Read img tag of cover file if cover is (x)html

for example
‘<item id="cover" href="Text/cover.xhtml" media-type="application/xhtml+xml”/>’
This commit is contained in:
林檎 2017-03-01 13:18:14 +08:00
parent 90aa269e92
commit 7972b987d4

View file

@ -54,6 +54,16 @@ def get_epub_info(tmp_file_path, original_file_name, original_file_extension):
else:
coversection = tree.xpath("/pkg:package/pkg:manifest/pkg:item[@id='cover']/@href", namespaces=ns)
if len(coversection) > 0:
filetype = coversection[0].rsplit('.',1)[-1]
if filetype == "xhtml" or filetype == "html": #if cover is (x)html format
markup = zip.read(os.path.join(coverpath,coversection[0]))
markupTree = etree.fromstring(markup)
#no matter xhtml or html with no namespace
imgsrc = markupTree.xpath( "//*[local-name() = 'img']/@src")
#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(zip, filename, "", tmp_file_path)
else:
coverfile = extractCover(zip, coversection[0], coverpath, tmp_file_path)
else:
coverfile = None