lochmueller / sourceopt

Classes/Manipulation/ManipulationInterface.php

Indentation +7 added lines, -7 removed lines

@@ -13,11 +13,11 @@
 interface ManipulationInterface
 {
 
-    /**
-     * @param string $html The original HTML
-     * @param array $configuration Configuration
-     *
-     * @return string the manipulated HTML
-     */
-    public function manipulate($html, array $configuration = []);
+	/**
+	 * @param string $html The original HTML
+	 * @param array $configuration Configuration
+	 *
+	 * @return string the manipulated HTML
+	 */
+	public function manipulate($html, array $configuration = []);
 }

Classes/Manipulation/RemoveComments.php

Indentation +57 added lines, -57 removed lines

@@ -13,65 +13,65 @@
 class RemoveComments implements ManipulationInterface
 {
 
-    /**
-     * Patterns for white-listing comments inside content
-     *
-     * @var array
-     */
-    protected $whiteListCommentsPatterns = [];
+	/**
+	 * Patterns for white-listing comments inside content
+	 *
+	 * @var array
+	 */
+	protected $whiteListCommentsPatterns = [];
 
-    /**
-     * @param string $html          The original HTML
-     * @param array  $configuration Configuration
-     *
-     * @return string the manipulated HTML
-     */
-    public function manipulate($html, array $configuration = [])
-    {
-        if (isset($configuration['keep.'])) {
-            $this->whiteListCommentsPatterns = $configuration['keep.'];
-        }
+	/**
+	 * @param string $html          The original HTML
+	 * @param array  $configuration Configuration
+	 *
+	 * @return string the manipulated HTML
+	 */
+	public function manipulate($html, array $configuration = [])
+	{
+		if (isset($configuration['keep.'])) {
+			$this->whiteListCommentsPatterns = $configuration['keep.'];
+		}
 
-        // match all styles, scripts and comments
-        $matches = [];
-        preg_match_all(
-            '/(?s)((<!--.*?-->)|(<[ \n\r]*style[^>]*>.*?<[ \n\r]*\/style[^>]*>)|(<[ \n\r]*script[^>]*>.*?<[ \n\r]*\/script[^>]*>))/im',
-            $html,
-            $matches
-        );
-        foreach ($matches[0] as $tag) {
-            if ($this->keepComment($tag) === false) {
-                $html = str_replace($tag, '', $html);
-            }
-        }
-        return $html;
-    }
+		// match all styles, scripts and comments
+		$matches = [];
+		preg_match_all(
+			'/(?s)((<!--.*?-->)|(<[ \n\r]*style[^>]*>.*?<[ \n\r]*\/style[^>]*>)|(<[ \n\r]*script[^>]*>.*?<[ \n\r]*\/script[^>]*>))/im',
+			$html,
+			$matches
+		);
+		foreach ($matches[0] as $tag) {
+			if ($this->keepComment($tag) === false) {
+				$html = str_replace($tag, '', $html);
+			}
+		}
+		return $html;
+	}
 
-    /**
-     * Check if a comment is defined to be kept in a pattern whiteListOfComments
-     *
-     * @param string $commentHtml
-     *
-     * @return boolean
-     */
-    protected function keepComment($commentHtml)
-    {
-        // if not even a comment, skip this
-        if (!preg_match('/^\<\!\-\-(.*?)\-\-\>$/usi', $commentHtml)) {
-            return true;
-        }
+	/**
+	 * Check if a comment is defined to be kept in a pattern whiteListOfComments
+	 *
+	 * @param string $commentHtml
+	 *
+	 * @return boolean
+	 */
+	protected function keepComment($commentHtml)
+	{
+		// if not even a comment, skip this
+		if (!preg_match('/^\<\!\-\-(.*?)\-\-\>$/usi', $commentHtml)) {
+			return true;
+		}
 
-        // if not defined in white list
-        if (!empty($this->whiteListCommentsPatterns)) {
-            $commentHtml = str_replace("<!--", "", $commentHtml);
-            $commentHtml = str_replace("-->", "", $commentHtml);
-            $commentHtml = trim($commentHtml);
-            foreach ($this->whiteListCommentsPatterns as $pattern) {
-                if (!empty($pattern) && preg_match($pattern, $commentHtml)) {
-                    return true;
-                }
-            }
-        }
-        return false;
-    }
+		// if not defined in white list
+		if (!empty($this->whiteListCommentsPatterns)) {
+			$commentHtml = str_replace("<!--", "", $commentHtml);
+			$commentHtml = str_replace("-->", "", $commentHtml);
+			$commentHtml = trim($commentHtml);
+			foreach ($this->whiteListCommentsPatterns as $pattern) {
+				if (!empty($pattern) && preg_match($pattern, $commentHtml)) {
+					return true;
+				}
+			}
+		}
+		return false;
+	}
 }

Classes/Manipulation/RemoveGenerator.php

Indentation +11 added lines, -11 removed lines

@@ -13,15 +13,15 @@
 class RemoveGenerator implements ManipulationInterface
 {
 
-    /**
-     * @param string $html The original HTML
-     * @param array $configuration Configuration
-     *
-     * @return string the manipulated HTML
-     */
-    public function manipulate($html, array $configuration = [])
-    {
-        $regex = '<meta name=["\']?generator["\']? [^>]+>';
-        return preg_replace('/' . $regex . '/is', '', $html);
-    }
+	/**
+	 * @param string $html The original HTML
+	 * @param array $configuration Configuration
+	 *
+	 * @return string the manipulated HTML
+	 */
+	public function manipulate($html, array $configuration = [])
+	{
+		$regex = '<meta name=["\']?generator["\']? [^>]+>';
+		return preg_replace('/' . $regex . '/is', '', $html);
+	}
 }

Spacing +1 added lines, -1 removed lines

@@ -22,6 +22,6 @@
     public function manipulate($html, array $configuration = [])
     {
         $regex = '<meta name=["\']?generator["\']? [^>]+>';
-        return preg_replace('/' . $regex . '/is', '', $html);
+        return preg_replace('/'.$regex.'/is', '', $html);
     }
 }

