Brian Daurelle

Hi, I have a patch which counts through a pattern of notes stored in a Coll object. It can undergo various modifications as it plays, including counting through (and modifying) only certain sections. It's easy enough to reverse the entire list, but I'd like to also figure out how to reverse just the segment that's being counted through at any given moment. For example, if it's set to count through entries 2 through 5, I'd like this function to renumber the entries to 

Hi, I have a patch which counts through a pattern of notes stored in a Coll object. It can undergo various modifications as it plays, including counting through (and modifying) only certain sections. It's easy enough to reverse the entire list, but I'd like to also figure out how to reverse just the segment that's being counted through at any given moment. For example, if it's set to count through entries 2 through 5, I'd like this function to renumber the entries to 1 5 4 3 2 6 7 etc... 

Hi, I have a patch which counts through a pattern of notes stored in a Coll object. It can undergo various modifications as it plays, includ

how-to-reverse-a-segment-of-a-list-in-coll

do want to write it back or only play it live?

and you are looking for zl slice / zl ecils.


do want to write it back or only play it live?

and you are looking for zl slice / zl ecils.


@Roman I'm not sure how to answer... I believe both? Not sure what "write it back" means, I guess. The patch is designed for live use, any changes to the data in the coll are triggered when the pattern loops around, so that it seems to evolve as it plays.

To clarify: I'm not having trouble generating that string of numbers, but rather editing the coll. It seems there's no message to replace individual index numbers of entries? My first thought was "nsub $1 0 $2", but this does't seem to work.

This is important because each 'note' in the pattern contains several entries (pitch, velocity, etc) which all need to 'move' together when the renumbering occurs.

@Roman I'm not sure how to answer... I believe both? Not sure what "write it back" means, I guess. The patch is designed for live use, any changes to the data in the coll are triggered when the pattern loops around, so that it seems to evolve as it plays. 

To clarify: I'm not having trouble generating that string of numbers, but rather editing the coll. It seems there's no message to replace individual index numbers of entries? My first thought was "nsub $1 0 $2", but this does't seem to work. 

yeah that is what i meant, writing back into the coll where it comes from.

so... if i dont miss anytyhing more obvious... it seems best to break it up into two tasks: how to write it back into coll, and the processes. 

i personally would literally write the code for those two tasks separated from each other, and then stick them together when they both work.

behind the coll there would be a gate 2, live output vs "to be processed", and afaik the easiest way would be to

to read out and process the content of index 3,

but it depends on the context and the rest of the patch which method you will like. there is also "remove 3", which offers a different scenario.

"delete" only woks with numeric indexes, because only those will be automatically renumbered.


yeah that is what i meant, writing back into the coll where it comes from.

so... if i dont miss anytyhing more obvious... it seems best to break it up into two tasks: how to write it back into coll, and the processes. 
i personally would literally write the code for those two tasks separated from each other, and then stick them together when they both work.

behind the coll there would be a gate 2, live output vs "to be processed", and afaik the easiest way would be to

"3"

to read out and process the content of index 3,
then

"delete 3"

to remove the old version
then

"insert 3 a b c d e f g h"

to write the new version into index 3.

but it depends on the context and the rest of the patch which method you will like. there is also "remove 3", which offers a different scenario.

"delete" only woks with numeric indexes, because only those will be automatically renumbered.




I am not sure if I understand this correctly

Are you trying to renumber indexes only ?

I am not sure if I understand this correctly
Are you trying to renumber indexes only ?

in simplest form, reversing 2 - 3
1, 1 1 1 1;
2, 2 2 2 2;
3, 3 3 3 3;

becomes
1, 1 1 1 1; 
2, 3 3 3 3;
3, 2 2 2 2; 

is that correct ?


@Source Audio yes, that's basically what I'm trying to do. (This example might be too simple, though, because with two adjacent entries, the swap function would work fine, but for a longer segment, there's no simple command)

I see no problem reversing even large number of indexes,

but do you really want to destructively renumber data ?

then again 11 - 40, original coll will get all messed up.

Alternative would be to leave coll as is, and

But here is one way to do what you want :

I see no problem reversing even large number of indexes,
but do you really want to destructively renumber data ?
Let's say you reverse 5 - 22,
then again 11 - 40, original coll will get all messed up.

Alternative would be to leave coll as is, and
only modify counter.

aahhhh, I suspected it might take a second coll to store data temporarily, but I hoped there was a simpler way. 

And, thanks for your concern for the integrity of my data, but changing the underlying pattern is exactly the point! I have a mechanism built in to save and refer back to previous states of the coll, but the main purpose of the patch is to create a pattern that can undergo various kinds of transformations.

 aahhhh, I suspected it might take a second coll to store data temporarily, but I hoped there was a simpler way.  


And, thanks for your concern for the integrity of my data, but changing the underlying pattern is exactly the point! I have a mechanism built in to save and refer back to previous states of the coll, but the main purpose of the patch is to create a pattern that can undergo various kinds of transformations. 

But it is your party, enjoy it the way you prefer it

just add it to previous patch
it reverses indexes for counter
But it is your party, enjoy it the way you prefer it


oh, this is also very impressive! your original suggestion is actually more "in the spirit" of what the patch should do (cumulative destructive edits are, as i said, the whole point of the patch!), but this is very cool and useful as well! thanks

then you could randomise the list instead of simply reversing it,

then you could randomise the list instead of simply reversing it,
would couse more chaos


How to reverse a segment of a list (in coll)