Regexp is always used for regular expressions

dg · Jan 21, 2025 · d628696 · d628696
1 parent 2c9fa6a
commit d628696
Show file tree

Hide file tree

Showing 13 changed files with 39 additions and 32 deletions.
diff --git a/src/Bridges/Latte/TexyMacro.php b/src/Bridges/Latte/TexyMacro.php
@@ -10,6 +10,7 @@
 namespace Texy\Bridges\Latte;
 
 use Latte;
+use Texy\Regexp;
 use Texy\Texy;
 
 
@@ -45,7 +46,7 @@ public function texyOpened(Latte\MacroNode $node)
 	public function texyClosed(Latte\MacroNode $node)
 	{
 		$text = $node->content;
-		if (preg_match('#^([\t ]*)\S#m', $text, $m)) { // remove & restore indentation
+		if ($m = Regexp::match($text, '#^([\t ]*)\S#m')) { // remove & restore indentation
 			$text = str_replace(["\r", "\n" . $m[1]], ['', "\n"], $text);
 			$this->texy->htmlOutputModule->baseIndent = strlen($m[1]);
 		}

diff --git a/src/Texy/Helpers.php b/src/Texy/Helpers.php
@@ -114,7 +114,7 @@ public static function webalize(string $s, string $charlist = ''): string
 	{
 		$s = self::toAscii($s);
 		$s = strtolower($s);
-		$s = Regexp::replace($s, '#[^a-z0-9' . preg_quote($charlist, '#') . ']+#', '-');
+		$s = Regexp::replace($s, '#[^a-z0-9' . Regexp::quote($charlist) . ']+#', '-');
 		$s = trim($s, '-');
 		return $s;
 	}
@@ -149,7 +149,7 @@ public static function outdent(string $s, bool $firstLine = false): string
 	public static function isRelative(string $URL): bool
 	{
 		// check for scheme, or absolute path, or absolute URL
-		return !preg_match('#[a-z][a-z0-9+.-]{0,20}:|[\#/?]#Ai', $URL);
+		return !Regexp::match($URL, '#[a-z][a-z0-9+.-]{0,20}:|[\#/?]#Ai');
 	}
 
 

diff --git a/src/Texy/Modifier.php b/src/Texy/Modifier.php
@@ -62,7 +62,7 @@ public function setProperties(?string $s): void
 			$ch = $s[$p];
 
 			if ($ch === '(') { // title
-				preg_match('#(?:\\\\\)|[^)\n])++\)#', $s, $m, 0, $p);
+				$m = Regexp::match($s, '#(?:\\\\\)|[^)\n])++\)#', offset: $p);
 				$this->title = Helpers::unescapeHtml(str_replace('\)', ')', trim(substr($m[0], 1, -1))));
 				$p += strlen($m[0]);
 

diff --git a/src/Texy/Modules/BlockModule.php b/src/Texy/Modules/BlockModule.php
@@ -74,7 +74,7 @@ public function pattern(Texy\BlockParser $parser, array $matches): HtmlElement|s
 		// [4] => ... content
 
 		$mod = new Texy\Modifier($mMod);
-		$parts = preg_split('#\s+#u', $mParam, 2);
+		$parts = Texy\Regexp::split($mParam, '#\s+#u', limit: 2);
 		$blocktype = empty($parts[0]) ? 'block/default' : 'block/' . $parts[0];
 		$param = empty($parts[1]) ? null : $parts[1];
 

diff --git a/src/Texy/Modules/EmoticonModule.php b/src/Texy/Modules/EmoticonModule.php
@@ -60,7 +60,7 @@ private function beforeParse(): void
 
 		$pattern = [];
 		foreach ($this->icons as $key => $foo) {
-			$pattern[] = preg_quote($key, '#') . '+'; // last char can be repeated
+			$pattern[] = Texy\Regexp::quote($key) . '+'; // last char can be repeated
 		}
 
 		$this->texy->registerLinePattern(

diff --git a/src/Texy/Modules/HtmlModule.php b/src/Texy/Modules/HtmlModule.php
@@ -12,6 +12,7 @@
 use Texy;
 use Texy\HtmlElement;
 use Texy\Patterns;
+use Texy\Regexp;
 
 
 /**
@@ -163,7 +164,7 @@ private function solveComment(Texy\HandlerInvocation $invocation, string $conten
 		}
 
 		// sanitize comment
-		$content = Texy\Regexp::replace($content, '#-{2,}#', ' - ');
+		$content = Regexp::replace($content, '#-{2,}#', ' - ');
 		$content = trim($content, '-');
 
 		return $this->texy->protect('<!--' . $content . '-->', Texy\Texy::CONTENT_MARKUP);
@@ -272,7 +273,7 @@ private function validateAttrs(HtmlElement $el, Texy\Texy $texy): bool
 
