MarkdownTestHelper::normalizeElementContent() - Code Metrics - ylvarw/module - Measure and Improve Code Quality continuously with Scrutinizer

MarkdownTestHelper::normalizeElementContent() F
last analyzed 2021-01-12 17:27 UTC

↳ Parent: MarkdownTestHelper

Complexity

Conditions	55
Paths	8820

Size

Total Lines	116
Code Lines	89

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
cc	55
eloc	89
nc	8820
nop	2
dl	0
loc	116
rs	0
c	0
b	0
f	0

How to fix Long Method Complexity

<?php

use PHPUnit\Framework\TestCase;

class MarkdownTestHelper
{
	/**
	 * Takes an input directory containing .text and .(x)html files, and returns an array
	 * of .text files and the corresponding output xhtml or html file. Can be used in a unit test data provider.
	 *
	 * @param string $directory Input directory
	 *
	 * @return array
	 */
	public static function getInputOutputPaths($directory) {
		$iterator = new RecursiveIteratorIterator(new RecursiveDirectoryIterator($directory));
		$regexIterator = new RegexIterator(
			$iterator,
			'/^.+\.text$/',
			RecursiveRegexIterator::GET_MATCH
		);

		$dataValues = [];

		/** @var SplFileInfo $inputFile */
		foreach ($regexIterator as $inputFiles) {
			foreach ($inputFiles as $inputMarkdownPath) {
				$xhtml = true;
				$expectedHtmlPath = substr($inputMarkdownPath, 0, -4) . 'xhtml';
				if (!file_exists($expectedHtmlPath)) {
					$expectedHtmlPath = substr($inputMarkdownPath, 0, -4) . 'html';
					$xhtml = false;
				}
				$dataValues[] = [$inputMarkdownPath, $expectedHtmlPath, $xhtml];
			}
		}

		return $dataValues;
	}

	/**
	 * Applies PHPUnit's assertSame after normalizing both strings (e.g. ignoring whitespace differences).
	 * Uses logic found originally in MDTest.
	 *
	 * @param string $string1
	 * @param string $string2
	 * @param string $message Positive message to print when test fails (e.g. "String1 matches String2")
	 * @param bool $xhtml
	 */
	public static function assertSameNormalized($string1, $string2, $message, $xhtml = true) {

		$t_result = $string1;
		$t_output = $string2;

		// DOMDocuments
		if ($xhtml) {
			$document = new DOMDocument();
			$doc_result = $document->loadXML('<!DOCTYPE html>' .
				"<html xmlns='http://www.w3.org/1999/xhtml'>" .
				"<body>$t_result</body></html>");

			$document2 = new DOMDocument();
			$doc_output = $document2->loadXML('<!DOCTYPE html>' .
				"<html xmlns='http://www.w3.org/1999/xhtml'>" .
				"<body>$t_output</body></html>");

			if ($doc_result) {
				static::normalizeElementContent($document->documentElement, false);
				$n_result = $document->saveXML();
			} else {
				$n_result = '--- Expected Result: XML Parse Error ---';
			}
			if ($doc_output) {
				static::normalizeElementContent($document2->documentElement, false);
				$n_output = $document2->saveXML();
			} else {
				$n_output = '--- Output: XML Parse Error ---';
			}
		} else {

			// '@' suppressors used because some tests have invalid HTML (multiple elements with the same id attribute)
			// Perhaps isolate to a separate test and remove this?

			$document = new DOMDocument();
			$doc_result = @$document->loadHTML($t_result);

			$document2 = new DOMDocument();
			$doc_output = @$document2->loadHTML($t_output);

			if ($doc_result) {
				static::normalizeElementContent($document->documentElement, false);
				$n_result = $document->saveHTML();
			} else {
				$n_result = '--- Expected Result: HTML Parse Error ---';
			}

			if ($doc_output) {
				static::normalizeElementContent($document2->documentElement, false);
				$n_output = $document2->saveHTML();
			} else {
				$n_output = '--- Output: HTML Parse Error ---';
			}
		}

		$n_result = preg_replace('{^.*?<body>|</body>.*?$}is', '', $n_result);
		$n_output = preg_replace('{^.*?<body>|</body>.*?$}is', '', $n_output);

		$c_result = $n_result;
		$c_output = $n_output;

		$c_result = trim($c_result) . "\n";
		$c_output = trim($c_output) . "\n";

