mirror of
https://github.com/janeczku/calibre-web
synced 2024-10-31 15:20:28 +00:00
184 lines
7.3 KiB
Python
184 lines
7.3 KiB
Python
# -*- coding: utf-8 -*-
|
|
|
|
# This file is part of the Calibre-Web (https://github.com/janeczku/calibre-web)
|
|
# Copyright (C) 2018 lemmsh, Kennyl, Kyosfonica, matthazinski
|
|
#
|
|
# This program is free software: you can redistribute it and/or modify
|
|
# it under the terms of the GNU General Public License as published by
|
|
# the Free Software Foundation, either version 3 of the License, or
|
|
# (at your option) any later version.
|
|
#
|
|
# This program is distributed in the hope that it will be useful,
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
# GNU General Public License for more details.
|
|
#
|
|
# You should have received a copy of the GNU General Public License
|
|
# along with this program. If not, see <http://www.gnu.org/licenses/>.
|
|
|
|
import os
|
|
import zipfile
|
|
from lxml import etree
|
|
|
|
from . import isoLanguages, cover
|
|
from .helper import split_authors
|
|
from .constants import BookMeta
|
|
|
|
|
|
def _extract_cover(zip_file, cover_file, cover_path, tmp_file_name):
|
|
if cover_file is None:
|
|
return None
|
|
else:
|
|
cf = extension = None
|
|
zip_cover_path = os.path.join(cover_path, cover_file).replace('\\', '/')
|
|
|
|
prefix = os.path.splitext(tmp_file_name)[0]
|
|
tmp_cover_name = prefix + '.' + os.path.basename(zip_cover_path)
|
|
ext = os.path.splitext(tmp_cover_name)
|
|
if len(ext) > 1:
|
|
extension = ext[1].lower()
|
|
if extension in cover.COVER_EXTENSIONS:
|
|
cf = zip_file.read(zip_cover_path)
|
|
return cover.cover_processing(tmp_file_name, cf, extension)
|
|
|
|
|
|
def get_epub_info(tmp_file_path, original_file_name, original_file_extension):
|
|
ns = {
|
|
'n': 'urn:oasis:names:tc:opendocument:xmlns:container',
|
|
'pkg': 'http://www.idpf.org/2007/opf',
|
|
'dc': 'http://purl.org/dc/elements/1.1/'
|
|
}
|
|
|
|
epub_zip = zipfile.ZipFile(tmp_file_path)
|
|
|
|
txt = epub_zip.read('META-INF/container.xml')
|
|
tree = etree.fromstring(txt)
|
|
cf_name = tree.xpath('n:rootfiles/n:rootfile/@full-path', namespaces=ns)[0]
|
|
cf = epub_zip.read(cf_name)
|
|
tree = etree.fromstring(cf)
|
|
|
|
cover_path = os.path.dirname(cf_name)
|
|
|
|
p = tree.xpath('/pkg:package/pkg:metadata', namespaces=ns)[0]
|
|
|
|
epub_metadata = {}
|
|
|
|
for s in ['title', 'description', 'creator', 'language', 'subject', 'publisher', 'date']:
|
|
tmp = p.xpath('dc:%s/text()' % s, namespaces=ns)
|
|
if len(tmp) > 0:
|
|
if s == 'creator':
|
|
epub_metadata[s] = ' & '.join(split_authors(tmp))
|
|
elif s == 'subject':
|
|
epub_metadata[s] = ', '.join(tmp)
|
|
elif s == 'date':
|
|
epub_metadata[s] = tmp[0][:10]
|
|
else:
|
|
epub_metadata[s] = tmp[0]
|
|
else:
|
|
epub_metadata[s] = 'Unknown'
|
|
|
|
if epub_metadata['subject'] == 'Unknown':
|
|
epub_metadata['subject'] = ''
|
|
|
|
if epub_metadata['publisher'] == u'Unknown':
|
|
epub_metadata['publisher'] = ''
|
|
|
|
if epub_metadata['date'] == u'Unknown':
|
|
epub_metadata['date'] = ''
|
|
|
|
if epub_metadata['description'] == u'Unknown':
|
|
description = tree.xpath("//*[local-name() = 'description']/text()")
|
|
if len(description) > 0:
|
|
epub_metadata['description'] = description
|
|
else:
|
|
epub_metadata['description'] = ""
|
|
|
|
lang = epub_metadata['language'].split('-', 1)[0].lower()
|
|
epub_metadata['language'] = isoLanguages.get_lang3(lang)
|
|
|
|
epub_metadata = parse_epub_series(ns, tree, epub_metadata)
