Valid XML for Plugin

[htb]

Hi zip,
great job with V0.6.

Many sites does not have a rss/atom/... interface. It is only possible to search their index via webbrowser (html data).
My idea was to add such query web-page as feed-url, plugin says "yep I know such an url" and the plugin extract all the date from the html-document.

For testing I've added a catch-all plugin (exractorMatches reaturns always true) ([PluginCompilerThread] Added plugin ARDMediathekExctractor (ARDMediathek.groovy)) and add the site "http://www.ardmediathek.de/ard/servlet/content/3517006?s=DTM" via Serviio-Console. But serviio says

  Code:

java.io.IOException: Cannot parse feed from http://www.ardmediathek.de/ard/servlet/content/3517006?s=DTM. Message: Error during feed parsing : Invalid XML: Error on line 375: The element type "li" must be terminated by the matching end-tag "</li>".

But IMHO Serviio should not be interested in this (html)content if there is a valid plugin. Serviio must only pass the url to the plugin.
Is it indented? Is it my fault?

bye

[zip]

Serviio always expects a valid XML feed. It parses it, retireves name, links, thumbnail, etc and then invokes the plugin for each feed entry.

Something that you want to accomplish is not possible yet. There might be a new type (other than RSS/Atom feed), e.g. Webpage, that would then need a plugin implementing a different interface.

[htb]

So we have to create a ticket
bye

I would also be very interested in support for something like this. Some sites simply don't publish rss-feeds for(all) their content.

A new type of online source "Webpage" sounds like it could be a hit. To make it simple it could just have a name-function, a validator-function(to see if it handles a page) and one function getting passed the page and which returns a list/collection of ExtendedContentURLContainer (extended with title/description).

This was just a rough idea, but it would eliminate the need for intermediate generation of rss-feeds, which is what I'm struggling with for the SvtPlay-plugin to handle the good stuff on that site.

[zip]

yep, something like that. I'll see if I can sqeeze it in 0.6.1

https://bitbucket.org/xnejp03/serviio/i ... -page-urls

Comment on the ticket if you want something specific to be implemented

[kairoh]

yep, something like that. I'll see if I can sqeeze it in 0.6.1

This can be a source of inspiration.