regexp: uses unmatched as null (BC break)

dg · Jan 21, 2025 · e2749bf · e2749bf
1 parent 1d845e8
commit e2749bf
Show file tree

Hide file tree

Showing 14 changed files with 55 additions and 56 deletions.
diff --git a/src/Texy/Modules/BlockQuoteModule.php b/src/Texy/Modules/BlockQuoteModule.php
@@ -64,7 +64,7 @@ public function pattern(Texy\BlockParser $parser, array $matches): Texy\HtmlElem
 			}
 			$content .= $mContent . "\n";
 
-			if (!$parser->next("~^>(?: | ([\\ \\t]{1,$spaces} | :) (.*))()$~mA", $matches)) {
+			if (!$parser->next("~^>(?: | ([\\ \\t]{1,$spaces} | :) (.*))$~mA", $matches)) {
 				break;
 			}
 

diff --git a/src/Texy/Modules/FigureModule.php b/src/Texy/Modules/FigureModule.php
@@ -56,7 +56,7 @@ public function __construct(Texy\Texy $texy)
 		   		\ ++ \*\*\* \ ++                  # separator
 		   		(.{0,2000})                       # figure content
 		   		' . Patterns::MODIFIER_H . '?
-	   		()$~mU',
+	   		$~mU',
 			'figure',
 		);
 	}

diff --git a/src/Texy/Modules/HeadingModule.php b/src/Texy/Modules/HeadingModule.php
@@ -79,7 +79,7 @@ public function __construct(Texy\Texy $texy)
 				( \#{2,}+ | ={2,}+ )             # opening characters
 				(.+)                             # heading text
 				' . Texy\Patterns::MODIFIER_H . '?
-			()$~mU',
+			$~mU',
 			'heading/surrounded',
 		);
 	}

diff --git a/src/Texy/Modules/HorizLineModule.php b/src/Texy/Modules/HorizLineModule.php
@@ -36,7 +36,7 @@ public function __construct(Texy\Texy $texy)
 	   			( \*{3,}+ | -{3,}+ )          # three or more * or -
 	   			[\ \t]*                       # optional spaces
 	   			' . Texy\Patterns::MODIFIER . '?
-	   		()$~mU',
+	   		$~mU',
 			'horizline',
 		);
 	}

diff --git a/src/Texy/Modules/HtmlModule.php b/src/Texy/Modules/HtmlModule.php
@@ -320,7 +320,7 @@ private function parseAttributes(string $attrs): array
 						[^'"\s]+               # unquoted value
 					)
 				)?
-				()~is
+				~is
 				X,
 		);
 

diff --git a/src/Texy/Modules/HtmlOutputModule.php b/src/Texy/Modules/HtmlOutputModule.php
@@ -71,7 +71,7 @@ private function postProcess(Texy\Texy $texy, string &$s): void
 			'~
 				( [^<]*+ )
 				< (?: (!--.*--) | (/?) ([a-z][a-z0-9._:-]*) (|[ \n].*) \s* (/?) ) >
-			()~Uis',
+			~Uis',
 			$this->cb(...),
 		);
 

diff --git a/src/Texy/Modules/ImageModule.php b/src/Texy/Modules/ImageModule.php
@@ -63,7 +63,7 @@ public function __construct(Texy\Texy $texy)
 		   		(?:
 		   			:(' . Patterns::LINK_URL . ' | : ) # link or just colon
 		   		)??
-	   		()~U',
+	   		~U',
 			'image',
 		);
 
@@ -88,7 +88,7 @@ private function beforeParse(Texy\Texy $texy, &$text): void
 			   		[\ \t]*
 			   		' . Patterns::MODIFIER . '?
 			   		\s*
-		   		()$~mU',
+		   		$~mU',
 				$this->patternReferenceDef(...),
 			);
 		}
@@ -167,7 +167,7 @@ public function getReference(string $name): ?Image
 	/**
 	 * Parses image's syntax. Input: small.jpg 80x13 || linked.jpg
 	 */
-	public function factoryImage(string $content, string $mod, bool $tryRef = true): Image
+	public function factoryImage(string $content, ?string $mod, bool $tryRef = true): Image
 	{
 		$image = $tryRef ? $this->getReference(trim($content)) : null;
 
