From 411c13977f030566bdf9505dfce63f5b580f40f9 Mon Sep 17 00:00:00 2001 From: Ozzie Isaacs Date: Sun, 6 Feb 2022 16:22:28 +0100 Subject: [PATCH] Fix for #2294 (cover files in epubs with property as property in manifest are now found) --- .github/FUNDING.yml | 1 + cps/epub.py | 73 ++++++++++++++++++++++++--------------------- 2 files changed, 40 insertions(+), 34 deletions(-) create mode 100644 .github/FUNDING.yml diff --git a/.github/FUNDING.yml b/.github/FUNDING.yml new file mode 100644 index 00000000..c3d0309e --- /dev/null +++ b/.github/FUNDING.yml @@ -0,0 +1 @@ +custom: ["https://PayPal.Me/calibreweb",] diff --git a/cps/epub.py b/cps/epub.py index cbbdcbbd..b436a755 100644 --- a/cps/epub.py +++ b/cps/epub.py @@ -25,15 +25,14 @@ from .helper import split_authors from .constants import BookMeta - -def extractCover(zipFile, coverFile, coverpath, tmp_file_name): - if coverFile is None: +def extract_cover(zip_file, cover_file, cover_path, tmp_file_name): + if cover_file is None: return None else: - zipCoverPath = os.path.join(coverpath, coverFile).replace('\\', '/') - cf = zipFile.read(zipCoverPath) + zip_cover_path = os.path.join(cover_path, cover_file).replace('\\', '/') + cf = zip_file.read(zip_cover_path) 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.write(cf) 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/' } - 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) 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) 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() 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']: 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, title=title.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'], tags=epub_metadata['subject'].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'], 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) - coverfile = None - if len(coversection) > 0: - coverfile = extractCover(epubZip, coversection[0], coverpath, tmp_file_path) + +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: + cover_file = extract_cover(epub_zip, cover_section[0], cover_path, tmp_file_path) else: meta_cover = tree.xpath("/pkg:package/pkg:metadata/pkg:meta[@name='cover']/@content", namespaces=ns) 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: - coversection = tree.xpath("/pkg:package/pkg:guide/pkg:reference/@href", namespaces=ns) - if len(coversection) > 0: - filetype = coversection[0].rsplit('.', 1)[-1] + cover_section = tree.xpath("/pkg:package/pkg:guide/pkg:reference/@href", namespaces=ns) + if len(cover_section) > 0: + filetype = cover_section[0].rsplit('.', 1)[-1] if filetype == "xhtml" or filetype == "html": # if cover is (x)html format - markup = epubZip.read(os.path.join(coverpath, coversection[0])) - markupTree = etree.fromstring(markup) + markup = epub_zip.read(os.path.join(cover_path, cover_section[0])) + markup_tree = etree.fromstring(markup) # 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 - if not len(imgsrc): - imgsrc = markupTree.xpath("//attribute::*[contains(local-name(), 'href')]") - if len(imgsrc): - # imgsrc maybe startwith "../"" so fullpath join then relpath to cwd - filename = os.path.relpath(os.path.join(os.path.dirname(os.path.join(coverpath, coversection[0])), - imgsrc[0])) - coverfile = extractCover(epubZip, filename, "", tmp_file_path) + 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: - coverfile = extractCover(epubZip, coversection[0], coverpath, tmp_file_path) - return coverfile + cover_file = extract_cover(epub_zip, cover_section[0], cover_path, tmp_file_path) + 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) if len(series) > 0: epub_metadata['series'] = series[0]