eserozvataf /
scabbia2-yaml
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 Inline 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. You can also have a look at the cohesion graph to spot any un-connected, or weakly-connected components.
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.
While breaking up the class, it is a good idea to analyze how other classes use Inline, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 43 | class Inline |
||
| 44 | { |
||
| 45 | /** @type string REGEX_QUOTED_STRING a regular expression pattern to match quoted strings */ |
||
| 46 | const REGEX_QUOTED_STRING = "(?:\"([^\"\\\\]*(?:\\\\.[^\"\\\\]*)*)\"|'([^']*(?:''[^']*)*)')"; |
||
| 47 | |||
| 48 | |||
| 49 | /** |
||
| 50 | * Converts a YAML string to a PHP array |
||
| 51 | * |
||
| 52 | * @param string $value A YAML string |
||
| 53 | * @param array $references Mapping of variable names to values |
||
| 54 | * |
||
| 55 | * @throws ParseException If the YAML is not valid |
||
| 56 | * @return array A PHP array representing the YAML string |
||
| 57 | */ |
||
| 58 | public static function parse($value, $references = []) |
||
| 59 | { |
||
| 60 | $value = trim($value); |
||
| 61 | |||
| 62 | if (strlen($value) === 0) { |
||
| 63 | return ""; |
||
| 64 | } |
||
| 65 | |||
| 66 | $i = 0; |
||
| 67 | if ($value[0] === "[") { |
||
| 68 | $result = self::parseSequence($value, $i, $references); |
||
| 69 | ++$i; |
||
| 70 | } elseif ($value[0] === "{") { |
||
| 71 | $result = self::parseMapping($value, $i, $references); |
||
| 72 | ++$i; |
||
| 73 | } else { |
||
| 74 | $result = self::parseScalar($value, null, ["\"", "'"], $i, true, $references); |
||
| 75 | } |
||
| 76 | |||
| 77 | // some comments are allowed at the end |
||
| 78 | if (preg_replace("/\\s+#.*$/A", "", substr($value, $i))) { |
||
| 79 | throw new ParseException(sprintf("Unexpected characters near \"%s\".", substr($value, $i))); |
||
| 80 | } |
||
| 81 | |||
| 82 | return $result; |
||
| 83 | } |
||
| 84 | |||
| 85 | /** |
||
| 86 | * Parses a scalar to a YAML string |
||
| 87 | * |
||
| 88 | * @param scalar $scalar |
||
| 89 | * @param string $delimiters |
||
| 90 | * @param array $stringDelimiters |
||
| 91 | * @param int &$i |
||
| 92 | * @param bool $evaluate |
||
| 93 | * @param array $references |
||
| 94 | * |
||
| 95 | * @throws ParseException When malformed inline YAML string is parsed |
||
| 96 | * @return string A YAML string |
||
| 97 | * |
||
| 98 | * @internal |
||
| 99 | */ |
||
| 100 | public static function parseScalar( |
||
| 101 | $scalar, |
||
| 102 | $delimiters = null, |
||
| 103 | array $stringDelimiters = ["\"", "'"], |
||
| 104 | &$i = 0, |
||
| 105 | $evaluate = true, |
||
| 106 | $references = [] |
||
| 107 | ) { |
||
| 108 | if (in_array($scalar[$i], $stringDelimiters)) { |
||
| 109 | // quoted scalar |
||
| 110 | $output = self::parseQuotedScalar($scalar, $i); |
||
| 111 | |||
| 112 | if ($delimiters !== null) { |
||
| 113 | $tmp = ltrim(substr($scalar, $i), " "); |
||
| 114 | if (!in_array($tmp[0], $delimiters)) { |
||
| 115 | throw new ParseException(sprintf("Unexpected characters (%s).", substr($scalar, $i))); |
||
| 116 | } |
||
| 117 | } |
||
| 118 | |||
| 119 | return $output; |
||
| 120 | } |
||
| 121 | |||
| 122 | // "normal" string |
||
| 123 | if (!$delimiters) { |
||
|
|
|||
| 124 | $output = substr($scalar, $i); |
||
| 125 | $i += strlen($output); |
||
| 126 | |||
| 127 | // remove comments |
||
| 128 | if (preg_match("/[ \t]+#/", $output, $match, PREG_OFFSET_CAPTURE)) { |
