PHysIO #180 #166: Removing fieldtrip paths instead of spm

[nbeliy]

Hi,
I was looking into bypassing issues #180 and #166,
and found that PhysIO already tries to remove sub-directories of spm, but, probably due to outdated version it doesn't work on my installation (my FirledTrip don't have ft_check_path).

In this pull request, I replace ft_check_path by ft_defaults, which is setting paths function,which exists since 2010, and
I remove paths related to all installations of fieldtrip:

oldPath = path();

pathThis = fileparts(mfilename('fullpath'));
ft_paths = which('ft_defaults', '-all')

try
for p = 1:size(ft_paths, 1)
    to_remove = fileparts(ft_paths{p});
    warning('off', 'MATLAB:rmpath:DirNotFound')
    rmpath(genpath(to_remove));
    warning('on', 'MATLAB:rmpath:DirNotFound')
end
ft_paths = which('ft_defaults', '-all')

After execution, the original paths are restored.

Alternative, is to remove all paths, using restoredefaultpaths, to inshure that PhysIO will run only on bare matlab.

I also added an otherwise with error in switch for the list of vendors.

[mrikasper]

Dear Nikita,

Thank you for suggesting these changes. I made two comments to your implementation, hope you can see them.

Furthermore, could you please also add a version of the Contributor License Agreement file to this pull request that has a table row with your name and info inserted. I have explained the rationale for that here and here.

You might want to wait a few days with that (so that you don't have to do it twice), because we're just in the process of finalizing a release, and it's probably easier to start from the new version.

All the best, and looking forward to your changes,
Lars

[nbeliy]

It's not urgent, tell me when it could be merged.
I left in he code some print-outs, I will obviously remove them.

[nbeliy]

Hi Lars, as requested a ping after the new release)

In current version, the fieldtrip paths are removed with message:

Warning: Removing FieldTrip paths (https://github.com/translationalneuromodeling/tapas/issues/166)
	/home/beliy/Documents/MATLAB/Add-Ons/spm/SPM12/external/fieldtrip

You can restore FieldTrip path using:
	addpath('/home/beliy/Documents/MATLAB/Add-Ons/spm/SPM12/external/fieldtrip');
	ft_defaults;

Second message appears only with verbosity >= 3.

About the reported conflict, the only revelant part of conflict is addpath:

% Include subfolders of code to path as well
pathThis = fileparts(mfilename('fullpath'));
addpath(genpath(pathThis)); 

For me it should be done during installation of PhysIO, and crosschecked in tapas_physio_init.m
But if you find it necessary, I can restore it.
The spm check are replaced by fieldtrip checks.

And the matlabbatch check is not important: if tapas_main is called via matlabbatch, then PhysIO is integrated into spm.
If main is called not via matlabbatch, then the check is not useful.

For me, the full tapas_physio_check_spm.m becomes obsolete.

Tell me, further modifications are required.

[mrikasper]

Dear Nikita,

My apologies, but I didn't have time to include your PR into the upcoming TAPAS release. I am working with a student on a summer project to improve PhysIO, but it will take a couple of weeks to overhaul everything. I will let you know, once we integrate your PR.

All the best,
Lars