0x2620
/
openmedialibrary
1
0
Fork 0
Code Issues 131 Pull Requests Releases Wiki Activity

epub: use metadata.cover if set

This commit is contained in:
j 2016-01-05 15:20:47 +05:30
parent 78c9c5443f
commit ca3888869b
1 changed files with 51 additions and 42 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

93
oml/media/epub.py
View File

@ -26,50 +26,59 @@ def cover(path):
except zipfile.BadZipFile: except zipfile.BadZipFile:
logger.debug('invalid epub file %s', path) logger.debug('invalid epub file %s', path)
return data return data
def use(filename):
logger.debug('using %s', filename)
return z.read(filename)
for f in z.filelist: for f in z.filelist:
if 'cover' in f.filename.lower() and f.filename.split('.')[-1] in ('jpg', 'jpeg', 'png'): if 'cover' in f.filename.lower() and f.filename.split('.')[-1] in ('jpg', 'jpeg', 'png'):
logger.debug('using %s', f.filename) return use(f.filename)
data = z.read(f.filename) files = [f.filename for f in z.filelist]
break opf = [f for f in files if f.endswith('opf')]
if not data: if opf:
files = [f.filename for f in z.filelist] #logger.debug('opf: %s', z.read(opf[0]).decode())
opf = [f for f in files if f.endswith('opf')] info = ET.fromstring(z.read(opf[0]))
if opf: metadata = info.findall('{http://www.idpf.org/2007/opf}metadata')[0]
info = ET.fromstring(z.read(opf[0])) for e in metadata.getchildren():
manifest = info.findall('{http://www.idpf.org/2007/opf}manifest')[0] if e.tag == '{http://www.idpf.org/2007/opf}meta' and e.attrib['name'] == 'cover':
images = [e for e in manifest.getchildren() if 'image' in e.attrib['media-type']] filename = e.attrib['content']
if images: filename = [f for f in files if filename in f]
image_data = [] if filename:
for e in images: return use(filename[0])
filename = unquote(e.attrib['href']) manifest = info.findall('{http://www.idpf.org/2007/opf}manifest')[0]
filename = os.path.normpath(os.path.join(os.path.dirname(opf[0]), filename)) images = [e for e in manifest.getchildren() if 'image' in e.attrib['media-type']]
if filename in files: if images:
image_data.append((filename, z.read(filename))) image_data = []
if image_data: for e in images:
image_data.sort(key=lambda i: len(i[1])) filename = unquote(e.attrib['href'])
data = image_data[-1][1] filename = os.path.normpath(os.path.join(os.path.dirname(opf[0]), filename))
if not data: if filename in files:
for e in manifest.getchildren(): image_data.append((filename, z.read(filename)))
if 'html' in e.attrib['media-type']: if image_data:
filename = unquote(e.attrib['href']) image_data.sort(key=lambda i: len(i[1]))
filename = os.path.normpath(os.path.join(os.path.dirname(opf[0]), filename)) data = image_data[-1][1]
html = z.read(filename).decode('utf-8', 'ignore') logger.debug('using %s', image_data[-1][0])
img = re.compile('<img.*?src="(.*?)"').findall(html) return data
#svg image for e in manifest.getchildren():
img += re.compile('<image.*?href="(.*?)"').findall(html) if 'html' in e.attrib['media-type']:
if img: filename = unquote(e.attrib['href'])
img = unquote(img[0]) filename = os.path.normpath(os.path.join(os.path.dirname(opf[0]), filename))
img = os.path.normpath(os.path.join(os.path.dirname(filename), img)) html = z.read(filename).decode('utf-8', 'ignore')
if img in files: img = re.compile('<img.*?src="(.*?)"').findall(html)
logger.debug('using %s', img) #svg image
data = z.read(img) img += re.compile('<image.*?href="(.*?)"').findall(html)
break if img:
if not data: img = unquote(img[0])
img = Image.new('RGB', (80, 128)) img = os.path.normpath(os.path.join(os.path.dirname(filename), img))
o = BytesIO() if img in files:
img.save(o, format='jpeg') return use(img)
data = o.getvalue() # fallback return black cover
o.close() img = Image.new('RGB', (80, 128))
o = BytesIO()
img.save(o, format='jpeg')
data = o.getvalue()
o.close()
return data return data
def info(epub): def info(epub):