Switches SPM version from dev to stable r7219

[miykael]

As discussed here (ReproNim/neurodocker#228), I'm curious if it's not better if we switch to a stable SPM12 version in the tutorial. I'm worried that otherwise users will use the dev version for their scientific analysis. Which would be bad, as those versions might just disappear at one point.

What do others think?

[miykael]

The checks are not all passing. Not sure why it gives an "all green". But the last issue should have been taken care of with ReproNim/neurodocker#228.
Will send a commit with the update in a second.

[miykael]

It seems that this setup/environment doesn't work with octave (which we use in the create interface section.

If you try to use octave you get the following error:

octave: X11 DISPLAY environment variable not set
octave: disabling GUI features
/usr/bin/octave-cli: /opt/matlabmcr-2010a/v713/sys/os/glnxa64/libgcc_s.so.1: version `GCC_4.3.0' not found (required by /usr/lib/x86_64-linux-gnu/libgfortran.so.3)

The error can be found in this circleci log .

I assume that the version of octave doesn't work with the MATLAB_MCR version 2010a.

I therefore decided to close this PR again. The change from SPM12 version dev to r7219 was just to prevent that users use dev for their scientific analysis. The functionality should otherwise be the same.

[miykael]

Thanks to @kaczmarj's help I was able to solve this issue. I just had to force LD_LIBRARY_PATH to take the libgcc_s.so.1 from /opt/miniconda-latest/envs/neuro/lib.

[miykael]

The advanced_create_interfaces.ipynb notebook is passing. This is the only notebook that needs octave. Therefore the commit 0284f21 took care of the libgcc_s.so.1 library issue.

I will merge this commit to master.

[miykael]

Not sure where to post this best, but switching SPM version from dev to r7219 reduced the size of the docker image from 7.46GB to 5.29GB. This is due to the fact that MCR 2018B is 2.8GB and MCR version 2010a is only 520MB big.

@kaczmarj, this might be a "good to know" fact for you.

[kaczmarj]

thanks @miykael - this is great to know!