Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
97.06% |
165 / 170 |
|
70.00% |
7 / 10 |
CRAP | |
0.00% |
0 / 1 |
| Parser | |
97.06% |
165 / 170 |
|
70.00% |
7 / 10 |
58 | |
0.00% |
0 / 1 |
| parseHeaders | |
100.00% |
24 / 24 |
|
100.00% |
1 / 1 |
6 | |||
| parseMediaType | |
100.00% |
18 / 18 |
|
100.00% |
1 / 1 |
4 | |||
| parseDataByContentType | |
96.00% |
24 / 25 |
|
0.00% |
0 / 1 |
16 | |||
| parseXml | |
100.00% |
18 / 18 |
|
100.00% |
1 / 1 |
4 | |||
| parseResponseFromUri | |
100.00% |
10 / 10 |
|
100.00% |
1 / 1 |
1 | |||
| parseResponseFromString | |
100.00% |
15 / 15 |
|
100.00% |
1 / 1 |
2 | |||
| encodeData | |
100.00% |
13 / 13 |
|
100.00% |
1 / 1 |
7 | |||
| decodeData | |
100.00% |
16 / 16 |
|
100.00% |
1 / 1 |
9 | |||
| decodeChunkedData | |
86.36% |
19 / 22 |
|
0.00% |
0 / 1 |
6.09 | |||
| isOfficeDocument | |
88.89% |
8 / 9 |
|
0.00% |
0 / 1 |
3.01 | |||
| 1 | <?php |
| 2 | declare(strict_types=1); |
| 3 | /** |
| 4 | * Pop PHP Framework (https://www.popphp.org/) |
| 5 | * |
| 6 | * @link https://github.com/popphp/popphp-framework |
| 7 | * @author Nick Sagona, III <dev@noladev.com> |
| 8 | * @copyright Copyright (c) 2009-2027 NOLA Interactive, LLC. |
| 9 | * @license https://www.popphp.org/license New BSD License |
| 10 | */ |
| 11 | |
| 12 | /** |
| 13 | * @namespace |
| 14 | */ |
| 15 | namespace Pop\Http; |
| 16 | |
| 17 | use Pop\Mime\Part\Header; |
| 18 | |
| 19 | /** |
| 20 | * HTTP response parser class |
| 21 | * |
| 22 | * @category Pop |
| 23 | * @package Pop\Http |
| 24 | * @author Nick Sagona, III <dev@noladev.com> |
| 25 | * @copyright Copyright (c) 2009-2027 NOLA Interactive, LLC. |
| 26 | * @license https://www.popphp.org/license New BSD License |
| 27 | * @version 6.0.0 |
| 28 | */ |
| 29 | class Parser |
| 30 | { |
| 31 | |
| 32 | /** |
| 33 | * Encoding constants |
| 34 | * @var string |
| 35 | */ |
| 36 | const BASE64 = 'BASE64'; |
| 37 | const QUOTED = 'QUOTED'; |
| 38 | const URL = 'URL'; |
| 39 | const RAW_URL = 'RAW_URL'; |
| 40 | const GZIP = 'GZIP'; |
| 41 | const DEFLATE = 'DEFLATE'; |
| 42 | |
| 43 | /** |
| 44 | * Parse headers |
| 45 | * |
| 46 | * @param mixed $headers |
| 47 | * @return array |
| 48 | */ |
| 49 | public static function parseHeaders(mixed $headers): array |
| 50 | { |
| 51 | $httpVersion = null; |
| 52 | $httpCode = null; |
| 53 | $httpMessage = null; |
| 54 | $headerObjects = []; |
| 55 | |
| 56 | $headers = (is_string($headers)) ? |
| 57 | array_map('trim', explode("\n", $headers)) : (array)$headers; |
| 58 | |
| 59 | foreach ($headers as $header) { |
| 60 | if (str_contains($header, 'HTTP/')) { |
| 61 | $httpVersion = substr($header, 0, strpos($header, ' ')); |
| 62 | $httpVersion = substr($httpVersion, (strpos($httpVersion, '/') + 1)); |
| 63 | |
| 64 | $match = []; |
| 65 | preg_match('/\d\d\d/', trim($header), $match); |
| 66 | |
| 67 | if (isset($match[0])) { |
| 68 | $httpCode = $match[0]; |
| 69 | $httpMessage = trim(substr($header, strpos($header, ' ' . $httpCode . ' ') + 5)); |
| 70 | } |
| 71 | } else if (str_contains($header, ':')) { |
| 72 | $headerObject = Header::parse($header); |
| 73 | $headerObjects[$headerObject->getName()] = $headerObject; |
| 74 | } |
| 75 | } |
| 76 | |
| 77 | return [ |
| 78 | 'version' => $httpVersion, |
| 79 | 'code' => $httpCode, |
| 80 | 'message' => $httpMessage, |
| 81 | 'headers' => $headerObjects |
| 82 | ]; |
| 83 | } |
| 84 | |
| 85 | /** |
| 86 | * Parse a Content-Type header value into its structured parts |
| 87 | * |
| 88 | * @param string $contentType |
| 89 | * @return array |
| 90 | */ |
| 91 | public static function parseMediaType(string $contentType): array |
| 92 | { |
| 93 | $parts = explode(';', $contentType); |
| 94 | $fullType = strtolower(trim(array_shift($parts))); |
