ENH: Don't enforce BIDS & Raw channel name equivalence in read_raw_bids()

[hoechenberger]

Fixes #697

Please note that the same issue still exists for iEEG data in dig._handle_electrodes_reading(), but I don't fully understand the code and therefore currently don't dare to touch it. @adam2392 might want to give this one a shot 😇

cc @MaelysSolal

Merge checklist

Maintainer, please confirm the following before merging:

• All comments resolved
• This is not your own PR
• All CIs are happy
• PR title starts with [MRG]
• whats_new.rst is updated
• PR description includes phrase "closes <#issue-number>"

[agramfort]

to me it's the right thing to do !

[sappelhoff]

to me it's the right thing to do !

agreed, and eventually the validator should check that channel names in raw and BIDS are the same.

[agramfort]

validator would then be required to read EEG/MEG formats. It's too much I think. For me we should not enforce that the names are the same. We take the raw files as they are and the sidecar take precedence.

…

[hoechenberger]

to me it's the right thing to do !

agreed, and eventually the validator should check that channel names in raw and BIDS are the same.

Yes I'm confused here, @sappelhoff said he agrees with you, but then suggests something that would enforce the opposite :) So just to clear up the confusion: Do you think this PR is doing the right thing, @sappelhoff?

[sappelhoff]

Yes it's the right thing to take the sidecar files as the truth.

The validator checking that raw and sidecar info is aligned would be more of a help to dataset curators to get their story straight (less room for error). --> but that's future talk ... and as Alex says, it may even never happen because we have nobody developing dataset readers in JS.