def convert(self, stream, opts, file_ext, log,
accelerators):
basedir = os.getcwd()
self.opts = opts
fname = None
if hasattr(stream, 'name'):
basedir = os.path.dirname(stream.name)
fname = os.path.basename(stream.name)
if file_ext != 'opf':
if opts.dont_package:
raise ValueError('The --dont-package option is not supported for an HTML input file')
from ebook_converter.ebooks.metadata.html import get_metadata
mi = get_metadata(stream)
if fname:
from ebook_converter.ebooks.metadata.meta import metadata_from_filename
fmi = metadata_from_filename(fname)
fmi.smart_update(mi)
mi = fmi
oeb = self.create_oebbook(stream.name, basedir, opts, log, mi)
return oeb
from ebook_converter.ebooks.conversion.plumber import create_oebbook
return create_oebbook(log, stream.name, opts,
encoding=opts.input_encoding)
def do_rebuild(opf, dest_path):
plumber = Plumber(opf, dest_path, default_log)
plumber.setup_options()
inp = plugin_for_input_format('azw3')
outp = plugin_for_output_format('azw3')
plumber.opts.mobi_passthrough = True
oeb = create_oebbook(default_log, opf, plumber.opts)
set_cover(oeb)
outp.convert(oeb, dest_path, inp, plumber.opts, default_log)
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
def convert(self, stream, options, file_ext, log,
accelerators):
from ebook_converter.ebooks.lit.reader import LitReader
from ebook_converter.ebooks.conversion.plumber import create_oebbook
self.log = log
return create_oebbook(log, stream, options, reader=LitReader)
def convert(self, stream, options, file_ext, log,
accelerators):
import uuid
from ebook_converter.ebooks.oeb.base import DirContainer
from ebook_converter.ebooks.snb.snbfile import SNBFile
log.debug("Parsing SNB file...")
snbFile = SNBFile()
try:
snbFile.Parse(stream)
except Exception:
raise ValueError("Invalid SNB file")
if not snbFile.IsValid():
log.debug("Invalid SNB file")
raise ValueError("Invalid SNB file")
log.debug("Handle meta data ...")
from ebook_converter.ebooks.conversion.plumber import create_oebbook
oeb = create_oebbook(log, None, options,
encoding=options.input_encoding, populate=False)
meta = snbFile.GetFileStream('snbf/book.snbf')
if meta is not None:
meta = etree.fromstring(meta)
item_map = {'title': './/head/name',
'creator': './/head/author',
'language': './/head/language',
'generator': './/head/generator',
'publisher': './/head/publisher',
'cover': './/head/cover'}
d = {}
for key, item in item_map.items():
node = meta.find(item)
if node is not None:
d[key] = node.text if node.text is not None else ''
else:
d[key] = ''
oeb.metadata.add('title', d['title'])
oeb.metadata.add('creator', d['creator'], attrib={'role': 'aut'})
oeb.metadata.add('language',
d['language'].lower().replace('_', '-'))
oeb.metadata.add('generator', d['generator'])
oeb.metadata.add('publisher', d['publisher'])
if d['cover'] != '':
oeb.guide.add('cover', 'Cover', d['cover'])
bookid = str(uuid.uuid4())
oeb.metadata.add('identifier', bookid, id='uuid_id', scheme='uuid')
for ident in oeb.metadata.identifier:
if 'id' in ident.attrib:
oeb.uid = oeb.metadata.identifier[0]
break
with TemporaryDirectory('_snb2oeb', keep=True) as tdir:
log.debug('Process TOC ...')
toc = snbFile.GetFileStream('snbf/toc.snbf')
oeb.container = DirContainer(tdir, log)
if toc is not None:
toc = etree.fromstring(toc)
i = 1
for ch in toc.find('.//body'):
chapterName = ch.text
chapterSrc = ch.get('src')
fname = 'ch_%d.htm' % i
data = snbFile.GetFileStream('snbc/' + chapterSrc)
if data is None:
continue
snbc = etree.fromstring(data)
lines = []
for line in snbc.find('.//body'):
if line.tag == 'text':
lines.append('<p>%s</p>' % html_encode(line.text))
elif line.tag == 'img':
lines.append('<p><img src="%s" /></p>' %
html_encode(line.text))
with open(os.path.join(tdir, fname), 'wb') as f:
f.write((HTML_TEMPLATE %
(chapterName,
'\n'.join(lines))).encode('utf-8',
'replace'))
oeb.toc.add(ch.text, fname)
id, href = oeb.manifest.generate(
id='html', href=ascii_filename(fname))
item = oeb.manifest.add(id, href, 'text/html')
item.html_input_href = fname
oeb.spine.add(item, True)
i = i + 1
imageFiles = snbFile.OutputImageFiles(tdir)
for f, m in imageFiles:
id, href = oeb.manifest.generate(id='image',
href=ascii_filename(f))
item = oeb.manifest.add(id, href, m)
item.html_input_href = f
return oeb