Gunes Bayir | ead57c2 | 2023-08-11 21:46:23 +0100 | [diff] [blame] | 1 | #!/usr/bin/env python3 |
| 2 | # -*- coding: utf-8 -*- |
| 3 | # |
| 4 | # Copyright (c) 2023 Arm Limited. |
| 5 | # |
| 6 | # SPDX-License-Identifier: MIT |
| 7 | # |
| 8 | # Permission is hereby granted, free of charge, to any person obtaining a copy |
| 9 | # of this software and associated documentation files (the "Software"), to |
| 10 | # deal in the Software without restriction, including without limitation the |
| 11 | # rights to use, copy, modify, merge, publish, distribute, sublicense, and/or |
| 12 | # sell copies of the Software, and to permit persons to whom the Software is |
| 13 | # furnished to do so, subject to the following conditions: |
| 14 | # |
| 15 | # The above copyright notice and this permission notice shall be included in all |
| 16 | # copies or substantial portions of the Software. |
| 17 | # |
| 18 | # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR |
| 19 | # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, |
| 20 | # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE |
| 21 | # AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER |
| 22 | # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, |
| 23 | # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE |
| 24 | # SOFTWARE. |
| 25 | |
| 26 | import argparse |
| 27 | from typing import List, Tuple |
| 28 | import logging |
| 29 | import re |
| 30 | |
| 31 | logger = logging.getLogger("check_header_guards") |
| 32 | |
| 33 | def find_code_boundaries(lines: List[str]) -> (int, int): |
| 34 | inside_comment : bool = False |
| 35 | |
| 36 | start = len(lines) |
| 37 | end = -1 |
| 38 | line_num = 0 |
| 39 | for line in lines: |
| 40 | stripped_line : str = line.strip() |
| 41 | if stripped_line.startswith("/*"): # block comment start |
| 42 | inside_comment = True |
| 43 | |
| 44 | if not inside_comment and not stripped_line.startswith("//") and stripped_line != "": |
| 45 | start = min(line_num, start) |
| 46 | end = line_num |
| 47 | |
| 48 | if inside_comment and stripped_line.endswith("*/"): |
| 49 | inside_comment = False |
| 50 | |
| 51 | line_num += 1 |
| 52 | |
| 53 | return start, end |
| 54 | |
| 55 | |
| 56 | def is_define(line: str) -> bool: |
| 57 | return line.strip().startswith("#define") |
| 58 | |
| 59 | def is_endif(line: str) -> bool: |
| 60 | return line.strip().startswith("#endif") |
| 61 | |
| 62 | def is_ifndef(line: str) -> bool: |
| 63 | return line.strip().startswith("#ifndef") |
| 64 | |
| 65 | # Strips the given line from // and /* */ blocks |
| 66 | def strip_comments(line: str) -> str: |
| 67 | line = re.sub(r"/\*.*\*/", "", line) |
| 68 | line = re.sub(r"//.*", "", line) |
| 69 | return line.strip() |
| 70 | |
| 71 | # If the line |
| 72 | # 1) startswith #ifndef |
| 73 | # 2) is all uppercase |
| 74 | # 3) does not start with double underscore, i.e. __ |
| 75 | # Then |
| 76 | # It "looks" like a header guard |
| 77 | def looks_like_header_guard(line: str) -> bool: |
| 78 | sline = line.strip() |
| 79 | guard_candidate = strip_comments(sline[len("#ifndef"):]) |
| 80 | |
| 81 | return is_ifndef(sline) and not guard_candidate.startswith("__") and guard_candidate.isupper() |
| 82 | |
| 83 | |
| 84 | def fix_header_guard(lines: List[str], expected_header_guard: str, comment_style: str) -> Tuple[List[str], bool]: |
| 85 | start_line, next_line, last_line = "", "", "" |
| 86 | start_index, last_index = find_code_boundaries(lines) |
| 87 | guards_updated: bool = True |
| 88 | |
| 89 | if start_index < len(lines): |
| 90 | # if not, the file is full of comments |
| 91 | start_line = lines[start_index] |
| 92 | |
| 93 | if start_index + 1 < len(lines): |
| 94 | # if not, the file has only one line of code |
| 95 | next_line = lines[start_index + 1] |
| 96 | |
| 97 | if last_index < len(lines) and last_index > start_index + 1: |
| 98 | # if not, either the file is full of comments OR it has less than three code lines |
| 99 | last_line = lines[last_index] |
| 100 | |
| 101 | expected_start_line = f"#ifndef {expected_header_guard}\n" |
| 102 | expected_next_line = f"#define {expected_header_guard}\n" |
| 103 | |
| 104 | if comment_style == 'double_slash': |
| 105 | expected_last_line = f"#endif // {expected_header_guard}\n" |
| 106 | elif comment_style == 'slash_asterix': |
| 107 | expected_last_line = f"#endif /* {expected_header_guard} */\n" |
| 108 | |
| 109 | empty_line = "\n" |
| 110 | |
| 111 | if looks_like_header_guard(start_line) and is_define(next_line) and is_endif(last_line): |
