| Total Complexity | 196 |
| Total Lines | 834 |
| Duplicated Lines | 4.08 % |
| Changes | 0 | ||
Duplicate code is one of the most pungent code smells. A rule that is often used is to re-structure code once it is duplicated in three or more places.
Common duplication problems, and corresponding solutions are:
Complex classes like utils.fix_rules often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
| 1 | #!/usr/bin/python3 |
||
| 2 | |||
| 3 | from __future__ import print_function |
||
| 4 | |||
| 5 | import sys |
||
| 6 | import os |
||
| 7 | import jinja2 |
||
| 8 | import argparse |
||
| 9 | import json |
||
| 10 | import re |
||
| 11 | import random |
||
| 12 | |||
| 13 | from ssg import yaml, cce, products |
||
| 14 | from ssg.shims import input_func |
||
| 15 | from ssg.utils import read_file_list |
||
| 16 | import ssg |
||
| 17 | import ssg.products |
||
| 18 | import ssg.rules |
||
| 19 | import ssg.rule_yaml |
||
| 20 | |||
| 21 | |||
| 22 | SSG_ROOT = os.path.abspath(os.path.join(os.path.dirname(__file__), "..")) |
||
| 23 | TO_SORT = ['identifiers', 'references'] |
||
| 24 | |||
| 25 | |||
| 26 | _COMMANDS = dict() |
||
| 27 | |||
| 28 | |||
| 29 | def command(name, description): |
||
| 30 | def wrapper(wrapped): |
||
| 31 | _COMMANDS[name] = wrapped |
||
| 32 | wrapped.description = description |
||
| 33 | return wrapped |
||
| 34 | return wrapper |
||
| 35 | |||
| 36 | |||
| 37 | def has_empty_identifier(rule_path, rule, rule_lines): |
||
| 38 | if 'identifiers' in rule and rule['identifiers'] is None: |
||
| 39 | return True |
||
| 40 | |||
| 41 | if 'identifiers' in rule and rule['identifiers'] is not None: |
||
| 42 | for _, value in rule['identifiers'].items(): |
||
| 43 | if str(value).strip() == "": |
||
| 44 | return True |
||
| 45 | return False |
||
| 46 | |||
| 47 | |||
| 48 | def has_no_cce(yaml_file, product_yaml=None): |
||
| 49 | rule = yaml.open_and_macro_expand(yaml_file, product_yaml) |
||
| 50 | product = product_yaml["product"] |
||
| 51 | if "prodtype" in rule and product not in rule["prodtype"]: |
||
| 52 | return False |
||
| 53 | if 'identifiers' in rule and rule['identifiers'] is None: |
||
| 54 | return True |
||
| 55 | |||
| 56 | if 'identifiers' in rule and rule['identifiers'] is not None: |
||
| 57 | for ident in rule['identifiers']: |
||
| 58 | if ident == "cce@" + product: |
||
| 59 | return False |
||
| 60 | return True |
||
| 61 | |||
| 62 | |||
| 63 | def has_empty_references(rule_path, rule, rule_lines): |
||
| 64 | if 'references' in rule and rule['references'] is None: |
||
| 65 | return True |
||
| 66 | |||
| 67 | if 'references' in rule and rule['references'] is not None: |
||
| 68 | for _, value in rule['references'].items(): |
||
| 69 | if str(value).strip() == "": |
||
| 70 | return True |
||
| 71 | return False |
||
| 72 | |||
| 73 | |||
| 74 | def has_prefix_cce(rule_path, rule, rule_lines): |
||
| 75 | if 'identifiers' in rule and rule['identifiers'] is not None: |
||
| 76 | for i_type, i_value in rule['identifiers'].items(): |
||
| 77 | if i_type[0:3] == 'cce': |
||
| 78 | has_prefix = i_value[0:3].upper() == 'CCE' |
||
| 79 | remainder_valid = cce.is_cce_format_valid("CCE-" + i_value[3:]) |
||
| 80 | remainder_valid |= cce.is_cce_format_valid("CCE-" + i_value[4:]) |
||
| 81 | return has_prefix and remainder_valid |
||
| 82 | return False |
||
| 83 | |||
| 84 | |||
| 85 | def has_invalid_cce(rule_path, rule, rule_lines): |
||
| 86 | if 'identifiers' in rule and rule['identifiers'] is not None: |
||
| 87 | for i_type, i_value in rule['identifiers'].items(): |
||
| 88 | if i_type[0:3] == 'cce': |
||
| 89 | if not cce.is_cce_value_valid("CCE-" + str(i_value)): |
||
| 90 | return True |
||
| 91 | return False |
||
| 92 | |||
| 93 | |||
| 94 | def has_int_identifier(rule_path, rule, rule_lines): |
||
| 95 | if 'identifiers' in rule and rule['identifiers'] is not None: |
||
| 96 | for _, value in rule['identifiers'].items(): |
||
| 97 | if type(value) != str: |
||
| 98 | return True |
||
| 99 | return False |
||
| 100 | |||
| 101 | |||
| 102 | def has_int_reference(rule_path, rule, rule_lines): |
||
| 103 | if 'references' in rule and rule['references'] is not None: |
||
| 104 | for _, value in rule['references'].items(): |
||
