From 389f6e95c5e5da3221c56ccecf77beaf6434e583 Mon Sep 17 00:00:00 2001 From: Julia Medina Date: Mon, 4 May 2015 17:11:05 -0300 Subject: [PATCH] Add deprecation warning to HtmlParserLinkExtractor --- scrapy/linkextractors/htmlparser.py | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/scrapy/linkextractors/htmlparser.py b/scrapy/linkextractors/htmlparser.py index fff9eabe6..5d263ac2c 100644 --- a/scrapy/linkextractors/htmlparser.py +++ b/scrapy/linkextractors/htmlparser.py @@ -2,6 +2,7 @@ HTMLParser-based link extractor """ +import warnings from HTMLParser import HTMLParser from six.moves.urllib.parse import urljoin @@ -9,12 +10,20 @@ from w3lib.url import safe_url_string from scrapy.link import Link from scrapy.utils.python import unique as unique_list +from scrapy.exceptions import ScrapyDeprecationWarning + class HtmlParserLinkExtractor(HTMLParser): def __init__(self, tag="a", attr="href", process=None, unique=False): HTMLParser.__init__(self) + warnings.warn( + "HtmlParserLinkExtractor is deprecated and will be removed in " + "future releases. Please use scrapy.linkextractors.LinkExtractor", + ScrapyDeprecationWarning + ) + self.scan_tag = tag if callable(tag) else lambda t: t == tag self.scan_attr = attr if callable(attr) else lambda a: a == attr self.process_attr = process if callable(process) else lambda v: v