Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
98.85% |
259 / 262 |
|
91.43% |
32 / 35 |
CRAP | |
0.00% |
0 / 1 |
| NameParser | |
98.85% |
259 / 262 |
|
91.43% |
32 / 35 |
113 | |
0.00% |
0 / 1 |
| getSalutation | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getFirstname | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getMiddlename | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getNickname | |
100.00% |
3 / 3 |
|
100.00% |
1 / 1 |
3 | |||
| getInitials | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getLastnamePrefix | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getLastname | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getSuffix | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getGivenName | |
100.00% |
2 / 2 |
|
100.00% |
1 / 1 |
2 | |||
| getFullName | |
100.00% |
2 / 2 |
|
100.00% |
1 / 1 |
2 | |||
| hasSalutation | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| hasFirstname | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| hasMiddlename | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| hasNickname | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| hasInitials | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| hasLastnamePrefix | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| hasLastname | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| hasSuffix | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| parse | |
100.00% |
33 / 33 |
|
100.00% |
1 / 1 |
9 | |||
| toArray | |
100.00% |
10 / 10 |
|
100.00% |
1 / 1 |
1 | |||
| clean | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| tokenize | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| normalizeCase | |
100.00% |
6 / 6 |
|
100.00% |
1 / 1 |
3 | |||
| normalizeWords | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| parseCommaMode | |
96.00% |
24 / 25 |
|
0.00% |
0 / 1 |
6 | |||
| absorbLeftovers | |
83.33% |
5 / 6 |
|
0.00% |
0 / 1 |
3.04 | |||
| finalizeInitials | |
100.00% |
8 / 8 |
|
100.00% |
1 / 1 |
4 | |||
| extractNickname | |
95.83% |
23 / 24 |
|
0.00% |
0 / 1 |
8 | |||
| extractSalutation | |
100.00% |
28 / 28 |
|
100.00% |
1 / 1 |
10 | |||
| extractSuffix | |
100.00% |
21 / 21 |
|
100.00% |
1 / 1 |
8 | |||
| extractInitials | |
100.00% |
24 / 24 |
|
100.00% |
1 / 1 |
14 | |||
| extractLastname | |
100.00% |
34 / 34 |
|
100.00% |
1 / 1 |
17 | |||
| extractFirstname | |
100.00% |
5 / 5 |
|
100.00% |
1 / 1 |
2 | |||
| extractMiddlename | |
100.00% |
4 / 4 |
|
100.00% |
1 / 1 |
2 | |||
| __toString | |
100.00% |
9 / 9 |
|
100.00% |
1 / 1 |
1 | |||
| 1 | <?php |
| 2 | /** |
| 3 | * Pop PHP Framework (https://www.popphp.org/) |
| 4 | * |
| 5 | * @link https://github.com/popphp/popphp-framework |
| 6 | * @author Nick Sagona, III <dev@noladev.com> |
| 7 | * @copyright Copyright (c) 2009-2026 NOLA Interactive, LLC. |
| 8 | * @license https://www.popphp.org/license New BSD License |
| 9 | */ |
| 10 | |
| 11 | /** |
| 12 | * @namespace |
| 13 | */ |
| 14 | namespace Pop\Parser\Name; |
| 15 | |
| 16 | use Pop\Parser\AbstractParser; |
| 17 | use Pop\Parser\Exception; |
| 18 | |
| 19 | /** |
| 20 | * Name parser class |
| 21 | * |
| 22 | * @category Pop |
| 23 | * @package Pop\Parser |
| 24 | * @author Nick Sagona, III <dev@noladev.com> |
| 25 | * @copyright Copyright (c) 2009-2026 NOLA Interactive, LLC. |
| 26 | * @license https://www.popphp.org/license New BSD License |
| 27 | * @version 1.0.0 |
| 28 | */ |
| 29 | class NameParser extends AbstractParser |
| 30 | { |
| 31 | |
| 32 | /** |
| 33 | * Salutation |
| 34 | * @var ?string |
| 35 | * */ |
| 36 | protected ?string $salutation = null; |
| 37 | |
| 38 | /** |
| 39 | * First name |
| 40 | * @var ?string |
| 41 | * */ |
| 42 | protected ?string $firstname = null; |
| 43 | |
| 44 | /** |
| 45 | * Middle name |
| 46 | * @var ?string |
| 47 | * */ |
| 48 | protected ?string $middlename = null; |
| 49 | |
| 50 | /** |
| 51 | * Nickname |
| 52 | * @var ?string |
| 53 | * */ |
| 54 | protected ?string $nickname = null; |
| 55 | |
| 56 | /** |
| 57 | * Initials |
