Skip to content

git-export: Reset local content to remote content on fresh clones #1144

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
leplatrem merged 2 commits into from
Jan 21, 2026
Merged

git-export: Reset local content to remote content on fresh clones #1144

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
21d9b0b
git-export: Reset local content to remote content on fresh clones
leplatrem Jan 20, 2026
c8bab55
Do not create local branches in tests setup
leplatrem Jan 20, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
43 changes: 29 additions & 14 deletions cronjobs/src/commands/_git_export_git_tools.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,32 +35,47 @@ def clone_or_fetch(
print("Head was at", repo.head.target)
print(f"Fetching from {repo_url}...")
remote.fetch(callbacks=callbacks, prune=True)
reset_repo(repo, callbacks=callbacks)
else:
# Clone remote repository into work dir.
print(f"Clone {repo_url} into {repo_path}...")
pygit2.clone_repository(repo_url, repo_path, callbacks=callbacks)
repo = pygit2.Repository(repo_path)
reset_repo(repo, callbacks=callbacks)
return repo


def reset_repo(repo: pygit2.Repository, callbacks: pygit2.RemoteCallbacks):
print("Reset local content to remote content...")
# Reset all local branches to their remote
for branch_name in repo.branches.local:
remote_ref_name = f"{REMOTE_NAME}/{branch_name}"
if remote_ref_name not in repo.branches:
print(f"Delete local branch {branch_name}")
repo.branches.delete(branch_name)
else:
local_branch = repo.branches[branch_name]
remote_branch = repo.branches[remote_ref_name]
if local_branch.target != remote_branch.target:
# Reset local branch to remote target
# If the repo is freshly cloned, the remotes branches do not exist locally. Create them.
# If the repo was cloned from previous run, reset the local branches to the remote targets.
# Remote always wins.

for branch_name in repo.branches.remote:
branch = repo.branches.remote[branch_name]
assert branch.remote_name == REMOTE_NAME
remote_target = branch.target
local_branch_name = branch_name.removeprefix(f"{REMOTE_NAME}/")
local_refname = f"refs/heads/{local_branch_name}"
if local_refname in repo.references:
local_ref = repo.lookup_reference(local_refname)
if local_ref.target != remote_target:
print(
f"Resetting local branch {branch_name} to remote {remote_ref_name}"
f"Resetting local branch {local_branch_name} to remote {branch_name}"
)
local_ref.set_target(
remote_target,
f"reset to {REMOTE_NAME}/{branch_name}",
)
local_branch.set_target(remote_branch.target)
else:
repo.create_reference(local_refname, remote_target)

# Delete local branches that are not on remote
for branch_name in repo.branches.local:
remote_branch_name = f"{REMOTE_NAME}/{branch_name}"
if remote_branch_name not in repo.branches.remote:
ref = repo.lookup_reference(f"refs/heads/{branch_name}")
print(f"Delete local branch {branch_name}")
ref.delete()

# Delete local tags that are not on remote
origin = repo.remotes[REMOTE_NAME]
Expand Down
21 changes: 11 additions & 10 deletions cronjobs/tests/commands/test_git_export.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -223,7 +223,7 @@ def init_fake_repo(path):
return repo


def create_branch_with_empty_commit(repo, branch_name):
def create_branch_with_empty_commit(repo, branch_name, set_as_repo_head=False):
author = pygit2.Signature("Test User", "test@example.com")
builder = repo.TreeBuilder()
tree = builder.write()
Expand All @@ -237,10 +237,10 @@ def create_branch_with_empty_commit(repo, branch_name):
)
commit = repo[commit_id]

repo.branches.local.create(branch_name, commit)
refname = f"refs/remotes/origin/{branch_name}"
repo.references.create(refname, commit.id)
repo.set_head(f"refs/heads/{branch_name}")
if set_as_repo_head:
repo.set_head(branch_name)


def simulate_pushed(repo, mock_ls_remotes):
Expand Down Expand Up @@ -282,6 +282,7 @@ def test_remote_is_clone_if_dir_missing(
mock_truncate_branch,
mock_github_lfs,
mock_git_push,
mock_ls_remotes,
):
def _fake_clone(url, path, *args, **kwargs):
return init_fake_repo(path)
Expand Down Expand Up @@ -353,7 +354,7 @@ def test_repo_sync_does_nothing_if_up_to_date(
mock_github_lfs,
mock_git_push,
):
create_branch_with_empty_commit(repo, "v1/common")
create_branch_with_empty_commit(repo, "v1/common", set_as_repo_head=True)
create_branch_with_empty_commit(repo, "v1/buckets/bid1")
create_branch_with_empty_commit(repo, "v1/buckets/bid2")

Expand Down Expand Up @@ -381,7 +382,7 @@ def test_repo_sync_can_be_forced_even_if_up_to_date(
mock_github_lfs,
mock_git_push,
):
create_branch_with_empty_commit(repo, "v1/common")
create_branch_with_empty_commit(repo, "v1/common", set_as_repo_head=True)
create_branch_with_empty_commit(repo, "v1/buckets/bid1")
create_branch_with_empty_commit(repo, "v1/buckets/bid2")

