IDNAEncoder.php 11.1 KB
Newer Older
imac's avatar
imac committed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388
<?php

/**
 * IDNA URL encoder
 *
 * Note: Not fully compliant, as nameprep does nothing yet.
 *
 * @package Requests
 * @subpackage Utilities
 * @see https://tools.ietf.org/html/rfc3490 IDNA specification
 * @see https://tools.ietf.org/html/rfc3492 Punycode/Bootstrap specification
 */
class Requests_IDNAEncoder {
	/**
	 * ACE prefix used for IDNA
	 *
	 * @see https://tools.ietf.org/html/rfc3490#section-5
	 * @var string
	 */
	const ACE_PREFIX = 'xn--';

	/**#@+
	 * Bootstrap constant for Punycode
	 *
	 * @see https://tools.ietf.org/html/rfc3492#section-5
	 * @var int
	 */
	const BOOTSTRAP_BASE         = 36;
	const BOOTSTRAP_TMIN         = 1;
	const BOOTSTRAP_TMAX         = 26;
	const BOOTSTRAP_SKEW         = 38;
	const BOOTSTRAP_DAMP         = 700;
	const BOOTSTRAP_INITIAL_BIAS = 72;
	const BOOTSTRAP_INITIAL_N    = 128;
	/**#@-*/

	/**
	 * Encode a hostname using Punycode
	 *
	 * @param string $string Hostname
	 * @return string Punycode-encoded hostname
	 */
	public static function encode($string) {
		$parts = explode('.', $string);
		foreach ($parts as &$part) {
			$part = self::to_ascii($part);
		}
		return implode('.', $parts);
	}

	/**
	 * Convert a UTF-8 string to an ASCII string using Punycode
	 *
	 * @throws Requests_Exception Provided string longer than 64 ASCII characters (`idna.provided_too_long`)
	 * @throws Requests_Exception Prepared string longer than 64 ASCII characters (`idna.prepared_too_long`)
	 * @throws Requests_Exception Provided string already begins with xn-- (`idna.provided_is_prefixed`)
	 * @throws Requests_Exception Encoded string longer than 64 ASCII characters (`idna.encoded_too_long`)
	 *
	 * @param string $string ASCII or UTF-8 string (max length 64 characters)
	 * @return string ASCII string
	 */
	public static function to_ascii($string) {
		// Step 1: Check if the string is already ASCII
		if (self::is_ascii($string)) {
			// Skip to step 7
			if (strlen($string) < 64) {
				return $string;
			}

			throw new Requests_Exception('Provided string is too long', 'idna.provided_too_long', $string);
		}

		// Step 2: nameprep
		$string = self::nameprep($string);

		// Step 3: UseSTD3ASCIIRules is false, continue
		// Step 4: Check if it's ASCII now
		if (self::is_ascii($string)) {
			// Skip to step 7
			if (strlen($string) < 64) {
				return $string;
			}

			throw new Requests_Exception('Prepared string is too long', 'idna.prepared_too_long', $string);
		}

		// Step 5: Check ACE prefix
		if (strpos($string, self::ACE_PREFIX) === 0) {
			throw new Requests_Exception('Provided string begins with ACE prefix', 'idna.provided_is_prefixed', $string);
		}

		// Step 6: Encode with Punycode
		$string = self::punycode_encode($string);

		// Step 7: Prepend ACE prefix
		$string = self::ACE_PREFIX . $string;

		// Step 8: Check size
		if (strlen($string) < 64) {
			return $string;
		}

		throw new Requests_Exception('Encoded string is too long', 'idna.encoded_too_long', $string);
	}

	/**
	 * Check whether a given string contains only ASCII characters
	 *
	 * @internal (Testing found regex was the fastest implementation)
	 *
	 * @param string $string
	 * @return bool Is the string ASCII-only?
	 */
	protected static function is_ascii($string) {
		return (preg_match('/(?:[^\x00-\x7F])/', $string) !== 1);
	}

	/**
	 * Prepare a string for use as an IDNA name
	 *
	 * @todo Implement this based on RFC 3491 and the newer 5891
	 * @param string $string
	 * @return string Prepared string
	 */
	protected static function nameprep($string) {
		return $string;
	}

	/**
	 * Convert a UTF-8 string to a UCS-4 codepoint array
	 *
	 * Based on Requests_IRI::replace_invalid_with_pct_encoding()
	 *
	 * @throws Requests_Exception Invalid UTF-8 codepoint (`idna.invalidcodepoint`)
	 * @param string $input
	 * @return array Unicode code points
	 */
	protected static function utf8_to_codepoints($input) {
		$codepoints = array();

