Podcast Support

[Mathias Panzenböck]

On 11/22/2009 10:42 AM, Bart Cerneels wrote:
> On Sun, Nov 22, 2009 at 00:48, Mathias Panzenböck
> <grosser.meister.morti at gmx.net> wrote:
>> 
>> Well, I think it would be better to have:
>> enum ContentType { TextContent, HtmlContent, XHtmlContent };
> 
> Yeah, probably better. How are we going to handle XHtml though? Why
> would it be different then normal HTML?
> 

Indeed, it isn't handled different. I'm jsut able to detect that it is xhtml
(ok, with maybe some false negatives that say it's html when escaped xhtml is
used in <description>). :)

>> I don't want to convert TO html. I want to convert html to plain text (strip
>> tags, resolve entity refs), because in Atom even elements like title can contain
>> (x)html!
> 
> Ah, I didn't know that. I have a feeling though that iTunes does not
> support non-cleartext for any of the tags. So we should be ok with
> just assuming it's clear text for all tags except rss:description and
> atom:summary. For safety we should also check the content type of
> itunes:summary.
> 
> I think there are Qt functions to convert HTML named refs and the XML
> numeric character refs should be converted by
> QXmlStreamReader::text(). Not sure about that last one though.
> 

I couldn't find such a function and Qt mailing list entries I've found say there
isn't. After all it's not always xhtml -> plain text but also html -> plain text
(which is not well formed xml).

>> Mhm. I guess no subtitle from RSS feeds then.
> 
> There is itunes:subtitle, so yes: RSS also has subtitle and we *will* use it.

Already use this in my latest commit. :)

	-panzi