| 105 | if type(value) != str: |
||
| 106 | return True |
||
| 107 | return False |
||
| 108 | |||
| 109 | |||
| 110 | def has_duplicated_subkeys(rule_path, rule, rule_lines): |
||
| 111 | return ssg.rule_yaml.has_duplicated_subkeys(rule_path, rule_lines, TO_SORT) |
||
| 112 | |||
| 113 | |||
| 114 | def has_unordered_sections(rule_path, rule, rule_lines): |
||
| 115 | if 'references' in rule or 'identifiers' in rule: |
||
| 116 | new_lines = ssg.rule_yaml.sort_section_keys(rule_path, rule_lines, TO_SORT) |
||
| 117 | |||
| 118 | # Compare string representations to avoid issues with references being |
||
| 119 | # different. |
||
| 120 | return "\n".join(rule_lines) != "\n".join(new_lines) |
||
| 121 | |||
| 122 | return False |
||
| 123 | |||
| 124 | |||
| 125 | def rule_data_generator(args): |
||
| 126 | # Iterates over all know rules in the build system (according to |
||
| 127 | # rule_dir_json.py) and attempts to load the resulting YAML files. |
||
| 128 | # If they parse correctly, yield them as a result. |
||
| 129 | # |
||
| 130 | # Note: this has become a generator rather than returning a list of |
||
| 131 | # results. |
||
| 132 | |||
| 133 | product_yamls = dict() |
||
| 134 | |||
| 135 | rule_dirs = json.load(open(args.json)) |
||
| 136 | for rule_id in rule_dirs: |
||
| 137 | rule_obj = rule_dirs[rule_id] |
||
| 138 | |||
| 139 | if 'products' not in rule_obj or not rule_obj['products']: |
||
| 140 | print(rule_id, rule_obj) |
||
| 141 | assert rule_obj['products'] |
||
| 142 | product = rule_obj['products'][0] |
||
| 143 | |||
| 144 | if product not in product_yamls: |
||
| 145 | product_path = ssg.products.product_yaml_path(args.root, product) |
||
| 146 | product_yaml = ssg.products.load_product_yaml(product_path) |
||
| 147 | properties_directory = os.path.join(args.root, "product_properties") |
||
| 148 | product_yaml.read_properties_from_directory(properties_directory) |
||
| 149 | product_yamls[product] = product_yaml |
||
| 150 | |||
| 151 | local_env_yaml = dict(cmake_build_type='Debug') |
||
| 152 | local_env_yaml.update(product_yamls[product]) |
||
| 153 | local_env_yaml['rule_id'] = rule_id |
||
| 154 | |||
| 155 | rule_path = ssg.rules.get_rule_dir_yaml(rule_obj['dir']) |
||
| 156 | try: |
||
| 157 | rule = yaml.open_and_macro_expand(rule_path, local_env_yaml) |
||
| 158 | rule_lines = read_file_list(rule_path) |
||
| 159 | yield rule_path, rule, rule_lines, product_path, local_env_yaml |
||
|
|
|||
| 160 | except jinja2.exceptions.UndefinedError as ue: |
||
| 161 | msg = "Failed to parse file {0} (with product.yml: {1}). Skipping. {2}" |
||
| 162 | msg = msg.format(rule_path, product_path, ue) |
||
| 163 | print(msg, file=sys.stderr) |
||
| 164 | |||
| 165 | |||
| 166 | def find_rules_generator(args, func): |
||
| 167 | for item in rule_data_generator(args): |
||
| 168 | rule_path, rule, rule_lines, product_path, local_env_yaml = item |
||
| 169 | if func(rule_path, rule, rule_lines): |
||
| 170 | yield (rule_path, product_path, local_env_yaml) |
||
| 171 | |||
| 172 | |||
| 173 | def find_rules(args, func): |
||
| 174 | # Returns find_rules_generator as a list |
||
| 175 | return list(find_rules_generator(args, func)) |
||
| 176 | |||
| 177 | |||
| 178 | def print_file(file_contents): |
||
| 179 | for line_num in range(0, len(file_contents)): |
||
| 180 | print("%d: %s" % (line_num, file_contents[line_num])) |
||
| 181 | |||
| 182 | |||
| 183 | def find_section_lines(file_contents, sec): |
||
| 184 | # Hack to find a global key ("section"/sec) in a YAML-like file. |
||
| 185 | # All indented lines until the next global key are included in the range. |
||
| 186 | # For example: |
||
| 187 | # |
||
| 188 | # 0: not_it: |
||
| 189 | # 1: - value |
||
| 190 | # 2: this_one: |
||
| 191 | # 3: - 2 |
||
| 192 | # 4: - 5 |
||
| 193 | # 5: |
||
| 194 | # 6: nor_this: |
||
| 195 | # |
||
| 196 | # for the section "this_one", the result [(2, 5)] will be returned. |
||
| 197 | # Note that multiple sections may exist in a file and each will be |
||
| 198 | # identified and returned. |
||
| 199 | sec_ranges = [] |
||
| 200 | |||
| 201 | sec_id = sec + ":" |
||
| 202 | sec_len = len(sec_id) |
||
| 203 | end_num = len(file_contents) |
||
| 204 | line_num = 0 |
||
| 205 | |||
