HtmlMin - Code Metrics - Inspection of "[+]: protect inline styles / scripts v2" - voku/HtmlMin - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( 634334...8dad79 )

by Lars

created 2016-10-17 09:16 UTC

HtmlMin D

↳ Parent: Project

Complexity

Total Complexity

Size/Duplication

Total Lines	525
Duplicated Lines	0 %

Coupling/Cohesion

Components	1
Dependencies	4

Test Coverage

Coverage

93.75%

Importance

Changes

Metric	Value
wmc	82
lcom	1
cbo	4
dl	0
loc	525
ccs	120
cts	128
cp	0.9375
rs	4.8717
c	0
b	0
f	0

11 Methods

Rating	Name	Size	Complexity
A	__construct()	8	1
B	minify()	97	4
B	protectTagsInDom()	31	6
A	optimizeAttributesInDom()	10	2
A	removeCommentsInDom()	14	3
B	trimTagsInDom()	29	6
B	removeWhitespaceInDom()	26	5
A	restoreProtectedHtml()	13	3
C	optimizeAttributes()	43	13
C	optimizeAttributesFilters()	54	33
B	sortCssClasses()	23	6

How to fix Complexity

<?php

namespace voku\helper;

/**
 * Class HtmlMin
 *
 * Inspired by:
 * - JS: https://github.com/kangax/html-minifier/blob/gh-pages/src/htmlminifier.js
 * - PHP: https://github.com/searchturbine/phpwee-php-minifier
 * - PHP: https://github.com/WyriHaximus/HtmlCompress
 * - PHP: https://github.com/zaininnari/html-minifier
 * - Java: https://code.google.com/archive/p/htmlcompressor/
 *
 * @package voku\helper
 */
class HtmlMin
{
  /**
   * // https://mathiasbynens.be/demo/javascript-mime-type
   * // https://developer.mozilla.org/en/docs/Web/HTML/Element/script#attr-type
   *
   * @var array
   */
  private static $executableScriptsMimeTypes = array(
      'text/javascript',
      'text/ecmascript',
      'text/jscript',
      'application/javascript',
      'application/x-javascript',
      'application/ecmascript',
  );

  /**
   * @var array
   */
  private static $booleanAttributes = array(
      'allowfullscreen',
      'async',
      'autofocus',
      'autoplay',
      'checked',
      'compact',
      'controls',
      'declare',
      'default',
      'defaultchecked',
      'defaultmuted',
      'defaultselected',
      'defer',
      'disabled',
      'enabled',
      'formnovalidate',
      'hidden',
      'indeterminate',
      'inert',
      'ismap',
      'itemscope',
      'loop',
      'multiple',
      'muted',
      'nohref',
      'noresize',
      'noshade',
      'novalidate',
      'nowrap',
      'open',
      'pauseonexit',
      'readonly',
      'required',
      'reversed',
      'scoped',
      'seamless',
      'selected',
      'sortable',
      'truespeed',
      'typemustmatch',
      'visible',
  );

  /**
   * An random md5-hash, generated via "random_bytes()".
   *
   * @var string
   */
  private $randomHash;

  /**
   * @var array
   */
  private $protectedChildNodes;

  /**
   * @var array
   */
  private static $skipTagsForRemoveWhitespace = array('style', 'pre', 'code', 'script', 'textarea');

  /**
   * @var string
   */
  private $protectedChildNodesHelper;

  /**
   * @var string
   */
  private $booleanAttributesHelper;

  /**
   * HtmlMin constructor.
   */
  public function __construct()
  {
    $this->protectedChildNodes = array();
    $this->randomHash = md5(Bootup::get_random_bytes(16));

    $this->protectedChildNodesHelper = 'html-min--saved-content-' . $this->randomHash;
    $this->booleanAttributesHelper = 'html-min--delete-this-' . $this->randomHash;
  }

  /**
   * @param string $html
   *
   * @return string
   */
  public function minify($html)
  {
    $html = (string)$html;
    if (!isset($html[0])) {
      return '';
    }

    $html = trim($html);
    if (!$html) {
      return '';
    }

    // init
    $this->protectedChildNodes = array();
    $origHtml = $html;
    $origHtmlLength = UTF8::strlen($html);

