def process_contributors_data(self, contributors: List[str]):
"""Pre process of data for contributors in a project repository."""
pr_ids = sorted([int(k) for k in self.pull_requests.keys()])
contributors_reviews_data: Dict[str, Any] = {}
contributors_reviews_data["reviewers"] = []
contributors_reviews_data["created_dts"] = []
interactions = {}
for contributor in contributors:
contributor_interaction = dict.fromkeys(contributors, 0)
interactions[contributor] = contributor_interaction
for pr_id in pr_ids:
pr = self.pull_requests[str(pr_id)]
self._analyze_pr_for_contributor_data(pr_id=pr_id, pr=pr, extracted_data=contributors_reviews_data)
self._analyze_contributors_interaction(
pr_interactions=pr["interactions"], pr_author=pr["created_by"], interactions_data=interactions
)
for reviewer in contributors_reviews_data["reviewers"]:
number_reviews = 0
reviews_length = []
time_reviews = []
for reviews in contributors_reviews_data[reviewer]["reviews"].values():
number_reviews += len(reviews)
review_words = 0
for review in reviews:
review_words += review["words_count"]
time_reviews.append(review["submitted_at"])
reviews_length.append(review_words)
last_review_dt = max(time_reviews)
contributors_reviews_data[reviewer]["number_reviews"] = number_reviews
contributors_reviews_data[reviewer]["median_review_length"] = np.median(reviews_length)
contributors_reviews_data[reviewer]["last_review_time"] = last_review_dt
# Encode Pull Request sizes for the contributor
contributor_prs_size_encoded = [
convert_score2num(label=pr_size) for pr_size in contributors_reviews_data[reviewer]["PRs_size"]
]
contributor_pr_median_size, contributor_relative_score = convert_num2label(
score=np.median(contributor_prs_size_encoded)
)
contributors_reviews_data[reviewer]["median_pr_length"] = contributor_pr_median_size
contributors_reviews_data[reviewer]["median_pr_length_score"] = contributor_relative_score
contributors_reviews_data[reviewer]["interactions"] = interactions[reviewer]
return contributors_reviews_data
def process_prs_project_data(self):
"""Pre process of data for a given project repository."""
if not self.pull_requests:
return {}
ids = sorted([int(k) for k in self.pull_requests.keys()])
project_reviews_data = {}
project_reviews_data["contributors"] = []
project_reviews_data["ids"] = []
project_reviews_data["created_dts"] = []
project_reviews_data["reviews_dts"] = []
project_reviews_data["TTFR"] = [] # Time to First Review (TTFR) [hr]
project_reviews_data["MTTFR"] = [] # Median TTFR [hr]
project_reviews_data["TTR"] = [] # Time to Review (TTR) [hr]
project_reviews_data["MTTR"] = [] # Median TTR [hr]
project_reviews_data["MTTCI"] = [] # Median TTCI [hr]
project_reviews_data["PRs_size"] = [] # Pull Request length
# Pull Request length encoded
project_reviews_data["encoded_PRs_size"] = []
for id in ids:
id = str(id)
if self.pull_requests[id]["closed_at"] is None:
continue
pr = self.pull_requests[str(id)]
if pr["created_by"] not in project_reviews_data["contributors"]:
project_reviews_data["contributors"].append(pr["created_by"])
self._analyze_pr_for_project_data(
pr_id=id, pr=pr, extracted_data=project_reviews_data)
project_reviews_data["last_review_time"] = max(
project_reviews_data["reviews_dts"])
# Encode Pull Request sizes for the contributor
project_pr_median_size, project_length_score = convert_num2label(
score=np.median(project_reviews_data["encoded_PRs_size"]))
project_reviews_data["median_pr_length"] = project_pr_median_size
project_reviews_data["median_pr_length_score"] = project_length_score
return project_reviews_data
def pre_process_project_data(data: Dict[str, Any]):
"""Pre process of data for a given project repository."""
if not data:
return {}
pr_ids = sorted([int(k) for k in data.keys()])
project_reviews_data = {}
project_reviews_data["contributors"] = []
project_reviews_data["ids"] = []
project_reviews_data["created_dts"] = []
project_reviews_data["reviews_dts"] = []
project_reviews_data["TTFR"] = [] # Time to First Review (TTFR) [hr]
project_reviews_data["MTTFR"] = [] # Median TTFR [hr]
project_reviews_data["TTR"] = [] # Time to Review (TTR) [hr]
project_reviews_data["MTTR"] = [] # Median TTR [hr]
project_reviews_data["PRs_size"] = [] # Pull Request length
project_reviews_data["encoded_PRs_size"] = [
] # Pull Request length encoded
for pr_id in pr_ids:
pr = data[str(pr_id)]
if pr["created_by"] not in project_reviews_data["contributors"]:
project_reviews_data["contributors"].append(pr["created_by"])
analyze_pr_for_project_data(pr_id=pr_id,
pr=pr,
extracted_data=project_reviews_data)
project_reviews_data["last_review_time"] = max(
project_reviews_data["reviews_dts"])
# Encode Pull Request sizes for the contributor
project_pr_median_size, project_length_score = convert_num2label(
score=np.median(project_reviews_data["encoded_PRs_size"]))
project_reviews_data["median_pr_length"] = project_pr_median_size
project_reviews_data["median_pr_length_score"] = project_length_score
return project_reviews_data