← All changes | vendor_prefixed/twig/symfony/polyfill-mbstring/Mbstring.php +39 -89 4.3.0-beta3 → 4.2.1 View file →

		@@ -75,9 +75,8 @@
75	75	private const SIMPLE_CASE_FOLD = [['µ', 'ſ', "ͅ", 'ς', "ϐ", "ϑ", "ϕ", "ϖ", "ϰ", "ϱ", "ϵ", "ẛ", "ι"], ['μ', 's', 'ι', 'σ', 'β', 'θ', 'φ', 'π', 'κ', 'ρ', 'ε', "ṡ", 'ι']];
76	76	private static $encodingList = ['ASCII', 'UTF-8'];
77	77	private static $language = 'neutral';
78	78	private static $internalEncoding = 'UTF-8';
79		- private static $iconvSupportsIgnore;
80	79	public static function mb_convert_encoding($s, $toEncoding, $fromEncoding = null)
81	80	{
82	81	if (\is_array($s)) {
83	82	$r = [];
		@@ -103,43 +102,17 @@
103	102	if ('HTML-ENTITIES' === $fromEncoding || 'HTML' === $fromEncoding) {
104	103	$fromEncoding = 'Windows-1252';
105	104	}
106	105	if ('UTF-8' !== $fromEncoding) {
107		- $s = self::iconv($fromEncoding, 'UTF-8', $s);
	106	+ $s = \iconv($fromEncoding, 'UTF-8//IGNORE', $s);
108	107	}
109	108	return \preg_replace_callback('/[\\x80-\\xFF]+/', [__CLASS__, 'html_encoding_callback'], $s);
110	109	}
111	110	if ('HTML-ENTITIES' === $fromEncoding) {
112		- $decodeControlChars = static function ($m) {
113		- $code = '' !== ($m[2] ?? '') ? \hexdec($m[2]) : (int) $m[1];
114		- if ($code < 32 || 127 === $code) {
115		- return \chr($code);
116		- }
117		- if (128 <= $code && $code <= 159) {
118		- return "\xc2" . \chr(0x80 | $code & 0x3f);
119		- }
120		- return $m[0];
121		- };
122		- if (\PHP_VERSION_ID >= 70400) {
123		- $s = \html_entity_decode($s, \ENT_QUOTES, 'UTF-8');
124		- // html_entity_decode() leaves numeric entities for C0/C1 control
125		- // characters as-is (HTML spec), but mb_convert_encoding() decodes
126		- // them. Catch what html_entity_decode() missed.
127		- if (\false !== \strpos($s, '&#')) {
128		- $s = \preg_replace_callback('/&#(?:0*([0-9]++)|[xX]0*([0-9a-fA-F]++));/', $decodeControlChars, $s);
129		- }
130		- } else {
131		- // PHP < 7.4: html_entity_decode() truncates strings at NUL bytes,
132		- // so decode the control character entities first then call
133		- // html_entity_decode() on each NUL-delimited chunk independently.
134		- $s = \preg_replace_callback('/&#(?:0*([0-9]++)|[xX]0*([0-9a-fA-F]++));/', $decodeControlChars, $s);
135		- $s = \implode("\x00", \array_map(static function ($chunk) {
136		- return \html_entity_decode($chunk, \ENT_QUOTES, 'UTF-8');
137		- }, \explode("\x00", $s)));
138		- }
	111	+ $s = \html_entity_decode($s, \ENT_COMPAT, 'UTF-8');
139	112	$fromEncoding = 'UTF-8';
140	113	}
141		- return self::iconv($fromEncoding, $toEncoding, $s);
	114	+ return \iconv($fromEncoding, $toEncoding . '//IGNORE', $s);
142	115	}
143	116	public static function mb_convert_variables($toEncoding, $fromEncoding, &...$vars)
144	117	{
145	118	$ok = \true;
		@@ -179,12 +152,12 @@
179	152	$encoding = self::getEncoding($encoding);
180	153	if ('UTF-8' === $encoding) {
181	154	$encoding = null;
182	155	if (!\preg_match('//u', $s)) {
183		- $s = @self::iconv('UTF-8', 'UTF-8', $s);
	156	+ $s = @\iconv('UTF-8', 'UTF-8//IGNORE', $s);
184	157	}
185	158	} else {
186		- $s = self::iconv($encoding, 'UTF-8', $s);
	159	+ $s = \iconv($encoding, 'UTF-8//IGNORE', $s);
187	160	}
188	161	$cnt = \floor(\count($convmap) / 4) * 4;
189	162	for ($i = 0; $i < $cnt; $i += 4) {
