mirror of
https://github.com/espressif/esp-idf.git
synced 2024-10-05 20:47:46 -04:00
00c3e0e8ec
The previous version used an incorrect /dir1/dir2 pattern to match the content of dir2. The correct pattern should be /dir1/dir2/ (with the trailing slash). This commit fixes these patterns. Regarding codeowners.py: 'git ls-files' can not be used to correctly implement the logic of CODEOWNERS file patterns, since it doesn't distinguish between /path/* and /path/. The former pattern in CODEOWNERS file should only match the files inside /path/, while the latter also matches files in nested directories. Because of this, the logic for evaluating patterns is re-implemented, by converting CODEOWNERS patterns into regular expressions. Gitlab CODEOWNERS parsing code was used as a reference, in addition to the approach for converting glob patterns into regular expressions proposed in https://stackoverflow.com/a/29354254.
209 lines
6.9 KiB
Python
Executable File
209 lines
6.9 KiB
Python
Executable File
#!/usr/bin/env python
|
|
#
|
|
# Utility script for ESP-IDF developers to work with the CODEOWNERS file.
|
|
#
|
|
# Copyright 2020 Espressif Systems (Shanghai) PTE LTD
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
import argparse
|
|
import os
|
|
import re
|
|
import subprocess
|
|
import sys
|
|
|
|
|
|
CODEOWNERS_PATH = os.path.join(os.path.dirname(__file__), "..", ".gitlab", "CODEOWNERS")
|
|
CODEOWNER_GROUP_PREFIX = "@esp-idf-codeowners/"
|
|
|
|
|
|
def get_all_files():
|
|
"""
|
|
Get list of all file paths in the repository.
|
|
"""
|
|
idf_root = os.path.join(os.path.dirname(__file__), "..")
|
|
# only split on newlines, since file names may contain spaces
|
|
return subprocess.check_output(["git", "ls-files"], cwd=idf_root).decode("utf-8").strip().split('\n')
|
|
|
|
|
|
def pattern_to_regex(pattern):
|
|
"""
|
|
Convert the CODEOWNERS path pattern into a regular expression string.
|
|
"""
|
|
orig_pattern = pattern # for printing errors later
|
|
|
|
# Replicates the logic from normalize_pattern function in Gitlab ee/lib/gitlab/code_owners/file.rb:
|
|
if not pattern.startswith('/'):
|
|
pattern = '/**/' + pattern
|
|
if pattern.endswith('/'):
|
|
pattern = pattern + '**/*'
|
|
|
|
# Convert the glob pattern into a regular expression:
|
|
# first into intermediate tokens
|
|
pattern = (pattern.replace('**/', ':REGLOB:')
|
|
.replace('**', ':INVALID:')
|
|
.replace('*', ':GLOB:')
|
|
.replace('.', ':DOT:')
|
|
.replace('?', ':ANY:'))
|
|
|
|
if pattern.find(':INVALID:') >= 0:
|
|
raise ValueError("Likely invalid pattern '{}': '**' should be followed by '/'".format(orig_pattern))
|
|
|
|
# then into the final regex pattern:
|
|
re_pattern = (pattern.replace(':REGLOB:', '(?:.*/)?')
|
|
.replace(':GLOB:', '[^/]*')
|
|
.replace(':DOT:', '[.]')
|
|
.replace(':ANY:', '.') + '$')
|
|
if re_pattern.startswith('/'):
|
|
re_pattern = '^' + re_pattern
|
|
|
|
return re_pattern
|
|
|
|
|
|
def files_by_regex(all_files, regex):
|
|
"""
|
|
Return all files in the repository matching the given regular expresion.
|
|
"""
|
|
return [file for file in all_files if regex.search('/' + file)]
|
|
|
|
|
|
def files_by_pattern(all_files, pattern=None):
|
|
"""
|
|
Return all the files in the repository matching the given CODEOWNERS pattern.