| 206 | View Code Duplication | while line_num < end_num: |
|
| 207 | if len(file_contents[line_num]) >= sec_len: |
||
| 208 | if file_contents[line_num][0:sec_len] == sec_id: |
||
| 209 | begin = line_num |
||
| 210 | line_num += 1 |
||
| 211 | while line_num < end_num: |
||
| 212 | if len(file_contents[line_num]) > 0 and file_contents[line_num][0] != ' ': |
||
| 213 | break |
||
| 214 | line_num += 1 |
||
| 215 | |||
| 216 | end = line_num - 1 |
||
| 217 | sec_ranges.append((begin, end)) |
||
| 218 | line_num += 1 |
||
| 219 | return sec_ranges |
||
| 220 | |||
| 221 | |||
| 222 | def remove_lines(file_contents, lines): |
||
| 223 | # Returns a series of lines and returns a new copy |
||
| 224 | new_file = [] |
||
| 225 | for line_num in range(0, len(file_contents)): |
||
| 226 | if line_num not in lines: |
||
| 227 | new_file.append(file_contents[line_num]) |
||
| 228 | |||
| 229 | return new_file |
||
| 230 | |||
| 231 | |||
| 232 | def remove_section_keys(file_contents, yaml_contents, section, removed_keys): |
||
| 233 | # Remove a series of keys from a section. Refuses to operate if there is more |
||
| 234 | # than one instance of the section. If the section is empty (because all keys |
||
| 235 | # are removed), then the section is also removed. Otherwise, only matching keys |
||
| 236 | # are removed. Note that all instances of the keys will be removed, if it appears |
||
| 237 | # more than once. |
||
| 238 | sec_ranges = find_section_lines(file_contents, section) |
||
| 239 | if len(sec_ranges) != 1: |
||
| 240 | raise RuntimeError("Refusing to fix file: %s -- could not find one section: %d" |
||
| 241 | % (path, sec_ranges)) |
||
| 242 | |||
| 243 | begin, end = sec_ranges[0] |
||
| 244 | r_lines = set() |
||
| 245 | |||
| 246 | if (yaml_contents[section] is None or len(yaml_contents[section].keys()) == len(removed_keys)): |
||
| 247 | r_lines = set(range(begin, end+1)) |
||
| 248 | print("Removing entire section since all keys are empty") |
||
| 249 | else: |
||
| 250 | # Don't include section header |
||
| 251 | for line_num in range(begin+1, end+1): |
||
| 252 | line = file_contents[line_num].strip() |
||
| 253 | len_line = len(line) |
||
| 254 | |||
| 255 | for key in removed_keys: |
||
| 256 | k_l = len(key)+1 |
||
| 257 | k_i = key + ":" |
||
| 258 | if len_line >= k_l and line[0:k_l] == k_i: |
||
| 259 | r_lines.add(line_num) |
||
| 260 | break |
||
| 261 | |||
| 262 | return remove_lines(file_contents, r_lines) |
||
| 263 | |||
| 264 | |||
| 265 | def rewrite_value_int_str(line): |
||
| 266 | # Rewrites a key's value to explicitly be a string. Assumes it starts |
||
| 267 | # as an integer. Takes a line. |
||
| 268 | key_end = line.index(':') |
||
| 269 | key = line[0:key_end] |
||
| 270 | value = line[key_end+1:].strip() |
||
| 271 | str_value = '"' + value + '"' |
||
| 272 | return key + ": " + str_value |
||
| 273 | |||
| 274 | |||
| 275 | def rewrite_keyless_section(file_contents, yaml_contents, section, content): |
||
| 276 | new_contents = file_contents[:] |
||
| 277 | |||
| 278 | sec_ranges = find_section_lines(file_contents, section) |
||
| 279 | if len(sec_ranges) != 1: |
||
| 280 | raise RuntimeError("Refusing to fix file: %s -- could not find one section: %d" |
||
| 281 | % (path, sec_ranges)) |
||
| 282 | |||
| 283 | if len(sec_ranges[0]) != 2: |
||
| 284 | raise RuntimeError("Section has more than one line") |
||
| 285 | |||
| 286 | new_contents[sec_ranges[0][0]] = "{section}: {content}".format(section=section, content=content) |
||
| 287 | |||
| 288 | return new_contents |
||
| 289 | |||
| 290 | |||
| 291 | def rewrite_value_remove_prefix(line): |
||
| 292 | # Rewrites a key's value to remove a "CCE" prefix. |
||
| 293 | key_end = line.index(':') |
||
| 294 | key = line[0:key_end] |
||
| 295 | value = line[key_end+1:].strip() |
||
| 296 | new_value = value |
||
| 297 | if cce.is_cce_format_valid("CCE-" + value[3:]): |
||
| 298 | new_value = value[3:] |
||
| 299 | elif cce.is_cce_format_valid("CCE-" + value[4:]): |
||
| 300 | new_value = value[4:] |
||
| 301 | return key + ": " + new_value |
||
| 302 | |||
| 303 | |||
| 304 | def add_to_the_section(file_contents, yaml_contents, section, new_keys): |
||
| 305 | to_insert = [] |
||
| 306 | |||