| 95 | $params = []; |
| 96 | |
| 97 | foreach ($parts as $part) { |
| 98 | if (str_contains($part, '=')) { |
| 99 | [$key, $value] = array_map('trim', explode('=', $part, 2)); |
| 100 | $params[strtolower($key)] = trim($value, '"'); |
| 101 | } |
| 102 | } |
| 103 | |
| 104 | [$type, $subtypeAndSuffix] = array_pad(explode('/', $fullType, 2), 2, ''); |
| 105 | $suffix = null; |
| 106 | $subtype = $subtypeAndSuffix; |
| 107 | |
| 108 | if (str_contains($subtypeAndSuffix, '+')) { |
| 109 | [$subtype, $suffix] = explode('+', $subtypeAndSuffix, 2); |
| 110 | } |
| 111 | |
| 112 | return [ |
| 113 | 'type' => $type, |
| 114 | 'subtype' => $subtype, |
| 115 | 'suffix' => $suffix, |
| 116 | 'params' => $params, |
| 117 | ]; |
| 118 | } |
| 119 | |
| 120 | /** |
| 121 | * Parse request or response data based on content type |
| 122 | * |
| 123 | * @param string $rawData |
| 124 | * @param ?string $contentType |
| 125 | * @param ?string $encoding |
| 126 | * @param bool $chunked |
| 127 | * @throws Exception |
| 128 | * @return mixed |
| 129 | */ |
| 130 | public static function parseDataByContentType( |
| 131 | string $rawData, ?string $contentType = null, ?string $encoding = null, bool $chunked = false |
| 132 | ): mixed |
| 133 | { |
| 134 | if ($encoding !== null) { |
| 135 | $encoding = strtoupper($encoding); |
| 136 | } |
| 137 | |
| 138 | if ($contentType === null) { |
| 139 | return self::decodeData($rawData, $encoding, $chunked); |
| 140 | } |
| 141 | |
| 142 | $media = self::parseMediaType($contentType); |
| 143 | $rawData = self::decodeData($rawData, $encoding, $chunked); |
| 144 | |
| 145 | if (isset($media['params']['charset']) && (strtoupper($media['params']['charset']) !== 'UTF-8')) { |
| 146 | $converted = @mb_convert_encoding($rawData, 'UTF-8', $media['params']['charset']); |
| 147 | if ($converted !== false) { |
| 148 | $rawData = $converted; |
| 149 | } |
| 150 | } |
| 151 | |
| 152 | // JSON: application/json, or a structured +json suffix (e.g. application/vnd.api+json) |
| 153 | if (($media['subtype'] === 'json') || ($media['suffix'] === 'json')) { |
| 154 | return json_decode($rawData, true); |
| 155 | } |
| 156 | |
| 157 | // Generic XML only: application/xml or text/xml specifically - NOT a |
| 158 | // +xml structured suffix (e.g. application/xhtml+xml, application/rdf+xml), |
| 159 | // and not an office document that merely contains "xml" in its name. |
| 160 | if ((($media['subtype'] === 'xml') && ($media['suffix'] === null)) && |
| 161 | !self::isOfficeDocument($contentType)) { |
| 162 | return self::parseXml($rawData); |
| 163 | } |
| 164 | |
| 165 | if (($media['type'] === 'application') && ($media['subtype'] === 'x-www-form-urlencoded')) { |
| 166 | $parsedResult = []; |
| 167 | parse_str($rawData, $parsedResult); |
| 168 | return $parsedResult; |
| 169 | } |
| 170 | |
| 171 | if (($media['type'] === 'multipart') && ($media['subtype'] === 'form-data')) { |
| 172 | $boundary = $media['params']['boundary'] ?? null; |
| 173 | if ($boundary === null) { |
| 174 | throw new Exception('Error: The multipart/form-data content type is missing its boundary parameter.'); |
| 175 | } |
| 176 | return \Pop\Http\Body\Multipart::parse($rawData, $boundary); |
| 177 | } |
| 178 | |
| 179 | return $rawData; |
| 180 | } |
| 181 | |
| 182 | /** |
| 183 | * Parse an XML string into an array, throwing on malformed input instead |
| 184 | * of silently returning a nonsense value. |
| 185 | * |
| 186 | * @param string $rawData |
| 187 | * @throws Exception |
| 188 | * @return array |
| 189 | */ |
| 190 | protected static function parseXml(string $rawData): array |
| 191 | { |
| 192 | $matches = []; |
| 193 | preg_match_all('/<!\[cdata\[(.*?)\]\]>/is', $rawData, $matches); |
