HtmlMin::__construct() - Code Metrics - Inspection of "Merge remote-tracking branch 'origin/master'" - voku/HtmlMin - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( c442f2...6b8e64 )

by Lars

created 2016-10-16 04:07 UTC

HtmlMin::__construct() A

↳ Parent: HtmlMin

Complexity

Conditions	1
Paths	1

Size

Total Lines	4
Code Lines	2

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	3
CRAP Score	1

Importance

Changes

Metric	Value
dl	0
loc	4
ccs	3
cts	3
cp	1
rs	10
c	0
b	0
f	0
cc	1
eloc	2
nc	1
nop	0
crap	1

<?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/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
   */
  protected $randomHash;

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

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

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

    $origHtml = $html;
    $origHtmlLength = UTF8::strlen($html);

    $dom = new HtmlDomParser();
    $dom->getDocument()->preserveWhiteSpace = false;
    $dom->getDocument()->formatOutput = false;

    $dom->loadHtml($html);
    $xpath = new \DOMXPath($dom->getDocument());

    foreach ($xpath->query('//comment()') as $comment) {
      $val = $comment->nodeValue;
      if (strpos($val, '[') !== 0) {
        $comment->parentNode->removeChild($comment);
      }
    }

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

    $textnodes = $xpath->query('//text()');
    $skip = array('style', 'pre', 'code', 'script', 'textarea');
    foreach ($textnodes as $t) {
      /* @var $t \DOMNode */
      $xp = $t->getNodePath();

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

      if ($doSkip) {
        continue;
      }

      $t->nodeValue = preg_replace("/\s{2,}/", ' ', $t->nodeValue);
    }

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

    $divnodes = $xpath->query('//div|//p|//nav|//footer|//article|//script|//hr|//br');
    foreach ($divnodes as $d) {
      $candidates = array();

      if (count($d->childNodes)) {
        $candidates[] = $d->firstChild;
        $candidates[] = $d->lastChild;
        $candidates[] = $d->previousSibling;
        $candidates[] = $d->nextSibling;
      }

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

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

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

    $elements = $dom->find('*');
    foreach ($elements as $element) {
      if (count($element) > 1) {
        foreach ($element as $e) {
          $this->optimizeAttributes($e);
        }
      } else {
        $this->optimizeAttributes($element);
      }
    }

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

    // ------------------------------------

    $html = UTF8::cleanup($dom->html());
    // final clean-up
    $html = str_replace(
        array(
            'html>' . "\n",
            "\n" . '<html',
            '<!doctype',
            '="delete-this-' . $this->randomHash . '"',
        ),
        array(
            'html>',
            '<html',
            '<!DOCTYPE',
            '',
        ),
        $html
    );

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

    // 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);

    return $html;
  }

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

    if (!$attributs) {
      return false;
    }

    /*

    if (
        ($element->tag === 'script' || $element->tag === 'style')
        &&
        !isset($attributs['src'])
    ) {
      // TODO: protect inline css / js
    }
    */

