[FieldTrip] Weird trialfun_general error

[Rodolphe Nenert]

Dear fieldtrippers,

i just noticed and found the origin of a weird error obtained with the
trialfun_general function, i thought it could be useful to send it here.
I use the ft_definetrial function on a file obtained using the merge
function in EEGlab. Apparently, this function is adding some triggers with
the value 'boundary' in the new file, indicating the old frontier between
the dataset that have been merged.
The funny part is that the trialfun_general() function in fieltrip is using
the intersect() matlab function to match event values. Moreover, im using 27
different event values to indicate point of interest
(112,114,116,122,etc...)
All of my event values dont create any problems except the value 114
because intersect('boundary',114) = 114 whereas it gives an empty result
with any other of my values...
Therefore, for that particular event value, there was a dimension mismatch
between trl and val in trialfun_general.

I just removed events with 'boundary' values and it worked.

Best regards,

Rodolphe Nenert, Ph.D.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20110606/63401cda/attachment-0001.html>