From 534f5edb53c80bc116b41dd3e07d282c947c53d0 Mon Sep 17 00:00:00 2001 From: Ed Bartosh Date: Tue, 30 May 2023 00:49:30 +0300 Subject: [PATCH] Cleanup boilerpate.py - reformatted with https://github.com/psf/black - fixed some of the https://github.com/pylint-dev/pylint warnings --- hack/boilerplate/boilerplate.py | 128 +++++++++++++++++--------------- 1 file changed, 68 insertions(+), 60 deletions(-) diff --git a/hack/boilerplate/boilerplate.py b/hack/boilerplate/boilerplate.py index eaf758b9fe5..d5604f4a9c6 100755 --- a/hack/boilerplate/boilerplate.py +++ b/hack/boilerplate/boilerplate.py @@ -24,23 +24,22 @@ import sys parser = argparse.ArgumentParser() parser.add_argument( - "filenames", - help="list of files to check, all files if unspecified", - nargs='*') + "filenames", help="list of files to check, all files if unspecified", nargs="*" +) rootdir = os.path.dirname(__file__) + "/../../" rootdir = os.path.abspath(rootdir) -parser.add_argument( - "--rootdir", default=rootdir, help="root directory to examine") +parser.add_argument("--rootdir", default=rootdir, help="root directory to examine") default_boilerplate_dir = os.path.join(rootdir, "hack/boilerplate") -parser.add_argument( - "--boilerplate-dir", default=default_boilerplate_dir) +parser.add_argument("--boilerplate-dir", default=default_boilerplate_dir) parser.add_argument( - "-v", "--verbose", + "-v", + "--verbose", help="give verbose output regarding why a file does not pass", - action="store_true") + action="store_true", +) args = parser.parse_args() @@ -53,31 +52,26 @@ def get_refs(): for path in glob.glob(os.path.join(args.boilerplate_dir, "boilerplate.*.txt")): extension = os.path.basename(path).split(".")[1] - ref_file = open(path, 'r') - ref = ref_file.read().splitlines() - ref_file.close() - refs[extension] = ref + with open(path, "r") as ref_file: + refs[extension] = ref_file.read().splitlines() return refs -def is_generated_file(filename, data, regexs): - p = regexs["generated"] - return p.search(data) +def is_generated_file(data, regexs): + return regexs["generated"].search(data) def file_passes(filename, refs, regexs): try: - f = open(filename, 'r') - except Exception as exc: - print("Unable to open %s: %s" % (filename, exc), file=verbose_out) + with open(filename) as stream: + data = stream.read() + except OSError as exc: + print(f"Unable to open {filename}: {exc}", file=verbose_out) return False - data = f.read() - f.close() - # determine if the file is automatically generated - generated = is_generated_file(filename, data, regexs) + generated = is_generated_file(data, regexs) basename = os.path.basename(filename) extension = file_extension(filename) @@ -91,51 +85,55 @@ def file_passes(filename, refs, regexs): ref = refs[basename] # remove extra content from the top of files - if extension == "go" or extension == "generatego": - p = regexs["go_build_constraints"] - (data, found) = p.subn("", data, 1) + if extension in ("go", "generatego"): + data, found = regexs["go_build_constraints"].subn("", data, 1) elif extension in ["sh", "py"]: - p = regexs["shebang"] - (data, found) = p.subn("", data, 1) + data, found = regexs["shebang"].subn("", data, 1) data = data.splitlines() # if our test file is smaller than the reference it surely fails! if len(ref) > len(data): - print('File %s smaller than reference (%d < %d)' % - (filename, len(data), len(ref)), - file=verbose_out) + print( + f"File {filename} smaller than reference ({len(data)} < {len(ref)})", + file=verbose_out, + ) return False # trim our file to the same number of lines as the reference file - data = data[:len(ref)] + data = data[: len(ref)] - p = regexs["year"] - for d in data: - if p.search(d): + pattern = regexs["year"] + for line in data: + if pattern.search(line): if generated: - print('File %s has the