def test_checkout_consecutive_commits(repo: Git):
repo.checkout('a7053a4dcd627f5f4f213dc9aa002eb1caf926f8')
repo.checkout('f0dd1308bd904a9b108a6a40865166ee962af3d4')
repo.checkout('9e71dd5726d775fb4a5f08506a539216e878adbb')
files3 = repo.files()
assert len(files3) == 3
repo.reset()
def test_commit_in_master_branch(repo: Git):
assert repo.get_head().hash == '29e929fbc5dc6a2e9c620069b24e2a143af4285f'
repo.checkout('8986af2a679759e5a15794f6d56e6d46c3f302f1')
git_to_change_head = Git('test-repos/branches_merged')
commit = git_to_change_head.get_commit('8169f76a3d7add54b4fc7bca7160d1f1eede6eda')
assert commit.in_main_branch is False
repo.reset()
assert repo.get_head().hash == '29e929fbc5dc6a2e9c620069b24e2a143af4285f'
def test_checkout_with_commit_not_fully_merged_to_master(repo: Git):
repo.checkout('developing')
files1 = repo.files()
assert len(files1) == 2
repo.reset()
files2 = repo.files()
assert len(files2) == 1
repo.checkout('developing')
files1 = repo.files()
assert len(files1) == 2
repo.reset()
def test_checkout_with_commit_not_fully_merged_to_master(repo: Git):
repo.checkout('developing')
files1 = repo.files()
assert len(files1) == 2
repo.reset()
assert 4, "temp branch should be cleared." == len(repo.repo.branches)
files2 = repo.files()
assert len(files2) == 1
repo.checkout('developing')
files1 = repo.files()
assert len(files1) == 2
repo.reset()
def extract(self,
labeled_files: List[FailureProneFile],
product: bool = True,
process: bool = True,
delta: bool = False):
""" Extract metrics from labeled files.
Parameters
----------
labeled_files : List[FailureProneFile]
The list of FailureProneFile objects that are used to label a script as failure-prone (1) or clean (0).
product: bool
Whether to extract product metrics.
process: bool
Whether to extract process metrics.
delta: bool
Whether to extract delta metrics between two successive releases or commits.
"""
self.dataset = pd.DataFrame()
git_repo = Git(self.path_to_repo)
metrics_previous_release = dict() # Values for iac metrics in the last release
for commit in Repository(self.path_to_repo, order='date-order', num_workers=1).traverse_commits():
# To handle renaming in metrics_previous_release
for modified_file in commit.modified_files:
old_path = modified_file.old_path
new_path = modified_file.new_path
if old_path != new_path and old_path in metrics_previous_release:
# Rename key old_path wit new_path
metrics_previous_release[new_path] = metrics_previous_release.pop(old_path)
if commit.hash not in self.commits_at:
continue
# Else
git_repo.checkout(commit.hash)
process_metrics = {}
if process:
# Extract process metrics
i = self.commits_at.index(commit.hash)
from_previous_commit = commit.hash if i == 0 else self.commits_at[i - 1]
to_current_commit = commit.hash # = self.commits_at[i]
process_metrics = self.get_process_metrics(from_previous_commit, to_current_commit)
for filepath in self.get_files():
file_content = get_content(os.path.join(self.path_to_repo, filepath))
if not file_content or self.ignore_file(filepath, file_content):
continue
tmp = FailureProneFile(filepath=filepath, commit=commit.hash, fixing_commit='')
if tmp not in labeled_files:
label = 0 # clean
else:
label = 1 # failure-prone
metrics = dict(
filepath=filepath,
commit=commit.hash,
committed_at=str(commit.committer_date),
failure_prone=label
)
if process_metrics:
metrics['change_set_max'] = process_metrics['dict_change_set_max']
metrics['change_set_avg'] = process_metrics['dict_change_set_avg']
metrics['code_churn_count'] = process_metrics['dict_code_churn_count'].get(filepath, 0)
metrics['code_churn_max'] = process_metrics['dict_code_churn_max'].get(filepath, 0)
metrics['code_churn_avg'] = process_metrics['dict_code_churn_avg'].get(filepath, 0)
metrics['commits_count'] = process_metrics['dict_commits_count'].get(filepath, 0)
metrics['contributors_count'] = process_metrics['dict_contributors_count'].get(filepath, 0)
metrics['minor_contributors_count'] = process_metrics['dict_minor_contributors_count'].get(filepath, 0)
metrics['highest_contributor_experience'] = process_metrics[
'dict_highest_contributor_experience'].get(filepath, 0)
metrics['hunks_median'] = process_metrics['dict_hunks_median'].get(filepath, 0)
metrics['additions'] = process_metrics['dict_additions'].get(filepath, 0)
metrics['additions_max'] = process_metrics['dict_additions_max'].get(filepath, 0)
metrics['additions_avg'] = process_metrics['dict_additions_avg'].get(filepath, 0)
metrics['deletions'] = process_metrics['dict_deletions'].get(filepath, 0)
metrics['deletions_max'] = process_metrics['dict_deletions_max'].get(filepath, 0)
metrics['deletions_avg'] = process_metrics['dict_deletions_avg'].get(filepath, 0)
if product:
metrics.update(self.get_product_metrics(file_content))
if delta:
delta_metrics = dict()
previous = metrics_previous_release.get(filepath, dict())
for metric, value in previous.items():
if metric in ('filepath', 'commit', 'committed_at', 'failure_prone'):
continue
difference = metrics.get(metric, 0) - value
delta_metrics[f'delta_{metric}'] = round(difference, 3)
metrics_previous_release[filepath] = metrics.copy()
metrics.update(delta_metrics)
self.dataset = self.dataset.append(metrics, ignore_index=True)
git_repo.reset()