||
| 129 | $output = substr($output, 0, $match[0][1]); |
||
| 130 | } |
||
| 131 | } elseif (preg_match("/^(.+?)(" . implode("|", $delimiters) . ")/", substr($scalar, $i), $match)) { |
||
| 132 | $output = $match[1]; |
||
| 133 | $i += strlen($output); |
||
| 134 | } else { |
||
| 135 | throw new ParseException(sprintf("Malformed inline YAML string (%s).", $scalar)); |
||
| 136 | } |
||
| 137 | |||
| 138 | // a non-quoted string cannot start with @ or ` (reserved) nor with a scalar indicator (| or >) |
||
| 139 | if ($output && ($output[0] === "@" || $output[0] === "`" || $output[0] === "|" || $output[0] === ">")) { |
||
| 140 | throw new ParseException(sprintf("The reserved indicator \"%s\" cannot start a plain scalar; you need to quote the scalar.", $output[0])); |
||
| 141 | } |
||
| 142 | |||
| 143 | if ($evaluate) { |
||
| 144 | return self::evaluateScalar($output, $references); |
||
| 145 | } |
||
| 146 | |||
| 147 | return $output; |
||
| 148 | } |
||
| 149 | |||
| 150 | /** |
||
| 151 | * Parses a quoted scalar to YAML |
||
| 152 | * |
||
| 153 | * @param string $scalar |
||
| 154 | * @param int &$i |
||
| 155 | * |
||
| 156 | * @throws ParseException When malformed inline YAML string is parsed |
||
| 157 | * @return string A YAML string |
||
| 158 | */ |
||
| 159 | protected static function parseQuotedScalar($scalar, &$i) |
||
| 160 | { |
||
| 161 | if (!preg_match("/" . self::REGEX_QUOTED_STRING . "/Au", substr($scalar, $i), $match)) { |
||
| 162 | throw new ParseException(sprintf("Malformed inline YAML string (%s).", substr($scalar, $i))); |
||
| 163 | } |
||
| 164 | |||
| 165 | $output = substr($match[0], 1, strlen($match[0]) - 2); |
||
| 166 | |||
| 167 | $escaper = new Escaper(); |
||
| 168 | if ($scalar[$i] == "\"") { |
||
| 169 | $output = $escaper->unescapeDoubleQuotedString($output); |
||
| 170 | } else { |
||
| 171 | $output = $escaper->unescapeSingleQuotedString($output); |
||
| 172 | } |
||
| 173 | |||
| 174 | $i += strlen($match[0]); |
||
| 175 | |||
| 176 | return $output; |
||
| 177 | } |
||
| 178 | |||
| 179 | /** |
||
| 180 | * Parses a sequence to a YAML string |
||
| 181 | * |
||
| 182 | * @param string $sequence |
||
| 183 | * @param int &$i |
||
| 184 | * @param array $references |
||
| 185 | * |
||
| 186 | * @throws ParseException When malformed inline YAML string is parsed |
||
| 187 | * @return string A YAML string |
||
| 188 | */ |
||
| 189 | protected static function parseSequence($sequence, &$i = 0, $references = []) |
||
| 190 | { |
||
| 191 | $output = []; |
||
| 192 | $len = strlen($sequence); |
||
| 193 | ++$i; |
||
| 194 | |||
| 195 | // [foo, bar, ...] |
||
| 196 | while ($i < $len) { |
||
| 197 | if ($sequence[$i] === "[") { |
||
| 198 | // nested sequence |
||
| 199 | $output[] = self::parseSequence($sequence, $i, $references); |
||
| 200 | } elseif ($sequence[$i] === "{") { |
||
| 201 | // nested mapping |
||
| 202 | $output[] = self::parseMapping($sequence, $i, $references); |
||
| 203 | } elseif ($sequence[$i] === "]") { |
||
| 204 | return $output; |
||
| 205 | } elseif ($sequence[$i] !== "," && $sequence[$i] !== " ") { |
||
| 206 | $isQuoted = in_array($sequence[$i], ["\"", "'"]); |
||
| 207 | $value = self::parseScalar($sequence, [",", "]"], ["\"", "'"], $i, true, $references); |
||
| 208 | |||
| 209 | // the value can be an array if a reference has been resolved to an array var |
||
| 210 | if (!is_array($value) && !$isQuoted && strpos($value, ": ") !== false) { |
||
| 211 | // embedded mapping? |
||