Classes/Middleware/CleanHtmlMiddleware.php

Indentation +39 added lines, -39 removed lines

@@ -17,43 +17,43 @@
  */
 class CleanHtmlMiddleware implements MiddlewareInterface
 {
-    /**
-     * @var CleanHtmlService
-     */
-    protected $cleanHtmlService = null;
-
-    public function __construct()
-    {
-        $this->cleanHtmlService = GeneralUtility::makeInstance(CleanHtmlService::class);
-    }
-
-    /**
-     * Clean the HTML output
-     *
-     * @param ServerRequestInterface $request
-     * @param RequestHandlerInterface $handler
-     * @return ResponseInterface
-     */
-    public function process(ServerRequestInterface $request, RequestHandlerInterface $handler): ResponseInterface
-    {
-        $response = $handler->handle($request);
-
-        if (
-            !($response instanceof NullResponse)
-            && $GLOBALS['TSFE'] instanceof TypoScriptFrontendController
-            && $GLOBALS['TSFE']->isOutputting()) {
-
-            $processedHtml = $this->cleanHtmlService->clean(
-                $response->getBody()->__toString(),
-                $GLOBALS['TSFE']->config['config']['sourceopt.']
-            );
-
-            // Replace old body with $processedHtml
-            $responseBody = new Stream('php://temp', 'rw');
-            $responseBody->write($processedHtml);
-            $response = $response->withBody($responseBody);
-        }
-
-        return $response;
-    }
+	/**
+	 * @var CleanHtmlService
+	 */
+	protected $cleanHtmlService = null;
+
+	public function __construct()
+	{
+		$this->cleanHtmlService = GeneralUtility::makeInstance(CleanHtmlService::class);
+	}
+
+	/**
+	 * Clean the HTML output
+	 *
+	 * @param ServerRequestInterface $request
+	 * @param RequestHandlerInterface $handler
+	 * @return ResponseInterface
+	 */
+	public function process(ServerRequestInterface $request, RequestHandlerInterface $handler): ResponseInterface
+	{
+		$response = $handler->handle($request);
+
+		if (
+			!($response instanceof NullResponse)
+			&& $GLOBALS['TSFE'] instanceof TypoScriptFrontendController
+			&& $GLOBALS['TSFE']->isOutputting()) {
+
+			$processedHtml = $this->cleanHtmlService->clean(
+				$response->getBody()->__toString(),
+				$GLOBALS['TSFE']->config['config']['sourceopt.']
+			);
+
+			// Replace old body with $processedHtml
+			$responseBody = new Stream('php://temp', 'rw');
+			$responseBody->write($processedHtml);
+			$response = $response->withBody($responseBody);
+		}
+
+		return $response;
+	}
 }

