HtmlConverter::convert() - Code Metrics - Inspection of "Make autolinking configurable (#188)" - thephpleague/html-to-markdown - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( a1ddbf...2742fe )

by Colin

created 2020-07-01 00:32 UTC

HtmlConverter::convert() A

↳ Parent: HtmlConverter

Complexity

Conditions	3
Paths	3

Size

Total Lines

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	9
CRAP Score	3.009

Importance

Changes

Metric	Value
dl	0
loc	21
ccs	9
cts	10
cp	0.9
rs	9.584
c	0
b	0
f	0
cc	3
nc	3
nop	1
crap	3.009

<?php

namespace League\HTMLToMarkdown;

/**
 * Class HtmlConverter
 *
 * A helper class to convert HTML to Markdown.
 *
 * @author Colin O'Dell <[email protected]>
 * @author Nick Cernis <[email protected]>
 *
 * @link https://github.com/thephpleague/html-to-markdown/ Latest version on GitHub.
 *
 * @license http://www.opensource.org/licenses/mit-license.php MIT
 */
class HtmlConverter implements HtmlConverterInterface
{
    /**
     * @var Environment
     */
    protected $environment;

    /**
     * Constructor
     *
     * @param Environment|array $options Environment object or configuration options
     */
    public function __construct($options = array())
    {
        if ($options instanceof Environment) {
            $this->environment = $options;
        } elseif (is_array($options)) {
            $defaults = array(
                'header_style' => 'setext', // Set to 'atx' to output H1 and H2 headers as # Header1 and ## Header2
                'suppress_errors' => true, // Set to false to show warnings when loading malformed HTML
                'strip_tags' => false, // Set to true to strip tags that don't have markdown equivalents. N.B. Strips tags, not their content. Useful to clean MS Word HTML output.
                'bold_style' => '**', // DEPRECATED: Set to '__' if you prefer the underlined style
                'italic_style' => '*', // DEPRECATED: Set to '_' if you prefer the underlined style
                'remove_nodes' => '', // space-separated list of dom nodes that should be removed. example: 'meta style script'
                'hard_break' => false, // Set to true to turn <br> into `\n` instead of `  \n`
                'list_item_style' => '-', // Set the default character for each <li> in a <ul>. Can be '-', '*', or '+'
                'preserve_comments' => false, // Set to true to preserve comments, or set to an array of strings to preserve specific comments
                'use_autolinks' => true, // Set to true to use simple link syntax if possible. Will always use []() if set to false
            );

            $this->environment = Environment::createDefaultEnvironment($defaults);

            $this->environment->getConfig()->merge($options);
        }
    }

    /**
     * @return Environment
     */
    public function getEnvironment()
    {
        return $this->environment;
    }

    /**
     * @return Configuration
     */
    public function getConfig()
    {
        return $this->environment->getConfig();
    }

    /**
     * Convert
     *
     * @see HtmlConverter::convert
     *
     * @param string $html
     *
     * @return string The Markdown version of the html
     */
    public function __invoke($html)
    {
        return $this->convert($html);
    }

    /**
     * Convert
     *
     * Loads HTML and passes to getMarkdown()
     *
     * @param string $html
     *
     * @throws \InvalidArgumentException
     *
     * @return string The Markdown version of the html
     */
    public function convert($html)
    {
        if (trim($html) === '') {
            return '';
        }

        $document = $this->createDOMDocument($html);

        // Work on the entire DOM tree (including head and body)
        if (!($root = $document->getElementsByTagName('html')->item(0))) {
            throw new \InvalidArgumentException('Invalid HTML was provided');
        }

        $rootElement = new Element($root);
        $this->convertChildren($rootElement);

