Reading non-contiguous channels in openNSx no longer works

[sbrincat]

openNSx used to return only the channels that were requested via the channels argument, even if they were a non-contiguous sequence. Now it returns the entire block of channels from min(channels) : max(channels).

I think it is related to this (line 1117-1119):

%% Remove extra channels that were read, but weren't supposed to be read
% Commenting this section out since I think that previous code should
% capture this - DK 20230303

Example: nsx = openNSx(datafile, 'read', 'channels',1:2:8);

Previous behavior: <returned data from 1st, 3rd, 5th, and 7th channels in nsx> New behavior: <returns data from all channels from 1st through 8th channel in nsx>

Bug discovered by our user @jeffersonroy

[cboulay]

Please try out the PR in #62 to see if it fixes your issue.

[dkluger]

@sbrincat we only changed openNSx in the PR @cboulay mentioned. If you don't feel like recloning the repo and just want to try our fixes to openNSx.m, you can find that file here: https://github.com/BlackrockNeurotech/NPMK/blob/development/NPMK/openNSx.m

Please let us know if that change addressed your issues. We'll merge development to the main branch if you confirm this issue and #60 were properly addressed

[sskellis]

Hi @sbrincat, thank you for all of the feedback. Can you take a look at the most recent commit in #62?

[sbrincat]

Spencer -- #62 works well for us, our issues are solved. Thanks for the fix.

On Mon, Feb 5, 2024 at 2:49 PM Spencer Kellis @.***> wrote:

Hi @sbrincat https://github.com/sbrincat, thank you for all of the feedback. Can you take a look at the most recent commit in #62 https://github.com/BlackrockNeurotech/NPMK/pull/62?

— Reply to this email directly, view it on GitHub https://github.com/BlackrockNeurotech/NPMK/issues/61#issuecomment-1927956837, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADMUGZPF4YQMK7EFAB5E623YSEZVTAVCNFSM6AAAAABCQFUSJOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMRXHE2TMOBTG4 . You are receiving this because you were mentioned.Message ID: @.***>

[sskellis]

Excellent, glad to hear that. Don't hesitate to reach back out if any other issues crop up.