def get_duration_data(durations, owner_repo="edx/edx-platform", since=None):
"""
Update `durations`, a dict of dict of lists of pull requests.
`durations` has four lists of data, where each list contains pull requests:
internal open pull requests (all)
external open pull requests (all)
internal closed pull requests (since the `since` value)
external closed pull requests (since the `since` value)
These lists are organized into a dictionary that categorizes the lists
by position and state.
"""
open_issues_generator = itertools.izip(
get_pulls(owner_repo, state="open", org=True),
itertools.repeat("open"))
closed_issues_generator = itertools.izip(
get_pulls(owner_repo, state="closed", since=since, org=True),
itertools.repeat("closed"))
for issue, state in itertools.chain(open_issues_generator,
closed_issues_generator):
created_at = issue.created_at
if state == "open":
closed_at = datetime.utcnow()
else:
closed_at = issue.closed_at
issue.duration = closed_at - created_at
if DEBUG:
print("{pr.id}: {pr.intext} {state}".format(pr=issue, state=state),
file=sys.stderr)
durations[state][issue.intext].append(issue)
def get_bucket_data(buckets, repo_name, date_bucket_fn, start, by_size=False, lines=False, closed=False):
print(repo_name)
pull_details = "all" if (by_size or lines) else "list"
for pull in get_pulls(repo_name, state="all", pull_details=pull_details, org=True):
# print("{0.id}: {0.combinedstate} {0.intext}".format(pull))
ignore_ref = "(^release$|^rc/)"
if re.search(ignore_ref, pull.base_ref) or pull.title == "Update translations (autogenerated message)":
#print("Ignoring pull #{0.number}: {0.title}".format(pull))
continue
if by_size:
size = " " + size_of_pull(pull)
else:
size = ""
intext = pull.intext
if lines:
increment = lines_in_pull(pull)
else:
increment = 1
created = make_timezone_aware(pull.created_at)
if created >= start:
buckets[date_bucket_fn(created)]["opened " + intext + size] += increment
if pull.combinedstate == "merged":
merged = make_timezone_aware(pull.merged_at)
if merged >= start:
buckets[date_bucket_fn(merged)]["merged " + intext + size] += increment
elif closed and pull.combinedstate == "closed":
closed = make_timezone_aware(pull.closed_at)
buckets[date_bucket_fn(closed)]["closed " + intext + size] += increment
def get_duration_data(durations, owner_repo="edx/edx-platform", since=None):
"""
Update `durations`, a dict of dict of lists of pull requests.
`durations` has four lists of data, where each list contains pull requests:
internal open pull requests (all)
external open pull requests (all)
internal closed pull requests (since the `since` value)
external closed pull requests (since the `since` value)
These lists are organized into a dictionary that categorizes the lists
by position and state.
"""
open_issues_generator = itertools.izip(
get_pulls(owner_repo, state="open", org=True),
itertools.repeat("open")
)
closed_issues_generator = itertools.izip(
get_pulls(owner_repo, state="closed", since=since, org=True),
itertools.repeat("closed")
)
for issue, state in itertools.chain(open_issues_generator, closed_issues_generator):
created_at = issue.created_at
if state == "open":
closed_at = datetime.utcnow()
else:
closed_at = issue.closed_at
issue.duration = closed_at - created_at
if DEBUG:
print("{pr.id}: {pr.intext} {state}".format(
pr=issue, state=state
), file=sys.stderr)
durations[state][issue.intext].append(issue)
def get_bucket_data(buckets,
repo_name,
date_bucket_fn,
start,
lines=False,
internal=False):
print(repo_name)
pull_details = "all" if lines else "list"
for pull in get_pulls(repo_name,
state="all",
pull_details=pull_details,
org=True):
