[mythtv-users] Question about mythstream parsers

[Phill Edwards]

I'm trying to write a parser for mythstream to get the new videos
headlines and associated video URLs for Australian ABC News etc. I'm
not a programmer, but I have managed to write a perl script which
connects to the right URL, downloads some pages and extracts headline
and video URL info and outputs stuff like this:

<snip>
[item]
ABC News
At least 13 dead after plane crashes off Sicily
http://www.abc.net.au//broadbandnews/20050807_1935/hq6.asx
[emptystring]
[emptystring]

[item]
ABC News
Kylie's cancer scare prompts screening surge
http://www.abc.net.au//broadbandnews/20050807_1935/hq7.asx
[emptystring]
[emptystring]
</snip>

I thought if I put my harvester file (abcnews,pl) as the harvester in
streams.res then this information would be what mythstream needs to
display the headlines and select one to play the video. I have this in
my streams.res file:

[item]
Video Streams
Phills Test
http://www.abc.net.au//broadband/player/test.htm
Phill's Test Harvester
abcnews

But when I pick this in Mythstream it just says "no url's found". I'm
obviously misunderstanding something here about how Mythstream works,
and as I'm not a perl programmer I find it really hard to understand
what the example.pl etc harvesters are doing. Could anyone please
enlighten me as to what I need to do to get this working?

Regards,
Phill