Expand All @@ -408,13 +409,13 @@ def test_repo_sync_content_uses_previous_run_to_fetch_changes(
mock_github_lfs,
mock_git_push,
):
create_branch_with_empty_commit(repo, "v1/common")
create_branch_with_empty_commit(repo, "v1/common", set_as_repo_head=True)
create_branch_with_empty_commit(repo, "v1/buckets/bid1")
create_branch_with_empty_commit(repo, "v1/buckets/bid2")

repo.create_tag(
"v1/timestamps/common/1600000000000",
repo.lookup_reference("refs/heads/v1/common").target,
repo.head.target,
pygit2.GIT_OBJECT_COMMIT,
pygit2.Signature("Test User", "test@example.com"),
"Test tag at 1600000000000",
Expand Down Expand Up @@ -456,13 +457,13 @@ def test_repo_sync_content_ignores_previous_run_if_forced(
mock_github_lfs,
mock_git_push,
):
create_branch_with_empty_commit(repo, "v1/common")
create_branch_with_empty_commit(repo, "v1/common", set_as_repo_head=True)
create_branch_with_empty_commit(repo, "v1/buckets/bid1")
create_branch_with_empty_commit(repo, "v1/buckets/bid2")

repo.create_tag(
"v1/timestamps/common/1600000000000",
repo.lookup_reference("refs/heads/v1/common").target,
repo.head.target,
pygit2.GIT_OBJECT_COMMIT,
pygit2.Signature("Test User", "test@example.com"),
"Test tag at 1600000000000",
Expand Down Expand Up @@ -746,7 +747,7 @@ def test_repo_is_resetted_to_local_content_on_error(
mock_git_push,
mock_ls_remotes,
):
create_branch_with_empty_commit(repo, "v1/common")
create_branch_with_empty_commit(repo, "v1/common", set_as_repo_head=True)
create_branch_with_empty_commit(repo, "v1/buckets/bid1")
create_branch_with_empty_commit(repo, "v1/buckets/bid2")

Expand Down
85 changes: 85 additions & 0 deletions cronjobs/tests/commands/test_git_export_git_tools.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,11 +1,14 @@
import time
from unittest import mock

import pygit2
import pytest
from commands import git_export
from commands._git_export_git_tools import (
delete_old_tags,
iter_tree,
parse_lfs_pointer,
reset_repo,
tree_upsert_blobs,
truncate_branch,
)
Expand Down Expand Up @@ -33,9 +36,17 @@ def tmp_repo(tmp_path):
tree_oid,
[],
)
repo.remotes.create("origin", git_export.GIT_REMOTE_URL)
return repo


@pytest.fixture
def mock_ls_remotes():
with mock.patch("pygit2.Remote.ls_remotes") as mock_ls:
mock_ls.return_value = []
yield mock_ls


def test_valid_lfs_pointer():
data = b"""version https://git-lfs.github.com/spec/v1
oid sha256:1234567890abcdef1234567890abcdef1234567890abcdef1234567890abcdef
Expand Down Expand Up @@ -71,6 +82,80 @@ def test_iter_tree_single_file(tmp_repo):
]


def test_reset_repo_creates_local_branches(tmp_repo, mock_ls_remotes):
repo = tmp_repo
commit = repo.revparse_single("main")
# remote branch can be created via reference only.
repo.create_reference(
"refs/remotes/origin/v1/buckets/main",
commit.id,
force=True,
)
assert "v1/buckets/main" not in repo.branches.local

reset_repo(repo, callbacks=None)

local_ref = repo.lookup_reference("refs/heads/v1/buckets/main")
remote_ref = repo.lookup_reference("refs/remotes/origin/v1/buckets/main")
assert local_ref.target == remote_ref.target


def test_reset_repo_resets_local_branches_to_remote(tmp_repo, mock_ls_remotes):
repo = tmp_repo
commit = repo.revparse_single("main")
author = committer = pygit2.Signature("Test", "test@example.com")
repo.create_commit(
"refs/heads/v1/buckets/main",
author,
committer,
"diverging commit",
repo.TreeBuilder().write(),
[commit.id],
)
# remote branch can be created via reference only.
repo.create_reference(
"refs/remotes/origin/v1/buckets/main",
commit.id,
force=True,
)
# With this local commit, the branches have diverged.
local_ref = repo.lookup_reference("refs/heads/v1/buckets/main")
remote_ref = repo.lookup_reference("refs/remotes/origin/v1/buckets/main")
assert local_ref.target != remote_ref.target

reset_repo(repo, callbacks=None)

# Now they match.
local_ref = repo.lookup_reference("refs/heads/v1/buckets/main")
remote_ref = repo.lookup_reference("refs/remotes/origin/v1/buckets/main")
assert local_ref.target == remote_ref.target


def test_reset_repo_deletes_extra_local_branches_and_tags(tmp_repo, mock_ls_remotes):
repo = tmp_repo
some_target = repo.references["refs/heads/main"].target
repo.create_reference("refs/remotes/origin/v1/buckets/main", some_target)
repo.create_reference("refs/heads/v1/buckets/main", some_target)
assert "v1/buckets/main" in repo.branches.local
# Create another ref that wouldn't be repo's head (to allow delete)
author = committer = pygit2.Signature("Test", "test@example.com")
commit_id = repo.create_commit(
"refs/heads/v1/buckets/main",
author,
committer,
"initial commit",
repo.TreeBuilder().write(),
[some_target],
)
# Create an extra branch.
repo.create_reference("refs/heads/v1/buckets/unknown", commit_id)

reset_repo(repo, callbacks=None)

assert "v1/buckets/main" in repo.branches.local
assert "v1/buckets/unknown" not in repo.branches.local


def test_delete_old_tags(tmp_repo):
repo = tmp_repo
now_ts = int(time.time())
Expand Down
Loading