|
| 1 | +#!/usr/bin/env python3 |
| 2 | + |
| 3 | +# Copyright 2015 The Kubernetes Authors. |
| 4 | +# |
| 5 | +# Licensed under the Apache License, Version 2.0 (the "License"); |
| 6 | +# you may not use this file except in compliance with the License. |
| 7 | +# You may obtain a copy of the License at |
| 8 | +# |
| 9 | +# http://www.apache.org/licenses/LICENSE-2.0 |
| 10 | +# |
| 11 | +# Unless required by applicable law or agreed to in writing, software |
| 12 | +# distributed under the License is distributed on an "AS IS" BASIS, |
| 13 | +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 14 | +# See the License for the specific language governing permissions and |
| 15 | +# limitations under the License. |
| 16 | + |
| 17 | +# Verifies that all source files contain the necessary copyright boilerplate |
| 18 | +# snippet. |
| 19 | + |
| 20 | +import argparse |
| 21 | +import datetime |
| 22 | +import glob |
| 23 | +import os |
| 24 | +import re |
| 25 | +import sys |
| 26 | + |
| 27 | + |
| 28 | +def get_args(): |
| 29 | + parser = argparse.ArgumentParser() |
| 30 | + parser.add_argument( |
| 31 | + "filenames", |
| 32 | + help="list of files to check, all files if unspecified", |
| 33 | + nargs='*') |
| 34 | + |
| 35 | + rootdir = os.path.abspath('.') |
| 36 | + parser.add_argument("--rootdir", |
| 37 | + default=rootdir, |
| 38 | + help="root directory to examine") |
| 39 | + |
| 40 | + default_boilerplate_dir = os.path.join(rootdir, "hack/boilerplate") |
| 41 | + parser.add_argument("--boilerplate-dir", default=default_boilerplate_dir) |
| 42 | + |
| 43 | + parser.add_argument( |
| 44 | + '--skip', |
| 45 | + default=[ |
| 46 | + '_bin', |
| 47 | + 'external/bazel_tools', |
| 48 | + '.git', |
| 49 | + 'node_modules', |
| 50 | + '_output', |
| 51 | + 'third_party', |
| 52 | + 'vendor', |
| 53 | + 'hack/boilerplate/test', |
| 54 | + 'verify_boilerplate.py', |
| 55 | + '.python_virtual_env', |
| 56 | + ], |
| 57 | + action='append', |
| 58 | + help='Customize paths to avoid', |
| 59 | + ) |
| 60 | + return parser.parse_args() |
| 61 | + |
| 62 | + |
| 63 | +def get_refs(): |
| 64 | + refs = {} |
| 65 | + |
| 66 | + template_dir = ARGS.boilerplate_dir |
| 67 | + if not os.path.isdir(template_dir): |
| 68 | + template_dir = os.path.dirname(template_dir) |
| 69 | + for path in glob.glob(os.path.join(template_dir, "boilerplate.*.txt")): |
| 70 | + extension = os.path.basename(path).split(".")[1] |
| 71 | + |
| 72 | + # Pass the encoding parameter to avoid ascii decode error for some |
| 73 | + # platform. |
| 74 | + ref_file = open(path, 'r', encoding='utf-8') |
| 75 | + ref = ref_file.read().splitlines() |
| 76 | + ref_file.close() |
| 77 | + refs[extension] = ref |
| 78 | + |
| 79 | + return refs |
| 80 | + |
| 81 | +# given the file contents, return true if the file appears to be generated |
| 82 | +def is_generated(data): |
| 83 | + if re.search(r"^// Code generated by .*\. DO NOT EDIT\.$", data, re.MULTILINE): |
| 84 | + return True |
| 85 | + return False |
| 86 | + |
| 87 | + |
| 88 | +def file_passes(filename, refs, regexs): # pylint: disable=too-many-locals,too-many-return-statements,too-many-branches |
| 89 | + try: |
| 90 | + # Pass the encoding parameter to avoid ascii decode error for some |
| 91 | + # platform. |
| 92 | + with open(filename, 'r', encoding='utf-8') as fp: |
| 93 | + file_data = fp.read() |
| 94 | + except IOError: |
| 95 | + return False |
| 96 | + |
| 97 | + if not file_data: |
| 98 | + return True # Nothing to copyright in this empty file. |
| 99 | + |
| 100 | + basename = os.path.basename(filename) |
| 101 | + extension = file_extension(filename) |
| 102 | + if extension != "": |
| 103 | + ref = refs[extension] |
| 104 | + else: |
| 105 | + ref = refs[basename] |
| 106 | + |
| 107 | + ref = ref.copy() |
| 108 | + |
| 109 | + # remove build tags from the top of Go files |
| 110 | + if extension == "go": |
| 111 | + con = regexs["go_build_constraints"] |
| 112 | + (file_data, found) = con.subn("", file_data, 1) |
| 113 | + |
| 114 | + # remove shebang from the top of shell files |
| 115 | + if extension in ("sh", "py"): |
| 116 | + she = regexs["shebang"] |
| 117 | + (file_data, found) = she.subn("", file_data, 1) |
| 118 | + |
| 119 | + data = file_data.splitlines() |
| 120 | + |
| 121 | + # if our test file is smaller than the reference it surely fails! |
| 122 | + if len(ref) > len(data): |
| 123 | + return False |
| 124 | + |
| 125 | + # trim our file to the same number of lines as the reference file |
| 126 | + data = data[:len(ref)] |
| 127 | + |
| 128 | + # check if we encounter a 'YEAR' placeholder if the file is generated |
| 129 | + if is_generated(file_data): |
| 130 | + # pylint: disable=unused-variable |
| 131 | + for i, line in enumerate(data): |
| 132 | + if "Copyright YEAR" in line: |
| 133 | + return False |
| 134 | + return True |
| 135 | + |
| 136 | + year = regexs["year"] |
| 137 | + for datum in data: |
| 138 | + if year.search(datum): |
| 139 | + return False |
| 140 | + |
| 141 | + # Replace all occurrences of the regex "2017|2016|2015|2014" with "YEAR" |
| 142 | + when = regexs["date"] |
| 143 | + for idx, datum in enumerate(data): |
| 144 | + (data[idx], found) = when.subn('YEAR', datum) |
| 145 | + if found != 0: |
| 146 | + break |
| 147 | + |
| 148 | + # if we don't match the reference at this point, fail |
| 149 | + if ref != data: |
| 150 | + return False |
| 151 | + |
| 152 | + return True |
| 153 | + |
| 154 | + |
| 155 | +def file_extension(filename): |
| 156 | + return os.path.splitext(filename)[1].split(".")[-1].lower() |
| 157 | + |
| 158 | + |
| 159 | +# even when generated by bazel we will complain about some generated files |
| 160 | +# not having the headers. since they're just generated, ignore them |
| 161 | +IGNORE_HEADERS = ['// Code generated by go-bindata.'] |
| 162 | + |
| 163 | + |
| 164 | +def has_ignored_header(pathname): |
| 165 | + # Pass the encoding parameter to avoid ascii decode error for some |
| 166 | + # platform. |
| 167 | + with open(pathname, 'r', encoding='utf-8') as myfile: |
| 168 | + data = myfile.read() |
| 169 | + for header in IGNORE_HEADERS: |
| 170 | + if data.startswith(header): |
| 171 | + return True |
| 172 | + return False |
| 173 | + |
| 174 | + |
| 175 | +def normalize_files(files): |
| 176 | + newfiles = [] |
| 177 | + for pathname in files: |
| 178 | + if any(x in pathname for x in ARGS.skip): |
| 179 | + continue |
| 180 | + newfiles.append(pathname) |
| 181 | + for idx, pathname in enumerate(newfiles): |
| 182 | + if not os.path.isabs(pathname): |
| 183 | + newfiles[idx] = os.path.join(ARGS.rootdir, pathname) |
| 184 | + return newfiles |
| 185 | + |
| 186 | + |
| 187 | +def get_files(extensions): |
| 188 | + files = [] |
| 189 | + if ARGS.filenames: |
| 190 | + files = ARGS.filenames |
| 191 | + else: |
| 192 | + for root, dirs, walkfiles in os.walk(ARGS.rootdir): |
| 193 | + # don't visit certain dirs. This is just a performance improvement |
| 194 | + # as we would prune these later in normalize_files(). But doing it |
| 195 | + # cuts down the amount of filesystem walking we do and cuts down |
| 196 | + # the size of the file list |
| 197 | + for dpath in ARGS.skip: |
| 198 | + if dpath in dirs: |
| 199 | + dirs.remove(dpath) |
| 200 | + |
| 201 | + for name in walkfiles: |
| 202 | + pathname = os.path.join(root, name) |
| 203 | + files.append(pathname) |
| 204 | + |
| 205 | + files = normalize_files(files) |
| 206 | + outfiles = [] |
| 207 | + for pathname in files: |
| 208 | + basename = os.path.basename(pathname) |
| 209 | + extension = file_extension(pathname) |
| 210 | + if extension in extensions or basename in extensions: |
| 211 | + if not has_ignored_header(pathname): |
| 212 | + outfiles.append(pathname) |
| 213 | + return outfiles |
| 214 | + |
| 215 | + |
| 216 | +def get_dates(): |
| 217 | + years = datetime.datetime.now().year |
| 218 | + return '(%s)' % '|'.join((str(year) for year in range(2014, years + 1))) |
| 219 | + |
| 220 | + |
| 221 | +def get_regexs(): |
| 222 | + regexs = {} |
| 223 | + # Search for "YEAR" which exists in the boilerplate, but shouldn't in the real thing |
| 224 | + regexs["year"] = re.compile('YEAR') |
| 225 | + # dates can be any year between 2014 and the current year, company holder names can be anything |
| 226 | + regexs["date"] = re.compile(get_dates()) |
| 227 | + # strip // +build \n\n build constraints |
| 228 | + regexs["go_build_constraints"] = re.compile(r"^(//( \+build|go:build).*\n)+\n", |
| 229 | + re.MULTILINE) |
| 230 | + # strip #!.* from shell/python scripts |
| 231 | + regexs["shebang"] = re.compile(r"^(#!.*\n)\n*", re.MULTILINE) |
| 232 | + return regexs |
| 233 | + |
| 234 | + |
| 235 | +def nonconforming_lines(files): |
| 236 | + yield '%d files have incorrect boilerplate headers:' % len(files) |
| 237 | + for fp in files: |
| 238 | + yield os.path.relpath(fp, ARGS.rootdir) |
| 239 | + |
| 240 | + |
| 241 | +def main(): |
| 242 | + regexs = get_regexs() |
| 243 | + refs = get_refs() |
| 244 | + filenames = get_files(refs.keys()) |
| 245 | + nonconforming_files = [] |
| 246 | + for filename in sorted(filenames): |
| 247 | + if not file_passes(filename, refs, regexs): |
| 248 | + nonconforming_files.append(filename) |
| 249 | + |
| 250 | + if nonconforming_files: |
| 251 | + for line in nonconforming_lines(nonconforming_files): |
| 252 | + print(line) |
| 253 | + sys.exit(1) |
| 254 | + |
| 255 | + print("Verified %d files" % (len(filenames), )) |
| 256 | + |
| 257 | + |
| 258 | +if __name__ == "__main__": |
| 259 | + ARGS = get_args() |
| 260 | + main() |
0 commit comments