# print("{0.id}: {0.combinedstate} {0.intext}".format(pull))
intext = pull.intext # internal or external
# if internal is True, only want to look at "internal" PRs, and if
# internal is False, only want to look at "external" PRs.
if (internal and intext != 'internal') or (not internal
and intext != 'external'):
continue
ignore_ref = "(^release$|^rc/)"
if re.search(
ignore_ref, pull.base_ref
) or pull.title == "Update translations (autogenerated message)":
# print("Ignoring pull #{0.number}: {0.title}".format(pull))
continue
if lines:
increment = lines_in_pull(pull)
else:
increment = 1
created = make_timezone_aware(pull.created_at)
bucket_key = date_bucket_fn(created)
if created >= start:
buckets[bucket_key]["opened " + intext] += increment
# Bucket based on its current state
if pull.combinedstate == "merged":
buckets[bucket_key]["merged " + intext] += increment
elif pull.combinedstate == "closed":
buckets[bucket_key]["closed " + intext] += increment
else:
# PR is still open
print("Unresolved[{1}]: pull #{0.number}: {0.title}".format(
pull, bucket_key))
buckets[bucket_key]["unresolved " + intext] += increment
def get_bucket_data(buckets,
repo_name,
date_bucket_fn,
start,
by_size=False,
lines=False,
closed=False):
print(repo_name)
pull_details = "all" if (by_size or lines) else "list"
for pull in get_pulls(repo_name,
state="all",
pull_details=pull_details,
org=True):
# print("{0.id}: {0.combinedstate} {0.intext}".format(pull))
ignore_ref = "(^release$|^rc/)"
if re.search(ignore_ref, pull.base_ref):
#print("Ignoring pull #{0.number}: {0.title}".format(pull))
continue
if by_size:
size = " " + size_of_pull(pull)
else:
size = ""
intext = pull.intext
if lines:
increment = lines_in_pull(pull)
else:
increment = 1
created = make_timezone_aware(pull.created_at)
if created >= start:
buckets[date_bucket_fn(created)]["opened " + intext +
size] += increment
if pull.combinedstate == "merged":
merged = make_timezone_aware(pull.merged_at)
if merged >= start:
buckets[date_bucket_fn(merged)]["merged " + intext +
size] += increment
elif closed and pull.combinedstate == "closed":
closed = make_timezone_aware(pull.closed_at)
buckets[date_bucket_fn(closed)]["closed " + intext +
size] += increment
def get_bucket_data(buckets, repo_name, date_bucket_fn, start, lines=False, internal=False):
print(repo_name)
pull_details = "all" if lines else "list"
for pull in get_pulls(repo_name, state="all", pull_details=pull_details, org=True):
# print("{0.id}: {0.combinedstate} {0.intext}".format(pull))
intext = pull.intext # internal or external
# if internal is True, only want to look at "internal" PRs, and if
# internal is False, only want to look at "external" PRs.
if (internal and intext != 'internal') or (not internal and intext != 'external'):
continue
ignore_ref = "(^release$|^rc/)"
if re.search(ignore_ref, pull.base_ref) or pull.title == "Update translations (autogenerated message)":
# print("Ignoring pull #{0.number}: {0.title}".format(pull))
continue
if lines:
increment = lines_in_pull(pull)
else:
increment = 1
created = make_timezone_aware(pull.created_at)
bucket_key = date_bucket_fn(created)
if created >= start:
buckets[bucket_key]["opened " + intext] += increment
# Bucket based on its current state
if pull.combinedstate == "merged":
buckets[bucket_key]["merged " + intext] += increment
elif pull.combinedstate == "closed":
buckets[bucket_key]["closed " + intext] += increment
else:
# PR is still open
print("Unresolved[{1}]: pull #{0.number}: {0.title}".format(pull, bucket_key))
buckets[bucket_key]["unresolved " + intext] += increment
def main(argv):
parser = argparse.ArgumentParser(description="Summarize pull requests by organization.")
parser.add_argument(
"--since", metavar="DAYS", type=int,
help="Only consider pull requests closed in the past DAYS days"
)
parser.add_argument(
"--start", type=date_arg,
help="Date to start collecting, format is flexible: "
"20141225, Dec/25/2014, 2014-12-25, etc"
)
parser.add_argument(
"--end", type=date_arg,
help="Date to end collecting, format is flexible: "
"25/Dec/2014, 12/25/2014, 2014-12-25, etc"
)
parser.add_argument(
"--short", action="store_true",
help="Only show the short summary"
)
args = parser.parse_args(argv[1:])
since = None
if args.since:
since = make_timezone_aware(datetime.today() - timedelta(days=args.since))
if args.start:
if since is not None:
raise Exception("Can't use --since and --start")
since = args.start
repos = [ r for r in Repo.from_yaml() if r.track_pulls ]
by_org = collections.defaultdict(list)
for repo in repos:
for pull in get_pulls(repo.name, state="closed", pull_details="list", org=True, since=since):
# We only want external pull requests.
if pull.intext != "external":
continue
# We only want merged pull requests.
if pull.combinedstate != "merged":
continue
# Pull requests can be recently modified even if they were merged
# long ago, so only take things merged since our since date.
merged = make_timezone_aware(pull.merged_at)
if merged < since:
continue
if args.end is not None:
# We don't want to count things merged after our end date.
if merged >= args.end:
continue
pull.repo = repo.nick
by_org[pull.org].append(pull)
keys = sorted(by_org, key=lambda k: len(by_org[k]), reverse=True)
for key in keys:
print("{}: {}".format(key, len(by_org[key])))
fmt = "{pull.repo:4s} {pull.number:5d} {pull.user_login:>17s} {pull.title}"
if args.short:
if 'unsigned' in keys:
print("\n-- {} -------".format('unsigned'))
for pull in by_org['unsigned']:
print(fmt.format(pull=pull))
else:
for key in keys:
print("\n-- {} -------".format(key))
for pull in by_org[key]:
print(fmt.format(pull=pull))
def get_external_pulls(repo):
"""Produce a stream of external pull requests."""
for issue in get_pulls(repo, state="all", org=True):
if issue.intext == 'external':
yield issue
def get_filtered_pulls(repo, interesting):
"""Produce a stream of external pull requests."""
for issue in get_pulls(repo, state="all", org=True):
if interesting(issue):
yield issue
def get_filtered_pulls(repo, interesting):
"""Produce a stream of external pull requests."""
for issue in get_pulls(repo, state="all", org=True):
if interesting(issue):
yield issue