Apply match_regexps and filter_regexpr on feeds

[logan]

Hi,

match_regexps and filter_regexps seems to be applied only on html link resolution.
I need them in feeds resolution too since some of them have heterogeneous links:

<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:dct="http://purl.org/dc/terms/">
	<channel>
        ...
            <item>
                ...
				<link>http://www.liberation.fr/monde/0101318771-les-electeurs-israeliens-ont-choisi-le-pragmatisme</link>
				...
			</item>
			<item>
                ...
				<link>http://www.liberation.fr/monde/0101318514-repli</link>
                ...
			</item>
            <item>
                ...
                <!-- Item I want to ignore because it reference another website that isn't handled by my recipe -->
				<link>http://secretdefense.blogs.liberation.fr/defense/2009/02/un-officier-fra.html</link>
                ...
			</item>
        ...
    </channel>
</rss>

I'd like to put something like

  match_regexps = (r'http://www\.liberation\.fr/.*')

in my recipe to filter relevant item.

Is there another option make that happend or can I create a new ticket on trac ?

[kovidgoyal]

Implement the get_article_url method in your recipe. You can do arbitrarily powerful processing in it.