    $dom = new HtmlDomParser();
    $dom->getDocument()->preserveWhiteSpace = false; // remove redundant white space
    $dom->getDocument()->formatOutput = false; // do not formats output with indentation

    $dom->loadHtml($html);

    $dom = $this->protectTagsInDom($dom);
    $dom = $this->optimizeAttributesInDom($dom);
    $dom = $this->removeCommentsInDom($dom);
    $dom = $this->removeWhitespaceInDom($dom);
    $dom = $this->trimTagsInDom($dom);

    $html = $dom->html();

    // -------------------------------------------------------------------------
    // Trim whitespace from html-string. [protected html is still protected]
    // -------------------------------------------------------------------------

    // Remove spaces that are followed by either > or <
    $html = preg_replace('/ (>)/', '$1', $html);
    // Remove spaces that are preceded by either > or <
    $html = preg_replace('/(<) /', '$1', $html);
    // Remove spaces that are between > and <
    $html = preg_replace('/(>) (<)/', '>$2', $html);

    // -------------------------------------------------------------------------
    // Restore protected HTML-code.
    // -------------------------------------------------------------------------

    $html = preg_replace_callback(
        '/<(?<element>'. $this->protectedChildNodesHelper . ')(?<attributes> [^>]*)?>(?<value>.*?)<\/' . $this->protectedChildNodesHelper . '>/',
        array($this, 'restoreProtectedHtml'),
        $html
    );
    $html = $dom::putReplacedBackToPreserveHtmlEntities($html);

    // ------------------------------------
    // final clean-up
    // ------------------------------------

    $html = UTF8::cleanup($html);

    $html = str_replace(
        array(
            'html>' . "\n",
            "\n" . '<html',
            '<!doctype',
            '="' . $this->booleanAttributesHelper . '"',
            '</' . $this->protectedChildNodesHelper . '>',
        ),
        array(
            'html>',
            '<html',
            '<!DOCTYPE',
            '',
            '',
        ),
        $html
    );

    $html = preg_replace(
        array(
            '/<(?:' . $this->protectedChildNodesHelper . ')(:? [^>]*)?>/'
        ),
        array(
            ''
        ),
        $html
    );

    // ------------------------------------
    // check if compression worked
    // ------------------------------------

    if ($origHtmlLength < UTF8::strlen($html)) {
      $html = $origHtml;
    }

    return $html;
  }

  /**
   * Prevent changes of inline "styles" and "scripts".
   *
   * @param HtmlDomParser $dom
   *
   * @return HtmlDomParser
   */
  private function protectTagsInDom(HtmlDomParser $dom)
  {
    // init
    $i = 0;

    foreach ($dom->find('script, style') as $element) {

      // skip external links
      if ($element->tag === 'script' || $element->tag === 'style') {
        $attributs = $element->getAllAttributes();
        if (isset($attributs['src'])) {
          continue;
        }
      }

      $node = $element->getNode();
      while ($node->childNodes->length > 0) {
        $this->protectedChildNodes[$i][] = $node->firstChild->nodeValue;
        $node->removeChild($node->firstChild);
      }

      $child = new \DOMElement($this->protectedChildNodesHelper);
      $node = $element->getNode()->appendChild($child);
      /* @var $node \DOMElement */
      $node->setAttribute('data-html-min--saved-content', $i);

      ++$i;
    }

    return $dom;
  }

  /**
   * Optimize HTML-tag attributes in the dom.
   *
   * @param HtmlDomParser $dom
   *
   * @return HtmlDomParser
   */
  private function optimizeAttributesInDom(HtmlDomParser $dom)
  {
    foreach ($dom->find('*') as $element) {
      $attributs = $element->getAllAttributes();

      $this->optimizeAttributes($element, $attributs);
    }

    return $dom;
  }

  /**
   * Remove comments in the dom.
   *
   * @param HtmlDomParser $dom
   *
   * @return HtmlDomParser
   */
  private function removeCommentsInDom(HtmlDomParser $dom)
  {
    foreach ($dom->find('//comment()') as $commentWrapper) {
      $comment = $commentWrapper->getNode();
      $val = $comment->nodeValue;
      if (strpos($val, '[') !== 0) {
        $comment->parentNode->removeChild($comment);
      }
    }