| 194 | |
| 195 | foreach ($matches[0] as $match) { |
| 196 | $strip = str_replace( |
| 197 | ['<![CDATA[', ']]>', '<', '>'], |
| 198 | ['', '', '<', '>'], |
| 199 | $match |
| 200 | ); |
| 201 | $rawData = str_replace($match, $strip, $rawData); |
| 202 | } |
| 203 | |
| 204 | $previous = libxml_use_internal_errors(true); |
| 205 | libxml_clear_errors(); |
| 206 | $xml = simplexml_load_string($rawData); |
| 207 | $errors = libxml_get_errors(); |
| 208 | libxml_use_internal_errors($previous); |
| 209 | |
| 210 | if ($xml === false) { |
| 211 | $message = !empty($errors) ? trim($errors[0]->message) : 'unknown XML parse error'; |
| 212 | throw new Exception('Error: Unable to parse XML content - ' . $message); |
| 213 | } |
| 214 | |
| 215 | return json_decode(json_encode((array)$xml), true); |
| 216 | } |
| 217 | |
| 218 | /** |
| 219 | * Parse a response from a URI |
| 220 | * |
| 221 | * @param string $uri |
| 222 | * @param string $method |
| 223 | * @param string $mode |
| 224 | * @param array $options |
| 225 | * @param array $params |
| 226 | * @throws Client\Exception|Exception |
| 227 | * @return Server\Response |
| 228 | */ |
| 229 | public static function parseResponseFromUri( |
| 230 | string $uri, string $method = 'GET', string $mode = 'r', array $options = [], array $params = [] |
| 231 | ): Server\Response |
| 232 | { |
| 233 | $request = new Client\Request($uri, $method); |
| 234 | $handler = new Client\Handler\Stream($mode, $options, $params); |
| 235 | $response = $handler->prepare($request, null, false)->send(); |
| 236 | |
| 237 | return new Server\Response([ |
| 238 | 'code' => $response->getCode(), |
| 239 | 'headers' => $response->getHeaderObjects(), |
| 240 | 'body' => $response->getBody(), |
| 241 | 'message' => $response->getMessage(), |
| 242 | 'version' => $response->getVersion() |
| 243 | ]); |
| 244 | } |
| 245 | |
| 246 | /** |
| 247 | * Parse a response from a full response string |
| 248 | * |
| 249 | * @param string $responseString |
| 250 | * @return Server\Response |
| 251 | */ |
| 252 | public static function parseResponseFromString(string $responseString): Server\Response |
| 253 | { |
| 254 | $headerString = substr($responseString, 0, strpos($responseString, "\r\n\r\n")); |
| 255 | $bodyString = substr($responseString, (strpos($responseString, "\r\n\r\n") + 4)); |
| 256 | $parsedHeaders = self::parseHeaders($headerString); |
| 257 | |
| 258 | // If the body content is encoded, decode the body content |
| 259 | if (array_key_exists('Content-Encoding', $parsedHeaders['headers'])) { |
| 260 | $encoding = strtoupper((string)$parsedHeaders['headers']['Content-Encoding']->getValueAsString()); |
| 261 | $chunked = ($parsedHeaders['headers']['Transfer-Encoding'] == 'chunked'); |
| 262 | $body = self::decodeData($bodyString, $encoding, $chunked); |
| 263 | } else { |
| 264 | $body = $bodyString; |
| 265 | } |
| 266 | |
| 267 | return new Server\Response([ |
| 268 | 'code' => (int)$parsedHeaders['code'], |
| 269 | 'headers' => $parsedHeaders['headers'], |
| 270 | 'body' => $body, |
| 271 | 'message' => $parsedHeaders['message'], |
| 272 | 'version' => $parsedHeaders['version'] |
| 273 | ]); |
| 274 | } |
| 275 | |
| 276 | /** |
| 277 | * Encode data |
| 278 | * |
| 279 | * @param string $data |
| 280 | * @param ?string $encoding |
| 281 | * @return string |
| 282 | */ |
| 283 | public static function encodeData(string $data, ?string $encoding = null): string |
| 284 | { |
| 285 | switch ($encoding) { |
| 286 | case self::BASE64: |
| 287 | $data = base64_encode($data); |
| 288 | break; |
| 289 | case self::QUOTED: |
| 290 | $data = quoted_printable_encode($data); |
| 291 | break; |
| 292 | case self::URL: |
| 293 | $data = urlencode($data); |
| 294 | break; |