Classes/Service/CleanHtmlService.php

Indentation +416 added lines, -416 removed lines

@@ -16,420 +16,420 @@
 class CleanHtmlService implements SingletonInterface
 {
 
-    /**
-     * Enable Debug comment in footer
-     *
-     * @var boolean
-     */
-    protected $debugComment = false;
-
-    /**
-     * Format Type
-     *
-     * @var integer
-     */
-    protected $formatType = 0;
-
-    /**
-     * Tab character
-     *
-     * @var string
-     */
-    protected $tab = "\t";
-
-    /**
-     * Newline character
-     *
-     * @var string
-     */
-    protected $newline = "\n";
-
-    /**
-     * Configured extra header comment
-     *
-     * @var string
-     */
-    protected $headerComment = '';
-
-    /**
-     * Empty space char
-     * @var string
-     */
-    protected $emptySpaceChar = ' ';
-
-    /**
-     * Set variables based on given config
-     *
-     * @param array $config
-     *
-     * @return void
-     */
-    public function setVariables(array $config)
-    {
-        if (!empty($config)) {
-            if ($config['formatHtml'] && is_numeric($config['formatHtml'])) {
-                $this->formatType = (int)$config['formatHtml'];
-            }
-
-            if ($config['formatHtml.']['tabSize'] && is_numeric($config['formatHtml.']['tabSize'])) {
-                $this->tab = str_pad('', $config['formatHtml.']['tabSize'], ' ');
-            }
-
-            if (isset($config['formatHtml.']['debugComment'])) {
-                $this->debugComment = (bool)$config['formatHtml.']['debugComment'];
-            }
-
-            if (isset($config['headerComment'])) {
-                $this->headerComment = $config['headerComment'];
-            }
-
-            if (isset($config['dropEmptySpaceChar']) && (bool)$config['dropEmptySpaceChar']) {
-                $this->emptySpaceChar = '';
-            }
-        }
-    }
-
-    /**
-     * Clean given HTML with formatter
-     *
-     * @param string $html
-     * @param array $config
-     *
-     * @return string
-     */
-    public function clean($html, $config = [])
-    {
-        if (!empty($config)) {
-            if ((bool)$config['enabled'] === false) {
-                return $html;
-            }
-
-            $this->setVariables($config);
-        }
-        // convert line-breaks to UNIX
-        $this->convNlOs($html);
-
-        $manipulations = [];
-
-        if (isset($config['removeGenerator']) && (bool)$config['removeGenerator']) {
-            $manipulations['removeGenerator'] = GeneralUtility::makeInstance(RemoveGenerator::class);
-        }
-
-        if (isset($config['removeComments']) && (bool)$config['removeComments']) {
-            $manipulations['removeComments'] = GeneralUtility::makeInstance(RemoveComments::class);
-        }
-
-        if (!empty($this->headerComment)) {
-            $this->includeHeaderComment($html);
-        }
-
-        foreach ($manipulations as $key => $manipulation) {
-            /** @var ManipulationInterface $manipulation */
-            $configuration = isset($config[$key . '.']) && is_array($config[$key . '.']) ? $config[$key . '.'] : [];
-            $html = $manipulation->manipulate($html, $configuration);
-        }
-
-        // cleanup HTML5 self-closing elements
-        if(!isset($GLOBALS['TSFE']->config['config']['doctype']) || 'x' !== substr($GLOBALS['TSFE']->config['config']['doctype'],0,1)) {
-            $html = preg_replace('/<((?:area|base|br|col|embed|hr|img|input|link|meta|param|source|track|wbr)\s[^>]+?)\s?\/>/', '<$1>', $html);
-        }
-
-        if ($this->formatType > 0) {
-            $html = $this->formatHtml($html);
-        }
-        // remove white space after line ending
-        $this->rTrimLines($html);
-
-        // recover line-breaks
-        if (Environment::isWindows()) {
-          $html = str_replace($this->newline, "\r\n", $html);
-        }
-
-        return $html;
-    }
-
-    /**
-     * Formats the (X)HTML code:
-     *  - taps according to the hirarchy of the tags
-     *  - removes empty spaces between tags
-     *  - removes linebreaks within tags (spares where necessary: pre, textarea, comments, ..)
-     *  choose from five options:
-     *    0 => off
-     *    1 => no line break at all  (code in one line)
-     *    2 => minimalistic line breaks (structure defining box-elements)
-     *    3 => aesthetic line breaks (important box-elements)
-     *    4 => logic line breaks (all box-elements)
-     *    5 => max line breaks (all elements)
-     *
-     * @param string $html
-     *
-     * @return string
-     */
-    protected function formatHtml($html)
-    {
-        // Save original formated comments, pre, textarea, styles and java-scripts & replace them with markers
-        preg_match_all(
-            '/(?s)((<!--.*?-->)|(<[ \n\r]*pre[^>]*>.*?<[ \n\r]*\/pre[^>]*>)|(<[ \n\r]*textarea[^>]*>.*?<[ \n\r]*\/textarea[^>]*>)|(<[ \n\r]*style[^>]*>.*?<[ \n\r]*\/style[^>]*>)|(<[ \n\r]*script[^>]*>.*?<[ \n\r]*\/script[^>]*>))/im',
-            $html,
-            $matches
-        );
-        $noFormat = $matches[0]; // do not format these block elements
-        for ($i = 0; $i < count($noFormat); $i++) {
-            $html = str_replace($noFormat[$i], "\n<!-- ELEMENT $i -->", $html);
-        }
-
-        // define box elements for formatting
-        $trueBoxElements = 'address|blockquote|center|dir|div|dl|fieldset|form|h1|h2|h3|h4|h5|h6|hr|isindex|menu|noframes|noscript|ol|p|pre|table|ul|article|aside|details|figcaption|figure|footer|header|hgroup|menu|nav|section';
-        $functionalBoxElements = 'dd|dt|frameset|li|tbody|td|tfoot|th|thead|tr|colgroup';
-        $usableBoxElements = 'applet|button|del|iframe|ins|map|object|script';
-        $imagineBoxElements = 'html|body|head|meta|title|link|script|base|!--';
-        $allBoxLikeElements = '(?>' . $trueBoxElements . '|' . $functionalBoxElements . '|' . $usableBoxElements . '|' . $imagineBoxElements . ')';
-        $esteticBoxLikeElements = '(?>html|head|body|meta name|title|div|table|h1|h2|h3|h4|h5|h6|p|form|pre|center|!--)';
-        $structureBoxLikeElements = '(?>html|head|body|div|!--)';
-
-        // split html into it's elements
-        $htmlArrayTemp = preg_split(
-            '/(<(?:[^<>]+(?:"[^"]*"|\'[^\']*\')?)+>)/',
-            $html,
-            -1,
-            PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY
-        );
-
-        if ($htmlArrayTemp === false) {
-            // Restore saved comments, styles and java-scripts
-            for ($i = 0; $i < count($noFormat); $i++) {
-                $html = str_replace("<!-- ELEMENT $i -->", $noFormat[$i], $html);
-            }
-            return $html;
-        }
-        // remove empty lines
-        $htmlArray = [''];
-        $z = 1;
-        for ($x = 0; $x < count($htmlArrayTemp); $x++) {