| 58 | * @var ?string |
| 59 | * */ |
| 60 | protected ?string $initials = null; |
| 61 | |
| 62 | /** |
| 63 | * Lastname prefix |
| 64 | * @var ?string |
| 65 | * */ |
| 66 | protected ?string $lastnamePrefix = null; |
| 67 | |
| 68 | /** |
| 69 | * Last name |
| 70 | * @var ?string |
| 71 | * */ |
| 72 | protected ?string $lastname = null; |
| 73 | |
| 74 | /** |
| 75 | * Suffix |
| 76 | * @var ?string |
| 77 | * */ |
| 78 | protected ?string $suffix = null; |
| 79 | |
| 80 | /** |
| 81 | * Queue of raw tokens claimed as initials during extraction, before the first one is |
| 82 | * (potentially) promoted back to firstname by finalizeInitials() |
| 83 | * @var array |
| 84 | * */ |
| 85 | protected array $initialsQueue = []; |
| 86 | |
| 87 | /** |
| 88 | * Method to get salutation |
| 89 | * |
| 90 | * @return ?string |
| 91 | */ |
| 92 | public function getSalutation(): ?string |
| 93 | { |
| 94 | return $this->salutation; |
| 95 | } |
| 96 | |
| 97 | /** |
| 98 | * Method to get first name |
| 99 | * |
| 100 | * @return ?string |
| 101 | */ |
| 102 | public function getFirstname(): ?string |
| 103 | { |
| 104 | return $this->firstname; |
| 105 | } |
| 106 | |
| 107 | /** |
| 108 | * Method to get middle name |
| 109 | * |
| 110 | * @return ?string |
| 111 | */ |
| 112 | public function getMiddlename(): ?string |
| 113 | { |
| 114 | return $this->middlename; |
| 115 | } |
| 116 | |
| 117 | /** |
| 118 | * Method to get nickname |
| 119 | * |
| 120 | * @param bool $wrap |
| 121 | * @return ?string |
| 122 | */ |
| 123 | public function getNickname(bool $wrap = false): ?string |
| 124 | { |
| 125 | if (($this->nickname !== null) && $wrap) { |
| 126 | return '(' . $this->nickname . ')'; |
| 127 | } |
| 128 | |
| 129 | return $this->nickname; |
| 130 | } |
| 131 | |
| 132 | /** |
| 133 | * Method to get initials |
| 134 | * |
| 135 | * @return ?string |
| 136 | */ |
| 137 | public function getInitials(): ?string |
| 138 | { |
| 139 | return $this->initials; |
| 140 | } |
| 141 | |
| 142 | /** |
| 143 | * Method to get lastname prefix |
| 144 | * |
| 145 | * @return ?string |
| 146 | */ |
| 147 | public function getLastnamePrefix(): ?string |
| 148 | { |
| 149 | return $this->lastnamePrefix; |
| 150 | } |
| 151 | |
| 152 | /** |
| 153 | * Method to get last name |
| 154 | * |
| 155 | * @return ?string |
| 156 | */ |
| 157 | public function getLastname(): ?string |
| 158 | { |
| 159 | return $this->lastname; |
| 160 | } |
| 161 | |
| 162 | /** |
| 163 | * Method to get suffix |
| 164 | * |
| 165 | * @return ?string |
| 166 | */ |
| 167 | public function getSuffix(): ?string |
| 168 | { |
| 169 | return $this->suffix; |
| 170 | } |
| 171 | |
| 172 | /** |
| 173 | * Method to get the given name (first name, initials and middle name, in that order) |
| 174 | * |
| 175 | * @return ?string |
| 176 | */ |
| 177 | public function getGivenName(): ?string |
| 178 | { |
| 179 | $parts = array_filter([$this->firstname, $this->initials, $this->middlename]); |
| 180 | return !empty($parts) ? implode(' ', $parts) : null; |
| 181 | } |
| 182 | |
| 183 | /** |
| 184 | * Method to get the full name (given name plus lastname prefix and lastname) |
| 185 | * |
| 186 | * @return ?string |
| 187 | */ |
| 188 | public function getFullName(): ?string |
| 189 | { |
| 190 | $parts = array_filter([$this->getGivenName(), $this->lastnamePrefix, $this->lastname]); |
| 191 | return !empty($parts) ? implode(' ', $parts) : null; |
| 192 | } |
| 193 | |
| 194 | /** |
| 195 | * Has salutation |
| 196 | * |
| 197 | * @return bool |
| 198 | */ |
| 199 | public function hasSalutation(): bool |
| 200 | { |
| 201 | return !empty($this->salutation); |
| 202 | } |
| 203 | |
| 204 | /** |
| 205 | * Has first name |
| 206 | * |
| 207 | * @return bool |
| 208 | */ |
| 209 | public function hasFirstname(): bool |
| 210 | { |
| 211 | return !empty($this->firstname); |
| 212 | } |
| 213 | |
| 214 | /** |
| 215 | * Has middle name |
| 216 | * |
| 217 | * @return bool |
| 218 | */ |