    $dom->getDocument()->normalizeDocument();

    return $dom;
  }

  /**
   * Trim tags in the dom.
   *
   * @param HtmlDomParser $dom
   *
   * @return HtmlDomParser
   */
  private function trimTagsInDom(HtmlDomParser $dom) {
    $divnodes = $dom->find('//div|//p|//nav|//footer|//article|//script|//hr|//br');
    foreach ($divnodes as $divnodeWrapper) {
      $divnode = $divnodeWrapper->getNode();

      $candidates = array();
      /** @noinspection PhpParamsInspection */
      if (count($divnode->childNodes) > 0) {
        $candidates[] = $divnode->firstChild;
        $candidates[] = $divnode->lastChild;
        $candidates[] = $divnode->previousSibling;
        $candidates[] = $divnode->nextSibling;
      }

      foreach ($candidates as $candidate) {
        if ($candidate === null) {
          continue;
        }

        if ($candidate->nodeType === 3) {
          $candidate->nodeValue = trim($candidate->nodeValue);
        }
      }
    }

    $dom->getDocument()->normalizeDocument();

    return $dom;
  }

  /**
   * Remove whitespace from dom-nodes.
   *
   * @param HtmlDomParser $dom
   *
   * @return HtmlDomParser
   */
  private function removeWhitespaceInDom(HtmlDomParser $dom)
  {
    $textnodes = $dom->find('//text()');
    foreach ($textnodes as $textnodeWrapper) {
      $textnode = $textnodeWrapper->getNode();
      $xp = $textnode->getNodePath();

      $doSkip = false;
      foreach (self::$skipTagsForRemoveWhitespace as $pattern) {
        if (strpos($xp, "/$pattern") !== false) {
          $doSkip = true;
          break;
        }
      }

      if ($doSkip) {
        continue;
      }

      $textnode->nodeValue = preg_replace("/\s{2,}/", ' ', $textnode->nodeValue);
    }

    $dom->getDocument()->normalizeDocument();

    return $dom;
  }

  /**
   * Callback function for preg_replace_callback use.
   *
   * @param  array $matches PREG matches
   *
   * @return string
   */
  private function restoreProtectedHtml($matches)
  {
    preg_match('/.*"(?<id>\d*)"/', $matches['attributes'], $matchesInner);

    $htmlChild = '';
    if (isset($this->protectedChildNodes[$matchesInner['id']])) {
      foreach ($this->protectedChildNodes[$matchesInner['id']] as $childNode) {
        $htmlChild .= $childNode;
      }
    }

    return $htmlChild;
  }

  /**
   * Sort HTML-Attributes, so that gzip can do better work
   *  and remove some default attributes.
   *
   * @param SimpleHtmlDom $element
   * @param array         $attributs
   *
   * @return bool
   */
  private function optimizeAttributes(SimpleHtmlDom $element, &$attributs)
  {
    if (!$attributs) {

      return false;
    }

    $attrs = array();
    foreach ($attributs as $attrName => $attrValue) {

      if (in_array($attrName, self::$booleanAttributes, true)) {
        $attrs[$attrName] = $this->booleanAttributesHelper;
        $element->{$attrName} = null;
        continue;
      }

      if (
          ($attrName === 'href' || $attrName === 'src' || $attrName === 'action')
          &&
          !(isset($attributs['rel']) && $attributs['rel'] === 'external')
          &&
          !(isset($attributs['target']) && $attributs['target'] === '_blank')
      ) {
        $attrValue = str_replace('http://', '//', $attrValue);
      }

      if ($this->optimizeAttributesFilters($element->tag, $attrName, $attrValue, $attributs)) {
function acceptsInteger($int) { }

$x = '123'; // string "123"

// Instead of
acceptsInteger($x);

// we recommend to use
acceptsInteger((integer) $x);
        $element->{$attrName} = null;
        continue;
      }

      $attrValue = $this->sortCssClasses($attrName, $attrValue);