		// This will throw a test exception if the strings don't exactly match
		TestCase::assertSame($c_result, $c_output, $message);
	}

	/**
	 * @param DOMElement $element Modifies this element by reference
	 * @param bool $whitespace_preserve Preserve Whitespace
	 * @return void
	 */
	protected static function normalizeElementContent($element, $whitespace_preserve) {
		#
		# Normalize content of HTML DOM $element. The $whitespace_preserve
		# argument indicates that whitespace is significant and shouldn't be
		# normalized; it should be used for the content of certain elements like
		# <pre> or <script>.
		#
		$node_list = $element->childNodes;
		switch (strtolower($element->nodeName)) {
			case 'body':
			case 'div':
			case 'blockquote':
			case 'ul':
			case 'ol':
			case 'dl':
			case 'h1':
			case 'h2':
			case 'h3':
			case 'h4':
			case 'h5':
			case 'h6':
				$whitespace = "\n\n";
				break;

			case 'table':
				$whitespace = "\n";
				break;

			case 'pre':
			case 'script':
			case 'style':
			case 'title':
				$whitespace_preserve = true;
				$whitespace = "";
				break;

			default:
				$whitespace = "";
				break;
		}
		foreach ($node_list as $node) {
			switch ($node->nodeType) {
				case XML_ELEMENT_NODE:
					static::normalizeElementContent($node, $whitespace_preserve);
					static::normalizeElementAttributes($node);

					switch (strtolower($node->nodeName)) {
						case 'p':
						case 'div':
						case 'hr':
						case 'blockquote':
						case 'ul':
						case 'ol':
						case 'dl':
						case 'li':
						case 'address':
						case 'table':
						case 'dd':
						case 'pre':
						case 'h1':
						case 'h2':
						case 'h3':
						case 'h4':
						case 'h5':
						case 'h6':
							$whitespace = "\n\n";
							break;

						case 'tr':
						case 'td':
						case 'dt':
							$whitespace = "\n";
							break;

						default:
							$whitespace = "";
							break;
					}

					if (($whitespace === "\n\n" || $whitespace === "\n") &&
						$node->nextSibling &&
						$node->nextSibling->nodeType != XML_TEXT_NODE) {
						$element->insertBefore(new DOMText($whitespace), $node->nextSibling);
					}
					break;

				case XML_TEXT_NODE:
					if (!$whitespace_preserve) {
						if (trim($node->data) === "") {
							$node->data = $whitespace;
						}
						else {
							$node->data = preg_replace('{\s+}', ' ', $node->data);
						}
					}
					break;
			}
		}
		if (!$whitespace_preserve &&
			($whitespace === "\n\n" || $whitespace === "\n")) {
			if ($element->firstChild) {
				if ($element->firstChild->nodeType == XML_TEXT_NODE) {
					$element->firstChild->data =

						preg_replace('{^\s+}', "\n", $element->firstChild->data);
				}
				else {
					$element->insertBefore(new DOMText("\n"), $element->firstChild);
				}
			}
			if ($element->lastChild) {
				if ($element->lastChild->nodeType == XML_TEXT_NODE) {
					$element->lastChild->data =
						preg_replace('{\s+$}', "\n", $element->lastChild->data);
				}
				else {
					$element->insertBefore(new DOMText("\n"), null);
				}
			}
		}
	}

	/**
	 * @param DOMElement $element Modifies this element by reference
	 */
	protected static function normalizeElementAttributes (DOMElement $element)
	{
		#
		# Sort attributes by name.
		#
		// Gather the list of attributes as an array.
		$attr_list = array();
		foreach ($element->attributes as $attr_node) {
			$attr_list[$attr_node->name] = $attr_node;
		}

		// Sort attribute list by name.
		ksort($attr_list);

		// Remove then put back each attribute following sort order.
		foreach ($attr_list as $attr_node) {
			$element->removeAttributeNode($attr_node);
			$element->setAttributeNode($attr_node);
		}
	}
}