| 112 | # modify the current header guard if necessary |
| 113 | lines = lines[:start_index] + [expected_start_line, expected_next_line] + \ |
| 114 | lines[start_index+2:last_index] + [expected_last_line] + lines[last_index+1:] |
| 115 | |
| 116 | guards_updated = (start_line != expected_start_line) or (next_line != expected_next_line) \ |
| 117 | or (last_line != expected_last_line) |
| 118 | else: |
| 119 | # header guard could not be detected, add header guards |
| 120 | lines = lines[:start_index] + [empty_line, expected_start_line, expected_next_line] + \ |
| 121 | [empty_line] + lines[start_index:] + [empty_line, expected_last_line] |
| 122 | |
| 123 | |
| 124 | return lines, guards_updated |
| 125 | |
| 126 | |
| 127 | def find_expected_header_guard(filepath: str, prefix: str, add_extension: str, drop_outermost_subdir: str) -> str: |
| 128 | if drop_outermost_subdir: |
| 129 | arr : List[str] = filepath.split("/") |
| 130 | arr = arr[min(1, len(arr)-1):] |
| 131 | filepath = "/".join(arr) |
| 132 | |
| 133 | if not add_extension: |
| 134 | filepath = ".".join(filepath.split(".")[:-1]) |
| 135 | |
| 136 | guard = filepath.replace("/", "_").replace(".", "_").upper() # snake case full path |
| 137 | return prefix + "_" + guard |
| 138 | |
| 139 | |
| 140 | def skip_file(filepath: str, extensions: List[str], exclude: List[str], include: List[str]) -> bool: |
| 141 | extension = filepath.split(".")[-1] |
| 142 | |
| 143 | if extension.lower() not in extensions: |
| 144 | return True |
| 145 | |
| 146 | if exclude and any([filepath.startswith(exc) for exc in exclude]): |
| 147 | print(exclude) |
| 148 | return True |
| 149 | |
| 150 | if include: |
| 151 | return not any([filepath.startswith(inc) for inc in include]) |
| 152 | |
| 153 | return False |
| 154 | |
| 155 | |
| 156 | if __name__ == "__main__": |
| 157 | parser = argparse.ArgumentParser( |
| 158 | formatter_class=argparse.RawDescriptionHelpFormatter, |
| 159 | description="Header Guard Checker. It adds full path snake case header guards with or without extension.", |
| 160 | ) |
| 161 | |
| 162 | parser.add_argument("files", type=str, nargs="+", help="Files to check the header guards") |
| 163 | parser.add_argument("--extensions", type=str, help="Comma separated list of extensions to run the checks. \ |
| 164 | If the input file does not have any of the extensions, it'll be skipped", required=True) |
| 165 | parser.add_argument("--comment_style", choices=['double_slash', 'slash_asterix'], required=True) |
| 166 | parser.add_argument("--exclude", type=str, help="Comma separated list of paths to exclude from header guard checks", default="") |
| 167 | parser.add_argument("--include", type=str, help="Comma separated list of paths to include. Defaults to empty string, \ |
| 168 | which means all the paths are included", default="") |
| 169 | parser.add_argument("--prefix", help="Prefix to apply to header guards", required=True) |
| 170 | parser.add_argument("--add_extension", action="store_true", help="If true, it adds the file extension to the end of the guard") |
| 171 | parser.add_argument("--drop_outermost_subdir", action="store_true", help="If true, it'll not use the outermost folder in the path. \ |
| 172 | This is intended for using in subdirs with different rules") |
| 173 | |
| 174 | args = parser.parse_args() |
| 175 | |
| 176 | files = args.files |
| 177 | extensions = args.extensions.split(",") |
| 178 | exclude = args.exclude.split(",") if args.exclude != '' else [] |
| 179 | include = args.include.split(",") if args.include != '' else [] |
| 180 | prefix = args.prefix |
| 181 | add_extension = args.add_extension |
| 182 | drop_outermost_subdir = args.drop_outermost_subdir |
| 183 | comment_style = args.comment_style |
| 184 | |
| 185 | logging_level = logging.INFO |
| 186 | logging.basicConfig(level=logging_level) |
| 187 | |
| 188 | retval = 0 |
| 189 | for file in files: |
| 190 | if skip_file(file, extensions, exclude, include): |
| 191 | logger.info(f"File {file} is SKIPPED") |
| 192 | continue |
| 193 | |
| 194 | expected_header_guard : str = find_expected_header_guard(file, prefix, add_extension, drop_outermost_subdir) |
| 195 | |
| 196 | with open(file, "r") as fd: |
| 197 | lines: List = fd.readlines() |
| 198 | |
| 199 | new_lines, guards_updated = fix_header_guard(lines, expected_header_guard, comment_style) |
| 200 | |
| 201 | with open(file, "w") as fd: |
| 202 | fd.writelines([f"{line}" for line in new_lines]) |
| 203 | |
| 204 | if guards_updated: |
| 205 | logger.info("File has been modified") |
| 206 | retval = 1 |
| 207 | |
| 208 | exit(retval) |