      $attrs[$attrName] = $attrValue;
      $element->{$attrName} = null;
    }

    ksort($attrs);
    foreach ($attrs as $attrName => $attrValue) {
      $element->setAttribute($attrName, $attrValue, true);
    }

    return true;
  }

  /**
   * Check if the attribute (key / value) is default and can be skipped.
   *
   * @param string $tag
   * @param string $attrName
   * @param string $attrValue
   * @param string $allAttr
   *
   * @return bool
   */
  private function optimizeAttributesFilters($tag, $attrName, $attrValue, $allAttr)
  {
    // remove default
    if ($tag === 'script' && $attrName === 'language' && $attrValue === 'javascript') {
      return true;
    }

    // remove default
    if ($tag === 'form' && $attrName === 'method' && $attrValue === 'get') {
      return true;
    }

    // remove default
    if ($tag === 'input' && $attrName === 'type' && $attrValue === 'text') {
      return true;
    }

    // remove default
    if ($tag === 'area' && $attrName === 'shape' && $attrValue === 'rect') {
      return true;
    }

    // remove deprecated charset-attribute (the Browser will use the charset from the HTTP-Header, anyway)
    if ($tag === 'script' && $attrName === 'charset' && !isset($allAttr['src'])) {
      return true;
    }

    // remove deprecated anchor-jump
    if ($tag === 'a' && $attrName === 'name' && isset($allAttr['id'])) {
      return true;
    }

    // remove "type=text/css" for css links
    if ($tag === 'link' && $attrName === 'type' && $attrValue === 'text/css' && isset($allAttr['rel']) && $allAttr['rel'] === 'stylesheet') {
      return true;
    }

    // remove deprecated script-mime-types
    if ($tag === 'script' && $attrName === 'type' && isset($allAttr['src']) && in_array($attrValue, self::$executableScriptsMimeTypes, true)) {
      return true;
    }

    // remove empty value from <input>
    if ($tag === 'input' && $attrName === 'value' && $attrValue === '') {
      return true;
    }

    // remove some empty attribute
    if ($attrValue === '' && preg_match('/^(?:class|id|style|title|lang|dir|on(?:focus|blur|change|click|dblclick|mouse(?:down|up|over|move|out)|key(?:press|down|up)))$/', $attrName)) {
      return true;
    }

    return false;
  }

  /**
   * @param $attrName
   * @param $attrValue
   *
   * @return string
   */
  private function sortCssClasses($attrName, $attrValue)
  {
    if ($attrName !== 'class' || !$attrValue) {
      return $attrValue;
    }

    $classes = explode(' ', $attrValue);
    if (!$classes) {

      return '';
    }

    sort($classes);
    $attrValue = '';
    foreach ($classes as $class) {
      if (!$class) {
        continue;
      }
      $attrValue .= trim($class) . ' ';
    }
    $attrValue = trim($attrValue);

