Back to the main page.
Bug 3091 - merge the two versions of matlablt2010b
| Status | NEW |
| Reported | 2016-03-09 15:15:00 +0100 |
| Modified | 2016-03-09 15:38:31 +0100 |
| Product: | FieldTrip |
| Component: | core |
| Version: | unspecified |
| Hardware: | PC |
| Operating System: | Mac OS |
| Importance: | P5 normal |
| Assigned to: | Robert Oostenveld |
| URL: | |
| Tags: | |
| Depends on: | |
| Blocks: | |
| See also: |
Robert Oostenveld - 2016-03-09 15:15:06 +0100
I noticed that we have a version in fileio and one in compat. I would only expect it to be present in compat, not in fileio. There are some differences. mac011> diff compat/matlablt2010b fileio/matlablt2010b Only in fileio/matlablt2010b: @uint64 Only in compat/matlablt2010b: iscolumn.m Only in compat/matlablt2010b: isrow.m The files are not considered in the bin/synchronize-private.sh script.
Robert Oostenveld - 2016-03-09 15:15:49 +0100
(In reply to Robert Oostenveld from comment #0) I suggest to remove the one from fileio and only keep the one in compat. Agreed?
nno - 2016-03-09 15:36:16 +0100
(In reply to Robert Oostenveld from comment #1) > I suggest to remove the one from fileio and only keep the one in compat. Agreed? Thanks Robert; I would agree it seems best to store the files only in compat and remove the fileio/matlablt2010b directory (after moving the @uint64 directory). As a side note: it would be good to re-compile the mex files in @uint64, in particular max.c and min.c, as the following does still crash Matlab on my machine: cd fileio/matlablt2010b x=uint64(zeros(0,1)) max(x) where the fix for the .c (but not .mex*) files was contributed earlier in commit 3e02cc6eda9acfcb6bc865c188b6ae4373de6989 [1] part of github PR #78 [2]. [1] https://github.com/fieldtrip/fieldtrip/commit/3e02cc6e [2] https://github.com/fieldtrip/fieldtrip/pull/78
Robert Oostenveld - 2016-03-09 15:37:15 +0100
(In reply to nno from comment #2) your machine is 64 bit linux?