| 219 | public function hasMiddlename(): bool |
| 220 | { |
| 221 | return !empty($this->middlename); |
| 222 | } |
| 223 | |
| 224 | /** |
| 225 | * Has nickname |
| 226 | * |
| 227 | * @return bool |
| 228 | */ |
| 229 | public function hasNickname(): bool |
| 230 | { |
| 231 | return !empty($this->nickname); |
| 232 | } |
| 233 | |
| 234 | /** |
| 235 | * Has initials |
| 236 | * |
| 237 | * @return bool |
| 238 | */ |
| 239 | public function hasInitials(): bool |
| 240 | { |
| 241 | return !empty($this->initials); |
| 242 | } |
| 243 | |
| 244 | /** |
| 245 | * Has lastname prefix |
| 246 | * |
| 247 | * @return bool |
| 248 | */ |
| 249 | public function hasLastnamePrefix(): bool |
| 250 | { |
| 251 | return !empty($this->lastnamePrefix); |
| 252 | } |
| 253 | |
| 254 | /** |
| 255 | * Has last name |
| 256 | * |
| 257 | * @return bool |
| 258 | */ |
| 259 | public function hasLastname(): bool |
| 260 | { |
| 261 | return !empty($this->lastname); |
| 262 | } |
| 263 | |
| 264 | /** |
| 265 | * Has suffix |
| 266 | * |
| 267 | * @return bool |
| 268 | */ |
| 269 | public function hasSuffix(): bool |
| 270 | { |
| 271 | return !empty($this->suffix); |
| 272 | } |
| 273 | |
| 274 | /** |
| 275 | * Parse method |
| 276 | * |
| 277 | * @param ?string $name |
| 278 | * @throws Exception |
| 279 | * @return static |
| 280 | */ |
| 281 | public function parse(?string $name = null): static |
| 282 | { |
| 283 | if (empty($this->data) && empty($name)) { |
| 284 | throw new Exception('Error: You must pass a name string to the parser object.'); |
| 285 | } |
| 286 | |
| 287 | if ((null === $name) && !empty($this->data)) { |
| 288 | $name = $this->data; |
| 289 | } else if ((null !== $name) && empty($this->data)) { |
| 290 | $this->data = $name; |
| 291 | } |
| 292 | |
| 293 | $name = $this->clean($name); |
| 294 | |
| 295 | if ($name === '') { |
| 296 | throw new Exception('Error: You must pass a name string to the parser object.'); |
| 297 | } |
| 298 | |
| 299 | // Several fields below are built by concatenation as extraction proceeds (e.g. |
| 300 | // consecutive salutations, multiple trailing suffixes). Reset them here so a second |
| 301 | // parse() call on the same instance starts clean rather than appending onto results |
| 302 | // from a previous call. |
| 303 | $this->salutation = null; |
| 304 | $this->firstname = null; |
| 305 | $this->middlename = null; |
| 306 | $this->nickname = null; |
| 307 | $this->initials = null; |
| 308 | $this->lastnamePrefix = null; |
| 309 | $this->lastname = null; |
| 310 | $this->suffix = null; |
| 311 | $this->initialsQueue = []; |
| 312 | |
| 313 | $nameValues = new NameValues(); |
| 314 | |
| 315 | if (str_contains($name, ',')) { |
| 316 | $this->parseCommaMode($name, $nameValues); |
| 317 | } else { |
| 318 | $tokens = $this->tokenize($name); |
| 319 | $originalCount = count($tokens); |
| 320 | $tokens = $this->extractNickname($tokens, $nameValues); |
| 321 | $tokens = $this->extractSalutation($tokens, $nameValues); |
| 322 | $tokens = $this->extractSuffix($tokens, $nameValues, 2); |
| 323 | $tokens = $this->extractInitials($tokens, false, $nameValues); |
| 324 | $tokens = $this->extractLastname($tokens, $nameValues, false, $originalCount); |
| 325 | $tokens = $this->extractFirstname($tokens); |
| 326 | $tokens = $this->extractMiddlename($tokens); |
| 327 | $this->absorbLeftovers($tokens); |
| 328 | } |
| 329 | |
| 330 | $this->finalizeInitials(); |
| 331 | |
| 332 | return $this; |
| 333 | } |
| 334 | |
| 335 | /** |
| 336 | * To array method |
| 337 | * |
| 338 | * @return array |
| 339 | */ |
| 340 | public function toArray(): array |
| 341 | { |
| 342 | return [ |
| 343 | 'salutation' => $this->salutation, |
| 344 | 'firstname' => $this->firstname, |
| 345 | 'initials' => $this->initials, |
| 346 | 'middlename' => $this->middlename, |
| 347 | 'nickname' => $this->nickname, |
| 348 | 'lastnamePrefix' => $this->lastnamePrefix, |
| 349 | 'lastname' => $this->lastname, |
| 350 | 'suffix' => $this->suffix, |
| 351 | ]; |
| 352 | } |
| 353 | |