    return $attrValue;
  }
}


1		<?php
2
3		namespace voku\helper;
4
5		/**
6		* Class HtmlMin
7		*
8		* Inspired by:
9		* - JS: https://github.com/kangax/html-minifier/blob/gh-pages/src/htmlminifier.js
10		* - PHP: https://github.com/searchturbine/phpwee-php-minifier
11		* - PHP: https://github.com/WyriHaximus/HtmlCompress
12		* - PHP: https://github.com/zaininnari/html-minifier
13		* - Java: https://code.google.com/archive/p/htmlcompressor/
14		*
15		* @package voku\helper
16		*/
17		class HtmlMin
18		{
19		/**
20		* // https://mathiasbynens.be/demo/javascript-mime-type
21		* // https://developer.mozilla.org/en/docs/Web/HTML/Element/script#attr-type
22		*
23		* @var array
24		*/
25		private static $executableScriptsMimeTypes = array(
26		'text/javascript',
27		'text/ecmascript',
28		'text/jscript',
29		'application/javascript',
30		'application/x-javascript',
31		'application/ecmascript',
32		);
33
34		/**
35		* @var array
36		*/
37		private static $booleanAttributes = array(
38		'allowfullscreen',
39		'async',
40		'autofocus',
41		'autoplay',
42		'checked',
43		'compact',
44		'controls',
45		'declare',
46		'default',
47		'defaultchecked',
48		'defaultmuted',
49		'defaultselected',
50		'defer',
51		'disabled',
52		'enabled',
53		'formnovalidate',
54		'hidden',
55		'indeterminate',
56		'inert',
57		'ismap',
58		'itemscope',
59		'loop',
60		'multiple',
61		'muted',
62		'nohref',
63		'noresize',
64		'noshade',
65		'novalidate',
66		'nowrap',
67		'open',
68		'pauseonexit',
69		'readonly',
70		'required',
71		'reversed',
72		'scoped',
73		'seamless',
74		'selected',
75		'sortable',
76		'truespeed',
77		'typemustmatch',
78		'visible',
79		);
80
81		/**
82		* An random md5-hash, generated via "random_bytes()".
83		*
84		* @var string
85		*/
86		private $randomHash;
87
88		/**
89		* @var array
90	21	*/
91		private $protectedChildNodes;
92	21
93	21	/**
94		* @var array
95		*/
96		private static $skipTagsForRemoveWhitespace = array('style', 'pre', 'code', 'script', 'textarea');
97
98		/**
99		* @var string
100	21	*/
101		private $protectedChildNodesHelper;
102	21
103	21	/**
104		* @var string
105		*/
106		private $booleanAttributesHelper;
107	21
108	21	/**
109	2	* HtmlMin constructor.
110		*/
111		public function __construct()
112	19	{
113	19	$this->protectedChildNodes = array();
114		$this->randomHash = md5(Bootup::get_random_bytes(16));
115	19
116	19	$this->protectedChildNodesHelper = 'html-min--saved-content-' . $this->randomHash;
117	19	$this->booleanAttributesHelper = 'html-min--delete-this-' . $this->randomHash;
118		}
119	19
120	19	/**
121		* @param string $html
122	19	*
123	2	* @return string
124	2	*/
125	2	public function minify($html)
126		{
127		$html = (string)$html;
128		if (!isset($html[0])) {
129	19	return '';
130		}
131	19
132	19	$html = trim($html);
133	19	if (!$html) {
134		return '';
135	15	}
136
137	15	// init
138	15	$this->protectedChildNodes = array();
139	15	$origHtml = $html;
140	3	$origHtmlLength = UTF8::strlen($html);
141	15
142		$dom = new HtmlDomParser();
143		$dom->getDocument()->preserveWhiteSpace = false; // remove redundant white space
144		$dom->getDocument()->formatOutput = false; // do not formats output with indentation
145	15
146	3	$dom->loadHtml($html);
147
148		$dom = $this->protectTagsInDom($dom);
149	15	$dom = $this->optimizeAttributesInDom($dom);
150		$dom = $this->removeCommentsInDom($dom);
151		$dom = $this->removeWhitespaceInDom($dom);
152	19	$dom = $this->trimTagsInDom($dom);
153
154	19	$html = $dom->html();
155	19
156	7	// -------------------------------------------------------------------------
157		// Trim whitespace from html-string. [protected html is still protected]
158	7	// -------------------------------------------------------------------------
159	7
160	7	// Remove spaces that are followed by either > or <
161	7	$html = preg_replace('/ (>)/', '$1', $html);
162	7	// Remove spaces that are preceded by either > or <
163		$html = preg_replace('/(<) /', '$1', $html);
164		// Remove spaces that are between > and <