1			<?php
2
3			use PHPUnit\Framework\TestCase;
4
5			class MarkdownTestHelper
6			{
7			/**
8			* Takes an input directory containing .text and .(x)html files, and returns an array
9			* of .text files and the corresponding output xhtml or html file. Can be used in a unit test data provider.
10			*
11			* @param string $directory Input directory
12			*
13			* @return array
14			*/
15			public static function getInputOutputPaths($directory) {
16			$iterator = new RecursiveIteratorIterator(new RecursiveDirectoryIterator($directory));
17			$regexIterator = new RegexIterator(
18			$iterator,
19			'/^.+\.text$/',
20			RecursiveRegexIterator::GET_MATCH
21			);
22
23			$dataValues = [];
24
25			/** @var SplFileInfo $inputFile */
26			foreach ($regexIterator as $inputFiles) {
27			foreach ($inputFiles as $inputMarkdownPath) {
28			$xhtml = true;
29			$expectedHtmlPath = substr($inputMarkdownPath, 0, -4) . 'xhtml';
30			if (!file_exists($expectedHtmlPath)) {
31			$expectedHtmlPath = substr($inputMarkdownPath, 0, -4) . 'html';
32			$xhtml = false;
33			}
34			$dataValues[] = [$inputMarkdownPath, $expectedHtmlPath, $xhtml];
35			}
36			}
37
38			return $dataValues;
39			}
40
41			/**
42			* Applies PHPUnit's assertSame after normalizing both strings (e.g. ignoring whitespace differences).
43			* Uses logic found originally in MDTest.
44			*
45			* @param string $string1
46			* @param string $string2
47			* @param string $message Positive message to print when test fails (e.g. "String1 matches String2")
48			* @param bool $xhtml
49			*/
50			public static function assertSameNormalized($string1, $string2, $message, $xhtml = true) {
51
52			$t_result = $string1;
53			$t_output = $string2;
54
55			// DOMDocuments
56			if ($xhtml) {
57			$document = new DOMDocument();
58			$doc_result = $document->loadXML('<!DOCTYPE html>' .
59			"<html xmlns='http://www.w3.org/1999/xhtml'>" .
60			"<body>$t_result</body></html>");
61
62			$document2 = new DOMDocument();
63			$doc_output = $document2->loadXML('<!DOCTYPE html>' .
64			"<html xmlns='http://www.w3.org/1999/xhtml'>" .
65			"<body>$t_output</body></html>");
66
67			if ($doc_result) {
68			static::normalizeElementContent($document->documentElement, false);
69			$n_result = $document->saveXML();
70			} else {
71			$n_result = '--- Expected Result: XML Parse Error ---';
72			}
73			if ($doc_output) {
74			static::normalizeElementContent($document2->documentElement, false);
75			$n_output = $document2->saveXML();
76			} else {
77			$n_output = '--- Output: XML Parse Error ---';
78			}
79			} else {
80
81			// '@' suppressors used because some tests have invalid HTML (multiple elements with the same id attribute)
82			// Perhaps isolate to a separate test and remove this?
83
84			$document = new DOMDocument();
85			$doc_result = @$document->loadHTML($t_result);
86
87			$document2 = new DOMDocument();
88			$doc_output = @$document2->loadHTML($t_output);
89
90			if ($doc_result) {
91			static::normalizeElementContent($document->documentElement, false);
92			$n_result = $document->saveHTML();
93			} else {
94			$n_result = '--- Expected Result: HTML Parse Error ---';
95			}
96
97			if ($doc_output) {
98			static::normalizeElementContent($document2->documentElement, false);
99			$n_output = $document2->saveHTML();
100			} else {
101			$n_output = '--- Output: HTML Parse Error ---';
102			}
103			}
104
105			$n_result = preg_replace('{^.?<body>\|</body>.?$}is', '', $n_result);
106			$n_output = preg_replace('{^.?<body>\|</body>.?$}is', '', $n_output);
107
108			$c_result = $n_result;
109			$c_output = $n_output;
110
111			$c_result = trim($c_result) . "\n";
112			$c_output = trim($c_output) . "\n";
113
114			// This will throw a test exception if the strings don't exactly match
115			TestCase::assertSame($c_result, $c_output, $message);
116			}
117
118			/**
119			* @param DOMElement $element Modifies this element by reference
120			* @param bool $whitespace_preserve Preserve Whitespace
121			* @return void
122			*/
123			protected static function normalizeElementContent($element, $whitespace_preserve) {
124			#