 				$texy->summary['links'][] = $el->attrs['href'];
 			}
-		} elseif (preg_match('#^h[1-6]#i', $name)) {
+		} elseif (Regexp::match($name, '#^h[1-6]#i')) {
 			$texy->headingModule->TOC[] = [
 				'el' => $el,
 				'level' => (int) substr($name, 1),
@@ -286,12 +287,10 @@ private function validateAttrs(HtmlElement $el, Texy\Texy $texy): bool
 
 	private function parseAttributes(string $attrs): array
 	{
-		$matches = $res = [];
-		preg_match_all(
-			'#([a-z0-9\_:-]+)\s*(?:=\s*(\'[^\']*\'|"[^"]*"|[^\'"\s]+))?()#isu',
+		$res = [];
+		$matches = Regexp::matchAll(
 			$attrs,
-			$matches,
-			PREG_SET_ORDER,
+			'#([a-z0-9\_:-]+)\s*(?:=\s*(\'[^\']*\'|"[^"]*"|[^\'"\s]+))?()#isu',
 		);
 
 		foreach ($matches as $m) {

diff --git a/src/Texy/Modules/LinkModule.php b/src/Texy/Modules/LinkModule.php
@@ -14,6 +14,7 @@
 use Texy\LineParser;
 use Texy\Link;
 use Texy\Patterns;
+use Texy\Regexp;
 
 
 /**
@@ -334,7 +335,7 @@ private function checkLink(Link $link): void
 			// special supported case
 			$link->URL = 'http://' . $link->URL;
 
-		} elseif (preg_match('#' . self::$EMAIL . '$#Au', $link->URL)) {
+		} elseif (Regexp::match($link->URL, '#' . self::$EMAIL . '$#Au')) {
 			// email
 			$link->URL = 'mailto:' . $link->URL;
 
@@ -352,17 +353,17 @@ private function checkLink(Link $link): void
 	 */
 	private function textualUrl(Link $link): string
 	{
-		if ($this->texy->obfuscateEmail && preg_match('#^' . self::$EMAIL . '$#u', $link->raw)) { // email
+		if ($this->texy->obfuscateEmail && Regexp::match($link->raw, '#^' . self::$EMAIL . '$#u')) { // email
 			return str_replace('@', '&#64;<!-- -->', $link->raw);
 		}
 
-		if ($this->shorten && preg_match('#^(https?://|ftp://|www\.|/)#i', $link->raw)) {
+		if ($this->shorten && Regexp::match($link->raw, '#^(https?://|ftp://|www\.|/)#i')) {
 			$raw = strncasecmp($link->raw, 'www.', 4) === 0
 				? 'none://' . $link->raw
 				: $link->raw;
 
 			// parse_url() in PHP damages UTF-8 - use regular expression
-			if (!preg_match('~^(?:(?P<scheme>[a-z]+):)?(?://(?P<host>[^/?#]+))?(?P<path>(?:/|^)(?!/)[^?#]*)?(?:\?(?P<query>[^#]*))?(?:#(?P<fragment>.*))?()$~', $raw, $parts)) {
+			if (!($parts = Regexp::match($raw, '~^(?:(?P<scheme>[a-z]+):)?(?://(?P<host>[^/?#]+))?(?P<path>(?:/|^)(?!/)[^?#]*)?(?:\?(?P<query>[^#]*))?(?:#(?P<fragment>.*))?()$~'))) {
 				return $link->raw;
 			}
 

diff --git a/src/Texy/Modules/ListModule.php b/src/Texy/Modules/ListModule.php
@@ -14,6 +14,7 @@
 use Texy\HtmlElement;
 use Texy\Modifier;
 use Texy\Patterns;
+use Texy\Regexp;
 
 
 /**
@@ -91,7 +92,7 @@ public function patternList(BlockParser $parser, array $matches): ?HtmlElement
 
 		$bullet = $min = null;
 		foreach ($this->bullets as $type => $desc) {
-			if (preg_match('#' . $desc[0] . '#Au', $mBullet)) {
+			if (Regexp::match($mBullet, '#' . $desc[0] . '#Au')) {
 				$bullet = $desc[3] ?? $desc[0];
 				$min = isset($desc[3]) ? 2 : 1;
 				$el->setName($desc[1] ? 'ol' : 'ul');
@@ -151,7 +152,7 @@ public function patternDefList(BlockParser $parser, array $matches): HtmlElement
 
 		$bullet = null;
 		foreach ($this->bullets as $desc) {
-			if (preg_match('#' . $desc[0] . '#Au', $mBullet)) {
+			if (Regexp::match($mBullet, '#' . $desc[0] . '#Au')) {
 				$bullet = $desc[3] ?? $desc[0];
 				break;
 			}

diff --git a/src/Texy/Modules/LongWordsModule.php b/src/Texy/Modules/LongWordsModule.php
@@ -92,14 +92,12 @@ private function pattern(array $matches): string
 			return $mWord;
 		}
 
