extra "empy" column in participants.tsv of eyetracking_binocular: warning or error or nothing?

[Remi-Gau]

see
https://github.com/bids-standard/bids-examples/blob/master/eyetracking_binocular/participants.tsv

more clear when loading

>>> import pandas as pd
>>> pd.read_c
pd.read_clipboard(  pd.read_csv(        
>>> df= pd.read_csv('/home/remi-gau/github/bids/matlab/tests/bids-examples/eyetracking_binocular/participants.tsv', sep='\t')
>>> df
  participant_id  Unnamed: 1
0         sub-01         NaN

pandas is fine the but the TSV parser used by bids-matlab did not like this and crashed miserably

not sure if the validator warns about this type of edge case in TSV or if it should

We can say this is a downstream issue and so I can make the bids-matlab TSV parser more robust (which I may do anyway) but if not it may require a fix in the bids-eexample repo

[effigies]

	[WARNING] TSV_ADDITIONAL_COLUMNS_UNDEFINED A TSV file has extra columns which are not defined in its associated JSON sidecar
		/participants.tsv

But I think probably we should catch empty strings for headers. Not sure if there's a good way to specify this in the schema at the moment... Maybe match(columns, '.') == length(columns), but that would need us to define (and possibly implement) the semantics that an object passed to a function that accepts arrays interprets the keys of the object as an array. Works by default in Python, I suspect not in JS.

[effigies]

Anyway, +1 for fixing this in the examples, and also making whatever changes needed to the schema/validator to make this an error.

[Remi-Gau]

Anyway, +1 for fixing this in the examples, and also making whatever changes needed to the schema/validator to make this an error.

will open PR as draft on the example: I suspect we may want to confirm we get an error with the old example once the schema / validator have been adapted