| 307 | sec_ranges = find_section_lines(file_contents, section) |
||
| 308 | if len(sec_ranges) != 1: |
||
| 309 | raise RuntimeError("could not find one section: %s" |
||
| 310 | % section) |
||
| 311 | |||
| 312 | begin, end = sec_ranges[0] |
||
| 313 | |||
| 314 | assert end > begin, "We need at least one identifier there already" |
||
| 315 | template_line = str(file_contents[end - 1]) |
||
| 316 | leading_whitespace = re.match(r"^\s*", template_line).group() |
||
| 317 | for key, value in new_keys.items(): |
||
| 318 | to_insert.append(leading_whitespace + key + ": " + value) |
||
| 319 | |||
| 320 | new_contents = file_contents[:end] + to_insert + file_contents[end:] |
||
| 321 | return new_contents |
||
| 322 | |||
| 323 | |||
| 324 | def sort_section(file_contents, yaml_contents, section): |
||
| 325 | new_contents = ssg.rule_yaml.sort_section_keys(yaml_contents, file_contents, section) |
||
| 326 | return new_contents |
||
| 327 | |||
| 328 | |||
| 329 | def rewrite_section_value(file_contents, yaml_contents, section, keys, transform): |
||
| 330 | # For a given section, rewrite the keys in int_keys to be strings. Refuses to |
||
| 331 | # operate if the given section appears more than once in the file. Assumes all |
||
| 332 | # instances of key are an integer; all will get updated. |
||
| 333 | new_contents = file_contents[:] |
||
| 334 | |||
| 335 | sec_ranges = find_section_lines(file_contents, section) |
||
| 336 | if len(sec_ranges) != 1: |
||
| 337 | raise RuntimeError("Refusing to fix file: %s -- could not find one section: %d" |
||
| 338 | % (path, sec_ranges)) |
||
| 339 | |||
| 340 | begin, end = sec_ranges[0] |
||
| 341 | r_lines = set() |
||
| 342 | |||
| 343 | # Don't include section header |
||
| 344 | for line_num in range(begin+1, end+1): |
||
| 345 | line = file_contents[line_num].strip() |
||
| 346 | len_line = len(line) |
||
| 347 | |||
| 348 | for key in keys: |
||
| 349 | k_l = len(key)+1 |
||
| 350 | k_i = key + ":" |
||
| 351 | |||
| 352 | if len_line >= k_l and line[0:k_l] == k_i: |
||
| 353 | new_contents[line_num] = transform(file_contents[line_num]) |
||
| 354 | break |
||
| 355 | |||
| 356 | return new_contents |
||
| 357 | |||
| 358 | |||
| 359 | def rewrite_section_value_int_str(file_contents, yaml_contents, section, int_keys): |
||
| 360 | return rewrite_section_value(file_contents, yaml_contents, section, int_keys, |
||
| 361 | rewrite_value_int_str) |
||
| 362 | |||
| 363 | |||
| 364 | View Code Duplication | def fix_empty_identifier(file_contents, yaml_contents): |
|
| 365 | section = 'identifiers' |
||
| 366 | |||
| 367 | empty_identifiers = [] |
||
| 368 | if yaml_contents[section] is not None: |
||
| 369 | for i_type, i_value in yaml_contents[section].items(): |
||
| 370 | if str(i_value).strip() == "": |
||
| 371 | empty_identifiers.append(i_type) |
||
| 372 | |||
| 373 | return remove_section_keys(file_contents, yaml_contents, section, empty_identifiers) |
||
| 374 | |||
| 375 | |||
| 376 | View Code Duplication | def fix_empty_reference(file_contents, yaml_contents): |
|
| 377 | section = 'references' |
||
| 378 | |||
| 379 | empty_identifiers = [] |
||
| 380 | |||
| 381 | if yaml_contents[section] is not None: |
||
| 382 | for i_type, i_value in yaml_contents[section].items(): |
||
| 383 | if str(i_value).strip() == "": |
||
| 384 | empty_identifiers.append(i_type) |
||
| 385 | |||
| 386 | return remove_section_keys(file_contents, yaml_contents, section, empty_identifiers) |
||
| 387 | |||
| 388 | |||
| 389 | def fix_prefix_cce(file_contents, yaml_contents): |
||
| 390 | section = 'identifiers' |
||
| 391 | |||
| 392 | prefixed_identifiers = [] |
||
| 393 | |||
| 394 | if yaml_contents[section] is not None: |
||
| 395 | for i_type, i_value in yaml_contents[section].items(): |
||
| 396 | if i_type[0:3] == 'cce': |
||
| 397 | has_prefix = i_value[0:3].upper() == 'CCE' |
||
| 398 | remainder_valid = cce.is_cce_format_valid("CCE-" + str(i_value[3:])) |
||
| 399 | remainder_valid |= cce.is_cce_format_valid("CCE-" + str(i_value[4:])) |
||
| 400 | if has_prefix and remainder_valid: |
||
| 401 | prefixed_identifiers.append(i_type) |
||
| 402 | |||
| 403 | return rewrite_section_value(file_contents, yaml_contents, section, prefixed_identifiers, |
||
| 404 | rewrite_value_remove_prefix) |