165	7	$html = preg_replace('/(>) (<)/', '>$2', $html);
166	7
167	7	// -------------------------------------------------------------------------
168		// Restore protected HTML-code.
169		// -------------------------------------------------------------------------
170	7
171	7	$html = preg_replace_callback(
172		'/<(?<element>'. $this->protectedChildNodesHelper . ')(?<attributes> [^>])?>(?<value>.?)<\/' . $this->protectedChildNodesHelper . '>/',
173		array($this, 'restoreProtectedHtml'),
174		$html
175		);
176	19	$html = $dom::putReplacedBackToPreserveHtmlEntities($html);
177
178	19	// ------------------------------------
179	19	// final clean-up
180	19	// ------------------------------------
181
182		$html = UTF8::cleanup($html);
183
184		$html = str_replace(
185	19	array(
186		'html>' . "\n",
187		"\n" . '<html',
188		'<!doctype',
189	19	'="' . $this->booleanAttributesHelper . '"',
190		'</' . $this->protectedChildNodesHelper . '>',
191		),
192		array(
193	19	'html>',
194		'<html',
195	19	'<!DOCTYPE',
196		'',
197	19	'',
198		),
199	19	$html
200	19	);
201
202		$html = preg_replace(
203	19	array(
204		'/<(?:' . $this->protectedChildNodesHelper . ')(:? [^>]*)?>/'
205		),
206		array(
207		''
208		),
209		$html
210		);
211	19
212	3	// ------------------------------------
213		// check if compression worked
214		// ------------------------------------
215
216	19	if ($origHtmlLength < UTF8::strlen($html)) {
217		$html = $origHtml;
218	19	}
219
220	19	return $html;
221		}
222	19
223		/**
224		* Prevent changes of inline "styles" and "scripts".
225		*
226		* @param HtmlDomParser $dom
227		*
228		* @return HtmlDomParser
229		*/
230		private function protectTagsInDom(HtmlDomParser $dom)
231		{
232		// init
233	19	$i = 0;
234
235	19	foreach ($dom->find('script, style') as $element) {
236
237	19	// skip external links
238	19	if ($element->tag === 'script' \|\| $element->tag === 'style') {
239		$attributs = $element->getAllAttributes();
240		if (isset($attributs['src'])) {
241		continue;
242		}
243		}
244
245		$node = $element->getNode();
246		while ($node->childNodes->length > 0) {
247		$this->protectedChildNodes[$i][] = $node->firstChild->nodeValue;
248		$node->removeChild($node->firstChild);
249		}
250
251	8	$child = new \DOMElement($this->protectedChildNodesHelper);
252	8	$node = $element->getNode()->appendChild($child);
253		/* @var $node \DOMElement */
254	8	$node->setAttribute('data-html-min--saved-content', $i);
255	6
256	6	++$i;
257	6	}
258
259		return $dom;
260		}
261	8
262		/**
263	8	* Optimize HTML-tag attributes in the dom.
264		*
265	8	* @param HtmlDomParser $dom
266		*
267	3	* @return HtmlDomParser
268		*/
269		private function optimizeAttributesInDom(HtmlDomParser $dom)
270	8	{
271	2	foreach ($dom->find('*') as $element) {
272	2	$attributs = $element->getAllAttributes();
273
274		$this->optimizeAttributes($element, $attributs);
275	8	}
276
277	8	return $dom;
278	8	}
279
280		/**
281	8	* Remove comments in the dom.
282	8	*
283	8	* @param HtmlDomParser $dom
284		*
285		* @return HtmlDomParser
286	8	*/
287		private function removeCommentsInDom(HtmlDomParser $dom)
288		{
289		foreach ($dom->find('//comment()') as $commentWrapper) {
290		$comment = $commentWrapper->getNode();
291		$val = $comment->nodeValue;
292		if (strpos($val, '[') !== 0) {
293		$comment->parentNode->removeChild($comment);
294		}
295		}
296
297		$dom->getDocument()->normalizeDocument();
298
299	8	return $dom;
300		}
301
302	8	/**
303		* Trim tags in the dom.
304		*
305		* @param HtmlDomParser $dom
306		*
307	8	* @return HtmlDomParser
308	1	*/
309		private function trimTagsInDom(HtmlDomParser $dom) {
310		$divnodes = $dom->find('//div\|//p\|//nav\|//footer\|//article\|//script\|//hr\|//br');
311		foreach ($divnodes as $divnodeWrapper) {
312	8	$divnode = $divnodeWrapper->getNode();
313	1
314		$candidates = array();
315		/** @noinspection PhpParamsInspection */
316		if (count($divnode->childNodes) > 0) {
317	8	$candidates[] = $divnode->firstChild;
318		$candidates[] = $divnode->lastChild;
