Skip to content

lxml.clean cannot defang some raw text ("<3") #2

@jmoiron

Description

@jmoiron

lxml.html.clean cannot clean some raw text (in titles, descriptions, etc); particularly, text that might look like html but isn't. The first case I noticed was the tag "<title><3</title>" will fail with a ParserError('Document is empty'), which is likely an underlying libxml2 issue.

We cannot simply pass the text through, as <3<script>alert('foo');</script> will also raise this same error. Currently, there is a regression test "TestHeartParserError" which confirms this error in speedparser and confirms that feedparser will read this content.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions