ClickHouse/tests/ci/run_check.py

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

368 lines
12 KiB
Python
Raw Permalink Normal View History

2021-09-15 12:59:39 +00:00
#!/usr/bin/env python3
2021-09-15 13:04:29 +00:00
import logging
2024-08-03 08:40:12 +00:00
import re
import sys
from typing import Tuple
2021-11-26 14:00:09 +00:00
from build_download_helper import APIException
from ci_config import CI
from commit_status_helper import (
create_ci_report,
format_description,
get_commit,
post_commit_status,
post_labels,
remove_labels,
)
from env_helper import GITHUB_REPOSITORY, GITHUB_SERVER_URL
from get_robot_token import get_best_robot_token
from github_helper import GitHub
from pr_info import PRInfo
from report import FAILURE, PENDING, SUCCESS, StatusType
2021-09-15 12:59:39 +00:00
TRUSTED_ORG_IDS = {
54801242, # clickhouse
}
2024-08-03 08:40:12 +00:00
TRUSTED_CONTRIBUTORS = {
e.lower()
for e in [
"amosbird",
"azat", # SEMRush
"bharatnc", # Many contributions.
"cwurm", # ClickHouse, Inc
"den-crane", # Documentation contributor
"ildus", # adjust, ex-pgpro
"nvartolomei", # Seasoned contributor, CloudFlare
"taiyang-li",
"ucasFL", # Amos Bird's friend
"thomoco", # ClickHouse, Inc
"tonickkozlov", # Cloudflare
"tylerhannan", # ClickHouse, Inc
"tsolodov", # ClickHouse, Inc
"justindeguzman", # ClickHouse, Inc
"XuJia0210", # ClickHouse, Inc
]
}
OK_SKIP_LABELS = {CI.Labels.RELEASE, CI.Labels.PR_BACKPORT, CI.Labels.PR_CHERRYPICK}
2021-09-15 12:59:39 +00:00
2024-08-03 08:40:12 +00:00
LABEL_CATEGORIES = {
"pr-backward-incompatible": ["Backward Incompatible Change"],
"pr-bugfix": [
"Bug Fix",
"Bug Fix (user-visible misbehavior in an official stable release)",
"Bug Fix (user-visible misbehaviour in official stable or prestable release)",
"Bug Fix (user-visible misbehavior in official stable or prestable release)",
],
"pr-critical-bugfix": [
"Critical Bug Fix (crash, data loss, RBAC) or LOGICAL_ERROR"
],
2024-08-03 08:40:12 +00:00
"pr-build": [
"Build/Testing/Packaging Improvement",
"Build Improvement",
"Build/Testing Improvement",
"Build",
"Packaging Improvement",
],
"pr-documentation": [
"Documentation (changelog entry is not required)",
"Documentation",
],
"pr-feature": ["New Feature"],
"pr-improvement": ["Improvement"],
"pr-not-for-changelog": [
"Not for changelog (changelog entry is not required)",
"Not for changelog",
],
"pr-performance": ["Performance Improvement"],
"pr-ci": ["CI Fix or Improvement (changelog entry is not required)"],
}
CATEGORY_TO_LABEL = {
c: lb for lb, categories in LABEL_CATEGORIES.items() for c in categories
}
def check_pr_description(pr_body: str, repo_name: str) -> Tuple[str, str]:
"""The function checks the body to being properly formatted according to
.github/PULL_REQUEST_TEMPLATE.md, if the first returned string is not empty,
then there is an error."""
lines = list(map(lambda x: x.strip(), pr_body.split("\n") if pr_body else []))
lines = [re.sub(r"\s+", " ", line) for line in lines]
# Check if body contains "Reverts ClickHouse/ClickHouse#36337"
if [True for line in lines if re.match(rf"\AReverts {repo_name}#[\d]+\Z", line)]:
return "", LABEL_CATEGORIES["pr-not-for-changelog"][0]
category = ""
entry = ""
description_error = ""
i = 0
while i < len(lines):
if re.match(r"(?i)^[#>*_ ]*change\s*log\s*category", lines[i]):
i += 1
if i >= len(lines):
break
# Can have one empty line between header and the category
# itself. Filter it out.
if not lines[i]:
i += 1
if i >= len(lines):
break
category = re.sub(r"^[-*\s]*", "", lines[i])
i += 1
# Should not have more than one category. Require empty line
# after the first found category.
if i >= len(lines):
break
if lines[i]:
second_category = re.sub(r"^[-*\s]*", "", lines[i])
description_error = (
"More than one changelog category specified: "
f"'{category}', '{second_category}'"
)
return description_error, category
elif re.match(
r"(?i)^[#>*_ ]*(short\s*description|change\s*log\s*entry)", lines[i]
):
