fix eyelink.valid handling when .loc_useRawData == true

[dshimaoka]

When eyelink plugin is operated with .loc_useRawData = false, .valid stays always false even during blinking. This causes unintended behaviour of eyeMovement.isInWindow, always returning false.

The unintended behavior of .valid stems from its definition: o.valid = any(sample.gx(eyeNr+1)~=o.el.MISSING_DATA); When operated with .loc_useRawData = true, gx always returns el.MISSING_DATA (=-32768). Thus .valid stays false.

Instead, px returns sensible values during successful eye tracking and returns el.MISSING_DATA (=-32768) only during unsuccessful eye tracking such us during blinking.

Hence, a simple fix to this problem is to define .valid by px as:

if o.loc_useRawData
o.valid  = any(sample.px(eyeNr+1)~=o.el.MISSING_DATA);
end

[adammorrissirrommada]

This looks right to me, but I will leave it for @cnuahs to confirm.

[adammorrissirrommada]

@dshimaoka, thanks for this pull request. for future PRs, please add an assignee (Shaun, Bart, or me) and at least one reviewer (can be the same as assignee if you like) so that it is on someone's desk as a to-do.

[dshimaoka]

@dshimaoka, thanks for this pull request. for future PRs, please add an assignee (Shaun, Bart, or me) and at least one reviewer (can be the same as assignee if you like) so that it is on someone's desk as a to-do.

To @adammorrissirrommada I suspect I am not authorized to add an assignee (https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/proposing-changes-to-your-work-with-pull-requests/requesting-a-pull-request-review)

[adammorrissirrommada]

@dshimaoka,

Ah, I see. in that case, just tag one of us in a comment to ask us to assign it. thanks.

[adammorrissirrommada]

@bartkrekelberg we're OK with this change, but giving you final say over the merge because it affects a fundamental plugin.