def get_commits(project_name):
"""
Get commits in temporal order. Merge commits are excluded.
Parameters
----------
project_name (str):
Name of the project. "<project_name>.sqlite3" has to be in data folder.
Returns
-------
list:
Tuples of commit hash, author and date in temporal order.
For example, "[(commit1, author1, 12Oct2013), (commit2, author1, 19Oct2013)]"
"""
query_results = execute_db_query(
"data/{}.sqlite3".format(project_name),
"""
SELECT commit_hash, author, committed_date
FROM change_set
WHERE is_merge=0
ORDER BY committed_date
""",
)
return [(commit_hash, author, committed_date)
for commit_hash, author, committed_date in query_results]
def get_commit_to_codechanges(project_name):
"""
Get a mapping from commit hash to code changes.
Parameters
----------
project_name (str):
Name of the project. "<project_name>.sqlite3" has to be in data folder.
Returns
-------
dict:
Mapping from commit hash to code changes.
"""
query_results = execute_db_query(
"data/{}.sqlite3".format(project_name),
"""
SELECT commit_hash, file_path, change_type, sum_added_lines, sum_removed_lines
FROM code_change
""",
)
commit_to_codechanges = defaultdict(list)
for commit_hash, fpath, ctype, num_added, num_deleted in query_results:
if fpath.endswith(".java"):
fname = fpath[fpath.rfind("/") + 1:]
commit_to_codechanges[commit_hash].append(
(fpath, ctype, fname, num_added, num_deleted))
return commit_to_codechanges
def get_commit_to_issues(project_name):
"""
Get a mapping from commit hash to issue ids.
Parameters
----------
project_name (str):
Name of the project. "<project_name>.sqlite3" has to be in data folder.
Returns
-------
dict:
Mapping from commit hash to issue ids.
"""
query_results = execute_db_query(
"data/{}.sqlite3".format(project_name),
"""
SELECT issue_id, commit_hash
FROM change_set_link
""",
)
commit_to_issues = defaultdict(list)
for issue_id, commit_hash in query_results:
commit_to_issues[commit_hash].append(issue_id)
return commit_to_issues
def number_of_developers_before_preprocessing():
"""
Generate the number of all distint developers for each project before preprocessing.
For example, author name correction not applied yet.
"""
print("\n*** Number of Developers Before Preprocessing ***\n")
print(("{:<12}" * len(project_list)).format(*project_list))
for project_name in project_list:
num_devs = execute_db_query(
"data/{}.sqlite3".format(project_name),
"SELECT count(DISTINCT author) FROM change_set",
)[0][0]
print("{:<12}".format(num_devs), end="")
print("\n")
def generate_issue_to_commenters(project_name):
"""
Generate a mapping from issue ids to commenters of the issues.
Parameters
----------
project_name (str):
Name of the project.
Returns
--------
dict:
Mapping from issue ids to commenters of the issues.
"""
query_results = execute_db_query(
"data/{}.sqlite3".format(project_name),
"""
SELECT issue_id, display_name
FROM issue_comment
""",
)
issue_to_commenters = defaultdict(list)
for issue_id, commenter in query_results:
# Clear whitespaces and make lower case.
commenter = commenter.strip().lower()
# Check ignore commenters
if commenter in ignored_commenters:
continue
# Replace the commenters name if it is in author mapping
commenter = combined_author_mapping.get(commenter, commenter)
# New issue
issue_to_commenters[issue_id].append(commenter)
return issue_to_commenters