YEAR field, but it should not be in generated file' % - filename, file=verbose_out) + print( + f"File {filename} has the YEAR field, but it should not be in generated file", + file=verbose_out, + ) else: - print('File %s has the YEAR field, but missing the year of date' % - filename, file=verbose_out) + print( + "File {filename} has the YEAR field, but missing the year of date", + file=verbose_out, + ) return False if not generated: # Replace all occurrences of the regex "2014|2015|2016|2017|2018" with "YEAR" - p = regexs["date"] - for i, d in enumerate(data): - (data[i], found) = p.subn('YEAR', d) + pattern = regexs["date"] + for i, line in enumerate(data): + data[i], found = pattern.subn("YEAR", line) if found != 0: break # if we don't match the reference at this point, fail if ref != data: - print("Header in %s does not match reference, diff:" % - filename, file=verbose_out) + print(f"Header in {filename} does not match reference, diff:", file=verbose_out) if args.verbose: print(file=verbose_out) - for line in difflib.unified_diff(ref, data, 'reference', filename, lineterm=''): + for line in difflib.unified_diff( + ref, data, "reference", filename, lineterm="" + ): print(line, file=verbose_out) print(file=verbose_out) return False @@ -147,9 +145,17 @@ def file_extension(filename): return os.path.splitext(filename)[1].split(".")[-1].lower() -skipped_names = ['third_party', '_gopath', '_output', '.git', 'cluster/env.sh', - "vendor", "test/e2e/generated/bindata.go", "hack/boilerplate/test", - "staging/src/k8s.io/kubectl/pkg/generated/bindata.go"] +skipped_names = [ + "third_party", + "_gopath", + "_output", + ".git", + "cluster/env.sh", + "vendor", + "test/e2e/generated/bindata.go", + "hack/boilerplate/test", + "staging/src/k8s.io/kubectl/pkg/generated/bindata.go", +] def normalize_files(files): @@ -174,13 +180,13 @@ def get_files(extensions): # as we would prune these later in normalize_files(). But doing it # cuts down the amount of filesystem walking we do and cuts down # the size of the file list - for d in skipped_names: - if d in dirs: - dirs.remove(d) - for d in dirs: + for dname in skipped_names: + if dname in dirs: + dirs.remove(dname) + for dname in dirs: # dirs that start with __ are ignored - if re.match("^__", d): - dirs.remove(d) + if dname.startswith("__"): + dirs.remove(dname) for name in walkfiles: pathname = os.path.join(root, name) @@ -198,21 +204,23 @@ def get_files(extensions): def get_dates(): years = datetime.datetime.now().year - return '(%s)' % '|'.join((str(year) for year in range(2014, years+1))) + return "(%s)" % "|".join(str(year) for year in range(2014, years + 1)) def get_regexs(): regexs = {} # Search for "YEAR" which exists in the boilerplate, but shouldn't in the real thing - regexs["year"] = re.compile('YEAR') - # get_dates return 2014, 2015, 2016, 2017, or 2018 until the current year as a regex like: "(2014|2015|2016|2017|2018)"; + regexs["year"] = re.compile("YEAR") + # get_dates return 2014, 2015, 2016, 2017, or 2018 until the current year + # as a regex like: "(2014|2015|2016|2017|2018)"; # company holder names can be anything regexs["date"] = re.compile(get_dates()) # strip the following build constraints/tags: # //go:build # // +build \n\n regexs["go_build_constraints"] = re.compile( - r"^(//(go:build| \+build).*\n)+\n", re.MULTILINE) + r"^(//(go:build| \+build).*\n)+\n", re.MULTILINE + ) # strip #!.* from scripts regexs["shebang"] = re.compile(r"^(#!.*\n)\n*", re.MULTILINE) # Search for generated files @@ -223,11 +231,11 @@ def get_regexs(): def main(): regexs = get_regexs() refs = get_refs() - filenames = get_files(list(refs.keys())) + filenames = get_files(refs) for filename in filenames: if not file_passes(filename, refs, regexs): - print(filename, file=sys.stdout) + print(filename) return 0