319		$candidates[] = $divnode->previousSibling;
320		$candidates[] = $divnode->nextSibling;
321		}
322	8
323		foreach ($candidates as $candidate) {
324		if ($candidate === null) {
325		continue;
326		}
327	8
328		if ($candidate->nodeType === 3) {
329		$candidate->nodeValue = trim($candidate->nodeValue);
330		}
331		}
332	8	}
333	1
334		$dom->getDocument()->normalizeDocument();
335
336		return $dom;
337	8	}
338	1
339		/**
340		* Remove whitespace from dom-nodes.
341		*
342	8	* @param HtmlDomParser $dom
343	1	*
344		* @return HtmlDomParser
345		*/
346		private function removeWhitespaceInDom(HtmlDomParser $dom)
347	8	{
348	1	$textnodes = $dom->find('//text()');
349		foreach ($textnodes as $textnodeWrapper) {
350		$textnode = $textnodeWrapper->getNode();
351	8	$xp = $textnode->getNodePath();
352
353		$doSkip = false;
354		foreach (self::$skipTagsForRemoveWhitespace as $pattern) {
355		if (strpos($xp, "/$pattern") !== false) {
356		$doSkip = true;
357		break;
358		}
359		}
360	8
361		if ($doSkip) {
362	8	continue;
363	8	}
364
365		$textnode->nodeValue = preg_replace("/\s{2,}/", ' ', $textnode->nodeValue);
366	4	}
367	4
368		$dom->getDocument()->normalizeDocument();
369
370		return $dom;
371	4	}
372	4
373	4	/**
374	4	* Callback function for preg_replace_callback use.
375	1	*
376		* @param array $matches PREG matches
377	4	*
378		* @return string
379	4	*/
380		private function restoreProtectedHtml($matches)
381	4	{
382		preg_match('/."(?<id>\d)"/', $matches['attributes'], $matchesInner);
383
384		$htmlChild = '';
385		if (isset($this->protectedChildNodes[$matchesInner['id']])) {
386		foreach ($this->protectedChildNodes[$matchesInner['id']] as $childNode) {
387		$htmlChild .= $childNode;
388		}
389		}
390
391		return $htmlChild;
392		}
393
394		/**
395		* Sort HTML-Attributes, so that gzip can do better work
396		* and remove some default attributes.
397		*
398		* @param SimpleHtmlDom $element
399		* @param array $attributs
400		*
401		* @return bool
402		*/
403		private function optimizeAttributes(SimpleHtmlDom $element, &$attributs)
404		{
405		if (!$attributs) {
		0 ignored issues – show Bug Best Practice introduced 2016-10-17 09:18 UTC by Report Bug Copy Issue Report The expression `$attributs` of type `array` is implicitly converted to a boolean; are you sure this is intended? If so, consider using `empty($expr)` instead to make it clear that you intend to check for an array without elements. This check marks implicit conversions of arrays to boolean values in a comparison. While in PHP an empty array is considered to be equal (but not identical) to false, this is not always apparent. Consider making the comparison explicit by using `empty(..)` or `! empty(...)` instead. Loading history...
406		return false;
407		}
408
409		$attrs = array();
410		foreach ($attributs as $attrName => $attrValue) {
411
412		if (in_array($attrName, self::$booleanAttributes, true)) {
413		$attrs[$attrName] = $this->booleanAttributesHelper;
414		$element->{$attrName} = null;
415		continue;
416		}
417
418		if (
419		($attrName === 'href' \|\| $attrName === 'src' \|\| $attrName === 'action')
420		&&
421		!(isset($attributs['rel']) && $attributs['rel'] === 'external')
422		&&
423		!(isset($attributs['target']) && $attributs['target'] === '_blank')
424		) {
425		$attrValue = str_replace('http://', '//', $attrValue);
426		}
427
428		if ($this->optimizeAttributesFilters($element->tag, $attrName, $attrValue, $attributs)) {
		0 ignored issues – show Documentation introduced 2016-10-16 03:44 UTC by Report Bug Copy Issue Report `$attributs` is of type `array`, but the function expects a `string`. It seems like the type of the argument is not accepted by the function/method which you are calling. In some cases, in particular if PHP’s automatic type-juggling kicks in this might be fine. In other cases, however this might be a bug. We suggest to add an explicit type cast like in the following example: function acceptsInteger($int) { } $x = '123'; // string "123" // Instead of acceptsInteger($x); // we recommend to use acceptsInteger((integer) $x); Loading history...