| 354 | /** |
| 355 | * Clean method |
| 356 | * |
| 357 | * @param string $name |
| 358 | * @return string |
| 359 | */ |
| 360 | public function clean(string $name): string |
| 361 | { |
| 362 | return trim(preg_replace('/\s+/', ' ', $name)); |
| 363 | } |
| 364 | |
| 365 | /** |
| 366 | * Tokenize method |
| 367 | * |
| 368 | * @param string $name |
| 369 | * @return array |
| 370 | */ |
| 371 | protected function tokenize(string $name): array |
| 372 | { |
| 373 | return preg_split('/\s+/', trim($name)); |
| 374 | } |
| 375 | |
| 376 | /** |
| 377 | * Normalize the case of a single word: an all-uppercase or all-lowercase word gets |
| 378 | * title-cased ("MACDONALD" / "macdonald" -> "Macdonald"); a word with any existing |
| 379 | * mixed case (e.g. "MacDonald", "McDonald", "O'Brien") is left exactly as typed, since |
| 380 | * that mixed case is almost always deliberate. |
| 381 | * |
| 382 | * @param string $word |
| 383 | * @return string |
| 384 | */ |
| 385 | protected function normalizeCase(string $word): string |
| 386 | { |
| 387 | $stripped = str_replace('.', '', $word); |
| 388 | |
| 389 | if (($stripped === mb_strtoupper($stripped)) || ($stripped === mb_strtolower($stripped))) { |
| 390 | return preg_replace_callback('/\p{L}+/u', function ($matches) { |
| 391 | return mb_convert_case($matches[0], MB_CASE_TITLE); |
| 392 | }, $word); |
| 393 | } |
| 394 | |
| 395 | return $word; |
| 396 | } |
| 397 | |
| 398 | /** |
| 399 | * Normalize the case of each word in an array and join them with a space |
| 400 | * |
| 401 | * @param array $words |
| 402 | * @return string |
| 403 | */ |
| 404 | protected function normalizeWords(array $words): string |
| 405 | { |
| 406 | return implode(' ', array_map([$this, 'normalizeCase'], $words)); |
| 407 | } |
| 408 | |
| 409 | /** |
| 410 | * Parse comma-separated "Last, First Middle[, Suffix]" format |
| 411 | * |
| 412 | * @param string $name |
| 413 | * @param NameValues $nameValues |
| 414 | * @return void |
| 415 | */ |
| 416 | protected function parseCommaMode(string $name, NameValues $nameValues): void |
| 417 | { |
| 418 | $segments = array_map('trim', explode(',', $name)); |
| 419 | |
| 420 | // Segment 1 (before the first comma): the lastname segment. Salutation, suffix and |
| 421 | // lastname(-with-prefix) extraction run here. Whatever's left over (e.g. "Garcia" in |
| 422 | // "Garcia Marquez, Gabriel", or a lastname-prefix word extractLastname's ordinary |
| 423 | // guards wouldn't fold at position 0) is deliberately NOT assigned to firstname here |
| 424 | // - it's carried forward and absorbed only after segment 2 runs, so it can never |
| 425 | // silently overwrite or lose to segment 2's own firstname; absorbLeftovers() merges |
| 426 | // it into middlename once firstname is already set instead. |
| 427 | $segment1 = $this->tokenize($segments[0]); |
| 428 | $originalCount1 = count($segment1); |
| 429 | $segment1 = $this->extractSalutation($segment1, $nameValues); |
| 430 | $segment1 = $this->extractSuffix($segment1, $nameValues, 0, true); |
| 431 | $segment1 = $this->extractLastname($segment1, $nameValues, true, $originalCount1); |
| 432 | |
| 433 | // Segment 2 (between commas, or everything after the first comma): the given-name segment |
| 434 | if (isset($segments[1]) && ($segments[1] !== '')) { |
| 435 | $segment2 = $this->tokenize($segments[1]); |
| 436 | $segment2 = $this->extractSalutation($segment2, $nameValues); |
| 437 | $segment2 = $this->extractSuffix($segment2, $nameValues, 0, true, true); |
| 438 | $segment2 = $this->extractNickname($segment2, $nameValues); |
| 439 | $segment2 = $this->extractInitials($segment2, true, $nameValues); |
| 440 | $segment2 = $this->extractFirstname($segment2); |
| 441 | $segment2 = $this->extractMiddlename($segment2); |
| 442 | $this->absorbLeftovers($segment2); |
| 443 | } |
| 444 | |
| 445 | // Now that segment 2 has had its chance to claim the firstname slot: if it didn't |
| 446 | // (firstname is still null), segment 1's leftover IS the given name, so split it the |
