in reply to Re^3: Problems with HTML Encoding in RSS after moving to XML::RSS 1.48
in thread Problems with HTML Encoding in RSS after moving to XML::RSS 1.48

Maybe the "problem" is that the closing angle brackets are not respectively are HTML-encoded (> vs >)? My short forays into generating HTML content for RSS make me think that all HTML needs to be entity-encoded so the XML remains valid, at least I (actually, XML::RSS::SimpleGen) do so with my Image photo stream RSS. You can add this as as "Smart bookmarks" in Firefox, which I think is how one could use Firefox as RSS reader. But without deeper knowledge and experience of RSS, I don't know.

Personally, I consider the full-encoded RSS to be "more valid".

Replies are listed 'Best First'.
Re^5: Problems with HTML Encoding in RSS after moving to XML::RSS 1.48
by ikegami (Patriarch) on Oct 22, 2010 at 22:25 UTC

    XML doesn't require a ">" in text data to be encoded, but any character in text data can be encoded.

    Or maybe you're playing with the idea that it's not text data. If so, neither would work as both encode the "<" as if it was text data.

    Personally, I consider the full-encoded RSS to be "more valid".

    So you always escape "!", "?", "[", "]" and "-"? They are no more and no less special than ">". Each is a special character in the right context in XML, and each is perfectly ordinary in text data and attribute values.