def __call__(self, oeb, opts):
import css_parser
self.log = oeb.logger
self.opts = opts
self.oeb = oeb
for item in oeb.manifest.items:
self.current_item = item
if etree.iselement(item.data):
rewrite_links(self.current_item.data, self.url_replacer)
elif hasattr(item.data, 'cssText'):
css_parser.replaceUrls(item.data, self.url_replacer)
if self.oeb.guide:
for ref in self.oeb.guide.values():
href = urlnormalize(ref.href)
href, frag = urllib.parse.urldefrag(href)
replacement = self.rename_map.get(href, None)
if replacement is not None:
nhref = replacement
if frag:
nhref += '#' + frag
ref.href = nhref
if self.oeb.toc:
self.fix_toc_entry(self.oeb.toc)
def fix_links(self):
'''
Fix references to the split files in other content files.
'''
for item in self.oeb.manifest:
if etree.iselement(item.data):
self.current_item = item
base.rewrite_links(item.data, self.rewrite_links)
def mlize_spine(self, oeb_book):
output = ['']
for item in oeb_book.spine:
self.log.debug('Converting %s to Markdown formatted TXT...' % item.href)
self.rewrite_ids(item.data, item)
rewrite_links(item.data, partial(self.rewrite_link, page=item))
stylizer = Stylizer(item.data, item.href, oeb_book, self.opts, self.opts.output_profile)
output += self.dump_text(item.data.find(XHTML('body')), stylizer)
output.append('\n\n')
return ''.join(output)
def mlize_spine(self, oeb_book):
output = [
u'<html><head><meta http-equiv="Content-Type" content="text/html;charset=utf-8" /><title>%s</title></head><body>'
% (prepare_string_for_xml(self.book_title))
]
for item in oeb_book.spine:
self.log.debug('Converting %s to HTML...' % item.href)
self.rewrite_ids(item.data, item)
base.rewrite_links(item.data, partial(self.rewrite_link,
page=item))
stylizer = Stylizer(item.data, item.href, oeb_book, self.opts)
output += self.dump_text(item.data.find(base.tag('xhtml', 'body')),
stylizer, item)
output.append('\n\n')
output.append('</body></html>')
return ''.join(output)
def mlize_spine(self, oeb_book):
output = []
for item in oeb_book.spine:
self.log.debug('Converting %s to HTML...' % item.href)
self.rewrite_ids(item.data, item)
base.rewrite_links(item.data, partial(self.rewrite_link,
page=item))
stylizer = Stylizer(item.data, item.href, oeb_book, self.opts)
output += self.dump_text(item.data.find(base.tag('xhtml', 'body')),
stylizer, item)
output.append('\n\n')
if self.opts.htmlz_class_style == 'external':
css = u'<link href="style.css" rel="stylesheet" type="text/css" />'
else:
css = u'<style type="text/css">' + self.get_css(
oeb_book) + u'</style>'
title = u'<title>%s</title>' % prepare_string_for_xml(self.book_title)
output = [u'<html><head><meta http-equiv="Content-Type" content="text/html;charset=utf-8" />'] + \
[css] + [title, u'</head><body>'] + output + [u'</body></html>']
return ''.join(output)
def create_oebbook(self, htmlpath, basedir, opts, log, mi):
import uuid
from ebook_converter.ebooks.conversion.plumber import create_oebbook
from ebook_converter.ebooks.oeb.base import (DirContainer,
rewrite_links, urlnormalize, BINARY_MIME, OEB_STYLES,
xpath, urlquote)
from ebook_converter.ebooks.oeb.transforms.metadata import \
meta_info_to_oeb_metadata
from ebook_converter.ebooks.html.input import get_filelist
from ebook_converter.ebooks.metadata import string_to_authors
from ebook_converter.utils.localization import canonicalize_lang
import css_parser, logging
css_parser.log.setLevel(logging.WARN)
self.OEB_STYLES = OEB_STYLES
oeb = create_oebbook(log, None, opts, self,
encoding=opts.input_encoding, populate=False)
self.oeb = oeb
metadata = oeb.metadata
meta_info_to_oeb_metadata(mi, metadata, log)
if not metadata.language:
l = canonicalize_lang(getattr(opts, 'language', None))
if not l:
oeb.logger.warn('Language not specified')
l = get_lang().replace('_', '-')
metadata.add('language', l)
if not metadata.creator:
a = getattr(opts, 'authors', None)
if a:
a = string_to_authors(a)
if not a:
oeb.logger.warn('Creator not specified')
a = [self.oeb.translate('Unknown')]
for aut in a:
metadata.add('creator', aut)
if not metadata.title:
oeb.logger.warn('Title not specified')
metadata.add('title', self.oeb.translate('Unknown'))
bookid = str(uuid.uuid4())
metadata.add('identifier', bookid, id='uuid_id', scheme='uuid')
for ident in metadata.identifier:
if 'id' in ident.attrib:
self.oeb.uid = metadata.identifier[0]
break
filelist = get_filelist(htmlpath, basedir, opts, log)
filelist = [f for f in filelist if not f.is_binary]
htmlfile_map = {}
for f in filelist:
path = f.path
oeb.container = DirContainer(os.path.dirname(path), log,
ignore_opf=True)
bname = os.path.basename(path)
id, href = oeb.manifest.generate(id='html', href=sanitize_file_name(bname))
htmlfile_map[path] = href
item = oeb.manifest.add(id, href, 'text/html')
if path == htmlpath and '%' in path:
bname = urlquote(bname)
item.html_input_href = bname
oeb.spine.add(item, True)
self.added_resources = {}
self.log = log
self.log('Normalizing filename cases')
for path, href in htmlfile_map.items():
self.added_resources[path] = href
self.urlnormalize, self.DirContainer = urlnormalize, DirContainer
self.urldefrag = urllib.parse.urldefrag
self.BINARY_MIME = BINARY_MIME
self.log('Rewriting HTML links')
for f in filelist:
path = f.path
dpath = os.path.dirname(path)
oeb.container = DirContainer(dpath, log, ignore_opf=True)
href = htmlfile_map[path]
try:
item = oeb.manifest.hrefs[href]
except KeyError:
item = oeb.manifest.hrefs[urlnormalize(href)]
rewrite_links(item.data,
functools.partial(self.resource_adder, base=dpath))
for item in oeb.manifest.values():
if item.media_type in self.OEB_STYLES:
dpath = None
for path, href in self.added_resources.items():
if href == item.href:
dpath = os.path.dirname(path)
break
css_parser.replaceUrls(item.data,
functools.partial(self.resource_adder, base=dpath))
toc = self.oeb.toc
self.oeb.auto_generated_toc = True
titles = []
headers = []
for item in self.oeb.spine:
if not item.linear:
continue
html = item.data
title = ''.join(xpath(html, '/h:html/h:head/h:title/text()'))
title = re.sub(r'\s+', ' ', title.strip())
if title:
titles.append(title)
headers.append('(unlabled)')
for tag in ('h1', 'h2', 'h3', 'h4', 'h5', 'strong'):
expr = '/h:html/h:body//h:%s[position()=1]/text()'
header = ''.join(xpath(html, expr % tag))
header = re.sub(r'\s+', ' ', header.strip())
if header:
headers[-1] = header
break
use = titles
if len(titles) > len(set(titles)):
use = headers
for title, item in zip(use, self.oeb.spine):
if not item.linear:
continue
toc.add(title, item.href)
oeb.container = DirContainer(os.getcwd(), oeb.log, ignore_opf=True)
return oeb