		// Get number of bytes
		$strlen = strlen($input);

		for ($position = 0; $position < $strlen; $position++) {
			$value = ord($input[$position]);

			// One byte sequence:
			if ((~$value & 0x80) === 0x80) {
				$character = $value;
				$length = 1;
				$remaining = 0;
			}
			// Two byte sequence:
			elseif (($value & 0xE0) === 0xC0) {
				$character = ($value & 0x1F) << 6;
				$length = 2;
				$remaining = 1;
			}
			// Three byte sequence:
			elseif (($value & 0xF0) === 0xE0) {
				$character = ($value & 0x0F) << 12;
				$length = 3;
				$remaining = 2;
			}
			// Four byte sequence:
			elseif (($value & 0xF8) === 0xF0) {
				$character = ($value & 0x07) << 18;
				$length = 4;
				$remaining = 3;
			}
			// Invalid byte:
			else {
				throw new Requests_Exception('Invalid Unicode codepoint', 'idna.invalidcodepoint', $value);
			}

			if ($remaining > 0) {
				if ($position + $length > $strlen) {
					throw new Requests_Exception('Invalid Unicode codepoint', 'idna.invalidcodepoint', $character);
				}
				for ($position++; $remaining > 0; $position++) {
					$value = ord($input[$position]);

					// If it is invalid, count the sequence as invalid and reprocess the current byte:
					if (($value & 0xC0) !== 0x80) {
						throw new Requests_Exception('Invalid Unicode codepoint', 'idna.invalidcodepoint', $character);
					}

					$character |= ($value & 0x3F) << (--$remaining * 6);
				}
				$position--;
			}

			if (
				// Non-shortest form sequences are invalid
				   $length > 1 && $character <= 0x7F
				|| $length > 2 && $character <= 0x7FF
				|| $length > 3 && $character <= 0xFFFF
				// Outside of range of ucschar codepoints
				// Noncharacters
				|| ($character & 0xFFFE) === 0xFFFE
				|| $character >= 0xFDD0 && $character <= 0xFDEF
				|| (
					// Everything else not in ucschar
					   $character > 0xD7FF && $character < 0xF900
					|| $character < 0x20
					|| $character > 0x7E && $character < 0xA0
					|| $character > 0xEFFFD
				)
			) {
				throw new Requests_Exception('Invalid Unicode codepoint', 'idna.invalidcodepoint', $character);
			}

			$codepoints[] = $character;
		}

		return $codepoints;
	}

	/**
	 * RFC3492-compliant encoder
	 *
	 * @internal Pseudo-code from Section 6.3 is commented with "#" next to relevant code
	 * @throws Requests_Exception On character outside of the domain (never happens with Punycode) (`idna.character_outside_domain`)
	 *
	 * @param string $input UTF-8 encoded string to encode
	 * @return string Punycode-encoded string
	 */
	public static function punycode_encode($input) {
		$output = '';
#		let n = initial_n
		$n = self::BOOTSTRAP_INITIAL_N;
#		let delta = 0
		$delta = 0;
#		let bias = initial_bias
		$bias = self::BOOTSTRAP_INITIAL_BIAS;
#		let h = b = the number of basic code points in the input
		$h = $b = 0; // see loop
#		copy them to the output in order
		$codepoints = self::utf8_to_codepoints($input);
		$extended = array();

