note 70862 deleted from function.strip-tags by danbrown

Note Submitter: anonymous ---- A different approach to cleaning up HTML would be to first escape all unsafe characters: & to &amp; < to &lt; > to &gt; then to unescape matching pairs of tags back (e.g. "&lt;b&gt;hello&lt;/b&gt;" => "<b>hello</b>"), if it is identified safe. This backwards-approach should be safer because if a tag is not identified correctly, it is, at the end, in an escaped state. So if a user enters invalid html, or tags that are unsupported or unwanted, they are shown in plain text, and not stripped away. This is good, because the characters "<" and ">" might have been used in a different way (e.g. to make a text arrow: "a <=> b"). This is the case in most forums (apart from the fact that they use "[tag]"-tags instead of "<tag>"-tags)