-            $t = trim($htmlArrayTemp[$x]);
-            if ($t !== '') {
-                $htmlArray[$z] = $htmlArrayTemp[$x];
-                $z++;
-            } else {
-                $htmlArray[$z] = $this->emptySpaceChar;
-                $z++;
-            }
-        }
-
-        // rebuild html
-        $html = '';
-        $tabs = 0;
-        for ($x = 0; $x < count($htmlArray); $x++) {
-            // check if the element should stand in a new line
-            $newline = false;
-            if (substr($htmlArray[$x - 1], 0, 5) == '<?xml') {
-                $newline = true;
-            } elseif ($this->formatType == 2 && ( // minimalistic line break
-                    # this element has a line break before itself
-                    preg_match(
-                        '/<' . $structureBoxLikeElements . '(.*)>/Usi',
-                        $htmlArray[$x]
-                    ) || preg_match(
-                        '/<' . $structureBoxLikeElements . '(.*) \/>/Usi',
-                        $htmlArray[$x]
-                    ) || # one element before is a element that has a line break after
-                    preg_match(
-                        '/<\/' . $structureBoxLikeElements . '(.*)>/Usi',
-                        $htmlArray[$x - 1]
-                    ) || substr(
-                        $htmlArray[$x - 1],
-                        0,
-                        4
-                    ) == '<!--' || preg_match('/<' . $structureBoxLikeElements . '(.*) \/>/Usi', $htmlArray[$x - 1]))
-            ) {
-                $newline = true;
-            } elseif ($this->formatType == 3 && ( // aestetic line break
-                    # this element has a line break before itself
-                    preg_match(
-                        '/<' . $esteticBoxLikeElements . '(.*)>/Usi',
-                        $htmlArray[$x]
-                    ) || preg_match(
-                        '/<' . $esteticBoxLikeElements . '(.*) \/>/Usi',
-                        $htmlArray[$x]
-                    ) || # one element before is a element that has a line break after
-                    preg_match('/<\/' . $esteticBoxLikeElements . '(.*)>/Usi', $htmlArray[$x - 1]) || substr(
-                        $htmlArray[$x - 1],
-                        0,
-                        4
-                    ) == '<!--' || preg_match('/<' . $esteticBoxLikeElements . '(.*) \/>/Usi', $htmlArray[$x - 1]))
-            ) {
-                $newline = true;
-            } elseif ($this->formatType >= 4 && ( // logical line break
-                    # this element has a line break before itself
-                    preg_match(
-                        '/<' . $allBoxLikeElements . '(.*)>/Usi',
-                        $htmlArray[$x]
-                    ) || preg_match(
-                        '/<' . $allBoxLikeElements . '(.*) \/>/Usi',
-                        $htmlArray[$x]
-                    ) || # one element before is a element that has a line break after
-                    preg_match('/<\/' . $allBoxLikeElements . '(.*)>/Usi', $htmlArray[$x - 1]) || substr(
-                        $htmlArray[$x - 1],
-                        0,
-                        4
-                    ) == '<!--' || preg_match('/<' . $allBoxLikeElements . '(.*) \/>/Usi', $htmlArray[$x - 1]))
-            ) {
-                $newline = true;
-            }
-
-            // count down a tab
-            if (substr($htmlArray[$x], 0, 2) == '</') {
-                $tabs--;
-            }
-
-            // add tabs and line breaks in front of the current tag
-            if ($newline) {
-                $html .= $this->newline;
-                for ($y = 0; $y < $tabs; $y++) {
-                    $html .= $this->tab;
-                }
-            }
-
-            // remove white spaces and line breaks and add current tag to the html-string
-            if (substr($htmlArray[$x], 0, 9) == '<![CDATA[' // remove multiple white space in CDATA / XML
-                || substr($htmlArray[$x], 0, 5) == '<?xml'
-            ) {
-                $html .= $this->killWhiteSpace($htmlArray[$x]);
-            } else { // remove all line breaks
-                $html .= $this->killLineBreaks($htmlArray[$x]);
-            }
-
-            // count up a tab
-            if (substr($htmlArray[$x], 0, 1) == '<' && substr($htmlArray[$x], 1, 1) != '/') {
-                if (
-                    substr($htmlArray[$x], 1, 1) !== ' '
-                    && substr($htmlArray[$x], 1, 3) !== 'img'
-                    && substr($htmlArray[$x], 1, 6) !== 'source'
-                    && substr($htmlArray[$x], 1, 2) !== 'br'
-                    && substr($htmlArray[$x], 1, 2) !== 'hr'
-                    && substr($htmlArray[$x], 1, 5) !== 'input'
-                    && substr($htmlArray[$x], 1, 4) !== 'link'
-                    && substr($htmlArray[$x], 1, 4) !== 'meta'
-                    && substr($htmlArray[$x], 1, 4) !== 'col '
-                    && substr($htmlArray[$x], 1, 5) !== 'frame'
-                    && substr($htmlArray[$x], 1, 7) !== 'isindex'
-                    && substr($htmlArray[$x], 1, 5) !== 'param'
-                    && substr($htmlArray[$x], 1, 4) !== 'area'
-                    && substr($htmlArray[$x], 1, 4) !== 'base'
-                    && substr($htmlArray[$x], 0, 2) !== '<!'
-                    && substr($htmlArray[$x], 0, 5) !== '<?xml'
-                ) {
-                    $tabs++;
-                }
-            }
-        }
-
-        // Remove empty lines
-        if ($this->formatType > 1) {
-            $this->removeEmptyLines($html);
-        }
-
-        // Restore saved comments, styles and java-scripts
-        for ($i = 0; $i < count($noFormat); $i++) {
-            $html = str_replace("<!-- ELEMENT $i -->", $noFormat[$i], $html);
-        }
-
-        // include debug comment at the end
-        if ($tabs != 0 && $this->debugComment === true) {
-            $html .= "<!-- $tabs open elements found -->";
-        }
-
-        return $html;
-    }
-
-    /**
-     * Remove ALL line breaks and multiple white space
-     *
-     * @param string $html
-     *
-     * @return string
-     */
-    protected function killLineBreaks($html)
-    {
-        $html = str_replace($this->newline, '', $html);
-        $html = preg_replace('/\s\s+/u', ' ', $html);
-        return $html;
-        #? return preg_replace('/\n|\s+(\s)/u', '$1', $html);
-    }
-
-    /**
-     * Remove multiple white space, keeps line breaks
-     *
-     * @param string $html
-     *
-     * @return string
-     */
-    protected function killWhiteSpace($html)
-    {
-        $temp = explode($this->newline, $html);
-        for ($i = 0; $i < count($temp); $i++) {
-            if (!trim($temp[$i])) {
-                unset($temp[$i]);
-            } else {