||
| 405 | |||
| 406 | |||
| 407 | def fix_invalid_cce(file_contents, yaml_contents): |
||
| 408 | section = 'identifiers' |
||
| 409 | |||
| 410 | invalid_identifiers = [] |
||
| 411 | |||
| 412 | if yaml_contents[section] is not None: |
||
| 413 | for i_type, i_value in yaml_contents[section].items(): |
||
| 414 | if i_type[0:3] == 'cce': |
||
| 415 | if not cce.is_cce_value_valid("CCE-" + str(i_value)): |
||
| 416 | invalid_identifiers.append(i_type) |
||
| 417 | |||
| 418 | return remove_section_keys(file_contents, yaml_contents, section, invalid_identifiers) |
||
| 419 | |||
| 420 | |||
| 421 | def fix_prodtypes(file_contents, yaml_contents): |
||
| 422 | section = 'prodtype' |
||
| 423 | sorted_prodtypes = yaml_contents[section].split(",") |
||
| 424 | sorted_prodtypes.sort() |
||
| 425 | out = ",".join(sorted_prodtypes) |
||
| 426 | |||
| 427 | return rewrite_keyless_section(file_contents, yaml_contents, section, out) |
||
| 428 | |||
| 429 | |||
| 430 | def has_product_cce(yaml_contents, product): |
||
| 431 | section = 'identifiers' |
||
| 432 | |||
| 433 | invalid_identifiers = [] |
||
| 434 | |||
| 435 | if not yaml_contents[section]: |
||
| 436 | return False |
||
| 437 | |||
| 438 | for i_type, i_value in yaml_contents[section].items(): |
||
| 439 | if i_type[0:3] != 'cce' or "@" not in i_type: |
||
| 440 | continue |
||
| 441 | |||
| 442 | _, cce_product = i_type.split("@", 1) |
||
| 443 | if product == cce_product: |
||
| 444 | return True |
||
| 445 | |||
| 446 | return False |
||
| 447 | |||
| 448 | |||
| 449 | def add_product_cce(file_contents, yaml_contents, product, cce): |
||
| 450 | section = 'identifiers' |
||
| 451 | |||
| 452 | if section not in yaml_contents: |
||
| 453 | return file_contents |
||
| 454 | |||
| 455 | new_contents = add_to_the_section( |
||
| 456 | file_contents, yaml_contents, section, {"cce@{product}".format(product=product): cce}) |
||
| 457 | new_contents = sort_section(new_contents, yaml_contents, section) |
||
| 458 | return new_contents |
||
| 459 | |||
| 460 | |||
| 461 | def fix_int_identifier(file_contents, yaml_contents): |
||
| 462 | section = 'identifiers' |
||
| 463 | |||
| 464 | int_identifiers = [] |
||
| 465 | for i_type, i_value in yaml_contents[section].items(): |
||
| 466 | if type(i_value) != str: |
||
| 467 | int_identifiers.append(i_type) |
||
| 468 | |||
| 469 | return rewrite_section_value_int_str(file_contents, yaml_contents, section, int_identifiers) |
||
| 470 | |||
| 471 | |||
| 472 | def fix_int_reference(file_contents, yaml_contents): |
||
| 473 | section = 'references' |
||
| 474 | |||
| 475 | int_identifiers = [] |
||
| 476 | for i_type, i_value in yaml_contents[section].items(): |
||
| 477 | if type(i_value) != str: |
||
| 478 | int_identifiers.append(i_type) |
||
| 479 | |||
| 480 | return rewrite_section_value_int_str(file_contents, yaml_contents, section, int_identifiers) |
||
| 481 | |||
| 482 | |||
| 483 | def sort_rule_subkeys(file_contents, yaml_contents): |
||
| 484 | return ssg.rule_yaml.sort_section_keys(None, file_contents, TO_SORT) |
||
| 485 | |||
| 486 | |||
| 487 | def _fixed_file_contents(path, file_contents, product_yaml, func): |
||
| 488 | if file_contents[-1] == '': |
||
| 489 | file_contents = file_contents[:-1] |
||
| 490 | |||
| 491 | subst_dict = product_yaml |
||
| 492 | yaml_contents = yaml.open_and_macro_expand(path, subst_dict) |
||
| 493 | |||
| 494 | try: |
||
| 495 | new_file_contents = func(file_contents, yaml_contents) |
||
| 496 | except Exception as exc: |
||
| 497 | msg = "Refusing to fix file: {path}: {error}".format(path=path, error=str(exc)) |
||
| 498 | raise RuntimeError(msg) |
||
| 499 | |||
| 500 | return new_file_contents |
||
| 501 | |||
| 502 | |||
| 503 | def fix_file(path, product_yaml, func): |
||
| 504 | file_contents = open(path, 'r').read().split("\n") |
||
| 505 | |||
| 506 | new_file_contents = _fixed_file_contents(path, file_contents, product_yaml, func) |
||
| 507 | if file_contents == new_file_contents: |
||
| 508 | return False |
||
| 509 | |||
| 510 | with open(path, 'w') as f: |
||
| 511 | for line in new_file_contents: |
||
| 512 | print(line, file=f) |
||
| 513 | return True |
||
| 514 | |||
| 515 | |||
| 516 | def fix_file_prompt(path, product_yaml, func, args): |