		foreach ($codepoints as $char) {
			if ($char < 128) {
				// Character is valid ASCII
				// TODO: this should also check if it's valid for a URL
				$output .= chr($char);
				$h++;
			}
			// Check if the character is non-ASCII, but below initial n
			// This never occurs for Punycode, so ignore in coverage
			// @codeCoverageIgnoreStart
			elseif ($char < $n) {
				throw new Requests_Exception('Invalid character', 'idna.character_outside_domain', $char);
			}
			// @codeCoverageIgnoreEnd
			else {
				$extended[$char] = true;
			}
		}
		$extended = array_keys($extended);
		sort($extended);
		$b = $h;
#		[copy them] followed by a delimiter if b > 0
		if (strlen($output) > 0) {
			$output .= '-';
		}
#		{if the input contains a non-basic code point < n then fail}
#		while h < length(input) do begin
		while ($h < count($codepoints)) {
#			let m = the minimum code point >= n in the input
			$m = array_shift($extended);
			//printf('next code point to insert is %s' . PHP_EOL, dechex($m));
#			let delta = delta + (m - n) * (h + 1), fail on overflow
			$delta += ($m - $n) * ($h + 1);
#			let n = m
			$n = $m;
#			for each code point c in the input (in order) do begin
			for ($num = 0; $num < count($codepoints); $num++) {
				$c = $codepoints[$num];
#				if c < n then increment delta, fail on overflow
				if ($c < $n) {
					$delta++;
				}
#				if c == n then begin
				elseif ($c === $n) {
#					let q = delta
					$q = $delta;
#					for k = base to infinity in steps of base do begin
					for ($k = self::BOOTSTRAP_BASE; ; $k += self::BOOTSTRAP_BASE) {
#						let t = tmin if k <= bias {+ tmin}, or
#								tmax if k >= bias + tmax, or k - bias otherwise
						if ($k <= ($bias + self::BOOTSTRAP_TMIN)) {
							$t = self::BOOTSTRAP_TMIN;
						}
						elseif ($k >= ($bias + self::BOOTSTRAP_TMAX)) {
							$t = self::BOOTSTRAP_TMAX;
						}
						else {
							$t = $k - $bias;
						}
#						if q < t then break
						if ($q < $t) {
							break;
						}
#						output the code point for digit t + ((q - t) mod (base - t))
						$digit = $t + (($q - $t) % (self::BOOTSTRAP_BASE - $t));
						$output .= self::digit_to_char($digit);
#						let q = (q - t) div (base - t)
						$q = floor(($q - $t) / (self::BOOTSTRAP_BASE - $t));
#					end
					}
#					output the code point for digit q
					$output .= self::digit_to_char($q);
#					let bias = adapt(delta, h + 1, test h equals b?)
					$bias = self::adapt($delta, $h + 1, $h === $b);
#					let delta = 0
					$delta = 0;
#					increment h
					$h++;
#				end
				}
#			end
			}
#			increment delta and n
			$delta++;
			$n++;
#		end
		}

		return $output;
	}

	/**
	 * Convert a digit to its respective character
	 *
	 * @see https://tools.ietf.org/html/rfc3492#section-5
	 * @throws Requests_Exception On invalid digit (`idna.invalid_digit`)
	 *
	 * @param int $digit Digit in the range 0-35
	 * @return string Single character corresponding to digit
	 */
	protected static function digit_to_char($digit) {
		// @codeCoverageIgnoreStart
		// As far as I know, this never happens, but still good to be sure.
		if ($digit < 0 || $digit > 35) {
			throw new Requests_Exception(sprintf('Invalid digit %d', $digit), 'idna.invalid_digit', $digit);
		}
		// @codeCoverageIgnoreEnd
		$digits = 'abcdefghijklmnopqrstuvwxyz0123456789';
		return substr($digits, $digit, 1);
	}

	/**
	 * Adapt the bias
	 *
	 * @see https://tools.ietf.org/html/rfc3492#section-6.1
	 * @param int $delta
	 * @param int $numpoints
	 * @param bool $firsttime
	 * @return int New bias
	 */
	protected static function adapt($delta, $numpoints, $firsttime) {
#	function adapt(delta,numpoints,firsttime):
#		if firsttime then let delta = delta div damp
		if ($firsttime) {
			$delta = floor($delta / self::BOOTSTRAP_DAMP);
		}
#		else let delta = delta div 2
		else {
			$delta = floor($delta / 2);
		}
#		let delta = delta + (delta div numpoints)
		$delta += floor($delta / $numpoints);
#		let k = 0
		$k = 0;
#		while delta > ((base - tmin) * tmax) div 2 do begin
		$max = floor(((self::BOOTSTRAP_BASE - self::BOOTSTRAP_TMIN) * self::BOOTSTRAP_TMAX) / 2);
		while ($delta > $max) {
#			let delta = delta div (base - tmin)
			$delta = floor($delta / (self::BOOTSTRAP_BASE - self::BOOTSTRAP_TMIN));
#			let k = k + base
			$k += self::BOOTSTRAP_BASE;
#		end
		}
#		return k + (((base - tmin + 1) * delta) div (delta + skew))
		return $k + floor(((self::BOOTSTRAP_BASE - self::BOOTSTRAP_TMIN + 1) * $delta) / ($delta + self::BOOTSTRAP_SKEW));
	}
}