Open edgarcosta opened 6 years ago
I agree that now is a good time if we're going to do it, but forcing everyone to rebase their code is pretty annoying (and if anyone merges we lose the benefit. It's only a couple megabytes....
Anyway, I'm pretty neutral.
We have a couple of big files in our history, that were added by mistake and then removed.
You can generate the top 20 big files that don't exist anymore by doing:
git rev-list --objects --all | git cat-file --batch-check='%(objecttype) %(objectname) %(objectsize) %(rest)' | awk '/^blob/ {print substr($0,6)}' | sort --numeric-sort --key=2 | tail -n 1000 | xargs -n 3 bash -c 'if [[ ! -e ${2} ]]; then awk -v size=${1} -v file=${2} "BEGIN{printf \"%.2f %s\\n\", size/(1024*1024), file}"; fi' | tail -n 20
Shall we try to remove them? If so, I believe that now is a great time to rewrite history.
GitHub gives us two options to do this: https://help.github.com/articles/removing-sensitive-data-from-a-repository/