| 295 | case self::RAW_URL: |
| 296 | $data = rawurlencode($data); |
| 297 | break; |
| 298 | case self::GZIP: |
| 299 | $data = gzencode($data); |
| 300 | break; |
| 301 | case self::DEFLATE: |
| 302 | $data = gzdeflate($data); |
| 303 | break; |
| 304 | } |
| 305 | |
| 306 | return $data; |
| 307 | } |
| 308 | |
| 309 | /** |
| 310 | * Decode data |
| 311 | * |
| 312 | * @param string $data |
| 313 | * @param ?string $encoding |
| 314 | * @param bool $chunked |
| 315 | * @return string |
| 316 | */ |
| 317 | public static function decodeData(string $data, ?string $encoding = null, bool $chunked = false): string |
| 318 | { |
| 319 | if ($chunked) { |
| 320 | $data = self::decodeChunkedData($data); |
| 321 | } |
| 322 | |
| 323 | switch ($encoding) { |
| 324 | case self::BASE64: |
| 325 | $data = base64_decode($data); |
| 326 | break; |
| 327 | case self::QUOTED: |
| 328 | $data = quoted_printable_decode($data); |
| 329 | break; |
| 330 | case self::URL: |
| 331 | $data = urldecode($data); |
| 332 | break; |
| 333 | case self::RAW_URL: |
| 334 | $data = rawurldecode($data); |
| 335 | break; |
| 336 | case self::GZIP: |
| 337 | $data = gzdecode($data); |
| 338 | break; |
| 339 | case self::DEFLATE: |
| 340 | $zLib = unpack('n', substr($data, 0, 2)); |
| 341 | $data = ($zLib[1] % 31 == 0) ? gzuncompress($data) : gzinflate($data); |
| 342 | break; |
| 343 | } |
| 344 | |
| 345 | return $data; |
| 346 | } |
| 347 | |
| 348 | /** |
| 349 | * Decode a chunked transfer-encoded data and return the decoded data |
| 350 | * |
| 351 | * @param string $data |
| 352 | * @return string |
| 353 | */ |
| 354 | public static function decodeChunkedData(string $data): string |
| 355 | { |
| 356 | $decoded = ''; |
| 357 | $offset = 0; |
| 358 | $length = strlen($data); |
| 359 | |
| 360 | // Tracks an offset into the original $data instead of reassigning $data = substr($data, ...) |
| 361 | // every iteration - the latter copies the entire remaining tail of the buffer on each chunk, |
| 362 | // making decode time O(N*k) for k chunks instead of linear in the total data length. |
| 363 | while ($offset < $length) { |
| 364 | $lfPos = strpos($data, "\012", $offset); |
| 365 | |
| 366 | if ($lfPos === false) { |
| 367 | $decoded .= substr($data, $offset); |
| 368 | break; |
| 369 | } |
| 370 | |
| 371 | $chunkHex = trim(substr($data, $offset, $lfPos - $offset)); |
| 372 | $scPos = strpos($chunkHex, ';'); |
| 373 | |
| 374 | if ($scPos !== false) { |
| 375 | $chunkHex = substr($chunkHex, 0, $scPos); |
| 376 | } |
| 377 | |
| 378 | if ($chunkHex == '') { |
| 379 | $decoded .= substr($data, $offset, $lfPos - $offset); |
| 380 | $offset = $lfPos + 1; |
| 381 | continue; |
| 382 | } |
| 383 | |
| 384 | $chunkLength = hexdec($chunkHex); |
| 385 | |
| 386 | if ($chunkLength) { |
| 387 | $decoded .= substr($data, $lfPos + 1, $chunkLength); |
| 388 | $offset = $lfPos + 2 + $chunkLength; |
| 389 | } else { |
| 390 | break; |
| 391 | } |
| 392 | } |
| 393 | |
| 394 | return $decoded; |
| 395 | } |
| 396 | |
| 397 | /** |
| 398 | * Determine if the content-type is that of an office document |
| 399 | * |
| 400 | * @param string $contentType |
| 401 | * @return bool |
| 402 | */ |
| 403 | public static function isOfficeDocument(string $contentType): bool |
| 404 | { |
| 405 | $keywords = [ |
| 406 | 'open', 'office', 'ms-', 'word', 'excel', 'powerpoint', 'document', |
| 407 | 'presentation', 'sheet', 'template', 'slideshow', 'addin' |
| 408 | ]; |
| 409 | $contentType = strtolower($contentType); |
| 410 | |
| 411 | foreach ($keywords as $keyword) { |
| 412 | if (str_contains($contentType, $keyword)) { |
| 413 | return true; |
| 414 | } |
| 415 | } |
| 416 | |
| 417 | return false; |
| 418 | } |
| 419 | |
| 420 | } |