Re: [Rhythmbox-devel] Improved random patch (bug 163196 ish)

[Alexander Senger <neutronenspalter freenet de>]

Hi,

that's funny. I've investigated the same thing, though without getting so far as to submit a patch.
Especially I was interested in lowering the probability of very recent songs being replayed.

In the course I noted, that the "history" of a song isn't updated before the next song is played,
but after the next but one (in the SVN-checkout of rhythmbox 0.11.6). Let's give me an example: song
A is played and was never played before (rhythmdb_entry_get_ulong (entry, RHYTHMDB_PROP_LAST_PLAYED)
returns 0). When the function "rb_random_by_age_and_rating_get_entry_weight" is called in the
process to select the next song, the query "rhythmdb_entry_get_ulong (entry,
RHYTHMDB_PROP_LAST_PLAYED)" for song A returns 0 again! Only after the next song you get the correct
result. A reason might be, that the next (random) song is choosen before the database is updated,
but I never checked for this.

This behaviour makes it rather pointless to try to avoid a song being played twice in a row in
random play. I always wanted to improve on that but couldn't find the time to learn the principles
of bookkeeping in rhythmbox. Maybe someone could give a hint?

Anyway: I think the patch of Alan is a great idea. I would like to have something like this in
rhythmbox.
@ Alan: I have a little setup here for testing the "randomness" of songs being played which gives
nice little graphs as output (sweet ;) ). If you like I could run it with your patch applied. Please
give me a note if this is of any help.

Cheers,

Alex


Dr. David Alan Gilbert schrieb:
> Hi,
>   I've just posted an improved version of my random patch onto
> bug 163196:
> 
> http://bugzilla.gnome.org/attachment.cgi?id=122538
> 
> Unlike the previous version I believe the maths actually now
> work well to summarise:
> 
> 
>   1) It adds a random pane on the preferences dialog
>   2) This allows you to select the behaviour when just 'shuffle' is
>      selected - either random or shuffle (for those who don't like the
>      odd UI trick of selection based on shuffle+repeat)
>   3) It combines the random-by-age, by-rating, by-age-and-rating all
>      into one chunk of code and selects the behaviour by a series of
>      sliders - age bias and rating bias.
>   4) A further bias is used for 'very recent' plays - and you can
>      define that; so for example you can set it up for plays within the
>      last 2 days are heavily penalised.
> 
> The only thing I think could improve is how to set the initial
> values of the sliders - I believe I have them in the schema - can
> anyone suggest how to improve that?
> 
> I find with the age-bias set to about 25% of the way it works well;
> if it's on max then you'll get little other than your very oldest
> tracks (or ones never played) - I've tried this out and it seems
> to work.
> 
> The patch is against the SVN version of 0.11.6 in Ubuntu Intrepid;
> but should go on pretty easily.
> 
> Any chance of this making it to a release?
> 
> All comments welcome.
> 
> Dave