dmoonfire/dosage
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Activity

Register EXSLT extensions by default.

Browse source 
This allows comic module authors to use the full power of regular
expressions in XPath expression, see http://exslt.org/regexp/regexp.html
for usage. Please be aware that these use the prefix re: instead of
regexp: here.
This commit is contained in:
Tobias Gruetzmacher 2016-04-19 23:48:14 +02:00
parent 4204f5f1e4
commit df46907f39
1 changed files with 9 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

11
dosagelib/scraper.py
View file

@ -1,7 +1,9 @@
# -*- coding: utf-8 -*- # -*- coding: utf-8 -*-
# Copyright (C) 2004-2005 Tristan Seligmann and Jonathan Jacobs # Copyright (C) 2004-2005 Tristan Seligmann and Jonathan Jacobs
# Copyright (C) 2012-2014 Bastian Kleineidam # Copyright (C) 2012-2014 Bastian Kleineidam
# Copyright (C) 2014-2016 Tobias Gruetzmacher # Copyright (C) 2015-2016 Tobias Gruetzmacher
from __future__ import absolute_import, division, print_function
import time import time
import random import random
@ -428,6 +430,10 @@ class _ParserScraper(Scraper):
XML_DECL = re.compile( XML_DECL = re.compile(
r'^(<\?xml[^>]+)\s+encoding\s*=\s*["\'][^"\']*["\'](\s*\?>|)', re.U) r'^(<\?xml[^>]+)\s+encoding\s*=\s*["\'][^"\']*["\'](\s*\?>|)', re.U)
NS = {
"re": "http://exslt.org/regular-expressions"
}
# Switch between CSS and XPath selectors for this class. Since CSS needs # Switch between CSS and XPath selectors for this class. Since CSS needs
# another Python module, XPath is the default for now. # another Python module, XPath is the default for now.
css = False css = False
@ -455,7 +461,8 @@ class _ParserScraper(Scraper):
if cls.css: if cls.css:
searchFun = data.cssselect searchFun = data.cssselect
else: else:
searchFun = data.xpath def searchFun(s):
return data.xpath(s, namespaces=cls.NS)
searches = makeSequence(urlSearch) searches = makeSequence(urlSearch)
for search in searches: for search in searches:
for match in searchFun(search): for match in searchFun(search):