Fix for #2294 (cover files in epubs with property as property in manifest are now found)

This commit is contained in:
Ozzie Isaacs 2022-02-06 16:22:28 +01:00
parent 41f89af959
commit 411c13977f
2 changed files with 40 additions and 34 deletions

1
.github/FUNDING.yml vendored Normal file
View File

@ -0,0 +1 @@
custom: ["https://PayPal.Me/calibreweb",]

View File

@ -25,15 +25,14 @@ from .helper import split_authors
from .constants import BookMeta from .constants import BookMeta
def extract_cover(zip_file, cover_file, cover_path, tmp_file_name):
def extractCover(zipFile, coverFile, coverpath, tmp_file_name): if cover_file is None:
if coverFile is None:
return None return None
else: else:
zipCoverPath = os.path.join(coverpath, coverFile).replace('\\', '/') zip_cover_path = os.path.join(cover_path, cover_file).replace('\\', '/')
cf = zipFile.read(zipCoverPath) cf = zip_file.read(zip_cover_path)
prefix = os.path.splitext(tmp_file_name)[0] prefix = os.path.splitext(tmp_file_name)[0]
tmp_cover_name = prefix + '.' + os.path.basename(zipCoverPath) tmp_cover_name = prefix + '.' + os.path.basename(zip_cover_path)
image = open(tmp_cover_name, 'wb') image = open(tmp_cover_name, 'wb')
image.write(cf) image.write(cf)
image.close() image.close()
@ -47,12 +46,12 @@ def get_epub_info(tmp_file_path, original_file_name, original_file_extension):
'dc': 'http://purl.org/dc/elements/1.1/' 'dc': 'http://purl.org/dc/elements/1.1/'
} }
epubZip = zipfile.ZipFile(tmp_file_path) epub_zip = zipfile.ZipFile(tmp_file_path)
txt = epubZip.read('META-INF/container.xml') txt = epub_zip.read('META-INF/container.xml')
tree = etree.fromstring(txt) tree = etree.fromstring(txt)
cfname = tree.xpath('n:rootfiles/n:rootfile/@full-path', namespaces=ns)[0] cfname = tree.xpath('n:rootfiles/n:rootfile/@full-path', namespaces=ns)[0]
cf = epubZip.read(cfname) cf = epub_zip.read(cfname)
tree = etree.fromstring(cf) tree = etree.fromstring(cf)
coverpath = os.path.dirname(cfname) coverpath = os.path.dirname(cfname)
@ -86,9 +85,9 @@ def get_epub_info(tmp_file_path, original_file_name, original_file_extension):
lang = epub_metadata['language'].split('-', 1)[0].lower() lang = epub_metadata['language'].split('-', 1)[0].lower()
epub_metadata['language'] = isoLanguages.get_lang3(lang) epub_metadata['language'] = isoLanguages.get_lang3(lang)
epub_metadata = parse_epbub_series(ns, tree, epub_metadata) epub_metadata = parse_epub_series(ns, tree, epub_metadata)
coverfile = parse_ebpub_cover(ns, tree, epubZip, coverpath, tmp_file_path) cover_file = parse_epub_cover(ns, tree, epub_zip, coverpath, tmp_file_path)
if not epub_metadata['title']: if not epub_metadata['title']:
title = original_file_name title = original_file_name
@ -100,7 +99,7 @@ def get_epub_info(tmp_file_path, original_file_name, original_file_extension):
extension=original_file_extension, extension=original_file_extension,
title=title.encode('utf-8').decode('utf-8'), title=title.encode('utf-8').decode('utf-8'),
author=epub_metadata['creator'].encode('utf-8').decode('utf-8'), author=epub_metadata['creator'].encode('utf-8').decode('utf-8'),
cover=coverfile, cover=cover_file,
description=epub_metadata['description'], description=epub_metadata['description'],
tags=epub_metadata['subject'].encode('utf-8').decode('utf-8'), tags=epub_metadata['subject'].encode('utf-8').decode('utf-8'),
series=epub_metadata['series'].encode('utf-8').decode('utf-8'), series=epub_metadata['series'].encode('utf-8').decode('utf-8'),