| 212 | try { |
||
| 213 | $pos = 0; |
||
| 214 | $value = self::parseMapping("{" . $value . "}", $pos, $references); |
||
| 215 | } catch (\InvalidArgumentException $e) { |
||
| 216 | // no, it's not |
||
| 217 | } |
||
| 218 | } |
||
| 219 | |||
| 220 | $output[] = $value; |
||
| 221 | --$i; |
||
| 222 | } |
||
| 223 | |||
| 224 | ++$i; |
||
| 225 | } |
||
| 226 | |||
| 227 | throw new ParseException(sprintf("Malformed inline YAML string %s", $sequence)); |
||
| 228 | } |
||
| 229 | |||
| 230 | /** |
||
| 231 | * Parses a mapping to a YAML string |
||
| 232 | * |
||
| 233 | * @param string $mapping |
||
| 234 | * @param int &$i |
||
| 235 | * @param array $references |
||
| 236 | * |
||
| 237 | * @throws ParseException When malformed inline YAML string is parsed |
||
| 238 | * @return string A YAML string |
||
| 239 | */ |
||
| 240 | protected static function parseMapping($mapping, &$i = 0, $references = []) |
||
| 241 | { |
||
| 242 | $output = []; |
||
| 243 | $len = strlen($mapping); |
||
| 244 | ++$i; |
||
| 245 | |||
| 246 | // {foo: bar, bar:foo, ...} |
||
| 247 | while ($i < $len) { |
||
| 248 | if ($mapping[$i] === " " || $mapping[$i] === ",") { |
||
| 249 | ++$i; |
||
| 250 | continue; |
||
| 251 | } elseif ($mapping[$i] === "}") { |
||
| 252 | return $output; |
||
| 253 | } |
||
| 254 | |||
| 255 | // key |
||
| 256 | $key = self::parseScalar($mapping, [":", " "], ["\"", "'"], $i, false); |
||
| 257 | |||
| 258 | // value |
||
| 259 | $done = false; |
||
| 260 | while ($i < $len) { |
||
| 261 | if ($mapping[$i] === "[") { |
||
| 262 | // nested sequence |
||
| 263 | $output[$key] = self::parseSequence($mapping, $i, $references); |
||
| 264 | $done = true; |
||
| 265 | } elseif ($mapping[$i] === "{") { |
||
| 266 | // nested mapping |
||
| 267 | $output[$key] = self::parseMapping($mapping, $i, $references); |
||
| 268 | $done = true; |
||
| 269 | } elseif ($mapping[$i] !== ":" && $mapping[$i] !== " ") { |
||
| 270 | $output[$key] = self::parseScalar($mapping, [",", "}"], ["\"", "'"], $i, true, $references); |
||
| 271 | $done = true; |
||
| 272 | --$i; |
||
| 273 | } |
||
| 274 | |||
| 275 | ++$i; |
||
| 276 | |||
| 277 | if ($done) { |
||
| 278 | continue 2; |
||
| 279 | } |
||
| 280 | } |
||
| 281 | } |
||
| 282 | |||
| 283 | throw new ParseException(sprintf("Malformed inline YAML string %s", $mapping)); |
||
| 284 | } |
||
| 285 | |||
| 286 | /** |
||
| 287 | * Evaluates scalars and replaces magic values |
||
| 288 | * |
||
| 289 | * @param string $scalar |
||
| 290 | * @param array $references |
||
| 291 | * |
||
| 292 | * @throws ParseException when a reference could not be resolved |
||
| 293 | * @return string A YAML string |
||
| 294 | */ |
||
| 295 | protected static function evaluateScalar($scalar, $references = []) |
||
| 296 | { |
||
| 297 | $scalar = trim($scalar); |
||
| 298 | $scalarLower = strtolower($scalar); |
||
| 299 | |||
| 300 | if (strpos($scalar, "*") === 0) { |
||
| 301 | View Code Duplication | if (($pos = strpos($scalar, "#")) !== false) { |
|
| 302 | $value = substr($scalar, 1, $pos - 2); |
||
| 303 | } else { |
||
| 304 | $value = substr($scalar, 1); |
||
| 305 | } |
||
| 306 | |||
| 307 | // an unquoted * |
||
| 308 | if ($value === false || $value === "") { |
||
| 309 | throw new ParseException("A reference must contain at least one character."); |
||
| 310 | } |
||
| 311 | |||
| 312 | if (!array_key_exists($value, $references)) { |
||
| 313 | throw new ParseException(sprintf("Reference \"%s\" does not exist.", $value)); |