-                $temp[$i] = trim($temp[$i]);
-                $temp[$i] = preg_replace('/\s\s+/', ' ', $temp[$i]);
-            }
-        }
-        $html = implode($this->newline, $temp);
-        return $html;
-    }
-
-    /**
-     * Remove white space at the end of lines, keeps other white space and line breaks
-     *
-     * @param string $html
-     *
-     * @return string
-     */
-    protected function rTrimLines(&$html)
-    {
-        $html = preg_replace('/\s+$/m', '', $html);
-    }
-
-    /**
-     * Convert newlines according to the current OS
-     *
-     * @param string $html
-     *
-     * @return string
-     */
-    protected function convNlOs(&$html)
-    {
-        $html = preg_replace("(\r\n|\r)", $this->newline, $html);
-    }
-
-    /**
-     * Remove empty lines
-     *
-     * @param string $html
-     *
-     * @return void
-     */
-    protected function removeEmptyLines(&$html)
-    {
-        $temp = explode($this->newline, $html);
-        $result = [];
-        for ($i = 0; $i < count($temp); ++$i) {
-            if ('' == trim($temp[$i])) {
-                continue;
-            }
-            $result[] = $temp[$i];
-        }
-        $html = implode($this->newline, $result);
-    }
-
-    /**
-     * Include configured header comment in HTML content block
-     *
-     * @param $html
-     */
-    public function includeHeaderComment(&$html)
-    {
-        $html = preg_replace('/^(-->)$/m', "\n\t" . $this->headerComment . "\n$1", $html);
-    }
+	/**
+	 * Enable Debug comment in footer
+	 *
+	 * @var boolean
+	 */
+	protected $debugComment = false;
+
+	/**
+	 * Format Type
+	 *
+	 * @var integer
+	 */
+	protected $formatType = 0;
+
+	/**
+	 * Tab character
+	 *
+	 * @var string
+	 */
+	protected $tab = "\t";
+
+	/**
+	 * Newline character
+	 *
+	 * @var string
+	 */
+	protected $newline = "\n";
+
+	/**
+	 * Configured extra header comment
+	 *
+	 * @var string
+	 */
+	protected $headerComment = '';
+
+	/**
+	 * Empty space char
+	 * @var string
+	 */
+	protected $emptySpaceChar = ' ';
+
+	/**
+	 * Set variables based on given config
+	 *
+	 * @param array $config
+	 *
+	 * @return void
+	 */
+	public function setVariables(array $config)
+	{
+		if (!empty($config)) {
+			if ($config['formatHtml'] && is_numeric($config['formatHtml'])) {
+				$this->formatType = (int)$config['formatHtml'];
+			}
+
+			if ($config['formatHtml.']['tabSize'] && is_numeric($config['formatHtml.']['tabSize'])) {
+				$this->tab = str_pad('', $config['formatHtml.']['tabSize'], ' ');
+			}
+
+			if (isset($config['formatHtml.']['debugComment'])) {
+				$this->debugComment = (bool)$config['formatHtml.']['debugComment'];
+			}
+
+			if (isset($config['headerComment'])) {
+				$this->headerComment = $config['headerComment'];
+			}
+
+			if (isset($config['dropEmptySpaceChar']) && (bool)$config['dropEmptySpaceChar']) {
+				$this->emptySpaceChar = '';
+			}
+		}
+	}
+
+	/**
+	 * Clean given HTML with formatter
+	 *
+	 * @param string $html
+	 * @param array $config
+	 *
+	 * @return string
+	 */
+	public function clean($html, $config = [])
+	{
+		if (!empty($config)) {
+			if ((bool)$config['enabled'] === false) {
+				return $html;
+			}
+
+			$this->setVariables($config);
+		}
+		// convert line-breaks to UNIX
+		$this->convNlOs($html);
+
+		$manipulations = [];
+
+		if (isset($config['removeGenerator']) && (bool)$config['removeGenerator']) {
+			$manipulations['removeGenerator'] = GeneralUtility::makeInstance(RemoveGenerator::class);
+		}
+
+		if (isset($config['removeComments']) && (bool)$config['removeComments']) {
+			$manipulations['removeComments'] = GeneralUtility::makeInstance(RemoveComments::class);
+		}
+
+		if (!empty($this->headerComment)) {
+			$this->includeHeaderComment($html);
+		}
+
+		foreach ($manipulations as $key => $manipulation) {
+			/** @var ManipulationInterface $manipulation */
+			$configuration = isset($config[$key . '.']) && is_array($config[$key . '.']) ? $config[$key . '.'] : [];
+			$html = $manipulation->manipulate($html, $configuration);
+		}
+
+		// cleanup HTML5 self-closing elements
+		if(!isset($GLOBALS['TSFE']->config['config']['doctype']) || 'x' !== substr($GLOBALS['TSFE']->config['config']['doctype'],0,1)) {
+			$html = preg_replace('/<((?:area|base|br|col|embed|hr|img|input|link|meta|param|source|track|wbr)\s[^>]+?)\s?\/>/', '<$1>', $html);
+		}
+
+		if ($this->formatType > 0) {
+			$html = $this->formatHtml($html);
+		}
+		// remove white space after line ending
+		$this->rTrimLines($html);
+
+		// recover line-breaks
+		if (Environment::isWindows()) {
+		  $html = str_replace($this->newline, "\r\n", $html);
+		}
+
+		return $html;
+	}
+
+	/**
+	 * Formats the (X)HTML code:
+	 *  - taps according to the hirarchy of the tags
+	 *  - removes empty spaces between tags
+	 *  - removes linebreaks within tags (spares where necessary: pre, textarea, comments, ..)
+	 *  choose from five options:
+	 *    0 => off
+	 *    1 => no line break at all  (code in one line)
+	 *    2 => minimalistic line breaks (structure defining box-elements)
+	 *    3 => aesthetic line breaks (important box-elements)
+	 *    4 => logic line breaks (all box-elements)
+	 *    5 => max line breaks (all elements)
+	 *
+	 * @param string $html
+	 *
+	 * @return string
+	 */
+	protected function formatHtml($html)
+	{
+		// Save original formated comments, pre, textarea, styles and java-scripts & replace them with markers
+		preg_match_all(
+			'/(?s)((<!--.*?-->)|(<[ \n\r]*pre[^>]*>.*?<[ \n\r]*\/pre[^>]*>)|(<[ \n\r]*textarea[^>]*>.*?<[ \n\r]*\/textarea[^>]*>)|(<[ \n\r]*style[^>]*>.*?<[ \n\r]*\/style[^>]*>)|(<[ \n\r]*script[^>]*>.*?<[ \n\r]*\/script[^>]*>))/im',
+			$html,
+			$matches
+		);
+		$noFormat = $matches[0]; // do not format these block elements
+		for ($i = 0; $i < count($noFormat); $i++) {
+			$html = str_replace($noFormat[$i], "\n<!-- ELEMENT $i -->", $html);
+		}
+
+		// define box elements for formatting
+		$trueBoxElements = 'address|blockquote|center|dir|div|dl|fieldset|form|h1|h2|h3|h4|h5|h6|hr|isindex|menu|noframes|noscript|ol|p|pre|table|ul|article|aside|details|figcaption|figure|footer|header|hgroup|menu|nav|section';
+		$functionalBoxElements = 'dd|dt|frameset|li|tbody|td|tfoot|th|thead|tr|colgroup';
+		$usableBoxElements = 'applet|button|del|iframe|ins|map|object|script';
+		$imagineBoxElements = 'html|body|head|meta|title|link|script|base|!--';
+		$allBoxLikeElements = '(?>' . $trueBoxElements . '|' . $functionalBoxElements . '|' . $usableBoxElements . '|' . $imagineBoxElements . ')';
+		$esteticBoxLikeElements = '(?>html|head|body|meta name|title|div|table|h1|h2|h3|h4|h5|h6|p|form|pre|center|!--)';
+		$structureBoxLikeElements = '(?>html|head|body|div|!--)';
+
+		// split html into it's elements
+		$htmlArrayTemp = preg_split(
+			'/(<(?:[^<>]+(?:"[^"]*"|\'[^\']*\')?)+>)/',
+			$html,