i += 1
# Can have one empty line between header and the entry itself.
# Filter it out.
if i < len(lines) and not lines[i]:
i += 1
# All following lines until empty one are the changelog entry.
entry_lines = []
while i < len(lines) and lines[i]:
entry_lines.append(lines[i])
i += 1
entry = " ".join(entry_lines)
# Don't accept changelog entries like '...'.
entry = re.sub(r"[#>*_.\- ]", "", entry)
# Don't accept changelog entries like 'Close #12345'.
entry = re.sub(r"^[\w\-\s]{0,10}#?\d{5,6}\.?$", "", entry)
else:
i += 1
if not category:
description_error = "Changelog category is empty"
# Filter out the PR categories that are not for changelog.
elif "(changelog entry is not required)" in category:
pass # to not check the rest of the conditions
elif category not in CATEGORY_TO_LABEL:
description_error, category = f"Category '{category}' is not valid", ""
elif not entry:
description_error = f"Changelog entry required for category '{category}'"
return description_error, category
2021-09-15 12:59:39 +00:00
def pr_is_by_trusted_user(pr_user_login, pr_user_orgs):
2024-08-03 08:40:12 +00:00
if pr_user_login.lower() in TRUSTED_CONTRIBUTORS:
logging.info("User '%s' is trusted", pr_user_login)
2021-09-15 12:59:39 +00:00
return True
logging.info("User '%s' is not trusted", pr_user_login)
2021-09-15 12:59:39 +00:00
for org_id in pr_user_orgs:
if org_id in TRUSTED_ORG_IDS:
2022-01-13 11:06:50 +00:00
logging.info(
"Org '%s' is trusted; will mark user %s as trusted",
org_id,
pr_user_login,
)
2021-09-15 12:59:39 +00:00
return True
logging.info("Org '%s' is not trusted", org_id)
2021-09-15 12:59:39 +00:00
return False
2022-01-13 11:06:50 +00:00
2021-09-15 12:59:39 +00:00
# Returns whether we should look into individual checks for this PR. If not, it
# can be skipped entirely.
# Returns can_run, description
def should_run_ci_for_pr(pr_info: PRInfo) -> Tuple[bool, str]:
2021-09-15 12:59:39 +00:00
# Consider the labels and whether the user is trusted.
logging.info("Got labels: %s", pr_info.labels)
2021-09-15 12:59:39 +00:00
if OK_SKIP_LABELS.intersection(pr_info.labels):
return True, "Don't try new checks for release/backports/cherry-picks"
if CI.Labels.CAN_BE_TESTED not in pr_info.labels and not pr_is_by_trusted_user(
2022-01-13 11:06:50 +00:00
pr_info.user_login, pr_info.user_orgs
):
logging.info(
"PRs by untrusted users need the '%s' label - "
"please contact a member of the core team",
CI.Labels.CAN_BE_TESTED,
2022-09-07 08:51:02 +00:00
)
return False, "Needs 'can be tested' label"
2021-09-15 12:59:39 +00:00
return True, "No special conditions apply"
2021-09-15 12:59:39 +00:00
2021-11-26 14:00:09 +00:00
2023-04-06 09:03:32 +00:00
def main():
2021-09-15 12:59:39 +00:00
logging.basicConfig(level=logging.INFO)
fail_early = False
try:
pr_info = PRInfo(
need_orgs=True, pr_event_from_api=True, need_changed_files=True
)
except APIException as e:
logging.exception(
"Failed to receive the PRInfo, backport to a simple case and exit with error",
exc_info=e,
)
pr_info = PRInfo()
fail_early = True
# The case for special branches like backports and releases without created
# PRs, like merged backport branches that are reset immediately after merge
if pr_info.number == 0 or fail_early:
print("::notice ::Cannot run, no PR exists for the commit")
gh = GitHub(get_best_robot_token(), per_page=100)
commit = get_commit(gh, pr_info.sha)
post_commit_status(
commit,
FAILURE,
"",
"No PRs found for the commit, finished early",
CI.StatusNames.PR_CHECK,
pr_info,
)
sys.exit(1)
can_run, description = should_run_ci_for_pr(pr_info)
if can_run and OK_SKIP_LABELS.intersection(pr_info.labels):
print("::notice :: Early finish the check, running in a special PR")
sys.exit(0)
2023-02-27 12:11:17 +00:00
description = format_description(description)
gh = GitHub(get_best_robot_token(), per_page=100)
commit = get_commit(gh, pr_info.sha)
status = SUCCESS # type: StatusType