||
| 517 | file_contents = open(path, 'r').read().split("\n") |
||
| 518 | |||
| 519 | new_file_contents = _fixed_file_contents(path, file_contents, product_yaml, func) |
||
| 520 | changes = file_contents != new_file_contents |
||
| 521 | |||
| 522 | if not changes: |
||
| 523 | return changes |
||
| 524 | |||
| 525 | need_input = not args.assume_yes and not args.dry_run |
||
| 526 | |||
| 527 | if need_input: |
||
| 528 | print("====BEGIN BEFORE====") |
||
| 529 | print_file(file_contents) |
||
| 530 | print("====END BEFORE====") |
||
| 531 | |||
| 532 | if need_input: |
||
| 533 | print("====BEGIN AFTER====") |
||
| 534 | print_file(new_file_contents) |
||
| 535 | print("====END AFTER====") |
||
| 536 | |||
| 537 | response = 'n' |
||
| 538 | if need_input: |
||
| 539 | response = input_func("Confirm writing output to %s: (y/n): " % path) |
||
| 540 | |||
| 541 | if args.assume_yes or response.strip().lower() == 'y': |
||
| 542 | changes = True |
||
| 543 | with open(path, 'w') as f: |
||
| 544 | for line in new_file_contents: |
||
| 545 | print(line, file=f) |
||
| 546 | else: |
||
| 547 | changes = False |
||
| 548 | return changes |
||
| 549 | |||
| 550 | |||
| 551 | def add_cce(args, product_yaml): |
||
| 552 | directory = os.path.join(args.root, args.subdirectory) |
||
| 553 | cce_pool = cce.CCE_POOLS[args.cce_pool]() |
||
| 554 | return _add_cce(directory, cce_pool, args.rule, product_yaml, args) |
||
| 555 | |||
| 556 | |||
| 557 | def _add_cce(directory, cce_pool, rules, product_yaml, args): |
||
| 558 | product = product_yaml["product"] |
||
| 559 | |||
| 560 | def is_relevant_rule(rule_path, rule, rule_lines): |
||
| 561 | for r in rules: |
||
| 562 | if ( |
||
| 563 | rule_path.endswith("/{r}/rule.yml".format(r=r)) |
||
| 564 | and has_no_cce(rule_path, product_yaml)): |
||
| 565 | return True |
||
| 566 | return False |
||
| 567 | |||
| 568 | results = find_rules(args, is_relevant_rule) |
||
| 569 | |||
| 570 | for result in results: |
||
| 571 | rule_path = result[0] |
||
| 572 | |||
| 573 | cce = cce_pool.random_cce() |
||
| 574 | |||
| 575 | def fix_callback(file_contents, yaml_contents): |
||
| 576 | return add_product_cce(file_contents, yaml_contents, product_yaml["product"], cce) |
||
| 577 | |||
| 578 | try: |
||
| 579 | changes = fix_file(rule_path, product_yaml, fix_callback) |
||
| 580 | except RuntimeError as exc: |
||
| 581 | msg = ( |
||
| 582 | "Error adding CCE into {rule_path}: {exc}" |
||
| 583 | .format(rule_path=rule_path, exc=str(exc))) |
||
| 584 | raise RuntimeError(exc) |
||
| 585 | |||
| 586 | if changes: |
||
| 587 | cce_pool.remove_cce_from_file(cce) |
||
| 588 | |||
| 589 | |||
| 590 | def has_unsorted_prodtype(rule_path, rule, rule_lines): |
||
| 591 | if 'prodtype' in rule: |
||
| 592 | prodtypes = rule['prodtype'].split(',') |
||
| 593 | return prodtypes != sorted(prodtypes) |
||
| 594 | return False |
||
| 595 | |||
| 596 | |||
| 597 | @command("empty_identifiers", "check and fix rules with empty identifiers") |
||
| 598 | def fix_empty_identifiers(args, product_yaml): |
||
| 599 | results = find_rules(args, has_empty_identifier) |
||
| 600 | print("Number of rules with empty identifiers: %d" % len(results)) |
||
| 601 | |||
| 602 | for result in results: |
||
| 603 | rule_path = result[0] |
||
| 604 | |||
| 605 | product_yaml_path = result[2] |
||
| 606 | |||
| 607 | if product_yaml_path is not None: |
||
| 608 | product_yaml = yaml.open_raw(product_yaml_path) |
||
| 609 | |||
| 610 | if args.dry_run: |
||
| 611 | print(rule_path + " has one or more empty identifiers") |
||
| 612 | continue |
||
| 613 | |||
| 614 | fix_file_prompt(rule_path, product_yaml, fix_empty_identifier, args) |
||
| 615 | |||
| 616 | exit(int(len(results) > 0)) |
||
| 617 | |||
| 618 | |||
| 619 | @command("empty_references", "check and fix rules with empty references") |
||
| 620 | def fix_empty_references(args, product_yaml): |
||
| 621 | results = find_rules(args, has_empty_references) |
||
| 622 | print("Number of rules with empty references: %d" % len(results)) |
||
| 623 | |||
| 624 | for result in results: |
||
| 625 | rule_path = result[0] |
||
| 626 | product_yaml = result[2] |
||
| 627 | |||
| 628 | if args.dry_run: |
||
