From 4ce0f53f7edc558ce70ef7b08cdcfd57fcd8b9fd Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Daniel=20Gra=C3=B1a?= Date: Wed, 12 Aug 2015 14:21:55 -0300 Subject: [PATCH] Replace usage of deprecated class by its parsel\'s counterpart --- scrapy/linkextractors/__init__.py | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/scrapy/linkextractors/__init__.py b/scrapy/linkextractors/__init__.py index 8567dbb76..bb799e572 100644 --- a/scrapy/linkextractors/__init__.py +++ b/scrapy/linkextractors/__init__.py @@ -6,12 +6,14 @@ This package contains a collection of Link Extractors. For more info see docs/topics/link-extractors.rst """ import re -from six.moves.urllib.parse import urlparse -from scrapy.selector.csstranslator import ScrapyHTMLTranslator -from scrapy.utils.url import url_is_from_any_domain -from scrapy.utils.url import canonicalize_url, url_is_from_any_domain, url_has_any_extension +from six.moves.urllib.parse import urlparse +from parsel.csstranslator import HTMLTranslator + from scrapy.utils.misc import arg_to_iter +from scrapy.utils.url import ( + canonicalize_url, url_is_from_any_domain, url_has_any_extension, +) # common file extensions that are not followed if they occur in links @@ -42,7 +44,7 @@ _is_valid_url = lambda url: url.split('://', 1)[0] in set(['http', 'https', 'fil class FilteringLinkExtractor(object): - _csstranslator = ScrapyHTMLTranslator() + _csstranslator = HTMLTranslator() def __init__(self, link_extractor, allow, deny, allow_domains, deny_domains, restrict_xpaths, canonicalize, deny_extensions, restrict_css):