2022-01-13 11:08:31 +00:00
2024-08-03 08:40:12 +00:00
description_error, category = check_pr_description(pr_info.body, GITHUB_REPOSITORY)
2022-03-30 08:58:34 +00:00
pr_labels_to_add = []
pr_labels_to_remove = []
2022-03-29 17:50:06 +00:00
if (
2024-08-03 08:40:12 +00:00
category in CATEGORY_TO_LABEL
and CATEGORY_TO_LABEL[category] not in pr_info.labels
2022-03-29 17:50:06 +00:00
):
2024-08-03 08:40:12 +00:00
pr_labels_to_add.append(CATEGORY_TO_LABEL[category])
2022-03-30 08:58:34 +00:00
for label in pr_info.labels:
2022-03-30 09:19:11 +00:00
if (
2024-08-03 08:40:12 +00:00
label in CATEGORY_TO_LABEL.values()
and category in CATEGORY_TO_LABEL
and label != CATEGORY_TO_LABEL[category]
2022-03-30 09:19:11 +00:00
):
2022-03-30 08:58:34 +00:00
pr_labels_to_remove.append(label)
2022-03-29 13:48:57 +00:00
2022-03-29 17:28:18 +00:00
if pr_info.has_changes_in_submodules():
pr_labels_to_add.append(CI.Labels.SUBMODULE_CHANGED)
elif CI.Labels.SUBMODULE_CHANGED in pr_info.labels:
pr_labels_to_remove.append(CI.Labels.SUBMODULE_CHANGED)
2022-03-30 08:58:34 +00:00
if any(label in CI.Labels.AUTO_BACKPORT for label in pr_labels_to_add):
backport_labels = [CI.Labels.MUST_BACKPORT, CI.Labels.MUST_BACKPORT_CLOUD]
pr_labels_to_add += [
label for label in backport_labels if label not in pr_info.labels
]
print(
f"::notice :: Add backport labels [{backport_labels}] for a given PR category"
)
logging.info(
"Change labels: add %s, remove %s", pr_labels_to_add, pr_labels_to_remove
)
2022-03-30 08:58:34 +00:00
if pr_labels_to_add:
post_labels(gh, pr_info, pr_labels_to_add)
if pr_labels_to_remove:
remove_labels(gh, pr_info, pr_labels_to_remove)
2022-03-29 17:28:18 +00:00
# 1. Next three IFs are in a correct order. First - fatal error
2022-04-19 12:47:18 +00:00
if description_error:
2022-04-04 23:06:46 +00:00
print(
"::error ::Cannot run, PR description does not match the template: "
2022-04-19 12:47:18 +00:00
f"{description_error}"
2022-04-04 23:06:46 +00:00
)
2022-01-26 12:23:20 +00:00
logging.info(
2023-04-20 11:55:33 +00:00
"PR body doesn't match the template: (start)\n%s\n(end)\nReason: %s",
2022-04-04 23:06:46 +00:00
pr_info.body,
2022-04-19 12:47:18 +00:00
description_error,
2022-01-26 12:23:20 +00:00
)
2022-01-13 11:08:31 +00:00
url = (
f"{GITHUB_SERVER_URL}/{GITHUB_REPOSITORY}/"
"blob/master/.github/PULL_REQUEST_TEMPLATE.md?plain=1"
)
status = FAILURE
post_commit_status(
2023-04-20 11:55:33 +00:00
commit,
status,
2023-04-20 11:55:33 +00:00
url,
format_description(description_error),
CI.StatusNames.PR_CHECK,
2023-04-20 11:55:33 +00:00
pr_info,
2022-01-13 11:08:31 +00:00
)
sys.exit(1)
# 2. Then we check if the documentation is not created to fail the Mergeable check
if (
CI.Labels.PR_FEATURE in pr_info.labels
and not pr_info.has_changes_in_documentation()
):
print(
f"::error ::The '{CI.Labels.PR_FEATURE}' in the labels, "
"but there's no changed documentation"
2023-04-20 11:55:33 +00:00
)
status = FAILURE
description = f"expect adding docs for {CI.Labels.PR_FEATURE}"
# 3. But we allow the workflow to continue
# 4. And post only a single commit status on a failure
if not can_run:
post_commit_status(
commit,
status,
"",
description,
CI.StatusNames.PR_CHECK,
pr_info,
)
print("::error ::Cannot run")
sys.exit(1)
2024-02-28 12:37:07 +00:00
# The status for continue can be posted only one time, not more.
post_commit_status(
commit,
status,
"",
description,
CI.StatusNames.PR_CHECK,
pr_info,
)
ci_report_url = create_ci_report(pr_info, [])
print("::notice ::Can run")
2024-04-08 16:09:47 +00:00
2024-04-17 20:23:41 +00:00
if not pr_info.is_merge_queue:
2024-04-08 16:09:47 +00:00
# we need clean CI status for MQ to merge (no pending statuses)
post_commit_status(
commit,
PENDING,
ci_report_url,
description,
2024-06-10 09:18:03 +00:00
CI.StatusNames.CI,
2024-04-08 16:09:47 +00:00
pr_info,
)
2023-04-06 09:03:32 +00:00
if __name__ == "__main__":
main()