@@ -178,7 +178,7 @@ public function factoryImage(string $content, string $mod, bool $tryRef = true):
 
 			// dimensions
 			$matches = null;
-			if ($matches = Texy\Regexp::match($content[0], '~^(.*)\ (\d+|\?)\ *(X|x)\ *(\d+|\?)\ *()$~U')) {
+			if ($matches = Texy\Regexp::match($content[0], '~^(.*)\ (\d+|\?)\ *(X|x)\ *(\d+|\?)\ *$~U')) {
 				$image->URL = trim($matches[1]);
 				$image->asMax = $matches[3] === 'X';
 				$image->width = $matches[2] === '?' ? null : (int) $matches[2];

diff --git a/src/Texy/Modules/LinkModule.php b/src/Texy/Modules/LinkModule.php
@@ -119,7 +119,7 @@ private function beforeParse(Texy\Texy $texy, &$text): void
 			   		( [\ \t] .{1,1000} )?             # optional description
 			   		' . Patterns::MODIFIER . '?
 			   		\s*
-		   		()$~mU',
+		   		$~mU',
 				$this->patternReferenceDef(...),
 			);
 		}
@@ -138,7 +138,7 @@ private function patternReferenceDef(array $matches): string
 		// [4] => .(title)[class]{style}
 
 		$link = new Link($mLink);
-		$link->label = trim($mLabel);
+		$link->label = trim($mLabel ?? '');
 		$link->modifier->setProperties($mMod);
 		$this->checkLink($link);
 		$this->addReference($mRef, $link);
@@ -397,29 +397,29 @@ private function textualUrl(Link $link): string
 				(?P<path> (?: / | ^ ) (?! / ) [^?#]* )?
 				(?: \? (?P<query> [^#]* ) )?
 				(?: \# (?P<fragment> .* ) )?
-				()$
+				$
 			~'))) {
 				return $link->raw;
 			}
 
 			$res = '';
-			if ($parts['scheme'] !== '' && $parts['scheme'] !== 'none') {
+			if ($parts['scheme'] !== null && $parts['scheme'] !== 'none') {
 				$res .= $parts['scheme'] . '://';
 			}
 
-			if ($parts['host'] !== '') {
+			if ($parts['host'] !== null) {
 				$res .= $parts['host'];
 			}
 
-			if ($parts['path'] !== '') {
+			if ($parts['path'] !== null) {
 				$res .= (iconv_strlen($parts['path'], 'UTF-8') > 16 ? ("/\u{2026}" . iconv_substr($parts['path'], -12, 12, 'UTF-8')) : $parts['path']);
 			}
 
-			if ($parts['query'] !== '') {
+			if ($parts['query'] > '') {
 				$res .= iconv_strlen($parts['query'], 'UTF-8') > 4
 					? "?\u{2026}"
 					: ('?' . $parts['query']);
-			} elseif ($parts['fragment'] !== '') {
+			} elseif ($parts['fragment'] > '') {
 				$res .= iconv_strlen($parts['fragment'], 'UTF-8') > 4
 					? "#\u{2026}"
 					: ('#' . $parts['fragment']);

diff --git a/src/Texy/Modules/ListModule.php b/src/Texy/Modules/ListModule.php
@@ -180,7 +180,7 @@ public function patternDefList(BlockParser $parser, array $matches): HtmlElement
 	   		( \S .* )                         # term content
 	   		\: [\ \t]*                        # colon separator
 	   		' . Patterns::MODIFIER_H . '?
-	   	()$~mUA';
+	   	$~mUA';
 
 		while (true) {
 			if ($elItem = $this->patternItem($parser, $bullet, true, 'dd')) {
@@ -225,7 +225,7 @@ private function patternItem(BlockParser $parser, string $bullet, bool $indented
 	   		[ \\t]*
 	   		( \\S .* )?                              # content
 	   		" . Patterns::MODIFIER_H . '?
-	   	()$~mAU';
+	   	$~mAU';
 