| 447 | // normal way (first word -> firstname, the rest -> middlename) rather than joining it |
| 448 | // into one string. If segment 2 already provided a firstname, segment 1's leftover is |
| 449 | // secondary content - absorbLeftovers() merges the whole thing into middlename. |
| 450 | if ($this->firstname === null) { |
| 451 | $segment1 = $this->extractFirstname($segment1); |
| 452 | $segment1 = $this->extractMiddlename($segment1); |
| 453 | } |
| 454 | $this->absorbLeftovers($segment1); |
| 455 | |
| 456 | // Segment 3 onward (after a second comma, if present): suffix only per segment, with |
| 457 | // any non-suffix leftover absorbed rather than discarded (e.g. "Smith, John, Michael" |
| 458 | // must not lose "Michael" just because it isn't a recognized suffix). Looping over |
| 459 | // every remaining segment - not just $segments[2] - means a name with more than 3 |
| 460 | // comma-separated parts (e.g. "Smith, John, PhD, Esq") doesn't silently drop |
| 461 | // anything past the third. |
| 462 | foreach (array_slice($segments, 2) as $segment) { |
| 463 | if ($segment === '') { |
| 464 | continue; |
| 465 | } |
| 466 | $extraSegment = $this->tokenize($segment); |
| 467 | $extraSegment = $this->extractSuffix($extraSegment, $nameValues, 0, true); |
| 468 | $this->absorbLeftovers($extraSegment); |
| 469 | } |
| 470 | } |
| 471 | |
| 472 | /** |
| 473 | * Absorb any tokens no extraction step claimed: the first time this is called with a |
| 474 | * non-empty leftover, it becomes firstname (since every name needs one); after that, |
| 475 | * leftovers are appended to middlename. This is a deliberate difference from |
| 476 | * theiconic/name-parser, which can silently drop an unrecognized leading word (e.g. "The" |
| 477 | * in "The Rev. Mark Williams") - here nothing is ever discarded. |
| 478 | * |
| 479 | * @param array $tokens |
| 480 | * @return void |
| 481 | */ |
| 482 | protected function absorbLeftovers(array $tokens): void |
| 483 | { |
| 484 | if (empty($tokens)) { |
| 485 | return; |
| 486 | } |
| 487 | |
| 488 | $text = $this->normalizeWords($tokens); |
| 489 | |
| 490 | if ($this->firstname === null) { |
| 491 | $this->firstname = $text; |
| 492 | } else { |
| 493 | $this->middlename = trim(($this->middlename ?? '') . ' ' . $text); |
| 494 | } |
| 495 | } |
| 496 | |
| 497 | /** |
| 498 | * If no raw token ever became firstname but one or more initials were set aside, promote |
| 499 | * the FIRST claimed initial back to firstname - handles "J. B. Hunt" (firstname="J.", |
| 500 | * initials="B.", lastname="Hunt"), since a name consisting only of initials plus a |
| 501 | * lastname still needs a firstname. |
| 502 | * |
| 503 | * @return void |
| 504 | */ |
| 505 | protected function finalizeInitials(): void |
| 506 | { |
| 507 | if (empty($this->initialsQueue)) { |
| 508 | return; |
| 509 | } |
| 510 | |
| 511 | if ($this->firstname === null) { |
| 512 | $promoted = array_shift($this->initialsQueue); |
| 513 | $this->firstname = $this->normalizeCase($promoted); |
| 514 | } |
| 515 | |
| 516 | if (!empty($this->initialsQueue)) { |
| 517 | $this->initials = trim(($this->initials ?? '') . ' ' . $this->normalizeWords($this->initialsQueue)); |
| 518 | } |
| 519 | |
| 520 | $this->initialsQueue = []; |
| 521 | } |
| 522 | |
| 523 | /** |
| 524 | * Extract nickname |
| 525 | * |
| 526 | * Scans for a token starting with an opening delimiter and collects tokens until one |
| 527 | * ends with the matching closing delimiter (supports multi-word nicknames). |
| 528 | * |
| 529 | * @param array $tokens |
| 530 | * @param NameValues $nameValues |
| 531 | * @return array |
| 532 | */ |
| 533 | protected function extractNickname(array $tokens, NameValues $nameValues): array |
| 534 | { |
| 535 | $delimiters = $nameValues->getNicknameDelimiters(); |
| 536 | $openChars = array_keys($delimiters); |
| 537 | |
| 538 | foreach ($tokens as $start => $token) { |
| 539 | $firstChar = mb_substr($token, 0, 1); |
| 540 | if (!in_array($firstChar, $openChars, true)) { |