+			-1,
+			PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY
+		);
+
+		if ($htmlArrayTemp === false) {
+			// Restore saved comments, styles and java-scripts
+			for ($i = 0; $i < count($noFormat); $i++) {
+				$html = str_replace("<!-- ELEMENT $i -->", $noFormat[$i], $html);
+			}
+			return $html;
+		}
+		// remove empty lines
+		$htmlArray = [''];
+		$z = 1;
+		for ($x = 0; $x < count($htmlArrayTemp); $x++) {
+			$t = trim($htmlArrayTemp[$x]);
+			if ($t !== '') {
+				$htmlArray[$z] = $htmlArrayTemp[$x];
+				$z++;
+			} else {
+				$htmlArray[$z] = $this->emptySpaceChar;
+				$z++;
+			}
+		}
+
+		// rebuild html
+		$html = '';
+		$tabs = 0;
+		for ($x = 0; $x < count($htmlArray); $x++) {
+			// check if the element should stand in a new line
+			$newline = false;
+			if (substr($htmlArray[$x - 1], 0, 5) == '<?xml') {
+				$newline = true;
+			} elseif ($this->formatType == 2 && ( // minimalistic line break
+					# this element has a line break before itself
+					preg_match(
+						'/<' . $structureBoxLikeElements . '(.*)>/Usi',
+						$htmlArray[$x]
+					) || preg_match(
+						'/<' . $structureBoxLikeElements . '(.*) \/>/Usi',
+						$htmlArray[$x]
+					) || # one element before is a element that has a line break after
+					preg_match(
+						'/<\/' . $structureBoxLikeElements . '(.*)>/Usi',
+						$htmlArray[$x - 1]
+					) || substr(
+						$htmlArray[$x - 1],
+						0,
+						4
+					) == '<!--' || preg_match('/<' . $structureBoxLikeElements . '(.*) \/>/Usi', $htmlArray[$x - 1]))
+			) {
+				$newline = true;
+			} elseif ($this->formatType == 3 && ( // aestetic line break
+					# this element has a line break before itself
+					preg_match(
+						'/<' . $esteticBoxLikeElements . '(.*)>/Usi',
+						$htmlArray[$x]
+					) || preg_match(
+						'/<' . $esteticBoxLikeElements . '(.*) \/>/Usi',
+						$htmlArray[$x]
+					) || # one element before is a element that has a line break after
+					preg_match('/<\/' . $esteticBoxLikeElements . '(.*)>/Usi', $htmlArray[$x - 1]) || substr(
+						$htmlArray[$x - 1],
+						0,
+						4
+					) == '<!--' || preg_match('/<' . $esteticBoxLikeElements . '(.*) \/>/Usi', $htmlArray[$x - 1]))
+			) {
+				$newline = true;
+			} elseif ($this->formatType >= 4 && ( // logical line break
+					# this element has a line break before itself
+					preg_match(
+						'/<' . $allBoxLikeElements . '(.*)>/Usi',
+						$htmlArray[$x]
+					) || preg_match(
+						'/<' . $allBoxLikeElements . '(.*) \/>/Usi',
+						$htmlArray[$x]
+					) || # one element before is a element that has a line break after
+					preg_match('/<\/' . $allBoxLikeElements . '(.*)>/Usi', $htmlArray[$x - 1]) || substr(
+						$htmlArray[$x - 1],
+						0,
+						4
+					) == '<!--' || preg_match('/<' . $allBoxLikeElements . '(.*) \/>/Usi', $htmlArray[$x - 1]))
+			) {
+				$newline = true;
+			}
+
+			// count down a tab
+			if (substr($htmlArray[$x], 0, 2) == '</') {
+				$tabs--;
+			}
+
+			// add tabs and line breaks in front of the current tag
+			if ($newline) {
+				$html .= $this->newline;
+				for ($y = 0; $y < $tabs; $y++) {
+					$html .= $this->tab;
+				}
+			}
+
+			// remove white spaces and line breaks and add current tag to the html-string
+			if (substr($htmlArray[$x], 0, 9) == '<![CDATA[' // remove multiple white space in CDATA / XML
+				|| substr($htmlArray[$x], 0, 5) == '<?xml'
+			) {
+				$html .= $this->killWhiteSpace($htmlArray[$x]);
+			} else { // remove all line breaks
+				$html .= $this->killLineBreaks($htmlArray[$x]);
+			}
+
+			// count up a tab
+			if (substr($htmlArray[$x], 0, 1) == '<' && substr($htmlArray[$x], 1, 1) != '/') {
+				if (
+					substr($htmlArray[$x], 1, 1) !== ' '
+					&& substr($htmlArray[$x], 1, 3) !== 'img'
+					&& substr($htmlArray[$x], 1, 6) !== 'source'
+					&& substr($htmlArray[$x], 1, 2) !== 'br'
+					&& substr($htmlArray[$x], 1, 2) !== 'hr'
+					&& substr($htmlArray[$x], 1, 5) !== 'input'
+					&& substr($htmlArray[$x], 1, 4) !== 'link'
+					&& substr($htmlArray[$x], 1, 4) !== 'meta'
+					&& substr($htmlArray[$x], 1, 4) !== 'col '
+					&& substr($htmlArray[$x], 1, 5) !== 'frame'
+					&& substr($htmlArray[$x], 1, 7) !== 'isindex'
+					&& substr($htmlArray[$x], 1, 5) !== 'param'
+					&& substr($htmlArray[$x], 1, 4) !== 'area'
+					&& substr($htmlArray[$x], 1, 4) !== 'base'
+					&& substr($htmlArray[$x], 0, 2) !== '<!'
+					&& substr($htmlArray[$x], 0, 5) !== '<?xml'
+				) {
+					$tabs++;
+				}
+			}
+		}
+
+		// Remove empty lines
+		if ($this->formatType > 1) {
+			$this->removeEmptyLines($html);
+		}
+
+		// Restore saved comments, styles and java-scripts
+		for ($i = 0; $i < count($noFormat); $i++) {
+			$html = str_replace("<!-- ELEMENT $i -->", $noFormat[$i], $html);
+		}
+
+		// include debug comment at the end
+		if ($tabs != 0 && $this->debugComment === true) {
+			$html .= "<!-- $tabs open elements found -->";
+		}
+
+		return $html;
+	}
+
+	/**
+	 * Remove ALL line breaks and multiple white space
+	 *
+	 * @param string $html
+	 *
+	 * @return string
+	 */
+	protected function killLineBreaks($html)
+	{
+		$html = str_replace($this->newline, '', $html);
+		$html = preg_replace('/\s\s+/u', ' ', $html);
+		return $html;
+		#? return preg_replace('/\n|\s+(\s)/u', '$1', $html);
+	}
+
+	/**
+	 * Remove multiple white space, keeps line breaks
+	 *
+	 * @param string $html
+	 *
+	 * @return string
+	 */
+	protected function killWhiteSpace($html)
+	{
+		$temp = explode($this->newline, $html);
+		for ($i = 0; $i < count($temp); $i++) {
+			if (!trim($temp[$i])) {
+				unset($temp[$i]);
+			} else {
+				$temp[$i] = trim($temp[$i]);
+				$temp[$i] = preg_replace('/\s\s+/', ' ', $temp[$i]);
+			}
+		}
+		$html = implode($this->newline, $temp);
+		return $html;
+	}
+
+	/**
+	 * Remove white space at the end of lines, keeps other white space and line breaks
+	 *
+	 * @param string $html
+	 *
+	 * @return string
+	 */
+	protected function rTrimLines(&$html)
+	{
+		$html = preg_replace('/\s+$/m', '', $html);
+	}
+
+	/**
+	 * Convert newlines according to the current OS
+	 *
+	 * @param string $html
+	 *
+	 * @return string
+	 */
+	protected function convNlOs(&$html)
+	{
+		$html = preg_replace("(\r\n|\r)", $this->newline, $html);
+	}
+
+	/**
+	 * Remove empty lines
+	 *
+	 * @param string $html
+	 *
+	 * @return void
+	 */
+	protected function removeEmptyLines(&$html)
+	{
+		$temp = explode($this->newline, $html);
+		$result = [];
+		for ($i = 0; $i < count($temp); ++$i) {
+			if ('' == trim($temp[$i])) {
+				continue;
+			}
+			$result[] = $temp[$i];
+		}
+		$html = implode($this->newline, $result);
+	}
+
+	/**
+	 * Include configured header comment in HTML content block
+	 *
+	 * @param $html
+	 */
+	public function includeHeaderComment(&$html)
+	{
+		$html = preg_replace('/^(-->)$/m', "\n\t" . $this->headerComment . "\n$1", $html);
+	}
 }