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

      if (in_array($attrName, self::$booleanAttributes, true)) {
        $attrs[$attrName] = 'delete-this-' . $this->randomHash;
        $element->{$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/zaininnari/html-minifier
12		* - Java: https://code.google.com/archive/p/htmlcompressor/
13		*
14		* @package voku\helper
15		*/
16		class HtmlMin
17		{
18		/**
19		* // https://mathiasbynens.be/demo/javascript-mime-type
20		* // https://developer.mozilla.org/en/docs/Web/HTML/Element/script#attr-type
21		*
22		* @var array
23		*/
24		private static $executableScriptsMimeTypes = array(
25		'text/javascript',
26		'text/ecmascript',
27		'text/jscript',
28		'application/javascript',
29		'application/x-javascript',
30		'application/ecmascript',
31		);
32
33		/**
34		* @var array
35		*/
36		private static $booleanAttributes = array(
37		'allowfullscreen',
38		'async',
39		'autofocus',
40		'autoplay',
41		'checked',
42		'compact',
43		'controls',
44		'declare',
45		'default',
46		'defaultchecked',
47		'defaultmuted',
48		'defaultselected',
49		'defer',
50		'disabled',
51		'enabled',
52		'formnovalidate',
53		'hidden',
54		'indeterminate',
55		'inert',
56		'ismap',
57		'itemscope',
58		'loop',
59		'multiple',
60		'muted',
61		'nohref',
62		'noresize',
63		'noshade',
64		'novalidate',
65		'nowrap',
66		'open',
67		'pauseonexit',
68		'readonly',
69		'required',
70		'reversed',
71		'scoped',
72		'seamless',
73		'selected',
74		'sortable',
75		'truespeed',
76		'typemustmatch',
77		'visible',
78		);
79
80		/**
81		* An random md5-hash, generated via "random_bytes()".
82		*
83		* @var string
84		*/
85		protected $randomHash;
86
87		/**
88		* HtmlMin constructor.
89		*/
90	21	public function __construct()
91		{
92	21	$this->randomHash = md5(Bootup::get_random_bytes(16));
93	21	}
94
95		/**
96		* @param string $html
97		*
98		* @return string
99		*/
100	21	public function minify($html)
101		{
102	21	$html = (string)$html;
103	21	if (!isset($html[0])) {
104		return '';
105		}
106
107	21	$html = trim($html);
108	21	if (!$html) {
109	2	return '';
110		}
111
112	19	$origHtml = $html;
113	19	$origHtmlLength = UTF8::strlen($html);
114
115	19	$dom = new HtmlDomParser();
116	19	$dom->getDocument()->preserveWhiteSpace = false;
117	19	$dom->getDocument()->formatOutput = false;
118
119	19	$dom->loadHtml($html);
120	19	$xpath = new \DOMXPath($dom->getDocument());
121
122	19	foreach ($xpath->query('//comment()') as $comment) {
123	2	$val = $comment->nodeValue;
124	2	if (strpos($val, '[') !== 0) {
125	2	$comment->parentNode->removeChild($comment);
126	2	}
127	19	}
128
129	19	$dom->getDocument()->normalizeDocument();
130
131	19	$textnodes = $xpath->query('//text()');
132	19	$skip = array('style', 'pre', 'code', 'script', 'textarea');
133	19	foreach ($textnodes as $t) {
134		/* @var $t \DOMNode */
135	15	$xp = $t->getNodePath();
136
137	15	$doSkip = false;
138	15	foreach ($skip as $pattern) {
139	15	if (strpos($xp, "/$pattern") !== false) {
140	3	$doSkip = true;
141	3	break;
142		}
143	15	}
144
145	15	if ($doSkip) {
146	3	continue;
147		}
148
149	15	$t->nodeValue = preg_replace("/\s{2,}/", ' ', $t->nodeValue);
150	19	}
151
152	19	$dom->getDocument()->normalizeDocument();
153
154	19	$divnodes = $xpath->query('//div\|//p\|//nav\|//footer\|//article\|//script\|//hr\|//br');
155	19	foreach ($divnodes as $d) {
156	7	$candidates = array();
157
158	7	if (count($d->childNodes)) {
159	7	$candidates[] = $d->firstChild;
160	7	$candidates[] = $d->lastChild;
161	7	$candidates[] = $d->previousSibling;
162	7	$candidates[] = $d->nextSibling;
163	7	}
164
165	7	foreach ($candidates as $c) {
166	7	if ($c === null) {
167	7	continue;
168		}
169
170	7	if ($c->nodeType === 3) {
171	7	$c->nodeValue = trim($c->nodeValue);
172	7	}
173	7	}
174	19	}
175
176	19	$dom->getDocument()->normalizeDocument();
177
178	19	$elements = $dom->find('*');
179	19	foreach ($elements as $element) {
180	19	if (count($element) > 1) {
181		foreach ($element as $e) {
182		$this->optimizeAttributes($e);
183		}
184		} else {
185	19	$this->optimizeAttributes($element);
186		}
187	19	}
188
189	19	$dom->getDocument()->normalizeDocument();
190
191		// ------------------------------------
192
193	19	$html = UTF8::cleanup($dom->html());
194		// final clean-up
195	19	$html = str_replace(
196		array(
197	19	'html>' . "\n",
198	19	"\n" . '<html',
199	19	'<!doctype',
200	19	'="delete-this-' . $this->randomHash . '"',
201	19	),
202		array(
203	19	'html>',
204	19	'<html',
205	19	'<!DOCTYPE',
206	19	'',
207	19	),
208		$html
209	19	);
210
211	19	if ($origHtmlLength < UTF8::strlen($html)) {
212	3	$html = $origHtml;
213	3	}
214
215		// Remove spaces that are followed by either > or <
216	19	$html = preg_replace('/ (>)/', '$1', $html);
217		// Remove spaces that are preceded by either > or <
218	19	$html = preg_replace('/(<) /', '$1', $html);
219		// Remove spaces that are between > and <
220	19	$html = preg_replace('/(>) (<)/', '>$2', $html);
221
222	19	return $html;
223		}
224
225		/**