 		// first line with bullet
 		$matches = null;
@@ -250,7 +250,7 @@ private function patternItem(BlockParser $parser, string $bullet, bool $indented
 	   		' . Regexp::quote($mIndent) . '
 	   		([\ \t]{1,' . $spaces . '})
 	   		(.*)
-	   	()$~Am', $matches)) {
+	   	$~Am', $matches)) {
 			[, $mBlank, $mSpaces, $mContent] = $matches;
 			// [1] => blank line?
 			// [2] => spaces

diff --git a/src/Texy/Modules/PhraseModule.php b/src/Texy/Modules/PhraseModule.php
@@ -53,7 +53,7 @@ public function __construct(Texy\Texy $texy)
 		// UNIVERSAL
 		$texy->registerLinePattern(
 			array($this, 'patternPhrase'),
-			'~((?>([*+/^_"\~`-])+?))(?!\s)(.*(?!\2).)'.Texy\Patterns::MODIFIER.'?(?<!\s)\1(?!\2)(?::('.Texy\Patterns::LINK_URL.'))??()~Us',
+			'~((?>([*+/^_"\~`-])+?))(?!\s)(.*(?!\2).)'.Texy\Patterns::MODIFIER.'?(?<!\s)\1(?!\2)(?::('.Texy\Patterns::LINK_URL.'))??~Us',
 			'phrase/strong'
 		);
 		*/
@@ -71,7 +71,7 @@ public function __construct(Texy\Texy $texy)
 		   		\*\*\*
 		   		(?! \* )                          # not followed by *
 		   		(?: :(' . Patterns::LINK_URL . ') )??  # optional link
-	   		()~Us',
+	   		~Us',
 			'phrase/strong+em',
 		);
 
@@ -88,7 +88,7 @@ public function __construct(Texy\Texy $texy)
 		   		\*\*
 		   		(?! \* )                          # not followed by *
 		   		(?: :(' . Patterns::LINK_URL . ') )??  # optional link
-	   		()~Us',
+	   		~Us',
 			'phrase/strong',
 		);
 
@@ -105,7 +105,7 @@ public function __construct(Texy\Texy $texy)
 		   		\/\/
 		   		(?! \/ )                          # not followed by /
 		   		(?: :(' . Patterns::LINK_URL . ') )??  # optional link
-	   		()~Us',
+	   		~Us',
 			'phrase/em',
 		);
 
@@ -122,7 +122,7 @@ public function __construct(Texy\Texy $texy)
 		   		\*
 		   		(?! \* )                         # not followed by *
 		   		(?: :(' . Patterns::LINK_URL . ') )??  # optional link
-	   		()~Us',
+	   		~Us',
 			'phrase/em-alt',
 		);
 
@@ -139,7 +139,7 @@ public function __construct(Texy\Texy $texy)
 		   		\*
 		   		(?! [^\s.,;:<>()"?!\'-] )        # must be followed by these chars
 		   		(?: :(' . Patterns::LINK_URL . ') )??  # optional link
-	   		()~Us',
+	   		~Us',
 			'phrase/em-alt2',
 		);
 
@@ -155,7 +155,7 @@ public function __construct(Texy\Texy $texy)
 		   		(?<! [\s+] )                     # not preceded by space or +
 		   		\+\+
 		   		(?! \+ )                         # not followed by +
-	   		()~U',
+	   		~U',
 			'phrase/ins',
 		);
 
@@ -171,7 +171,7 @@ public function __construct(Texy\Texy $texy)
 		   		(?<! [\s<-] )                    # not preceded by space, < or -
 		   		\-\-
 		   		(?! [>-] )                       # not followed by > or -
-	   		()~U',
+	   		~U',
 			'phrase/del',
 		);
 
@@ -187,7 +187,7 @@ public function __construct(Texy\Texy $texy)
 		   		(?<! [\s^] )                     # not preceded by space or ^
 		   		\^\^
 		   		(?! \^ )                         # not followed by ^
-	   		()~U',
+	   		~U',
 			'phrase/sup',
 		);
 
@@ -215,7 +215,7 @@ public function __construct(Texy\Texy $texy)
 		   		(?<! [\s_] )                     # not preceded by space or _
 		   		\_\_
 		   		(?! \_ )                         # not followed by _
-	   		()~U',
+	   		~U',
 			'phrase/sub',
 		);
 
