fetchgit: split off deterministic-git functions

This commit is contained in:
Maciej Krüger 2021-12-04 14:24:17 +01:00
parent 7f96f015f3
commit ba83787f0d
No known key found for this signature in database
GPG key ID: 0D948CE19CF49C5F

View file

@ -0,0 +1,45 @@
#!/bin/sh
# some git commands print to stdout, which would contaminate our JSON output
clean_git(){
git "$@" >&2
}
# Remove all remote branches, remove tags not reachable from HEAD, do a full
# repack and then garbage collect unreferenced objects.
make_deterministic_repo(){
local repo="$1"
# run in sub-shell to not touch current working directory
(
cd "$repo"
# Remove files that contain timestamps or otherwise have non-deterministic
# properties.
rm -rf .git/logs/ .git/hooks/ .git/index .git/FETCH_HEAD .git/ORIG_HEAD \
.git/refs/remotes/origin/HEAD .git/config
# Remove all remote branches.
git branch -r | while read -r branch; do
clean_git branch -rD "$branch"
done
# Remove tags not reachable from HEAD. If we're exactly on a tag, don't
# delete it.
maybe_tag=$(git tag --points-at HEAD)
git tag --contains HEAD | while read -r tag; do
if [ "$tag" != "$maybe_tag" ]; then
clean_git tag -d "$tag"
fi
done
# Do a full repack. Must run single-threaded, or else we lose determinism.
clean_git config pack.threads 1
clean_git repack -A -d -f
rm -f .git/config
# Garbage collect unreferenced objects.
# Note: --keep-largest-pack prevents non-deterministic ordering of packs
# listed in .git/objects/info/packs by only using a single pack
clean_git gc --prune=all --keep-largest-pack
)
}