| 541 | continue; |
| 542 | } |
| 543 | |
| 544 | $closeChar = $delimiters[$firstChar]; |
| 545 | $end = null; |
| 546 | |
| 547 | for ($i = $start; $i < count($tokens); $i++) { |
| 548 | if (str_ends_with($tokens[$i], $closeChar) && (($i > $start) || (strlen($tokens[$i]) > 1))) { |
| 549 | $end = $i; |
| 550 | break; |
| 551 | } |
| 552 | } |
| 553 | |
| 554 | if ($end === null) { |
| 555 | continue; |
| 556 | } |
| 557 | |
| 558 | $span = array_slice($tokens, $start, $end - $start + 1); |
| 559 | $span[0] = ltrim($span[0], $firstChar); |
| 560 | $lastIndex = count($span) - 1; |
| 561 | $span[$lastIndex] = rtrim($span[$lastIndex], $closeChar); |
| 562 | $span = array_map(fn($word) => trim($word, '\'"'), $span); |
| 563 | $span = array_filter($span, fn($word) => $word !== ''); |
| 564 | |
| 565 | $this->nickname = $this->normalizeWords($span); |
| 566 | array_splice($tokens, $start, $end - $start + 1); |
| 567 | break; |
| 568 | } |
| 569 | |
| 570 | return array_values($tokens); |
| 571 | } |
| 572 | |
| 573 | /** |
| 574 | * Extract salutation |
| 575 | * |
| 576 | * Scans from the start of the tokens (bounded to roughly the first half) for matches |
| 577 | * against the salutation list, checked as both single tokens and multi-word phrases. |
| 578 | * Multiple consecutive salutations (e.g. "Rev. Dr John Doe") are all claimed. |
| 579 | * |
| 580 | * @param array $tokens |
| 581 | * @param NameValues $nameValues |
| 582 | * @return array |
| 583 | */ |
| 584 | protected function extractSalutation(array $tokens, NameValues $nameValues): array |
| 585 | { |
| 586 | $salutations = $nameValues->getSalutations(); |
| 587 | $claimed = []; |
| 588 | $index = 0; |
| 589 | |
| 590 | while (true) { |
| 591 | $max = !empty($tokens) ? max(1, (int) floor(count($tokens) / 2)) : 0; |
| 592 | if (($index >= $max) || ($index >= count($tokens))) { |
| 593 | break; |
| 594 | } |
| 595 | |
| 596 | $matchedLength = null; |
| 597 | $matchedValue = null; |
| 598 | |
| 599 | foreach ($salutations as $key => $display) { |
| 600 | $keyWords = explode(' ', $key); |
| 601 | $length = count($keyWords); |
| 602 | if (($index + $length) > count($tokens)) { |
| 603 | continue; |
| 604 | } |
| 605 | $subset = array_slice($tokens, $index, $length); |
| 606 | $subsetKeys = array_map(fn($word) => strtolower(str_replace('.', '', $word)), $subset); |
| 607 | if ($subsetKeys === $keyWords) { |
| 608 | $matchedLength = $length; |
| 609 | $matchedValue = $display; |
| 610 | break; |
| 611 | } |
| 612 | } |
| 613 | |
| 614 | if ($matchedLength === null) { |
| 615 | $index++; |
| 616 | continue; |
| 617 | } |
| 618 | |
| 619 | $claimed[] = $matchedValue; |
| 620 | array_splice($tokens, $index, $matchedLength); |
| 621 | } |
| 622 | |
| 623 | if (!empty($claimed)) { |
| 624 | $this->salutation = trim(($this->salutation ?? '') . ' ' . implode(' ', $claimed)); |
| 625 | } |
| 626 | |
| 627 | return array_values($tokens); |
| 628 | } |
| 629 | |
| 630 | /** |
| 631 | * Extract suffix |
| 632 | * |
| 633 | * Scans from the end backward while trailing tokens keep matching the suffix list, |
| 634 | * stopping before it would eat into the reserved leading tokens (or, in single-part |
| 635 | * mode, matches only when exactly one token remains). |
| 636 | * |
| 637 | * @param array $tokens |
| 638 | * @param NameValues $nameValues |
| 639 | * @param int $reservedParts |
| 640 | * @param bool $matchSinglePart |
| 641 | * @param bool $reserveLastToken |
| 642 | * @return array |
| 643 | */ |
| 644 | protected function extractSuffix( |
| 645 | array $tokens, |
| 646 | NameValues $nameValues, |
| 647 | int $reservedParts = 2, |
| 648 | bool $matchSinglePart = false, |
| 649 | bool $reserveLastToken = false |
| 650 | ): array |
| 651 | { |
| 652 | $suffixes = $nameValues->getSuffixes(); |
| 653 | |
| 654 | if ($matchSinglePart && (count($tokens) === 1)) { |
| 655 | $key = strtolower(str_replace('.', '', $tokens[0])); |
| 656 | if (isset($suffixes[$key])) { |
