On Thu, May 14, 2009 at 8:33 PM, Brandon Casci <[email protected]> wrote:
> all = random(weights=[1,8],strict=true,[ playlist_1, playlist_2 ])

Let me develop a bit what this does.

It does not make the playlists random. The random() operator doesn't
even know if its children are playlists or external streams. What
random() does is pick successive tracks from a random children. For
example if playlist_1 plays tracks p1, p2, p3... and playlist_2 plays
q1, q2, q3... then random([playlist_1,playlist_2]) might play tracks
p1,q1,p2,p3,q2,... The relative order of p2 and p3, for example, will
never be changed. If you want the playlist itself to be random, use
the "mode" parameter of playlist().

Now, in your example you pass weights=[1,8]: this means that
playlist_2 has height times more chances to be selected than
playlist_1.

When you pass strict=true, the random() is not random anymore, but
acts like a very deterministic quota operator. In you example it
means: play 1 file of playlist_1, then 8 of playlist_2, then start
again.

> If you don't want the playlists to be random, would you use prepend
> along with some custom selection logic to occasionally merge tracks
> from play tracks from both playlists?

I'm not sure to understand the question. Maybe you want to force two
successive tracks of one playlist to be played consecutively in the
random()? If so, the solution is indeed to merge those tracks before
the random() node so that they become one. Prepending is one way of
doing this. It should also be possible to play with transitions (using
cross()) depending on what you want exactly.

HTH
-- 
David

------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables 
unlimited royalty-free distribution of the report engine 
for externally facing server and web deployment. 
http://p.sf.net/sfu/businessobjects
_______________________________________________
Savonet-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/savonet-users

Reply via email to