|
|
"""
|
|
if not pattern:
|
|
return all_files
|
|
|
|
return files_by_regex(all_files, re.compile(pattern_to_regex(pattern)))
|
|
|
|
|
|
def action_identify(args):
|
|
best_match = []
|
|
all_files = get_all_files()
|
|
with open(CODEOWNERS_PATH) as f:
|
|
for line in f:
|
|
line = line.strip()
|
|
if not line or line.startswith("#"):
|
|
continue
|
|
tokens = line.split()
|
|
path_pattern = tokens[0]
|
|
owners = tokens[1:]
|
|
files = files_by_pattern(all_files, path_pattern)
|
|
if args.path in files:
|
|
best_match = owners
|
|
for owner in best_match:
|
|
print(owner)
|
|
|
|
|
|
def action_test_pattern(args):
|
|
re_pattern = pattern_to_regex(args.pattern)
|
|
|
|
if args.regex:
|
|
print(re_pattern)
|
|
return
|
|
|
|
files = files_by_regex(get_all_files(), re.compile(re_pattern))
|
|
for f in files:
|
|
print(f)
|
|
|
|
|
|
def action_ci_check(args):
|
|
errors = []
|
|
|
|
def add_error(msg):
|
|
errors.append("Error at CODEOWNERS:{}: {}".format(line_no, msg))
|
|
|
|
all_files = get_all_files()
|
|
prev_path_pattern = ""
|
|
with open(CODEOWNERS_PATH) as f:
|
|
for line_no, line in enumerate(f, start=1):
|
|
# Skip empty lines and comments
|
|
line = line.strip()
|
|
if line.startswith("# sort-order-reset"):
|
|
prev_path_pattern = ""
|
|
|
|
if not line or line.startswith("#"):
|
|
continue
|
|
|
|
# Each line has a form of "<path> <owners>+"
|
|
tokens = line.split()
|
|
path_pattern = tokens[0]
|
|
owners = tokens[1:]
|
|
if not owners:
|
|
add_error("no owners specified for {}".format(path_pattern))
|
|
|
|
# Check that the file is sorted by path patterns
|
|
path_pattern_for_cmp = path_pattern.replace("-", "_") # ignore difference between _ and - for ordering
|
|
if prev_path_pattern and path_pattern_for_cmp < prev_path_pattern:
|
|
add_error("file is not sorted: {} < {}".format(path_pattern_for_cmp, prev_path_pattern))
|
|
prev_path_pattern = path_pattern_for_cmp
|
|
|
|
# Check that the pattern matches at least one file
|
|
files = files_by_pattern(all_files, path_pattern)
|
|
if not files:
|
|
add_error("no files matched by pattern {}".format(path_pattern))
|
|
|
|
for o in owners:
|
|
# Sanity-check the owner group name
|
|
if not o.startswith(CODEOWNER_GROUP_PREFIX):
|
|
add_error("owner {} doesn't start with {}".format(o, CODEOWNER_GROUP_PREFIX))
|
|
|
|
if not errors:
|
|
print("No errors found.")
|
|
else:
|
|
print("Errors found!")
|
|
for e in errors:
|
|
print(e)
|
|
raise SystemExit(1)
|
|
|
|
|
|
def main():
|
|
parser = argparse.ArgumentParser(
|
|
sys.argv[0], description="Internal helper script for working with the CODEOWNERS file."
|
|
)
|
|
subparsers = parser.add_subparsers(dest="action")
|
|
|
|
identify = subparsers.add_parser(
|
|
"identify",
|
|
help="List the owners of the specified path within IDF."
|
|
"This command doesn't support files inside submodules, or files not added to git repository.",
|
|
)
|
|
identify.add_argument("path", help="Path of the file relative to the root of the repository")
|
|
|
|
subparsers.add_parser(
|
|
"ci-check",
|
|
help="Check CODEOWNERS file: every line should match at least one file, sanity-check group names, "
|
|
"check that the file is sorted by paths",
|
|
)
|
|
|
|
test_pattern = subparsers.add_parser(
|
|
"test-pattern",
|
|
help="Print files in the repository for a given CODEOWNERS pattern. Useful when adding new rules."
|
|
)
|
|
test_pattern.add_argument("--regex", action="store_true", help="Print the equivalent regular expression instead of the file list.")
|
|
test_pattern.add_argument("pattern", help="Path pattern to get the list of files for")
|
|
|
|
args = parser.parse_args()
|
|
|
|
if args.action is None:
|
|
parser.print_help()
|
|
parser.exit(1)
|
|
|
|
action_func_name = "action_" + args.action.replace("-", "_")
|
|
action_func = globals()[action_func_name]
|
|
action_func(args)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|