||
| 314 | } |
||
| 315 | |||
| 316 | return $references[$value]; |
||
| 317 | } |
||
| 318 | |||
| 319 | if ($scalarLower === "null" || $scalar === "" || $scalar === "~") { |
||
| 320 | return null; |
||
| 321 | } elseif ($scalarLower === "true") { |
||
| 322 | return true; |
||
| 323 | } elseif ($scalarLower === "false") { |
||
| 324 | return false; |
||
| 325 | } elseif ($scalar[0] === "+" || $scalar[0] === "-" || $scalar[0] === "." || $scalar[0] === "!" || |
||
| 326 | is_numeric($scalar[0])) { |
||
| 327 | // Optimise for returning strings. |
||
| 328 | if (strpos($scalar, "!str") === 0) { |
||
| 329 | return (string)substr($scalar, 5); |
||
| 330 | } elseif (strpos($scalar, "! ") === 0) { |
||
| 331 | return (int)self::parseScalar(substr($scalar, 2)); |
||
| 332 | } elseif (strpos($scalar, "!!php/object:") === 0) { |
||
| 333 | return unserialize(substr($scalar, 13)); |
||
| 334 | } elseif (strpos($scalar, "!!float ") === 0) { |
||
| 335 | return (float)substr($scalar, 8); |
||
| 336 | } elseif (ctype_digit($scalar)) { |
||
| 337 | $raw = $scalar; |
||
| 338 | $cast = (int)$scalar; |
||
| 339 | |||
| 340 | return $scalar[0] == "0" ? octdec($scalar) : (((string)$raw == (string)$cast) ? $cast : $raw); |
||
| 341 | } elseif ($scalar[0] === "-" && ctype_digit(substr($scalar, 1))) { |
||
| 342 | $raw = $scalar; |
||
| 343 | $cast = (int)$scalar; |
||
| 344 | |||
| 345 | return $scalar[1] == "0" ? octdec($scalar) : (((string)$raw == (string)$cast) ? $cast : $raw); |
||
| 346 | } elseif (is_numeric($scalar) || preg_match(self::getHexRegex(), $scalar)) { |
||
| 347 | return $scalar[0] . $scalar[1] == "0x" ? hexdec($scalar) : (float)$scalar; |
||
| 348 | } elseif ($scalarLower === ".inf" || $scalarLower === ".nan") { |
||
| 349 | return -log(0); |
||
| 350 | } elseif ($scalarLower === "-.inf") { |
||
| 351 | return log(0); |
||
| 352 | } elseif (preg_match("/^(-|\\+)?[0-9,]+(\\.[0-9]+)?$/", $scalar)) { |
||
| 353 | return (float)str_replace(",", "", $scalar); |
||
| 354 | } elseif (preg_match(self::getTimestampRegex(), $scalar)) { |
||
| 355 | return strtotime($scalar); |
||
| 356 | } |
||
| 357 | } else { |
||
| 358 | return (string)$scalar; |
||
| 359 | } |
||
| 360 | } |
||
| 361 | |||
| 362 | /** |
||
| 363 | * Gets a regex that matches a YAML date |
||
| 364 | * |
||
| 365 | * @return string The regular expression |
||
| 366 | * |
||
| 367 | * @see http://www.yaml.org/spec/1.2/spec.html#id2761573 |
||
| 368 | */ |
||
| 369 | public static function getTimestampRegex() |
||
| 370 | { |
||
| 371 | return <<<EOF |
||
| 372 | ~^ |
||
| 373 | (?P<year>[0-9][0-9][0-9][0-9]) |
||
| 374 | -(?P<month>[0-9][0-9]?) |
||
| 375 | -(?P<day>[0-9][0-9]?) |
||
| 376 | (?:(?:[Tt]|[ \t]+) |
||
| 377 | (?P<hour>[0-9][0-9]?) |
||
| 378 | :(?P<minute>[0-9][0-9]) |
||
| 379 | :(?P<second>[0-9][0-9]) |
||
| 380 | (?:\.(?P<fraction>[0-9]*))? |
||
| 381 | (?:[ \t]*(?P<tz>Z|(?P<tz_sign>[-+])(?P<tz_hour>[0-9][0-9]?) |
||
| 382 | (?::(?P<tz_minute>[0-9][0-9]))?))?)? |
||
| 383 | $~x |
||
| 384 | EOF; |
||
| 385 | } |
||
| 386 | |||
| 387 | /** |
||
| 388 | * Gets a regex that matches a YAML number in hexadecimal notation. |
||
| 389 | * |
||
| 390 | * @return string |
||
| 391 | */ |
||
| 392 | public static function getHexRegex() |
||
| 396 | } |
||
| 397 |
Loading history...
In PHP, under loose comparison (like
==, or!=, orswitchconditions), values of different types might be equal.For
stringvalues, the empty string''is a special case, in particular the following results might be unexpected: