Hm, originally I was thinking this:
[struct fifo float head float tail array a fifo-datum]

[struct fifo-datum float flag float f symbol s]

for a fifo of single-atom messages.

But for a fifo of Pd messages I'm talking about this:
[struct fifo float head float tail array a fifo-datum]

[struct fifo-datum text t]

-Jonathan







On Sunday, November 22, 2015 1:35 PM, Matt Barber <brbrofsvl@gmail.com> wrote:


​You'd store the number of elements in the list at the first index of a fifo entry and the number of elements in the symbol or float just before the data for each atom -- on read you'd accumulate the atoms of the list and stop. The next index would be the number of elements in the next fifo list entry.

On Sun, Nov 22, 2015 at 12:46 PM, Jonathan Wilkes via Pd-list <pd-list@lists.iem.at> wrote:
I might be reading it wrong, but Matt's description sounds like a way to queue atoms, not Pd messages.

With the "text" field of data structures it might be possible to add Pd messages to that design--
that is, you could have a data structure array where each element is a text field.  But
Pd is still going to have to allocate memory for each text field as you add it to the queue.

-Jonathan






On Sunday, November 22, 2015 11:50 AM, Roman Haefeli <reduzent@gmail.com> wrote:


On Sun, 2015-11-22 at 16:24 +0000, Jonathan Wilkes wrote:
> Are you trying to add Pd messages to a queue of Pd messages, or create
> a fifo
>
> for atoms?

I was/am interested in a FIFO implementation for Pd messages of any size
(containing symbols and/or floats).

Why are you asking? Do you know a way to works well specifically for
single atoms?


Roman

_______________________________________________
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list



_______________________________________________
Pd-list@lists.iem.at mailing list
UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list