| 657 | $this->suffix = trim(($this->suffix ?? '') . ' ' . $suffixes[$key]); |
| 658 | return []; |
| 659 | } |
| 660 | return $tokens; |
| 661 | } |
| 662 | |
| 663 | $claimed = []; |
| 664 | $stop = $reserveLastToken ? 1 : $reservedParts; |
| 665 | $index = count($tokens) - 1; |
| 666 | |
| 667 | while ($index >= $stop) { |
| 668 | $key = strtolower(str_replace('.', '', $tokens[$index])); |
| 669 | if (!isset($suffixes[$key])) { |
| 670 | break; |
| 671 | } |
| 672 | array_unshift($claimed, $suffixes[$key]); |
| 673 | $index--; |
| 674 | } |
| 675 | |
| 676 | if (!empty($claimed)) { |
| 677 | $count = count($claimed); |
| 678 | array_splice($tokens, count($tokens) - $count, $count); |
| 679 | $this->suffix = trim(($this->suffix ?? '') . ' ' . implode(' ', $claimed)); |
| 680 | } |
| 681 | |
| 682 | return array_values($tokens); |
| 683 | } |
| 684 | |
| 685 | /** |
| 686 | * Extract initials |
| 687 | * |
| 688 | * A remaining single letter (optionally with a trailing period) is an initial. An |
| 689 | * all-caps 2-letter run (e.g. "JR") is split into two separate initials first - unless |
| 690 | * that run is also a recognized lastname prefix ("DE", "LA", "ST", ...), in which case |
| 691 | * it's left alone so extractLastname() can fold it as a prefix; without this guard, |
| 692 | * all-caps input like "JAMES DE LUCA" would have "DE" shredded into two fake initials |
| 693 | * before extractLastname() ever saw it. The very last remaining token is never treated |
| 694 | * as an initial unless $matchLastPart is true. |
| 695 | * |
| 696 | * @param array $tokens |
| 697 | * @param bool $matchLastPart |
| 698 | * @param NameValues $nameValues |
| 699 | * @return array |
| 700 | */ |
| 701 | protected function extractInitials(array $tokens, bool $matchLastPart, NameValues $nameValues): array |
| 702 | { |
| 703 | $prefixes = $nameValues->getLastnamePrefixes(); |
| 704 | |
| 705 | $last = count($tokens) - 1; |
| 706 | for ($i = 0; $i < count($tokens); $i++) { |
| 707 | if (!$matchLastPart && ($i === $last)) { |
| 708 | continue; |
| 709 | } |
| 710 | $stripped = str_replace('.', '', $tokens[$i]); |
| 711 | if ((strlen($stripped) === 2) && ($stripped === strtoupper($stripped)) && ctype_alpha($stripped) |
| 712 | && !isset($prefixes[strtolower($stripped)])) { |
| 713 | array_splice($tokens, $i, 1, [$stripped[0], $stripped[1]]); |
| 714 | $last = count($tokens) - 1; |
| 715 | $i++; |
| 716 | } |
| 717 | } |
| 718 | |
| 719 | $last = count($tokens) - 1; |
| 720 | $claimedIndexes = []; |
| 721 | |
| 722 | foreach ($tokens as $i => $token) { |
| 723 | if (!$matchLastPart && ($i === $last)) { |
| 724 | continue; |
| 725 | } |
| 726 | $stripped = str_replace('.', '', $token); |
| 727 | if (strlen($stripped) === 1) { |
| 728 | $claimedIndexes[] = $i; |
| 729 | } |
| 730 | } |
| 731 | |
| 732 | foreach ($claimedIndexes as $i) { |
| 733 | $this->initialsQueue[] = $tokens[$i]; |
| 734 | } |
| 735 | foreach (array_reverse($claimedIndexes) as $i) { |
| 736 | array_splice($tokens, $i, 1); |
| 737 | } |
| 738 | |
| 739 | return array_values($tokens); |
| 740 | } |
| 741 | |
| 742 | /** |
| 743 | * Extract lastname (with prefix folding) |
| 744 | * |
| 745 | * Scans remaining tokens from the end backward, claiming them as lastname. A claimed run |
| 746 | * immediately preceded by a recognized lastname-prefix word, with at least one unclaimed |
| 747 | * token still before it, folds the prefix into lastnamePrefix. Stops once it hits a word |
| 748 | * long enough to look like a complete lastname on its own with more still unclaimed |
| 749 | * before it - what keeps a middle name from being swallowed into the lastname. |
| 750 | * |
| 751 | * $originalCount is the token count BEFORE any earlier extraction step ran; it (not the |
| 752 | * current, shrunk token count) determines whether there was ever more than one word in |
| 753 | * this name to begin with, since a name reduced to a single remaining token by earlier |
| 754 | * steps (e.g. "J. B. Hunt" -> "Hunt" once both initials are claimed) should still have |