| 629 | print(rule_path + " has one or more empty references") |
||
| 630 | continue |
||
| 631 | |||
| 632 | fix_file_prompt(rule_path, product_yaml, fix_empty_reference, args) |
||
| 633 | |||
| 634 | exit(int(len(results) > 0)) |
||
| 635 | |||
| 636 | |||
| 637 | @command("prefixed_identifiers", "check and fix rules with prefixed (CCE-) identifiers") |
||
| 638 | def find_prefix_cce(args): |
||
| 639 | results = find_rules(args, has_prefix_cce) |
||
| 640 | print("Number of rules with prefixed CCEs: %d" % len(results)) |
||
| 641 | |||
| 642 | for result in results: |
||
| 643 | rule_path = result[0] |
||
| 644 | product_yaml = result[2] |
||
| 645 | |||
| 646 | if args.dry_run: |
||
| 647 | print(rule_path + " has one or more CCE with CCE- prefix") |
||
| 648 | continue |
||
| 649 | |||
| 650 | fix_file_prompt(rule_path, product_yaml, fix_prefix_cce, args) |
||
| 651 | |||
| 652 | exit(int(len(results) > 0)) |
||
| 653 | |||
| 654 | |||
| 655 | @command("invalid_identifiers", "check and fix rules with invalid identifiers") |
||
| 656 | def find_invalid_cce(args, product_yamls): |
||
| 657 | results = find_rules(args, has_invalid_cce) |
||
| 658 | print("Number of rules with invalid CCEs: %d" % len(results)) |
||
| 659 | |||
| 660 | for result in results: |
||
| 661 | rule_path = result[0] |
||
| 662 | product_yaml = result[2] |
||
| 663 | |||
| 664 | if args.dry_run: |
||
| 665 | print(rule_path + " has one or more invalid CCEs") |
||
| 666 | continue |
||
| 667 | |||
| 668 | fix_file_prompt(rule_path, product_yaml, fix_invalid_cce, args) |
||
| 669 | exit(int(len(results) > 0)) |
||
| 670 | |||
| 671 | |||
| 672 | @command("int_identifiers", "check and fix rules with pseudo-integer identifiers") |
||
| 673 | def find_int_identifiers(args, product_yaml): |
||
| 674 | results = find_rules(args, has_int_identifier) |
||
| 675 | print("Number of rules with integer identifiers: %d" % len(results)) |
||
| 676 | |||
| 677 | for result in results: |
||
| 678 | rule_path = result[0] |
||
| 679 | product_yaml = result[2] |
||
| 680 | |||
| 681 | if args.dry_run: |
||
| 682 | print(rule_path + " has one or more integer references") |
||
| 683 | continue |
||
| 684 | |||
| 685 | fix_file_prompt(rule_path, product_yaml, fix_int_identifier, args) |
||
| 686 | |||
| 687 | exit(int(len(results) > 0)) |
||
| 688 | |||
| 689 | |||
| 690 | @command("int_references", "check and fix rules with pseudo-integer references") |
||
| 691 | def find_int_references(args, product_yaml): |
||
| 692 | results = find_rules(args, has_int_reference) |
||
| 693 | print("Number of rules with integer references: %d" % len(results)) |
||
| 694 | |||
| 695 | for result in results: |
||
| 696 | rule_path = result[0] |
||
| 697 | product_yaml = result[2] |
||
| 698 | |||
| 699 | if args.dry_run: |
||
| 700 | print(rule_path + " has one or more unsorted references") |
||
| 701 | continue |
||
| 702 | |||
| 703 | fix_file_prompt(rule_path, product_yaml, fix_int_reference, args) |
||
| 704 | |||
| 705 | exit(int(len(results) > 0)) |
||
| 706 | |||
| 707 | |||
| 708 | @command("duplicate_subkeys", "check for duplicated references and identifiers") |
||
| 709 | def duplicate_subkeys(args, product_yaml): |
||
| 710 | results = find_rules(args, has_duplicated_subkeys) |
||
| 711 | print("Number of rules with duplicated subkeys: %d" % len(results)) |
||
| 712 | |||
| 713 | for result in results: |
||
| 714 | print(result[0] + " has one or more duplicated subkeys") |
||
| 715 | |||
| 716 | exit(int(len(results) > 0)) |
||
| 717 | |||
| 718 | |||
| 719 | @command("sort_subkeys", "sort references and identifiers") |
||
| 720 | def sort_subkeys(args, product_yaml): |
||
| 721 | results = find_rules(args, has_unordered_sections) |
||
| 722 | print("Number of modified rules: %d" % len(results)) |
||
| 723 | |||
| 724 | for result in results: |
||
| 725 | rule_path = result[0] |
||
| 726 | product_yaml = result[2] |
||
| 727 | |||
| 728 | if args.dry_run: |
||
| 729 | print(rule_path + " has one or more unsorted references") |
||
| 730 | continue |
||
| 731 | |||
| 732 | fix_file_prompt(rule_path, product_yaml, sort_rule_subkeys, args) |
||
| 733 | |||
| 734 | exit(int(len(results) > 0)) |
||
| 735 | |||
| 736 | |||
| 737 | @command("sort_prodtypes", "sorts the products in the prodtype") |
