)
if len(toc) > 0:
parent = li.makeelement(XHTML('ul'))
parent.set('class', 'level%d' % (style_level))
li.append(parent)
a.tail = '\n\n' + (indent*(level+2))
parent.text = '\n'+(indent*(level+3))
parent.tail = '\n\n' + (indent*(level+1))
for child in toc:
process_node(parent, child, level+3, style_level=style_level + 1)
parent[-1].tail = '\n' + (indent*(level+2))
E = ElementMaker(namespace=XHTML_NS, nsmap={None:XHTML_NS})
html = E.html(
E.head(
E.title(title),
E.style('''
li { list-style-type: none; padding-left: 2em; margin-left: 0}
a { text-decoration: none }
a:hover { color: red }''', type='text/css'),
),
E.body(
E.h2(title),
E.ul(),
id="calibre_generated_inline_toc",
)
)
name = toc_name
ul = html[1][1]
ul.set('class', 'level1')
for child in toc:
process_node(ul, child)
if lang:
html.set('lang', lang)
pretty_html_tree(container, html)
raw = serialize(html, 'text/html')
if name is None:
name, c = 'toc.xhtml', 0
while container.has_name(name):
c += 1
name = 'toc%d.xhtml' % c
container.add_file(name, raw, spine_index=0)
else:
with container.open(name, 'wb') as f:
f.write(raw)
set_guide_item(container, 'toc', title, name, frag='calibre_generated_inline_toc')
return name
def pretty_css(container, name, raw):
' Pretty print the CSS represented as a string in raw '
sheet = container.parse_css(raw)
return serialize(sheet, 'text/css')
def pretty_html(container, name, raw):
root = container.parse_xhtml(raw)
pretty_html_tree(container, root)
return serialize(root, 'text/html')
def pretty_xml(container, name, raw):
root = container.parse_xml(raw)
if name == container.opf_name:
pretty_opf(root)
pretty_xml_tree(root)
return serialize(root, 'text/xml')
def create_book(mi,
path,
fmt='epub',
opf_name='metadata.opf',
html_name='start.xhtml',
toc_name='toc.ncx'):
''' Create an empty book in the specified format at the specified location. '''
if fmt not in valid_empty_formats:
raise ValueError('Cannot create empty book in the %s format' % fmt)
if fmt == 'txt':
with open(path, 'wb') as f:
if not mi.is_null('title'):
f.write(as_bytes(mi.title))
return
if fmt == 'docx':
from calibre.ebooks.conversion.plumber import Plumber
from calibre.ebooks.docx.writer.container import DOCX
from calibre.utils.logging import default_log
p = Plumber('a.docx', 'b.docx', default_log)
p.setup_options()
# Use the word default of one inch page margins
for x in 'left right top bottom'.split():
setattr(p.opts, 'margin_' + x, 72)
DOCX(p.opts, default_log).write(path, mi, create_empty_document=True)
return
path = os.path.abspath(path)
lang = 'und'
opf = metadata_to_opf(mi, as_string=False)
for l in opf.xpath('//*[local-name()="language"]'):
if l.text:
lang = l.text
break
lang = lang_as_iso639_1(lang) or lang
opfns = OPF_NAMESPACES['opf']
m = opf.makeelement('{%s}manifest' % opfns)
opf.insert(1, m)
i = m.makeelement('{%s}item' % opfns, href=html_name, id='start')
i.set('media-type', guess_type('a.xhtml'))
m.append(i)
i = m.makeelement('{%s}item' % opfns, href=toc_name, id='ncx')
i.set('media-type', guess_type(toc_name))
m.append(i)
s = opf.makeelement('{%s}spine' % opfns, toc="ncx")
opf.insert(2, s)
i = s.makeelement('{%s}itemref' % opfns, idref='start')
s.append(i)
CONTAINER = '''\
<?xml version="1.0"?>
<container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">
<rootfiles>
<rootfile full-path="{0}" media-type="application/oebps-package+xml"/>
</rootfiles>
</container>
'''.format(prepare_string_for_xml(opf_name, True)).encode('utf-8')
HTML = P('templates/new_book.html', data=True).decode('utf-8').replace(
'_LANGUAGE_', prepare_string_for_xml(lang, True)).replace(
'_TITLE_', prepare_string_for_xml(mi.title)).replace(
'_AUTHORS_',
prepare_string_for_xml(authors_to_string(
mi.authors))).encode('utf-8')
h = parse(HTML)
pretty_html_tree(None, h)
HTML = serialize(h, 'text/html')
ncx = etree.tostring(create_toc(mi, opf, html_name, lang),
encoding='utf-8',
xml_declaration=True,
pretty_print=True)
pretty_xml_tree(opf)
opf = etree.tostring(opf,
encoding='utf-8',
xml_declaration=True,
pretty_print=True)
if fmt == 'azw3':
with TemporaryDirectory('create-azw3') as tdir, CurrentDir(tdir):
for name, data in ((opf_name, opf), (html_name, HTML), (toc_name,
ncx)):
with open(name, 'wb') as f:
f.write(data)
c = Container(os.path.dirname(os.path.abspath(opf_name)), opf_name,
DevNull())
opf_to_azw3(opf_name, path, c)
else:
with ZipFile(path, 'w', compression=ZIP_STORED) as zf:
zf.writestr('mimetype',
b'application/epub+zip',
compression=ZIP_STORED)
zf.writestr('META-INF/', b'', 0o755)
zf.writestr('META-INF/container.xml', CONTAINER)
zf.writestr(opf_name, opf)
zf.writestr(html_name, HTML)
zf.writestr(toc_name, ncx)
def fix_html(container, raw):
' Fix any parsing errors in the HTML represented as a string in raw. Fixing is done using the HTML5 parsing algorithm. '
root = container.parse_xhtml(raw)
return serialize(root, 'text/html')
def pretty_css(container, name, raw):
sheet = container.parse_css(raw)
return serialize(sheet, 'text/css')
def pretty_xml(container, name, raw):
root = container.parse_xml(raw)
if name == container.opf_name:
pretty_opf(root)
pretty_xml_tree(root)
return serialize(root, 'text/xml')
def pretty_html(container, name, raw):
root = container.parse_xhtml(raw)
pretty_html_tree(container, root)
return serialize(root, 'text/html')
def fix_html(container, raw):
root = container.parse_xhtml(raw)
return serialize(root, 'text/html')
def create_inline_toc(container, title=None):
'''
Create an inline (HTML) Table of Contents from an existing NCX table of contents.
:param title: The title for this table of contents.
'''
lang = get_book_language(container)
default_title = 'Table of Contents'
if lang:
lang = lang_as_iso639_1(lang) or lang
default_title = translate(lang, default_title)
title = title or default_title
toc = get_toc(container)
if len(toc) == 0:
return None
toc_name = find_inline_toc(container)
def process_node(html_parent, toc, level=1, indent=' ', style_level=2):
li = html_parent.makeelement(XHTML('li'))
li.tail = '\n' + (indent * level)
html_parent.append(li)
name, frag = toc.dest, toc.frag
href = '#'
if name:
href = container.name_to_href(name, toc_name)
if frag:
href += '#' + frag
a = li.makeelement(XHTML('a'), href=href)
a.text = toc.title
li.append(a)
if len(toc) > 0:
parent = li.makeelement(XHTML('ul'))
parent.set('class', 'level%d' % (style_level))
li.append(parent)
a.tail = '\n\n' + (indent * (level + 2))
parent.text = '\n' + (indent * (level + 3))
parent.tail = '\n\n' + (indent * (level + 1))
for child in toc:
process_node(parent,
child,
level + 3,
style_level=style_level + 1)
parent[-1].tail = '\n' + (indent * (level + 2))
E = ElementMaker(namespace=XHTML_NS, nsmap={None: XHTML_NS})
html = E.html(
E.head(
E.title(title),
E.style('''
li { list-style-type: none; padding-left: 2em; margin-left: 0}
a { text-decoration: none }
a:hover { color: red }''',
type='text/css'),
), E.body(
E.h2(title),
E.ul(),
id="calibre_generated_inline_toc",
))
name = toc_name
ul = html[1][1]
ul.set('class', 'level1')
for child in toc:
process_node(ul, child)
if lang:
html.set('lang', lang)
pretty_html_tree(container, html)
raw = serialize(html, 'text/html')
if name is None:
name, c = 'toc.xhtml', 0
while container.has_name(name):
c += 1
name = 'toc%d.xhtml' % c
container.add_file(name, raw, spine_index=0)
else:
with container.open(name, 'wb') as f:
f.write(raw)
set_guide_item(container,
'toc',
title,
name,
frag='calibre_generated_inline_toc')
return name
def pretty_css(container, name, raw):
' Pretty print the CSS represented as a string in raw '
sheet = container.parse_css(raw)
return serialize(sheet, 'text/css')
def parse_html(raw):
root = parse(raw, decoder=lambda x: x.decode("utf-8"), line_numbers=True, linenumber_attribute="data-lnum")
return serialize(root, "text/html").encode("utf-8")
def pretty_html(container, name, raw):
' Pretty print the HTML represented as a string in raw '
root = container.parse_xhtml(raw)
pretty_html_tree(container, root)
return serialize(root, 'text/html')
def convert_epub3_nav(self, nav_path, opf, log, opts):
from lxml import etree
from calibre.ebooks.chardet import xml_to_unicode
from calibre.ebooks.oeb.polish.parsing import parse
from calibre.ebooks.oeb.base import EPUB_NS, XHTML, NCX_MIME, NCX, urlnormalize, urlunquote, serialize
from calibre.ebooks.oeb.polish.toc import first_child
from tempfile import NamedTemporaryFile
with lopen(nav_path, 'rb') as f:
raw = f.read()
raw = xml_to_unicode(raw, strip_encoding_pats=True, assume_utf8=True)[0]
root = parse(raw, log=log)
ncx = etree.fromstring('<ncx xmlns="http://www.daisy.org/z3986/2005/ncx/" version="2005-1" xml:lang="eng"><navMap/></ncx>')
navmap = ncx[0]
et = '{%s}type' % EPUB_NS
bn = os.path.basename(nav_path)
def add_from_li(li, parent):
href = text = None
for x in li.iterchildren(XHTML('a'), XHTML('span')):
text = etree.tostring(x, method='text', encoding=unicode, with_tail=False).strip() or ' '.join(x.xpath('descendant-or-self::*/@title')).strip()
href = x.get('href')
if href:
if href.startswith('#'):
href = bn + href
break
np = parent.makeelement(NCX('navPoint'))
parent.append(np)
np.append(np.makeelement(NCX('navLabel')))
np[0].append(np.makeelement(NCX('text')))
np[0][0].text = text
if href:
np.append(np.makeelement(NCX('content'), attrib={'src':href}))
return np
def process_nav_node(node, toc_parent):
for li in node.iterchildren(XHTML('li')):
child = add_from_li(li, toc_parent)
ol = first_child(li, XHTML('ol'))
if child is not None and ol is not None:
process_nav_node(ol, child)
for nav in root.iterdescendants(XHTML('nav')):
if nav.get(et) == 'toc':
ol = first_child(nav, XHTML('ol'))
if ol is not None:
process_nav_node(ol, navmap)
break
else:
return
with NamedTemporaryFile(suffix='.ncx', dir=os.path.dirname(nav_path), delete=False) as f:
f.write(etree.tostring(ncx, encoding='utf-8'))
ncx_href = os.path.relpath(f.name, os.getcwdu()).replace(os.sep, '/')
ncx_id = opf.create_manifest_item(ncx_href, NCX_MIME, append=True).get('id')
for spine in opf.root.xpath('//*[local-name()="spine"]'):
spine.set('toc', ncx_id)
opts.epub3_nav_href = urlnormalize(os.path.relpath(nav_path).replace(os.sep, '/'))
opts.epub3_nav_parsed = root
if getattr(self, 'removed_cover', None):
changed = False
base_path = os.path.dirname(nav_path)
for elem in root.xpath('//*[@href]'):
href, frag = elem.get('href').partition('#')[::2]
link_path = os.path.relpath(os.path.join(base_path, urlunquote(href)), base_path)
abs_href = urlnormalize(link_path)
if abs_href == self.removed_cover:
changed = True
elem.set('data-calibre-removed-titlepage', '1')
if changed:
with open(nav_path, 'wb') as f:
f.write(serialize(root, 'application/xhtml+xml'))
def fix_html(container, raw):
root = container.parse_xhtml(raw)
return serialize(root, 'text/html')
def fix_html(container, raw):
' Fix any parsing errors in the HTML represented as a string in raw. Fixing is done using the HTML5 parsing algorithm. '
root = container.parse_xhtml(raw)
return serialize(root, 'text/html')
def pretty_css(container, name, raw):
sheet = container.parse_css(raw)
return serialize(sheet, 'text/css')
def pretty_html(container, name, raw):
' Pretty print the HTML represented as a string in raw '
root = container.parse_xhtml(raw)
pretty_html_tree(container, root)
return serialize(root, 'text/html')
def fix_html(raw):
root = parse(raw)
return serialize(root, 'text/html').decode('utf-8')
def create_book(mi, path, fmt='epub', opf_name='metadata.opf', html_name='start.xhtml', toc_name='toc.ncx'):
''' Create an empty book in the specified format at the specified location. '''
if fmt not in valid_empty_formats:
raise ValueError('Cannot create empty book in the %s format' % fmt)
if fmt == 'txt':
with open(path, 'wb') as f:
if not mi.is_null('title'):
f.write(mi.title)
return
if fmt == 'docx':
from calibre.ebooks.conversion.plumber import Plumber
from calibre.ebooks.docx.writer.container import DOCX
from calibre.utils.logging import default_log
p = Plumber('a.docx', 'b.docx', default_log)
p.setup_options()
# Use the word default of one inch page margins
for x in 'left right top bottom'.split():
setattr(p.opts, 'margin_' + x, 72)
DOCX(p.opts, default_log).write(path, mi, create_empty_document=True)
return
path = os.path.abspath(path)
lang = 'und'
opf = metadata_to_opf(mi, as_string=False)
for l in opf.xpath('//*[local-name()="language"]'):
if l.text:
lang = l.text
break
lang = lang_as_iso639_1(lang) or lang
opfns = OPF_NAMESPACES['opf']
m = opf.makeelement('{%s}manifest' % opfns)
opf.insert(1, m)
i = m.makeelement('{%s}item' % opfns, href=html_name, id='start')
i.set('media-type', guess_type('a.xhtml'))
m.append(i)
i = m.makeelement('{%s}item' % opfns, href=toc_name, id='ncx')
i.set('media-type', guess_type(toc_name))
m.append(i)
s = opf.makeelement('{%s}spine' % opfns, toc="ncx")
opf.insert(2, s)
i = s.makeelement('{%s}itemref' % opfns, idref='start')
s.append(i)
CONTAINER = '''\
<?xml version="1.0"?>
<container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">
<rootfiles>
<rootfile full-path="{0}" media-type="application/oebps-package+xml"/>
</rootfiles>
</container>
'''.format(prepare_string_for_xml(opf_name, True)).encode('utf-8')
HTML = P('templates/new_book.html', data=True).decode('utf-8').replace(
'_LANGUAGE_', prepare_string_for_xml(lang, True)
).replace(
'_TITLE_', prepare_string_for_xml(mi.title)
).replace(
'_AUTHORS_', prepare_string_for_xml(authors_to_string(mi.authors))
).encode('utf-8')
h = parse(HTML)
pretty_html_tree(None, h)
HTML = serialize(h, 'text/html')
ncx = etree.tostring(create_toc(mi, opf, html_name, lang), encoding='utf-8', xml_declaration=True, pretty_print=True)
pretty_xml_tree(opf)
opf = etree.tostring(opf, encoding='utf-8', xml_declaration=True, pretty_print=True)
if fmt == 'azw3':
with TemporaryDirectory('create-azw3') as tdir, CurrentDir(tdir):
for name, data in ((opf_name, opf), (html_name, HTML), (toc_name, ncx)):
with open(name, 'wb') as f:
f.write(data)
c = Container(os.path.dirname(os.path.abspath(opf_name)), opf_name, DevNull())
opf_to_azw3(opf_name, path, c)
else:
with ZipFile(path, 'w', compression=ZIP_STORED) as zf:
zf.writestr('mimetype', b'application/epub+zip', compression=ZIP_STORED)
zf.writestr('META-INF/', b'', 0755)
zf.writestr('META-INF/container.xml', CONTAINER)
zf.writestr(opf_name, opf)
zf.writestr(html_name, HTML)
zf.writestr(toc_name, ncx)
def parse_html(raw):
root = parse(raw,
decoder=lambda x: x.decode('utf-8'),
line_numbers=True,
linenumber_attribute='data-lnum')
return serialize(root, 'text/html').encode('utf-8')
def parse_html(raw):
root = parse(raw, decoder=lambda x:x.decode('utf-8'), line_numbers=True, linenumber_attribute='data-lnum')
return serialize(root, 'text/html').encode('utf-8')