Spacing +23 added lines, -23 removed lines

@@ -68,7 +68,7 @@ 
     {
         if (!empty($config)) {
             if ($config['formatHtml'] && is_numeric($config['formatHtml'])) {
-                $this->formatType = (int)$config['formatHtml'];
+                $this->formatType = (int) $config['formatHtml'];
             }
 
             if ($config['formatHtml.']['tabSize'] && is_numeric($config['formatHtml.']['tabSize'])) {
@@ -76,14 +76,14 @@ 
             }
 
             if (isset($config['formatHtml.']['debugComment'])) {
-                $this->debugComment = (bool)$config['formatHtml.']['debugComment'];
+                $this->debugComment = (bool) $config['formatHtml.']['debugComment'];
             }
 
             if (isset($config['headerComment'])) {
                 $this->headerComment = $config['headerComment'];
             }
 
-            if (isset($config['dropEmptySpaceChar']) && (bool)$config['dropEmptySpaceChar']) {
+            if (isset($config['dropEmptySpaceChar']) && (bool) $config['dropEmptySpaceChar']) {
                 $this->emptySpaceChar = '';
             }
         }
@@ -100,7 +100,7 @@ 
     public function clean($html, $config = [])
     {
         if (!empty($config)) {
-            if ((bool)$config['enabled'] === false) {
+            if ((bool) $config['enabled'] === false) {
                 return $html;
             }
 
@@ -111,11 +111,11 @@ 
 
         $manipulations = [];
 
-        if (isset($config['removeGenerator']) && (bool)$config['removeGenerator']) {
+        if (isset($config['removeGenerator']) && (bool) $config['removeGenerator']) {
             $manipulations['removeGenerator'] = GeneralUtility::makeInstance(RemoveGenerator::class);
         }
 
-        if (isset($config['removeComments']) && (bool)$config['removeComments']) {
+        if (isset($config['removeComments']) && (bool) $config['removeComments']) {
             $manipulations['removeComments'] = GeneralUtility::makeInstance(RemoveComments::class);
         }
 
@@ -125,12 +125,12 @@ 
 
         foreach ($manipulations as $key => $manipulation) {
             /** @var ManipulationInterface $manipulation */
-            $configuration = isset($config[$key . '.']) && is_array($config[$key . '.']) ? $config[$key . '.'] : [];
+            $configuration = isset($config[$key.'.']) && is_array($config[$key.'.']) ? $config[$key.'.'] : [];
             $html = $manipulation->manipulate($html, $configuration);
         }
 