||
| 738 | def sort_prodtypes(args, product_yaml): |
||
| 739 | results = find_rules(args, has_unsorted_prodtype) |
||
| 740 | for result in results: |
||
| 741 | rule_path = result[0] |
||
| 742 | product_yaml = result[2] |
||
| 743 | |||
| 744 | if args.dry_run: |
||
| 745 | print(rule_path + " prodtype is unsorted") |
||
| 746 | continue |
||
| 747 | |||
| 748 | fix_file(rule_path, product_yaml, fix_prodtypes) |
||
| 749 | |||
| 750 | exit(int(len(results) > 0)) |
||
| 751 | |||
| 752 | |||
| 753 | @command("test_all", "Perform all checks on all rules") |
||
| 754 | def test_all(args, product_yaml): |
||
| 755 | result = 0 |
||
| 756 | checks = [ |
||
| 757 | (has_empty_identifier, "empty identifiers"), |
||
| 758 | (has_invalid_cce, "invalid CCEs"), |
||
| 759 | (has_int_identifier, "integer references"), |
||
| 760 | (has_empty_references, "empty references"), |
||
| 761 | (has_int_reference, "unsorted references"), |
||
| 762 | (has_duplicated_subkeys, "duplicated subkeys"), |
||
| 763 | (has_unordered_sections, "unsorted references"), |
||
| 764 | (has_unsorted_prodtype, "unsorted prodtype") |
||
| 765 | ] |
||
| 766 | for item in rule_data_generator(args): |
||
| 767 | rule_path, rule, rule_lines, _, _ = item |
||
| 768 | for func, msg in checks: |
||
| 769 | if func(rule_path, rule, rule_lines): |
||
| 770 | print("Rule '%s' has %s" % (rule_path, msg)) |
||
| 771 | result = 1 |
||
| 772 | exit(result) |
||
| 773 | |||
| 774 | |||
| 775 | def create_parser_from_functions(subparsers): |
||
| 776 | for name, function in _COMMANDS.items(): |
||
| 777 | subparser = subparsers.add_parser(name, description=function.description) |
||
| 778 | subparser.set_defaults(func=function) |
||
| 779 | |||
| 780 | |||
| 781 | def create_other_parsers(subparsers): |
||
| 782 | subparser = subparsers.add_parser("add-cce", description="Add CCE to rule files") |
||
| 783 | subparser.add_argument("rule", nargs="+") |
||
| 784 | subparser.add_argument("--subdirectory", default="linux_os") |
||
| 785 | subparser.add_argument( |
||
| 786 | "--cce-pool", "-p", default="redhat", choices=list(cce.CCE_POOLS.keys()), |
||
| 787 | ) |
||
| 788 | subparser.set_defaults(func=add_cce) |
||
| 789 | |||
| 790 | |||
| 791 | def parse_args(): |
||
| 792 | parser = argparse.ArgumentParser(formatter_class=argparse.RawDescriptionHelpFormatter, |
||
| 793 | description="Utility for fixing mistakes in rule files") |
||
| 794 | parser.add_argument( |
||
| 795 | "-y", "--assume-yes", default=False, action="store_true", |
||
| 796 | help="Assume yes and overwrite all files (no prompt)") |
||
| 797 | parser.add_argument( |
||
| 798 | "-d", "--dry-run", default=False, action="store_true", |
||
| 799 | help="Assume no and don't overwrite any files") |
||
| 800 | parser.add_argument( |
||
| 801 | "-j", "--json", type=str, action="store", |
||
| 802 | default="build/rule_dirs.json", help="File to read json " |
||
| 803 | "output of rule_dir_json.py from (defaults to " |
||
| 804 | "build/rule_dirs.json") |
||
| 805 | parser.add_argument( |
||
| 806 | "-r", "--root", default=SSG_ROOT, |
||
| 807 | help="Path to root of the project directory") |
||
| 808 | parser.add_argument("--product", "-p", help="Path to the main product.yml") |
||
| 809 | subparsers = parser.add_subparsers(title="command", help="What to perform.") |
||
| 810 | subparsers.required = True |
||
| 811 | create_parser_from_functions(subparsers) |
||
| 812 | create_other_parsers(subparsers) |
||
| 813 | return parser.parse_args() |
||
| 814 | |||
| 815 | |||
| 816 | def __main__(): |
||
| 817 | args = parse_args() |
||
| 818 | project_root = args.root |
||
| 819 | if not project_root: |
||
| 820 | project_root = os.path.join(os.path.dirname(os.path.abspath(__file__)), os.path.pardir) |
||
| 821 | |||
| 822 | subst_dict = dict() |
||
| 823 | if args.product: |
||
| 824 | subst_dict = dict() |
||
| 825 | product = products.load_product_yaml(args.product) |
||
| 826 | product.read_properties_from_directory(os.path.join(project_root, "product_properties")) |
||
| 827 | subst_dict.update(product) |
||
| 828 | |||
| 829 | args.func(args, subst_dict) |
||
| 830 | |||
| 831 | |||
| 832 | if __name__ == "__main__": |
||
| 833 | __main__() |
||
| 834 |