125			# Normalize content of HTML DOM $element. The $whitespace_preserve
126			# argument indicates that whitespace is significant and shouldn't be
127			# normalized; it should be used for the content of certain elements like
128			# <pre> or <script>.
129			#
130			$node_list = $element->childNodes;
131			switch (strtolower($element->nodeName)) {
132			case 'body':
133			case 'div':
134			case 'blockquote':
135			case 'ul':
136			case 'ol':
137			case 'dl':
138			case 'h1':
139			case 'h2':
140			case 'h3':
141			case 'h4':
142			case 'h5':
143			case 'h6':
144			$whitespace = "\n\n";
145			break;
146
147			case 'table':
148			$whitespace = "\n";
149			break;
150
151			case 'pre':
152			case 'script':
153			case 'style':
154			case 'title':
155			$whitespace_preserve = true;
156			$whitespace = "";
157			break;
158
159			default:
160			$whitespace = "";
161			break;
162			}
163			foreach ($node_list as $node) {
164			switch ($node->nodeType) {
165			case XML_ELEMENT_NODE:
166			static::normalizeElementContent($node, $whitespace_preserve);
167			static::normalizeElementAttributes($node);
168
169			switch (strtolower($node->nodeName)) {
170			case 'p':
171			case 'div':
172			case 'hr':
173			case 'blockquote':
174			case 'ul':
175			case 'ol':
176			case 'dl':
177			case 'li':
178			case 'address':
179			case 'table':
180			case 'dd':
181			case 'pre':
182			case 'h1':
183			case 'h2':
184			case 'h3':
185			case 'h4':
186			case 'h5':
187			case 'h6':
188			$whitespace = "\n\n";
189			break;
190
191			case 'tr':
192			case 'td':
193			case 'dt':
194			$whitespace = "\n";
195			break;
196
197			default:
198			$whitespace = "";
199			break;
200			}
201
202			if (($whitespace === "\n\n" \|\| $whitespace === "\n") &&
203			$node->nextSibling &&
204			$node->nextSibling->nodeType != XML_TEXT_NODE) {
205			$element->insertBefore(new DOMText($whitespace), $node->nextSibling);
206			}
207			break;
208
209			case XML_TEXT_NODE:
210			if (!$whitespace_preserve) {
211			if (trim($node->data) === "") {
212			$node->data = $whitespace;
213			}
214			else {
215			$node->data = preg_replace('{\s+}', ' ', $node->data);
216			}
217			}
218			break;
219			}
220			}
221			if (!$whitespace_preserve &&
222			($whitespace === "\n\n" \|\| $whitespace === "\n")) {
223			if ($element->firstChild) {
224			if ($element->firstChild->nodeType == XML_TEXT_NODE) {
225			$element->firstChild->data =
			0 ignored issues – show Bug introduced 2021-01-12 16:59 UTC by Report Bug Copy Issue Report The property `data` does not seem to exist on `DOMElement`. Loading history...
226			preg_replace('{^\s+}', "\n", $element->firstChild->data);
227			}
228			else {
229			$element->insertBefore(new DOMText("\n"), $element->firstChild);
230			}
231			}
232			if ($element->lastChild) {
233			if ($element->lastChild->nodeType == XML_TEXT_NODE) {
234			$element->lastChild->data =
235			preg_replace('{\s+$}', "\n", $element->lastChild->data);
236			}
237			else {
238			$element->insertBefore(new DOMText("\n"), null);
239			}
240			}
241			}
242			}
243
244			/**
245			* @param DOMElement $element Modifies this element by reference
246			*/
247			protected static function normalizeElementAttributes (DOMElement $element)
248			{
249			#
250			# Sort attributes by name.
251			#
252			// Gather the list of attributes as an array.
253			$attr_list = array();
254			foreach ($element->attributes as $attr_node) {
255			$attr_list[$attr_node->name] = $attr_node;
256			}
257
258			// Sort attribute list by name.
259			ksort($attr_list);
260
261			// Remove then put back each attribute following sort order.
262			foreach ($attr_list as $attr_node) {
263			$element->removeAttributeNode($attr_node);
264			$element->setAttributeNode($attr_node);
265			}
266			}
267			}
268

ylvarw / module

MarkdownTestHelper::normalizeElementContent() F last analyzed 2021-01-12 17:27 UTC

Complexity

Size

Duplication

Importance

How to fix Long Method Complexity

Long Method

Duplication Side-by-Side

Filter issues like

MarkdownTestHelper::normalizeElementContent() F
last analyzed 2021-01-12 17:27 UTC