         // cleanup HTML5 self-closing elements
-        if(!isset($GLOBALS['TSFE']->config['config']['doctype']) || 'x' !== substr($GLOBALS['TSFE']->config['config']['doctype'],0,1)) {
+        if (!isset($GLOBALS['TSFE']->config['config']['doctype']) || 'x' !== substr($GLOBALS['TSFE']->config['config']['doctype'], 0, 1)) {
             $html = preg_replace('/<((?:area|base|br|col|embed|hr|img|input|link|meta|param|source|track|wbr)\s[^>]+?)\s?\/>/', '<$1>', $html);
         }
 
@@ -183,7 +183,7 @@ 
         $functionalBoxElements = 'dd|dt|frameset|li|tbody|td|tfoot|th|thead|tr|colgroup';
         $usableBoxElements = 'applet|button|del|iframe|ins|map|object|script';
         $imagineBoxElements = 'html|body|head|meta|title|link|script|base|!--';
-        $allBoxLikeElements = '(?>' . $trueBoxElements . '|' . $functionalBoxElements . '|' . $usableBoxElements . '|' . $imagineBoxElements . ')';
+        $allBoxLikeElements = '(?>'.$trueBoxElements.'|'.$functionalBoxElements.'|'.$usableBoxElements.'|'.$imagineBoxElements.')';
         $esteticBoxLikeElements = '(?>html|head|body|meta name|title|div|table|h1|h2|h3|h4|h5|h6|p|form|pre|center|!--)';
         $structureBoxLikeElements = '(?>html|head|body|div|!--)';
 
@@ -192,7 +192,7 @@ 
             '/(<(?:[^<>]+(?:"[^"]*"|\'[^\']*\')?)+>)/',
             $html,
             -1,
-            PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY
+            PREG_SPLIT_DELIM_CAPTURE|PREG_SPLIT_NO_EMPTY
         );
 
         if ($htmlArrayTemp === false) {
@@ -227,52 +227,52 @@ 
             } elseif ($this->formatType == 2 && ( // minimalistic line break
                     # this element has a line break before itself
                     preg_match(
-                        '/<' . $structureBoxLikeElements . '(.*)>/Usi',
+                        '/<'.$structureBoxLikeElements.'(.*)>/Usi',
                         $htmlArray[$x]
                     ) || preg_match(
-                        '/<' . $structureBoxLikeElements . '(.*) \/>/Usi',
+                        '/<'.$structureBoxLikeElements.'(.*) \/>/Usi',
                         $htmlArray[$x]
                     ) || # one element before is a element that has a line break after
                     preg_match(
-                        '/<\/' . $structureBoxLikeElements . '(.*)>/Usi',
+                        '/<\/'.$structureBoxLikeElements.'(.*)>/Usi',
                         $htmlArray[$x - 1]
                     ) || substr(
                         $htmlArray[$x - 1],
                         0,
                         4
-                    ) == '<!--' || preg_match('/<' . $structureBoxLikeElements . '(.*) \/>/Usi', $htmlArray[$x - 1]))
+                    ) == '<!--' || preg_match('/<'.$structureBoxLikeElements.'(.*) \/>/Usi', $htmlArray[$x - 1]))
             ) {
                 $newline = true;
             } elseif ($this->formatType == 3 && ( // aestetic line break
                     # this element has a line break before itself
                     preg_match(
-                        '/<' . $esteticBoxLikeElements . '(.*)>/Usi',
+                        '/<'.$esteticBoxLikeElements.'(.*)>/Usi',
                         $htmlArray[$x]
                     ) || preg_match(
-                        '/<' . $esteticBoxLikeElements . '(.*) \/>/Usi',
+                        '/<'.$esteticBoxLikeElements.'(.*) \/>/Usi',
                         $htmlArray[$x]
                     ) || # one element before is a element that has a line break after
-                    preg_match('/<\/' . $esteticBoxLikeElements . '(.*)>/Usi', $htmlArray[$x - 1]) || substr(
+                    preg_match('/<\/'.$esteticBoxLikeElements.'(.*)>/Usi', $htmlArray[$x - 1]) || substr(
                         $htmlArray[$x - 1],
                         0,
                         4
-                    ) == '<!--' || preg_match('/<' . $esteticBoxLikeElements . '(.*) \/>/Usi', $htmlArray[$x - 1]))
+                    ) == '<!--' || preg_match('/<'.$esteticBoxLikeElements.'(.*) \/>/Usi', $htmlArray[$x - 1]))
             ) {
                 $newline = true;
             } elseif ($this->formatType >= 4 && ( // logical line break
                     # this element has a line break before itself
                     preg_match(
-                        '/<' . $allBoxLikeElements . '(.*)>/Usi',
+                        '/<'.$allBoxLikeElements.'(.*)>/Usi',
                         $htmlArray[$x]
                     ) || preg_match(
-                        '/<' . $allBoxLikeElements . '(.*) \/>/Usi',
+                        '/<'.$allBoxLikeElements.'(.*) \/>/Usi',
                         $htmlArray[$x]
                     ) || # one element before is a element that has a line break after
-                    preg_match('/<\/' . $allBoxLikeElements . '(.*)>/Usi', $htmlArray[$x - 1]) || substr(
+                    preg_match('/<\/'.$allBoxLikeElements.'(.*)>/Usi', $htmlArray[$x - 1]) || substr(
                         $htmlArray[$x - 1],
                         0,
                         4
-                    ) == '<!--' || preg_match('/<' . $allBoxLikeElements . '(.*) \/>/Usi', $htmlArray[$x - 1]))
+                    ) == '<!--' || preg_match('/<'.$allBoxLikeElements.'(.*) \/>/Usi', $htmlArray[$x - 1]))
             ) {
                 $newline = true;
             }
@@ -430,6 +430,6 @@ 
      */
     public function includeHeaderComment(&$html)
     {
-        $html = preg_replace('/^(-->)$/m', "\n\t" . $this->headerComment . "\n$1", $html);
+        $html = preg_replace('/^(-->)$/m', "\n\t".$this->headerComment."\n$1", $html);
     }
 }