190	163	// collector_decode_htmlnumericentity ignores $convmap[$i + 3]
		@@ -202,9 +175,9 @@
202	175	}, $s);
203	176	if (null === $encoding) {
204	177	return $s;
205	178	}
206		- return self::iconv('UTF-8', $encoding, $s);
	179	+ return \iconv('UTF-8', $encoding . '//IGNORE', $s);
207	180	}
208	181	public static function mb_encode_numericentity($s, $convmap, $encoding = null, $is_hex = \false)
209	182	{
210	183	if (null !== $s && !\is_scalar($s) && !(\is_object($s) && \method_exists($s, '__toString'))) {
		@@ -230,12 +203,12 @@
230	203	$encoding = self::getEncoding($encoding);
231	204	if ('UTF-8' === $encoding) {
232	205	$encoding = null;
233	206	if (!\preg_match('//u', $s)) {
234		- $s = @self::iconv('UTF-8', 'UTF-8', $s);
	207	+ $s = @\iconv('UTF-8', 'UTF-8//IGNORE', $s);
235	208	}
236	209	} else {
237		- $s = self::iconv($encoding, 'UTF-8', $s);
	210	+ $s = \iconv($encoding, 'UTF-8//IGNORE', $s);
238	211	}
239	212	static $ulenMask = ["\xc0" => 2, "\xd0" => 2, "\xe0" => 3, "\xf0" => 4];
240	213	$cnt = \floor(\count($convmap) / 4) * 4;
241	214	$i = 0;
		@@ -257,9 +230,9 @@
257	230	}
258	231	if (null === $encoding) {
259	232	return $result;
260	233	}
261		- return self::iconv('UTF-8', $encoding, $result);
	234	+ return \iconv('UTF-8', $encoding . '//IGNORE', $result);
262	235	}
263	236	public static function mb_convert_case($s, $mode, $encoding = null)
264	237	{
265	238	$s = (string) $s;
		@@ -269,12 +242,12 @@
269	242	$encoding = self::getEncoding($encoding);
270	243	if ('UTF-8' === $encoding) {
271	244	$encoding = null;
272	245	if (!\preg_match('//u', $s)) {
273		- $s = @self::iconv('UTF-8', 'UTF-8', $s);
	246	+ $s = @\iconv('UTF-8', 'UTF-8//IGNORE', $s);
274	247	}
275	248	} else {
276		- $s = self::iconv($encoding, 'UTF-8', $s);
	249	+ $s = \iconv($encoding, 'UTF-8//IGNORE', $s);
277	250	}
278	251	if (\MB_CASE_TITLE == $mode) {
279	252	static $titleRegexp = null;
280	253	if (null === $titleRegexp) {
		@@ -327,9 +300,9 @@
327	300	}
328	301	if (null === $encoding) {
329	302	return $s;
330	303	}
331		- return self::iconv('UTF-8', $encoding, $s);
	304	+ return \iconv('UTF-8', $encoding . '//IGNORE', $s);
332	305	}
333	306	public static function mb_internal_encoding($encoding = null)
334	307	{
335	308	if (null === $encoding) {
		@@ -455,15 +428,9 @@
455	428	$encoding = self::getEncoding($encoding);
456	429	if ('CP850' === $encoding || 'ASCII' === $encoding) {
457	430	return \strlen($s);
458	431	}
459		- if (\false !== ($len = @\iconv_strlen($s, $encoding))) {
460		- return $len;
461		- }
462		- if ('UTF-8' !== $encoding) {
463		- return $len;
464		- }
465		- return \preg_match_all('/[\\x00-\\x7F]|[\\xC0-\\xDF][\\x80-\\xBF]?|[\\xE0-\\xEF][\\x80-\\xBF]{0,2}|[\\xF0-\\xF7][\\x80-\\xBF]{0,3}|[\\xF8-\\xFB][\\x80-\\xBF]{0,4}|[\\xFC-\\xFD][\\x80-\\xBF]{0,5}|[\\x80-\\xBF\\xFE\\xFF]/s', $s);
	432	+ return @\iconv_strlen($s, $encoding);
466	433	}
467	434	public static function mb_strpos($haystack, $needle, $offset = 0, $encoding = null)
468	435	{
469	436	$encoding = self::getEncoding($encoding);
		@@ -647,9 +614,9 @@
647	614	public static function mb_strwidth($s, $encoding = null)
648	615	{
649	616	$encoding = self::getEncoding($encoding);
650	617	if ('UTF-8' !== $encoding) {
651		- $s = self::iconv($encoding, 'UTF-8', $s);
	618	+ $s = \iconv($encoding, 'UTF-8//IGNORE', $s);
652	619	}
