Washington Post news feed crash

[nickredding]

The Washington Post RSS feeds have some issues which involve empty articles in the index, causing the parser to crash. I've opened a ticket, and in the mean time the following code added to the recipe will work around the issue.

    def preprocess_html(self, soup):
        for tag in soup.findAll('font'):
            if tag['size']:
                if tag['size'] == '+2':
                    if tag.b:
                        return soup
        return None

[nickredding]

Kovid - thaks for the response to the ticket--since it is an XHTML defect in the Washington Post feed, perhaps you should put the workaround in the standard recipe (I've tested it and it works).

[nickredding]

Actually, the following is safer since the test tag['size'] doesn't work if there is no attribute 'size'.

    
def preprocess_html(self, soup):
        for tag in soup.findAll('font'):
            if tag.has_key('size'):
                if tag['size'] == '+2':
                    if tag.b:
                        return soup
        return None

[kovidgoyal]

done.