@ -108,37 +107,43 @@ def get_epub_info(tmp_file_path, original_file_name, original_file_extension):
languages=epub_metadata['language'], languages=epub_metadata['language'],
publisher="") publisher="")
def parse_ebpub_cover(ns, tree, epubZip, coverpath, tmp_file_path):
coversection = tree.xpath("/pkg:package/pkg:manifest/pkg:item[@id='cover-image']/@href", namespaces=ns) def parse_epub_cover(ns, tree, epub_zip, cover_path, tmp_file_path):
coverfile = None cover_section = tree.xpath("/pkg:package/pkg:manifest/pkg:item[@id='cover-image']/@href", namespaces=ns)
if len(coversection) > 0: cover_file = None
coverfile = extractCover(epubZip, coversection[0], coverpath, tmp_file_path) if len(cover_section) > 0:
cover_file = extract_cover(epub_zip, cover_section[0], cover_path, tmp_file_path)
else: else:
meta_cover = tree.xpath("/pkg:package/pkg:metadata/pkg:meta[@name='cover']/@content", namespaces=ns) meta_cover = tree.xpath("/pkg:package/pkg:metadata/pkg:meta[@name='cover']/@content", namespaces=ns)
if len(meta_cover) > 0: if len(meta_cover) > 0:
coversection = tree.xpath("/pkg:package/pkg:manifest/pkg:item[@id='"+meta_cover[0]+"']/@href", namespaces=ns) 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: else:
coversection = tree.xpath("/pkg:package/pkg:guide/pkg:reference/@href", namespaces=ns) cover_section = tree.xpath("/pkg:package/pkg:guide/pkg:reference/@href", namespaces=ns)
if len(coversection) > 0: if len(cover_section) > 0:
filetype = coversection[0].rsplit('.', 1)[-1] filetype = cover_section[0].rsplit('.', 1)[-1]
if filetype == "xhtml" or filetype == "html": # if cover is (x)html format if filetype == "xhtml" or filetype == "html": # if cover is (x)html format
markup = epubZip.read(os.path.join(coverpath, coversection[0])) markup = epub_zip.read(os.path.join(cover_path, cover_section[0]))
markupTree = etree.fromstring(markup) markup_tree = etree.fromstring(markup)
# no matter xhtml or html with no namespace # no matter xhtml or html with no namespace
imgsrc = markupTree.xpath("//*[local-name() = 'img']/@src") img_src = markup_tree.xpath("//*[local-name() = 'img']/@src")
# Alternative image source # Alternative image source
if not len(imgsrc): if not len(img_src):
imgsrc = markupTree.xpath("//attribute::*[contains(local-name(), 'href')]") img_src = markup_tree.xpath("//attribute::*[contains(local-name(), 'href')]")
if len(imgsrc): if len(img_src):
# imgsrc maybe startwith "../"" so fullpath join then relpath to cwd # 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(coverpath, coversection[0])), filename = os.path.relpath(os.path.join(os.path.dirname(os.path.join(cover_path, cover_section[0])),
imgsrc[0])) img_src[0]))
coverfile = extractCover(epubZip, filename, "", tmp_file_path) cover_file = extract_cover(epub_zip, filename, "", tmp_file_path)
else: else:
coverfile = extractCover(epubZip, coversection[0], coverpath, tmp_file_path) cover_file = extract_cover(epub_zip, cover_section[0], cover_path, tmp_file_path)
return coverfile return cover_file
def parse_epbub_series(ns, tree, epub_metadata):
def parse_epub_series(ns, tree, epub_metadata):
series = tree.xpath("/pkg:package/pkg:metadata/pkg:meta[@name='calibre:series']/@content", namespaces=ns) series = tree.xpath("/pkg:package/pkg:metadata/pkg:meta[@name='calibre:series']/@content", namespaces=ns)
if len(series) > 0: if len(series) > 0:
epub_metadata['series'] = series[0] epub_metadata['series'] = series[0]