|
|
|
|
cover_file = parse_epub_cover(ns, tree, epub_zip, cover_path, tmp_file_path)
|
|
|
|
identifiers = []
|
|
for node in p.xpath('dc:identifier', namespaces=ns):
|
|
identifier_name=node.attrib.values()[-1];
|
|
identifier_value=node.text;
|
|
if identifier_name in ('uuid','calibre'):
|
|
continue;
|
|
identifiers.append( [identifier_name, identifier_value] )
|
|
|
|
if not epub_metadata['title']:
|
|
title = original_file_name
|
|
else:
|
|
title = epub_metadata['title']
|
|
|
|
return BookMeta(
|
|
file_path=tmp_file_path,
|
|
extension=original_file_extension,
|
|
title=title.encode('utf-8').decode('utf-8'),
|
|
author=epub_metadata['creator'].encode('utf-8').decode('utf-8'),
|
|
cover=cover_file,
|
|
description=epub_metadata['description'],
|
|
tags=epub_metadata['subject'].encode('utf-8').decode('utf-8'),
|
|
series=epub_metadata['series'].encode('utf-8').decode('utf-8'),
|
|
series_id=epub_metadata['series_id'].encode('utf-8').decode('utf-8'),
|
|
languages=epub_metadata['language'],
|
|
publisher=epub_metadata['publisher'].encode('utf-8').decode('utf-8'),
|
|
pubdate=epub_metadata['date'],
|
|
identifiers=identifiers)
|
|
|
|
|
|
def parse_epub_cover(ns, tree, epub_zip, cover_path, tmp_file_path):
|
|
cover_section = tree.xpath("/pkg:package/pkg:manifest/pkg:item[@id='cover-image']/@href", namespaces=ns)
|
|
cover_file = None
|
|
# if len(cover_section) > 0:
|
|
for cs in cover_section:
|
|
cover_file = _extract_cover(epub_zip, cs, cover_path, tmp_file_path)
|
|
if cover_file:
|
|
break
|
|
if not cover_file:
|
|
meta_cover = tree.xpath("/pkg:package/pkg:metadata/pkg:meta[@name='cover']/@content", namespaces=ns)
|
|
if len(meta_cover) > 0:
|
|
cover_section = tree.xpath(
|
|
"/pkg:package/pkg:manifest/pkg:item[@id='"+meta_cover[0]+"']/@href", namespaces=ns)
|
|
if not cover_section:
|
|
cover_section = tree.xpath(
|
|
"/pkg:package/pkg:manifest/pkg:item[@properties='" + meta_cover[0] + "']/@href", namespaces=ns)
|
|
else:
|
|
cover_section = tree.xpath("/pkg:package/pkg:guide/pkg:reference/@href", namespaces=ns)
|
|
for cs in cover_section:
|
|
filetype = cs.rsplit('.', 1)[-1]
|
|
if filetype == "xhtml" or filetype == "html": # if cover is (x)html format
|
|
markup = epub_zip.read(os.path.join(cover_path, cs))
|
|
markup_tree = etree.fromstring(markup)
|
|
# no matter xhtml or html with no namespace
|
|
img_src = markup_tree.xpath("//*[local-name() = 'img']/@src")
|
|
# Alternative image source
|
|
if not len(img_src):
|
|
img_src = markup_tree.xpath("//attribute::*[contains(local-name(), 'href')]")
|
|
if len(img_src):
|
|
# img_src maybe start with "../"" so fullpath join then relpath to cwd
|
|
filename = os.path.relpath(os.path.join(os.path.dirname(os.path.join(cover_path, cover_section[0])),
|
|
img_src[0]))
|
|
cover_file = _extract_cover(epub_zip, filename, "", tmp_file_path)
|
|
else:
|
|
cover_file = _extract_cover(epub_zip, cs, cover_path, tmp_file_path)
|
|
if cover_file: break
|
|
return cover_file
|
|
|
|
|
|
def parse_epub_series(ns, tree, epub_metadata):
|
|
series = tree.xpath("/pkg:package/pkg:metadata/pkg:meta[@name='calibre:series']/@content", namespaces=ns)
|
|
if len(series) > 0:
|
|
epub_metadata['series'] = series[0]
|
|
else:
|
|
epub_metadata['series'] = ''
|
|
|
|
series_id = tree.xpath("/pkg:package/pkg:metadata/pkg:meta[@name='calibre:series_index']/@content", namespaces=ns)
|
|
if len(series_id) > 0:
|
|
epub_metadata['series_id'] = series_id[0]
|
|
else:
|
|
epub_metadata['series_id'] = '1'
|
|
return epub_metadata
|