226		* Sort HTML-Attributes, so that gzip can do better work
227		* and remove some default attributes.
228		*
229		* @param SimpleHtmlDom $element
230		*
231		* @return bool
232		*/
233	19	private function optimizeAttributes(SimpleHtmlDom $element)
234		{
235	19	$attributs = $element->getAllAttributes();
236
237	19	if (!$attributs) {
238	19	return false;
239		}
240
241		/*
		0 ignored issues – show Unused Code Comprehensibility introduced 2016-10-16 04:05 UTC by Report Bug Copy Issue Report `43%` of this comment could be valid code. Did you maybe forget this after debugging? Sometimes obsolete code just ends up commented out instead of removed. In this case it is better to remove the code once you have checked you do not need it. The code might also have been commented out for debugging purposes. In this case it is vital that someone uncomments it again or your project may behave in very unexpected ways in production. This check looks for comments that seem to be mostly valid code and reports them. Loading history...
242		if (
243		($element->tag === 'script' \|\| $element->tag === 'style')
244		&&
245		!isset($attributs['src'])
246		) {
247		// TODO: protect inline css / js
248		}
249		*/
250
251	8	$attrs = array();
252	8	foreach ((array)$attributs as $attrName => $attrValue) {
253
254	8	if (in_array($attrName, self::$booleanAttributes, true)) {
255	6	$attrs[$attrName] = 'delete-this-' . $this->randomHash;
256	6	$element->{$attrName} = null;
257	6	continue;
258		}
259
260		if (
261	8	($attrName === 'href' \|\| $attrName === 'src' \|\| $attrName === 'action')
262	8	&&
263	3	!(isset($attributs['rel']) && $attributs['rel'] === 'external')
264	8	&&
265	3	!(isset($attributs['target']) && $attributs['target'] === '_blank')
266	8	) {
267	3	$attrValue = str_replace('http://', '//', $attrValue);
268	3	}
269
270	8	if ($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...
271	2	$element->{$attrName} = null;
272	2	continue;
273		}
274
275	8	$attrValue = $this->sortCssClasses($attrName, $attrValue);
276
277	8	$attrs[$attrName] = $attrValue;
278	8	$element->{$attrName} = null;
279	8	}
280
281	8	ksort($attrs);
282	8	foreach ($attrs as $attrName => $attrValue) {
283	8	$element->setAttribute($attrName, $attrValue, true);
284	8	}
285
286	8	return true;
287		}
288
289		/**
290		* Check if the attribute (key / value) is default and can be skipped.
291		*
292		* @param string $tag
293		* @param string $attrName
294		* @param string $attrValue
295		* @param string $allAttr
296		*
297		* @return bool
298		*/
299	8	private function optimizeAttributesFilters($tag, $attrName, $attrValue, $allAttr)
300		{
301		// remove default
302	8	if ($tag === 'script' && $attrName === 'language' && $attrValue === 'javascript') {
303		return true;
304		}
305
306		// remove default
307	8	if ($tag === 'form' && $attrName === 'method' && $attrValue === 'get') {
308	1	return true;
309		}
310
311		// remove default
312	8	if ($tag === 'input' && $attrName === 'type' && $attrValue === 'text') {
313	1	return true;
314		}
315
316		// remove default
317	8	if ($tag === 'area' && $attrName === 'shape' && $attrValue === 'rect') {
318		return true;
319		}
320
321		// remove deprecated charset-attribute (the Browser will use the charset from the HTTP-Header, anyway)
322	8	if ($tag === 'script' && $attrName === 'charset' && !isset($allAttr['src'])) {
323		return true;
324		}
325
326		// remove deprecated anchor-jump
327	8	if ($tag === 'a' && $attrName === 'name' && isset($allAttr['id'])) {
328		return true;
329		}
330
331		// remove "type=text/css" for css links
332	8	if ($tag === 'link' && $attrName === 'type' && $attrValue === 'text/css' && isset($allAttr['rel']) && $allAttr['rel'] === 'stylesheet') {
333	1	return true;
334		}
335
336		// remove deprecated script-mime-types
337	8	if ($tag === 'script' && $attrName === 'type' && isset($allAttr['src']) && in_array($attrValue, self::$executableScriptsMimeTypes, true)) {
338	1	return true;
339		}
340
341		// remove empty value from <input>
342	8	if ($tag === 'input' && $attrName === 'value' && $attrValue === '') {
343	1	return true;
344		}
345
346		// remove some empty attribute
347	8	if ($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)) {
348	1	return true;
349		}
350
351	8	return false;
352		}
353
354		/**
355		* @param $attrName
356		* @param $attrValue
357		*
358		* @return string
359		*/
360	8	private function sortCssClasses($attrName, $attrValue)
361		{
362	8	if ($attrName !== 'class' \|\| !$attrValue) {
363	8	return $attrValue;
364		}
365
366	4	$classes = explode(' ', $attrValue);
367	4	if (!$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...
368		return '';
369		}
370
371	4	sort($classes);
372	4	$attrValue = '';
373	4	foreach ($classes as $class) {
374	4	if (!$class) {
375	1	continue;
376		}
377	4	$attrValue .= trim($class) . ' ';
378	4	}
379	4	$attrValue = trim($attrValue);
380
381	4	return $attrValue;
382		}
383		}
384

voku / HtmlMin

Push — master ( c442f2...6b8e64 )

HtmlMin::__construct() A

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like