`in set file` should be movd to EEG.datfile
neuromechanist opened this issue · comments
Concerning an issue with loading EEGLAB SET files in MNE-Python, when EEG.data does not point to the FDT file, I think the problem seems to root from the following lines:
eeglab/functions/popfunc/pop_loadset.m
Line 146 in f3570de
eeglab/functions/popfunc/pop_loadset.m
Line 152 in f3570de
Line 144 overwrites EEG.data if EEG.data is not numeric. But if EEG.data is not numeric, why should it be replaced with the in set file flag? And more importantly, where is the data?
Line 152 also poses a similar question. If EEG.datfile does not exist or is empty, EEG.data is overwritten with the in set file flag.
In both cases, it seems that the EEG.datfile should have been overwritten with the flag, not the EEG.data.
Overall, since this is only under info reading mode, which is not the default mode, and since there is another check for this flag a few lines down, leaving it as is should not cause much problem. The problem will be if the user loads the data in the info mode and saves the file right away (likely directly using Matlab, very unlikely).
This is for a load mode 'info' where the data stays on disk. I have looked at the lines but they seem OK.
There was a bug I fixed 6 months ago about ‘in set file’ and FDT file. I believe EEGLAB reads these files OK although they probably should be exported again for other software.
Closing this for now, but feel free to reopen.
I have also created an issue about ds003645