653	620	$s = \preg_replace('/[\\x{1100}-\\x{115F}\\x{2329}\\x{232A}\\x{2E80}-\\x{303E}\\x{3040}-\\x{A4CF}\\x{AC00}-\\x{D7A3}\\x{F900}-\\x{FAFF}\\x{FE10}-\\x{FE19}\\x{FE30}-\\x{FE6F}\\x{FF00}-\\x{FF60}\\x{FFE0}-\\x{FFE6}\\x{20000}-\\x{2FFFD}\\x{30000}-\\x{3FFFD}]/u', '', $s, -1, $wide);
654	621	return ($wide << 1) + \iconv_strlen($s, 'UTF-8');
655	622	}
		@@ -697,20 +664,17 @@
697	664	}
698	665	return $code;
699	666	}
700	667	/** @return string|false */
701		- public static function mb_scrub(?string $string, ?string $encoding = null) : string
	668	+ public static function mb_str_pad(string $string, int $length, string $pad_string = ' ', int $pad_type = \STR_PAD_RIGHT, ?string $encoding = null)
702	669	{
703		- if (null === $encoding) {
704		- $encoding = self::mb_internal_encoding();
705		- } elseif (!self::assertEncoding($encoding, 'mb_scrub(): Argument #2 ($encoding) must be a valid encoding, "%s" given')) {
706		- return \false;
	670	+ if (!\in_array($pad_type, [\STR_PAD_RIGHT, \STR_PAD_LEFT, \STR_PAD_BOTH], \true)) {
	671	+ if (\PHP_VERSION_ID < 80000) {
	672	+ \trigger_error('mb_str_pad(): Argument #4 ($pad_type) must be STR_PAD_LEFT, STR_PAD_RIGHT, or STR_PAD_BOTH', \E_USER_WARNING);
	673	+ return \false;
	674	+ }
	675	+ throw new \ValueError('mb_str_pad(): Argument #4 ($pad_type) must be STR_PAD_LEFT, STR_PAD_RIGHT, or STR_PAD_BOTH');
707	676	}
708		- return self::mb_convert_encoding((string) $string, $encoding, $encoding);
709		- }
710		- /** @return string|false */
711		- public static function mb_str_pad(string $string, int $length, string $pad_string = ' ', int $pad_type = \STR_PAD_RIGHT, ?string $encoding = null)
712		- {
713	677	if (null === $encoding) {
714	678	$encoding = self::mb_internal_encoding();
715	679	} elseif (!self::assertEncoding($encoding, 'mb_str_pad(): Argument #5 ($encoding) must be a valid encoding, "%s" given')) {
716	680	return \false;
		@@ -721,15 +685,8 @@
721	685	return \false;
722	686	}
723	687	throw new \ValueError('mb_str_pad(): Argument #3 ($pad_string) must be a non-empty string');
724	688	}
725		- if (!\in_array($pad_type, [\STR_PAD_RIGHT, \STR_PAD_LEFT, \STR_PAD_BOTH], \true)) {
726		- if (\PHP_VERSION_ID < 80000) {
727		- \trigger_error('mb_str_pad(): Argument #4 ($pad_type) must be STR_PAD_LEFT, STR_PAD_RIGHT, or STR_PAD_BOTH', \E_USER_WARNING);
728		- return \false;
729		- }
730		- throw new \ValueError('mb_str_pad(): Argument #4 ($pad_type) must be STR_PAD_LEFT, STR_PAD_RIGHT, or STR_PAD_BOTH');
731		- }
732	689	$paddingRequired = $length - self::mb_strlen($string, $encoding);
733	690	if ($paddingRequired < 1) {
734	691	return $string;
735	692	}
		@@ -767,23 +724,8 @@
767	724	$firstChar = \mb_substr($string, 0, 1, $encoding);
768	725	$firstChar = \mb_convert_case($firstChar, \MB_CASE_LOWER, $encoding);
769	726	return $firstChar . \mb_substr($string, 1, null, $encoding);
770	727	}
771		- /** @return string|false */
772		- public static function mb_trim(string $string, ?string $characters = null, ?string $encoding = null)
773		- {
774		- return self::mb_internal_trim('{^[%s]+|[%1$s]+$}Du', $string, $characters, $encoding, __FUNCTION__);
775		- }
776		- /** @return string|false */
777		- public static function mb_ltrim(string $string, ?string $characters = null, ?string $encoding = null)
778		- {
