def eval_regression(self):
results = {}
model = RegressionModel.load('regressionmodel')
for bug in bugzilla.get_bugs():
if self.is_regression(bug):
results[bug['id']] = True
elif self.is_feature(bug):
results[bug['id']] = False
else:
results[bug['id']] = True if model.classify(bug)[0] == 1 else False
with open('regression.json', 'w') as f:
json.dump(results, f)
def eval_regression(self):
results = {}
model = RegressionModel.load('regressionmodel')
for bug in bugzilla.get_bugs():
if self.is_regression(bug):
results[bug['id']] = True
elif self.is_feature(bug):
results[bug['id']] = False
else:
results[bug['id']] = True if model.classify(
bug)[0] == 1 else False
with open('regression.json', 'w') as f:
json.dump(results, f)
def __init__(self):
super(Regression, self).__init__()
self.model = RegressionModel.load(self.retrieve_model('regression'))
self.autofix_regression = []
def find_bug_fixing_commits(self):
logger.info("Downloading commits database...")
if db.is_old_version(
repository.COMMITS_DB) or not db.exists(repository.COMMITS_DB):
db.download(repository.COMMITS_DB, force=True)
logger.info("Downloading bugs database...")
if db.is_old_version(
bugzilla.BUGS_DB) or not db.exists(bugzilla.BUGS_DB):
db.download(bugzilla.BUGS_DB, force=True)
logger.info("Download previous classifications...")
if db.is_old_version(
BUG_FIXING_COMMITS_DB) or not db.exists(BUG_FIXING_COMMITS_DB):
db.download(BUG_FIXING_COMMITS_DB, force=True)
logger.info("Get previously classified commits...")
prev_bug_fixing_commits = list(db.read(BUG_FIXING_COMMITS_DB))
prev_bug_fixing_commits_nodes = set(
bug_fixing_commit["rev"]
for bug_fixing_commit in prev_bug_fixing_commits)
logger.info(
f"Already classified {len(prev_bug_fixing_commits)} commits...")
# TODO: Switch to the pure Defect model, as it's better in this case.
logger.info("Downloading defect/enhancement/task model...")
download_model("defectenhancementtask")
defect_model = DefectEnhancementTaskModel.load(
"defectenhancementtaskmodel")
logger.info("Downloading regression model...")
download_model("regression")
regression_model = RegressionModel.load("regressionmodel")
start_date = datetime.now() - RELATIVE_START_DATE
end_date = datetime.now() - RELATIVE_END_DATE
logger.info(
f"Gathering bug IDs associated to commits (since {start_date} and up to {end_date})..."
)
commit_map = defaultdict(list)
for commit in repository.get_commits():
if commit["node"] in prev_bug_fixing_commits_nodes:
continue
commit_date = dateutil.parser.parse(commit["pushdate"])
if commit_date < start_date or commit_date > end_date:
continue
commit_map[commit["bug_id"]].append(commit["node"])
logger.info(
f"{sum(len(commit_list) for commit_list in commit_map.values())} commits found, {len(commit_map)} bugs linked to commits"
)
assert len(commit_map) > 0
def get_relevant_bugs():
return (bug for bug in bugzilla.get_bugs()
if bug["id"] in commit_map)
bug_count = sum(1 for bug in get_relevant_bugs())
logger.info(
f"{bug_count} bugs in total, {len(commit_map) - bug_count} bugs linked to commits missing"
)
known_defect_labels = defect_model.get_labels()
known_regression_labels = regression_model.get_labels()
bug_fixing_commits = []
def append_bug_fixing_commits(bug_id, type_):
for commit in commit_map[bug_id]:
bug_fixing_commits.append({"rev": commit, "type": type_})
for bug in tqdm(get_relevant_bugs(), total=bug_count):
# Ignore bugs which are not linked to the commits we care about.
if bug["id"] not in commit_map:
continue
# If we know the label already, we don't need to apply the model.
if (bug["id"] in known_regression_labels
and known_regression_labels[bug["id"]] == 1):
append_bug_fixing_commits(bug["id"], "r")
continue
if bug["id"] in known_defect_labels:
if known_defect_labels[bug["id"]] == "defect":
append_bug_fixing_commits(bug["id"], "d")
else:
append_bug_fixing_commits(bug["id"], "e")
continue
if defect_model.classify(bug)[0] == "defect":
if regression_model.classify(bug)[0] == 1:
append_bug_fixing_commits(bug["id"], "r")
else:
append_bug_fixing_commits(bug["id"], "d")
else:
append_bug_fixing_commits(bug["id"], "e")
db.append(BUG_FIXING_COMMITS_DB, bug_fixing_commits)
zstd_compress(BUG_FIXING_COMMITS_DB)
bug_fixing_commits = prev_bug_fixing_commits + bug_fixing_commits
return [
bug_fixing_commit for bug_fixing_commit in bug_fixing_commits
if bug_fixing_commit["type"] in ["r", "d"]
]
def train_regression(self):
logger.info("Training *regression vs non-regression* model")
model = RegressionModel()
model.train()
self.compress_file("regressionmodel")
def train_regression(self):
logger.info("Training *regression vs non-regression* model")
model = RegressionModel()
model.train()
self.compress_file("regressionmodel")
parser.add_argument(
"--goal",
help="Goal of the labeler",
choices=["str", "regressionrange"],
default="str",
)
args = parser.parse_args()
if args.goal == "str":
from bugbug.models.bug import BugModel
model = BugModel.load("bugmodel")
elif args.goal == "regressionrange":
from bugbug.models.regression import RegressionModel
model = RegressionModel.load("regressionmodel")
file_path = os.path.join("bugbug", "labels", f"{args.goal}.csv")
with open(file_path, "r") as f:
reader = csv.reader(f)
next(reader)
labeled_comments = [(int(r[0]), int(r[1]), r[2]) for r in reader]
already_done = set((c[0], c[1]) for c in labeled_comments)
bugs = []
for bug in bugzilla.get_bugs():
# For the str and regressionrange problems, we don't care about test failures,
if ("intermittent-failure" in bug["keywords"]
or "stockwell" in bug["whiteboard"]
parser.add_argument(
"--goal",
help="Goal of the labeler",
choices=["str", "regressionrange"],
default="str",
)
args = parser.parse_args()
if args.goal == "str":
from bugbug.models.bug import BugModel
model = BugModel.load("bugmodel")
elif args.goal == "regressionrange":
from bugbug.models.regression import RegressionModel
model = RegressionModel.load("regressionmodel")
file_path = os.path.join("bugbug", "labels", f"{args.goal}.csv")
with open(file_path, "r") as f:
reader = csv.reader(f)
next(reader)
labeled_comments = [(int(r[0]), int(r[1]), r[2]) for r in reader]
already_done = set((c[0], c[1]) for c in labeled_comments)
bugs = []
for bug in bugzilla.get_bugs():
# For the str and regressionrange problems, we don't care about test failures,
if (
"intermittent-failure" in bug["keywords"]
def test_get_regression_labels():
model = RegressionModel()
classes, _ = model.get_labels()
assert classes[447581] == 0
assert classes[518272] == 1
def __init__(self):
super().__init__()
self.model = RegressionModel.load(self.retrieve_model())
self.autofix_regression = []