Skip to content
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.

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

Make fetching source tarball by git commit more robust #4680

Open
wants to merge 2 commits into
base: develop
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
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
18 changes: 15 additions & 3 deletions easybuild/tools/filetools.py
Original file line number Diff line number Diff line change
Expand Up @@ -2701,10 +2701,15 @@ def get_source_tarball_from_git(filename, targetdir, git_config):
git_cmd = 'git'
clone_cmd = [git_cmd, 'clone']

if not keep_git_dir and not commit:
is_short_commit_hash = commit and len(commit) != 40
if keep_git_dir or is_short_commit_hash:
depth_args = []
else:
# Speed up cloning by only fetching the most recent commit, not the whole history
# When we don't want to keep the .git folder there won't be a difference in the result
clone_cmd.extend(['--depth', '1'])
# We also can't use it if only a short commit hash is given, as `git fetch` only works with full commit hashe
depth_args = ['--depth=1']
clone_cmd.extend(depth_args)

if tag:
clone_cmd.extend(['--branch', tag])
Expand All @@ -2716,7 +2721,8 @@ def get_source_tarball_from_git(filename, targetdir, git_config):
# checkout is done separately below for specific commits
clone_cmd.append('--no-checkout')

clone_cmd.append('%s/%s.git' % (url, repo_name))
repo_url = '%s/%s.git' % (url, repo_name)
clone_cmd.append(repo_url)

if clone_into:
clone_cmd.append(clone_into)
Expand All @@ -2731,6 +2737,12 @@ def get_source_tarball_from_git(filename, targetdir, git_config):

# if a specific commit is asked for, check it out
if commit:
# The commit might not be reachable from the default branch that is fetched, so fetch it explicitely
# Only works for long commit hashes
if not is_short_commit_hash:
fetch_cmd = [git_cmd, 'fetch'] + depth_args + [repo_url, commit]
run.run_cmd(' '.join(fetch_cmd), log_all=True, simple=True, regexp=False, path=repo_name)

checkout_cmd = [git_cmd, 'checkout', commit]

if recursive or recurse_submodules:
Expand Down
43 changes: 30 additions & 13 deletions test/framework/filetools.py
Original file line number Diff line number Diff line change
Expand Up @@ -2865,7 +2865,7 @@ def run_check():
}
git_repo = {'git_repo': '[email protected]:easybuilders/testrepository.git'} # Just to make the below shorter
expected = '\n'.join([
r' running command "git clone --depth 1 --branch tag_for_tests %(git_repo)s"',
r' running command "git clone --depth=1 --branch tag_for_tests %(git_repo)s"',
r" \(in .*/tmp.*\)",
r' running command "tar cfvz .*/target/test.tar.gz --exclude .git testrepository"',
r" \(in .*/tmp.*\)",
Expand All @@ -2874,7 +2874,7 @@ def run_check():

git_config['clone_into'] = 'test123'
expected = '\n'.join([
r' running command "git clone --depth 1 --branch tag_for_tests %(git_repo)s test123"',
r' running command "git clone --depth=1 --branch tag_for_tests %(git_repo)s test123"',
r" \(in .*/tmp.*\)",
r' running command "tar cfvz .*/target/test.tar.gz --exclude .git test123"',
r" \(in .*/tmp.*\)",
Expand All @@ -2884,7 +2884,7 @@ def run_check():

git_config['recursive'] = True
expected = '\n'.join([
r' running command "git clone --depth 1 --branch tag_for_tests --recursive %(git_repo)s"',
r' running command "git clone --depth=1 --branch tag_for_tests --recursive %(git_repo)s"',
r" \(in .*/tmp.*\)",
r' running command "tar cfvz .*/target/test.tar.gz --exclude .git testrepository"',
r" \(in .*/tmp.*\)",
Expand All @@ -2893,7 +2893,7 @@ def run_check():

git_config['recurse_submodules'] = ['!vcflib', '!sdsl-lite']
expected = '\n'.join([
' running command "git clone --depth 1 --branch tag_for_tests --recursive'
' running command "git clone --depth=1 --branch tag_for_tests --recursive'
+ ' --recurse-submodules=\'!vcflib\' --recurse-submodules=\'!sdsl-lite\' %(git_repo)s"',
r" \(in .*/tmp.*\)",
r' running command "tar cfvz .*/target/test.tar.gz --exclude .git testrepository"',
Expand All @@ -2907,7 +2907,7 @@ def run_check():
]
expected = '\n'.join([
' running command "git -c submodule."fastahack".active=false -c submodule."sha1".active=false'
+ ' clone --depth 1 --branch tag_for_tests --recursive'
+ ' clone --depth=1 --branch tag_for_tests --recursive'
+ ' --recurse-submodules=\'!vcflib\' --recurse-submodules=\'!sdsl-lite\' %(git_repo)s"',
r" \(in .*/tmp.*\)",
r' running command "tar cfvz .*/target/test.tar.gz --exclude .git testrepository"',
Expand All @@ -2928,11 +2928,27 @@ def run_check():
del git_config['keep_git_dir']

del git_config['tag']
git_config['commit'] = '90366eac4408c5d615d69c5444ed784734b167c8'
git_repo['commit'] = git_config['commit']
expected = '\n'.join([
r' running command "git clone --depth=1 --no-checkout %(git_repo)s"',
r" \(in .*/tmp.*\)",
r' running command "git fetch --depth=1 %(git_repo)s %(commit)s"',
r" \(in testrepository\)",
r' running command "git checkout %(commit)s && git submodule update --init --recursive"',
r" \(in testrepository\)",
r' running command "tar cfvz .*/target/test.tar.gz --exclude .git testrepository"',
r" \(in .*/tmp.*\)",
]) % git_repo
run_check()

# depth=1 clone & fetch does not work for short hashes
git_config['commit'] = '8456f86'
git_repo['commit'] = git_config['commit']
expected = '\n'.join([
r' running command "git clone --no-checkout %(git_repo)s"',
r" \(in .*/tmp.*\)",
r' running command "git checkout 8456f86 && git submodule update --init --recursive"',
r' running command "git checkout %(commit)s && git submodule update --init --recursive"',
r" \(in testrepository\)",
r' running command "tar cfvz .*/target/test.tar.gz --exclude .git testrepository"',
r" \(in .*/tmp.*\)",
Expand Down Expand Up @@ -2999,13 +3015,14 @@ def run_check():
self.assertTrue(os.path.isfile(os.path.join(extracted_repo_dir, 'this-is-a-tag.txt')))

del git_config['tag']
git_config['commit'] = '90366ea'
res = ft.get_source_tarball_from_git('test2.tar.gz', target_dir, git_config)
test_file = os.path.join(target_dir, 'test2.tar.gz')
self.assertEqual(res, test_file)
self.assertTrue(os.path.isfile(test_file))
test_tar_gzs.append(os.path.basename(test_file))
self.assertEqual(sorted(os.listdir(target_dir)), test_tar_gzs)
for i, commit in enumerate(['90366eac4408c5d615d69c5444ed784734b167c8', '90366ea']):
git_config['commit'] = commit
test_file = os.path.join(target_dir, 'test2-%s.tar.gz' % i)
res = ft.get_source_tarball_from_git(os.path.basename(test_file), target_dir, git_config)
self.assertEqual(res, test_file)
self.assertTrue(os.path.isfile(test_file))
test_tar_gzs.append(os.path.basename(test_file))
self.assertEqual(sorted(os.listdir(target_dir)), test_tar_gzs)

git_config['keep_git_dir'] = True
res = ft.get_source_tarball_from_git('test3.tar.gz', target_dir, git_config)
Expand Down
Loading