779		- return self::mb_internal_trim('{^[%s]+}Du', $string, $characters, $encoding, __FUNCTION__);
780		- }
781		- /** @return string|false */
782		- public static function mb_rtrim(string $string, ?string $characters = null, ?string $encoding = null)
783		- {
784		- return self::mb_internal_trim('{[%s]+$}Du', $string, $characters, $encoding, __FUNCTION__);
785		- }
786	728	private static function getSubpart($pos, $part, $haystack, $encoding)
787	729	{
788	730	if (\false === $pos) {
789	731	return \false;
		@@ -847,16 +789,24 @@
847	789	return 'UTF-16BE';
848	790	}
849	791	return $encoding;
850	792	}
851		- private static function iconv($fromEncoding, $toEncoding, $s)
	793	+ /** @return string|false */
	794	+ public static function mb_trim(string $string, ?string $characters = null, ?string $encoding = null)
852	795	{
853		- if (null === self::$iconvSupportsIgnore) {
854		- self::$iconvSupportsIgnore = \false !== @\iconv('UTF-8', 'UTF-8//IGNORE', '');
855		- }
856		- return self::$iconvSupportsIgnore ? \iconv($fromEncoding, $toEncoding . '//IGNORE', $s) : \iconv($fromEncoding, $toEncoding, $s);
	796	+ return self::mb_internal_trim('{^[%s]+|[%1$s]+$}Du', $string, $characters, $encoding, __FUNCTION__);
857	797	}
858	798	/** @return string|false */
	799	+ public static function mb_ltrim(string $string, ?string $characters = null, ?string $encoding = null)
	800	+ {
	801	+ return self::mb_internal_trim('{^[%s]+}Du', $string, $characters, $encoding, __FUNCTION__);
	802	+ }
	803	+ /** @return string|false */
	804	+ public static function mb_rtrim(string $string, ?string $characters = null, ?string $encoding = null)
	805	+ {
	806	+ return self::mb_internal_trim('{[%s]+$}Du', $string, $characters, $encoding, __FUNCTION__);
	807	+ }
	808	+ /** @return string|false */
859	809	private static function mb_internal_trim(string $regex, string $string, ?string $characters, ?string $encoding, string $function)
860	810	{
861	811	if (null === $encoding) {
862	812	$encoding = self::mb_internal_encoding();
		@@ -868,17 +818,17 @@
868	818	}
869	819	if ('UTF-8' === $encoding) {
870	820	$encoding = null;
871	821	if (!\preg_match('//u', $string)) {
872		- $string = @self::iconv('UTF-8', 'UTF-8', $string);
	822	+ $string = @\iconv('UTF-8', 'UTF-8//IGNORE', $string);
873	823	}
874	824	if (null !== $characters && !\preg_match('//u', $characters)) {
875		- $characters = @self::iconv('UTF-8', 'UTF-8', $characters);
	825	+ $characters = @\iconv('UTF-8', 'UTF-8//IGNORE', $characters);
876	826	}
877	827	} else {
878		- $string = self::iconv($encoding, 'UTF-8', $string);
	828	+ $string = \iconv($encoding, 'UTF-8//IGNORE', $string);
879	829	if (null !== $characters) {
880		- $characters = self::iconv($encoding, 'UTF-8', $characters);
	830	+ $characters = \iconv($encoding, 'UTF-8//IGNORE', $characters);
881	831	}
882	832	}
883	833	if (null === $characters) {
884	834	$characters = "\\0 \f\n\r\t\v                　 ᠎";
		@@ -888,9 +838,9 @@
888	838	$string = \preg_replace(\sprintf($regex, $characters), '', $string);
889	839	if (null === $encoding) {
890	840	return $string;
891	841	}
892		- return self::iconv('UTF-8', $encoding, $string);
	842	+ return \iconv('UTF-8', $encoding . '//IGNORE', $string);
893	843	}
894	844	private static function assertEncoding(string $encoding, string $errorFormat) : bool
895	845	{
896	846	try {