429		$element->{$attrName} = null;
430		continue;
431		}
432
433		$attrValue = $this->sortCssClasses($attrName, $attrValue);
434
435		$attrs[$attrName] = $attrValue;
436		$element->{$attrName} = null;
437		}
438
439		ksort($attrs);
440		foreach ($attrs as $attrName => $attrValue) {
441		$element->setAttribute($attrName, $attrValue, true);
442		}
443
444		return true;
445		}
446
447		/**
448		* Check if the attribute (key / value) is default and can be skipped.
449		*
450		* @param string $tag
451		* @param string $attrName
452		* @param string $attrValue
453		* @param string $allAttr
454		*
455		* @return bool
456		*/
457		private function optimizeAttributesFilters($tag, $attrName, $attrValue, $allAttr)
458		{
459		// remove default
460		if ($tag === 'script' && $attrName === 'language' && $attrValue === 'javascript') {
461		return true;
462		}
463
464		// remove default
465		if ($tag === 'form' && $attrName === 'method' && $attrValue === 'get') {
466		return true;
467		}
468
469		// remove default
470		if ($tag === 'input' && $attrName === 'type' && $attrValue === 'text') {
471		return true;
472		}
473
474		// remove default
475		if ($tag === 'area' && $attrName === 'shape' && $attrValue === 'rect') {
476		return true;
477		}
478
479		// remove deprecated charset-attribute (the Browser will use the charset from the HTTP-Header, anyway)
480		if ($tag === 'script' && $attrName === 'charset' && !isset($allAttr['src'])) {
481		return true;
482		}
483
484		// remove deprecated anchor-jump
485		if ($tag === 'a' && $attrName === 'name' && isset($allAttr['id'])) {
486		return true;
487		}
488
489		// remove "type=text/css" for css links
490		if ($tag === 'link' && $attrName === 'type' && $attrValue === 'text/css' && isset($allAttr['rel']) && $allAttr['rel'] === 'stylesheet') {
491		return true;
492		}
493
494		// remove deprecated script-mime-types
495		if ($tag === 'script' && $attrName === 'type' && isset($allAttr['src']) && in_array($attrValue, self::$executableScriptsMimeTypes, true)) {
496		return true;
497		}
498
499		// remove empty value from <input>
500		if ($tag === 'input' && $attrName === 'value' && $attrValue === '') {
501		return true;
502		}
503
504		// remove some empty attribute
505		if ($attrValue === '' && preg_match('/^(?:class\|id\|style\|title\|lang\|dir\|on(?:focus\|blur\|change\|click\|dblclick\|mouse(?:down\|up\|over\|move\|out)\|key(?:press\|down\|up)))$/', $attrName)) {
506		return true;
507		}
508
509		return false;
510		}
511
512		/**
513		* @param $attrName
514		* @param $attrValue
515		*
516		* @return string
517		*/
518		private function sortCssClasses($attrName, $attrValue)
519		{
520		if ($attrName !== 'class' \|\| !$attrValue) {
521		return $attrValue;
522		}
523
524		$classes = explode(' ', $attrValue);
525		if (!$classes) {
		0 ignored issues – show Bug Best Practice introduced 2016-10-16 03:44 UTC by Report Bug Copy Issue Report The expression `$classes` of type `array` is implicitly converted to a boolean; are you sure this is intended? If so, consider using `empty($expr)` instead to make it clear that you intend to check for an array without elements. This check marks implicit conversions of arrays to boolean values in a comparison. While in PHP an empty array is considered to be equal (but not identical) to false, this is not always apparent. Consider making the comparison explicit by using `empty(..)` or `! empty(...)` instead. Loading history...
526		return '';
527		}
528
529		sort($classes);
530		$attrValue = '';
531		foreach ($classes as $class) {
532		if (!$class) {
533		continue;
534		}
535		$attrValue .= trim($class) . ' ';
536		}
537		$attrValue = trim($attrValue);
538
539		return $attrValue;
540		}
541		}
542

voku / HtmlMin

Push — master ( 634334...8dad79 )

HtmlMin D

Complexity

Size/Duplication

Coupling/Cohesion

Test Coverage

Importance

11 Methods

How to fix Complexity

Complex Class

Duplication Side-by-Side

Filter issues like