Hi,
CVS just saw three new abstractions for the [list]-abs collection, that have been inspired by the functional programming functions map, reduce and filter, known from Python and elsewhere. I attached the patches to this mail as well.
The basic idea of how to do this in Pd is by IOhannes and Thomas Musil. (IOhannes and I were discussing it at the LAC in Berlin.) Instead of having the function to map inside the list-abstraction, as I did it with [list-apply], now the operation to do on the list items is handled outside the abstraction through a cross-connection. You'll better understand what I mean by looking at the help files.
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__
Hallo, Frank Barknecht hat gesagt: // Frank Barknecht wrote:
CVS just saw three new abstractions for the [list]-abs collection, that have been inspired by the functional programming functions map, reduce and filter, known from Python and elsewhere. I attached the patches to this mail as well.
Att. is a small bugfix to list-filter.pd where an "a" was replaced by a "b" in a "t".
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__
hello frank, very nice abstractions. I recntly also started using the new features of lists. what I miss is a split -1 (a split with negative values, which would split by counting from the right side). I have some other ideas, like speedlim, which works different than the speedlim of thomas musil, it should receive input at any time, and spit it out in the same order, but timed like a metro. maybe you know, is there still a restriciton in the number of elements a list can carry? marius.
Frank Barknecht wrote:
Hallo, Frank Barknecht hat gesagt: // Frank Barknecht wrote:
CVS just saw three new abstractions for the [list]-abs collection, that have been inspired by the functional programming functions map, reduce and filter, known from Python and elsewhere. I attached the patches to this mail as well.
Att. is a small bugfix to list-filter.pd where an "a" was replaced by a "b" in a "t".
Ciao
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Hallo, marius schebella hat gesagt: // marius schebella wrote:
very nice abstractions. I recntly also started using the new features of lists. what I miss is a split -1 (a split with negative values, which would split by counting from the right side).
This is part of the [list]-abs collection under the name [list-splat]. It works exactly like [list split], but also allows negative indices for splitting off at the tail.
[list]-abs currently contains about 50 list processing objects, so you may want to take a look before starting to patch your own versions.
I have some other ideas, like speedlim, which works different than the speedlim of thomas musil, it should receive input at any time, and spit it out in the same order, but timed like a metro.
Maybe this is [list-fifo] from [list]-abs?
maybe you know, is there still a restriciton in the number of elements a list can carry?
Just try it: With [tabdump] you can easily create very large lists. list-map.pd contains an example that uses rather small 64-element lists, which work fine and reasonably fast.
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__
jesus, this was always in front of my nose! thnx! marius.
Frank Barknecht wrote:
Hallo, marius schebella hat gesagt: // marius schebella wrote:
very nice abstractions. I recntly also started using the new features of lists. what I miss is a split -1 (a split with negative values, which would split by counting from the right side).
This is part of the [list]-abs collection under the name [list-splat]. It works exactly like [list split], but also allows negative indices for splitting off at the tail.
[list]-abs currently contains about 50 list processing objects, so you may want to take a look before starting to patch your own versions.
I have some other ideas, like speedlim, which works different than the speedlim of thomas musil, it should receive input at any time, and spit it out in the same order, but timed like a metro.
Maybe this is [list-fifo] from [list]-abs?
maybe you know, is there still a restriciton in the number of elements a list can carry?
Just try it: With [tabdump] you can easily create very large lists. list-map.pd contains an example that uses rather small 64-element lists, which work fine and reasonably fast.
Ciao
-
Frank Barknecht -
marius schebella