My programs generally generate huge output files (~1 GB) which I do not want to be backing up to the git repository. So instead of being able to do
git add .
I have to do something like
git add *.c *.cc *.f *.F *.C *.h *.cu
which is a little bit cumbersome...
I feel fairly confident I could write a quicky perl script ls the directory contents into .gitignore and then remove files based on a .gitinclude (or some similar name) file, but that seems a little too hackish. Is there a better way?