[PD] Another bug in [pool]?

Phil Stone pkstone at ucdavis.edu
Sun Mar 16 19:33:02 CET 2008


Luke,

As I mess around with rradical, I'm starting to use [pool] a lot more, 
and this post of yours from last month was a little worrisome; was there 
any resolution?


Phil Stone
pkstonemusic.com



Luke Iannini (pd) wrote:
> Hi all,
> I've been staring at this patch for a couple hours with no luck, so
> I'm ready to ask for help.
>
> I seem to be running into another bug in pool.
>
> What's happening is I'm setting a key with 3 floats.  I then try to
> [get( that key, but can't.  If I [printall(, I can see the key is in
> the [pool].  But it won't come out.
>
> Even stranger, I can compose a new [set {key} 123 123 123( message and
> send it to the same [pool], and instead of replacing the key, it
> appears alongside it.  I can get the newly set key, but the old one
> remains in the [pool], inaccessible.
>
> Since the usual behavior of pool would be to overwrite the first key
> when a key of the same name is sent, I'm guessing I'm hitting a very
> obscure bug again.
>
> Here's the patch, I'll keep trying to isolate it further.
>
> Cheers
> Luke
>
> To repro: click "reset" at the top left, then click the [77 120(
> message, then [77 0( immediately after that.  At that point, a message
> like
>
>  , 777.868 , 77 120 539.864
>
> will be in the pool $0notePool (click the [getall( message to
> confirm).  Now, send a message [get 777.868( (or whatever the number
> is), and if you're anything like me, you'll see... nothing!
>
> Now, make a message [set 777.868 5 5 5( and send it to the pool, then
> click your [get 777.868( again.  Out will pop 5, 5, 5, but if you
> click [getall( again you'll see
>  , 777.868 , 5 5 5
>  , 777.868 , 77 120 539.864
>
> You'll now notice a [pool] of tears.
>   
> ------------------------------------------------------------------------
>
> _______________________________________________
> PD-list at iem.at mailing list
> UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
>   





More information about the Pd-list mailing list