@@ -244,7 +244,7 @@ public function __construct(Texy\Texy $texy)
 		   		\"
 		   		(?! \" )                         # not followed by "
 		   		(?: :(' . Patterns::LINK_URL . ') )??  # optional link
-	   		()~U',
+	   		~U',
 			'phrase/span',
 		);
 
@@ -261,7 +261,7 @@ public function __construct(Texy\Texy $texy)
 		   		\~
 		   		(?! \~ )
 		   		(?: :(' . Patterns::LINK_URL . ') )??  # optional link
-	   		()~U',
+	   		~U',
 			'phrase/span-alt',
 		);
 
@@ -278,7 +278,7 @@ public function __construct(Texy\Texy $texy)
 		   		\<\<
 		   		(?! \< )                         # not followed by
 		   		(?: :(' . Patterns::LINK_URL . ') )??  # optional link
-	   		()~U',
+	   		~U',
 			'phrase/quote',
 		);
 
@@ -297,7 +297,7 @@ public function __construct(Texy\Texy $texy)
 		   		\(\(
 		   		( .+ )                           # explanation
 		   		\)\)
-	   		()~U',
+	   		~U',
 			'phrase/acronym',
 		);
 
@@ -326,7 +326,7 @@ public function __construct(Texy\Texy $texy)
 		   		(?<! [\s\'] )                     # not preceded by space or quote
 		   		\'\'
 		   		(?! \' )                          # not followed by quote
-	   		()~U',
+	   		~U',
 			'phrase/notexy',
 		);
 
@@ -340,7 +340,7 @@ public function __construct(Texy\Texy $texy)
 		   		(?<! \s )                         # not preceded by space
 		   		\`
 		   		(?: : (' . Patterns::LINK_URL . ') )??  # optional link
-	   		()~U',
+	   		~U',
 			'phrase/code',
 		);
 
@@ -352,7 +352,7 @@ public function __construct(Texy\Texy $texy)
 		   		()                                 # empty group
 		   		(?= :\[ )                          # followed by :[
 		   		(?: : (' . Patterns::LINK_URL . ') )  # link
-	   		()~U',
+	   		~U',
 			'phrase/quicklink',
 		);
 
@@ -370,7 +370,7 @@ public function __construct(Texy\Texy $texy)
 		   		(?<! \s )                        # not preceded by space
 		   		\]
 		   		(?! \] )                         # not followed by ]
-	   		()~U',
+	   		~U',
 			'phrase/wikilink',
 		);
 
@@ -388,7 +388,7 @@ public function __construct(Texy\Texy $texy)
 		   		\(
 		   		( (?: [^' . Patterns::MARK . '\r )]++ | [ ] )+ )  # link
 		   		\)
-	   		()~U',
+	   		~U',
 			'phrase/markdown',
 		);
 
@@ -411,11 +411,10 @@ public function __construct(Texy\Texy $texy)
 	 */
 	public function patternPhrase(LineParser $parser, array $matches, string $phrase): Texy\HtmlElement|string|null
 	{
-		[, $mContent, $mMod, $mLink] = $matches;
-		// [1] => **
-		// [2] => ...
-		// [3] => .(title)[class]{style}
-		// [4] => LINK
+		[, $mContent, $mMod, $mLink] = $matches + [3 => null];
+		// [1] => ...
+		// [2] => .(title)[class]{style}
+		// [3] => LINK
 
 		if ($phrase === 'phrase/wikilink') {
 			[$mLink, $mMod] = [$mMod, $mLink];

diff --git a/src/Texy/Modules/ScriptModule.php b/src/Texy/Modules/ScriptModule.php
@@ -37,7 +37,7 @@ public function __construct(Texy\Texy $texy)
 					[}]                     # or single }
 				)+)
 				\}\}
-			()~U',
+			~U',
 			'script',
 		);
 	}
-Original file line number
+Diff line change
@@ Expand Up @@
     			}
     			$content .= $mContent . "\n";
-    			if (!$parser->next("~^>(?: | ([\\ \\t]{1,$spaces} | :) (.*))()$~mA", $matches)) {
+    			if (!$parser->next("~^>(?: | ([\\ \\t]{1,$spaces} | :) (.*))$~mA", $matches)) {
     				break;
     			}
@@ Expand Down @@