| 755 | * that token claimed as lastname. |
| 756 | * |
| 757 | * @param array $tokens |
| 758 | * @param NameValues $nameValues |
| 759 | * @param bool $singlePartOk |
| 760 | * @param ?int $originalCount |
| 761 | * @return array |
| 762 | */ |
| 763 | protected function extractLastname( |
| 764 | array $tokens, |
| 765 | NameValues $nameValues, |
| 766 | bool $singlePartOk = false, |
| 767 | ?int $originalCount = null |
| 768 | ): array |
| 769 | { |
| 770 | $originalCount ??= count($tokens); |
| 771 | |
| 772 | if ((!$singlePartOk && ($originalCount < 2)) || empty($tokens)) { |
| 773 | return $tokens; |
| 774 | } |
| 775 | |
| 776 | $prefixes = $nameValues->getLastnamePrefixes(); |
| 777 | $lastnameWords = []; |
| 778 | $prefixWords = []; |
| 779 | $index = count($tokens) - 1; |
| 780 | $claimedAny = false; |
| 781 | $lastClaimedWord = null; |
| 782 | |
| 783 | while ($index >= 0) { |
| 784 | $word = $tokens[$index]; |
| 785 | $key = strtolower(str_replace('.', '', $word)); |
| 786 | |
| 787 | // The "must be at index > 0" guard exists to always leave at least one token |
| 788 | // unclaimed for a firstname - but in singlePartOk mode (comma-mode segment 1), |
| 789 | // firstname comes from segment 2, not this segment, so there's nothing to |
| 790 | // reserve: allow folding and continued lastname-claiming all the way to index 0. |
| 791 | if ($claimedAny && isset($prefixes[$key]) && (($index > 0) || $singlePartOk)) { |
| 792 | array_unshift($prefixWords, $prefixes[$key]); |
| 793 | array_splice($tokens, $index, 1); |
| 794 | $index--; |
| 795 | continue; |
| 796 | } |
| 797 | |
| 798 | if ($claimedAny) { |
| 799 | if (!$singlePartOk && ($index < 1)) { |
| 800 | break; |
| 801 | } |
| 802 | if (strlen($lastClaimedWord) >= 3) { |
| 803 | break; |
| 804 | } |
| 805 | } |
| 806 | |
| 807 | array_unshift($lastnameWords, $word); |
| 808 | array_splice($tokens, $index, 1); |
| 809 | $claimedAny = true; |
| 810 | $lastClaimedWord = $word; |
| 811 | $index--; |
| 812 | |
| 813 | if (!$singlePartOk && ($index < 0)) { |
| 814 | break; |
| 815 | } |
| 816 | } |
| 817 | |
| 818 | if (!empty($lastnameWords)) { |
| 819 | $this->lastname = $this->normalizeWords($lastnameWords); |
| 820 | } |
| 821 | if (!empty($prefixWords)) { |
| 822 | $this->lastnamePrefix = implode(' ', $prefixWords); |
| 823 | } |
| 824 | |
| 825 | return array_values($tokens); |
| 826 | } |
| 827 | |
| 828 | /** |
| 829 | * Extract firstname |
| 830 | * |
| 831 | * If exactly one raw token remains, it's the firstname outright; otherwise the first |
| 832 | * remaining token becomes firstname. |
| 833 | * |
| 834 | * @param array $tokens |
| 835 | * @return array |
| 836 | */ |
| 837 | protected function extractFirstname(array $tokens): array |
| 838 | { |
| 839 | if (empty($tokens)) { |
| 840 | return $tokens; |
| 841 | } |
| 842 | |
| 843 | $this->firstname = $this->normalizeCase($tokens[0]); |
| 844 | array_splice($tokens, 0, 1); |
| 845 | |
| 846 | return array_values($tokens); |
| 847 | } |
| 848 | |
| 849 | /** |
| 850 | * Extract middlename |
| 851 | * |
| 852 | * Whatever raw tokens remain after firstname extraction join as middlename. |
| 853 | * |
| 854 | * @param array $tokens |
| 855 | * @return array |
| 856 | */ |
| 857 | protected function extractMiddlename(array $tokens): array |
| 858 | { |
| 859 | if (empty($tokens)) { |
| 860 | return $tokens; |
| 861 | } |
| 862 | |
| 863 | $this->middlename = trim(($this->middlename ?? '') . ' ' . $this->normalizeWords($tokens)); |
| 864 | |
| 865 | return []; |
| 866 | } |
| 867 | |
| 868 | /** |
| 869 | * To string method |
| 870 | * |
| 871 | * @return string |
| 872 | */ |
| 873 | public function __toString(): string |
| 874 | { |
| 875 | $parts = array_filter([ |
| 876 | $this->salutation, |
| 877 | $this->getGivenName(), |
| 878 | $this->getNickname(true), |
| 879 | $this->lastnamePrefix, |
| 880 | $this->lastname, |
| 881 | $this->suffix, |
| 882 | ]); |
| 883 | |
| 884 | return implode(' ', $parts); |
| 885 | } |
| 886 | |
| 887 | } |