ff21df596b
Maintaining the descriptions creates quite a bit of overhead (finding them, copying them, checking if they are still correct) for a minimal user benefit. PS: Viewing this diff should be easier in a difftool that shows changes in a line, for example kdiff3.
32 lines
1.1 KiB
Python
32 lines
1.1 KiB
Python
# -*- coding: iso-8859-1 -*-
|
|
# Copyright (C) 2004-2005 Tristan Seligmann and Jonathan Jacobs
|
|
# Copyright (C) 2012-2014 Bastian Kleineidam
|
|
|
|
from re import compile
|
|
|
|
from ..scraper import _BasicScraper
|
|
from ..helpers import bounceStarter
|
|
from ..util import tagre
|
|
|
|
|
|
class xkcd(_BasicScraper):
|
|
url = 'http://xkcd.com/'
|
|
starter = bounceStarter(url, compile(tagre("a", "href", r'(/\d+/)', before="next")))
|
|
stripUrl = url + '%s/'
|
|
firstStripUrl = stripUrl % '1'
|
|
imageSearch = compile(tagre("img", "src", r'(//imgs\.xkcd\.com/comics/[^"]+)'))
|
|
prevSearch = compile(tagre("a", "href", r'(/\d+/)', before="prev"))
|
|
help = 'Index format: n (unpadded)'
|
|
textSearch = compile(tagre("img", "title", r'([^"]+)', before=r'//imgs\.xkcd\.com/comics/'))
|
|
|
|
@classmethod
|
|
def namer(cls, imageUrl, pageUrl):
|
|
index = int(pageUrl.rstrip('/').rsplit('/', 1)[-1])
|
|
name = imageUrl.rsplit('/', 1)[-1].split('.')[0]
|
|
return '%03d-%s' % (index, name)
|
|
|
|
@classmethod
|
|
def imageUrlModifier(cls, url, data):
|
|
if url and '/large/' in data:
|
|
return url.replace(".png", "_large.png")
|
|
return url
|