Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

News system: Make handling of data urls in img tags more robust

Browse Source
This commit is contained in:
Kovid Goyal 2021-08-12 13:18:32 +05:30
parent f16d7fb89f
commit d7d329554b
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/calibre/web/fetch/simple.py
View File

@ -17,6 +17,7 @@ import sys
import threading
import time
import traceback
from urllib.request import urlopen
from calibre import browser, relpath, unicode_path
from calibre.constants import filesystem_encoding, iswindows
@ -28,7 +29,6 @@ from calibre.utils.img import image_from_data, image_to_data
from calibre.utils.imghdr import what
from calibre.utils.logging import Log
from calibre.web.fetch.utils import rescale_image
from polyglot.binary import from_base64_bytes
from polyglot.builtins import as_bytes, unicode_type
from polyglot.http_client import responses
from polyglot.urllib import (
@ -402,9 +402,9 @@ class RecursiveFetcher(object):
c = 0
for tag in soup.findAll('img', src=True):
iurl = tag['src']
if iurl.startswith('data:image/'):
if iurl.startswith('data:'):
try:
data = from_base64_bytes(iurl.partition(',')[-1])
data = urlopen(iurl).read()
except Exception:
self.log.exception('Failed to decode embedded image')
continue