-		$chars = [];
-		preg_match_all(
-			'#[' . Texy\Patterns::MARK . ']+|.#u',
+		$chars = Texy\Regexp::matchAll(
 			$mWord,
-			$chars,
+			'#[' . Texy\Patterns::MARK . ']+|.#u',
 		);
 
-		$chars = $chars[0];
+		$chars = array_column($chars, 0);
 		if (count($chars) < $this->wordLimit) {
 			return $mWord;
 		}

diff --git a/src/Texy/Modules/ParagraphModule.php b/src/Texy/Modules/ParagraphModule.php
@@ -28,8 +28,8 @@ public function __construct(Texy\Texy $texy)
 	public function process(Texy\BlockParser $parser, string $content, Texy\HtmlElement $el): void
 	{
 		$parts = $parser->isIndented()
-			? preg_split('#(\n(?!\ )|\n{2,})#', $content, -1, PREG_SPLIT_NO_EMPTY)
-			: preg_split('#(\n{2,})#', $content, -1, PREG_SPLIT_NO_EMPTY);
+			? Regexp::split($content, '#(\n(?!\ )|\n{2,})#', skipEmpty: true)
+			: Regexp::split($content, '#(\n{2,})#', skipEmpty: true);
 
 		foreach ($parts as $s) {
 			$s = trim($s);
@@ -90,7 +90,7 @@ private function solve(
 			// leave element p
 
 		// block contains text
-		} elseif (preg_match('#[^\s' . Texy\Patterns::MARK . ']#u', $content)) {
+		} elseif (Regexp::match($content, '#[^\s' . Texy\Patterns::MARK . ']#u')) {
 			// leave element p
 
 		// block contains only replaced element

diff --git a/src/Texy/Modules/ScriptModule.php b/src/Texy/Modules/ScriptModule.php
@@ -10,6 +10,7 @@
 namespace Texy\Modules;
 
 use Texy;
+use Texy\Regexp;
 
 
 /**
@@ -51,11 +52,11 @@ public function pattern(Texy\LineParser $parser, array $matches): Texy\HtmlEleme
 		$raw = null;
 		$args = [];
 		// function (arg, arg, ...) or function: arg, arg
-		if ($matches = Texy\Regexp::match($cmd, '#^([a-z_][a-z0-9_-]*)\s*(?:\(([^()]*)\)|:(.*))$#iu')) {
+		if ($matches = Regexp::match($cmd, '#^([a-z_][a-z0-9_-]*)\s*(?:\(([^()]*)\)|:(.*))$#iu')) {
 			$cmd = $matches[1];
 			$raw = trim($matches[3] ?? $matches[2]);
 			if ($raw !== '') {
-				$args = preg_split('#\s*' . preg_quote($this->separator, '#') . '\s*#u', $raw);
+				$args = Regexp::split($raw, '#\s*' . Regexp::quote($this->separator) . '\s*#u');
 			}
 		}
 

diff --git a/src/Texy/Regexp.php b/src/Texy/Regexp.php
@@ -105,6 +105,12 @@ public static function replace(
 	}
 
 
+	public static function quote(string $s): string
+	{
+		return preg_quote($s, '#');
+	}
+
+
 	/** @internal */
 	public static function pcre(string $func, array $args)
 	{

diff --git a/src/Texy/Texy.php b/src/Texy/Texy.php
@@ -246,7 +246,7 @@ final public function registerLinePattern(
 
 	final public function registerBlockPattern(callable $handler, string $pattern, string $name): void
 	{
-		// if (!preg_match('#(.)\^.*\$\1[a-z]*#is', $pattern)) die("Texy: Not a block pattern $name");
+		// if (!Regexp::match($pattern, '#(.)\^.*\$\1[a-z]*#is')) die("Texy: Not a block pattern $name");
 		if (!isset($this->allowed[$name])) {
 			$this->allowed[$name] = true;
 		}
@@ -519,8 +519,8 @@ final public function checkURL(string $URL, string $type): bool
 	{
 		// absolute URL with scheme? check scheme!
 		return empty($this->urlSchemeFilters[$type])
-			|| !preg_match('#\s*[a-z][a-z0-9+.-]{0,20}:#Ai', $URL) // http: | mailto:
-			|| preg_match($this->urlSchemeFilters[$type], $URL);
+			|| !Regexp::match($URL, '#\s*[a-z][a-z0-9+.-]{0,20}:#Ai') // http: | mailto:
+			|| Regexp::match($URL, $this->urlSchemeFilters[$type]);
 	}