        // Store the now-modified DOMDocument as a string
        $markdown = $document->saveHTML();

        return $this->sanitize($markdown);
    }

    /**
     * @param string $html
     *
     * @return \DOMDocument
     */
    private function createDOMDocument($html)
    {
        $document = new \DOMDocument();

        if ($this->getConfig()->getOption('suppress_errors')) {
            // Suppress conversion errors (from http://bit.ly/pCCRSX)
            libxml_use_internal_errors(true);
        }

        // Hack to load utf-8 HTML (from http://bit.ly/pVDyCt)
        $document->loadHTML('<?xml encoding="UTF-8">' . $html);
        $document->encoding = 'UTF-8';

        if ($this->getConfig()->getOption('suppress_errors')) {
            libxml_clear_errors();
        }

        return $document;
    }

    /**
     * Convert Children
     *
     * Recursive function to drill into the DOM and convert each node into Markdown from the inside out.
     *
     * Finds children of each node and convert those to #text nodes containing their Markdown equivalent,
     * starting with the innermost element and working up to the outermost element.
     *
     * @param ElementInterface $element
     */
    private function convertChildren(ElementInterface $element)
    {
        // Don't convert HTML code inside <code> and <pre> blocks to Markdown - that should stay as HTML
        // except if the current node is a code tag, which needs to be converted by the CodeConverter.
        if ($element->isDescendantOf(array('pre', 'code')) && $element->getTagName() !== 'code') {
            return;
        }

        // If the node has children, convert those to Markdown first
        if ($element->hasChildren()) {
            foreach ($element->getChildren() as $child) {
                $this->convertChildren($child);
            }
        }

        // Now that child nodes have been converted, convert the original node
        $markdown = $this->convertToMarkdown($element);

        // Create a DOM text node containing the Markdown equivalent of the original node

        // Replace the old $node e.g. '<h3>Title</h3>' with the new $markdown_node e.g. '### Title'
        $element->setFinalMarkdown($markdown);
<?php

function getDate($date)
{
    if ($date !== null) {
        return new DateTime($date);
    }

    return false;
}
    }

    /**
     * Convert to Markdown
     *
     * Converts an individual node into a #text node containing a string of its Markdown equivalent.
     *
     * Example: An <h3> node with text content of 'Title' becomes a text node with content of '### Title'
     *
     * @param ElementInterface $element
     *
     * @return string The converted HTML as Markdown
     */
    protected function convertToMarkdown(ElementInterface $element)
    {
        $tag = $element->getTagName();

        // Strip nodes named in remove_nodes
        $tags_to_remove = explode(' ', $this->getConfig()->getOption('remove_nodes'));
        if (in_array($tag, $tags_to_remove)) {
            return false;
        }

        $converter = $this->environment->getConverterByTag($tag);

        return $converter->convert($element);
    }

    /**
     * @param string $markdown
     *
     * @return string
     */
    protected function sanitize($markdown)
    {
        $markdown = html_entity_decode($markdown, ENT_QUOTES, 'UTF-8');
        $markdown = preg_replace('/<!DOCTYPE [^>]+>/', '', $markdown); // Strip doctype declaration
        $markdown = trim($markdown); // Remove blank spaces at the beggining of the html

        /*
         * Removing unwanted tags. Tags should be added to the array in the order they are expected.
         * XML, html and body opening tags should be in that order. Same case with closing tags
         */
        $unwanted = array('<?xml encoding="UTF-8">', '<html>', '</html>', '<body>', '</body>', '<head>', '</head>', '&#xD;');

        foreach ($unwanted as $tag) {
            if (strpos($tag, '/') === false) {
                // Opening tags
                if (strpos($markdown, $tag) === 0) {
                    $markdown = substr($markdown, strlen($tag));
                }
            } else {
                // Closing tags
                if (strpos($markdown, $tag) === strlen($markdown) - strlen($tag)) {
                    $markdown = substr($markdown, 0, -strlen($tag));
                }
            }
        }

        return trim($markdown, "\n\r\0\x0B");
    }

    /**
     * Pass a series of key-value pairs in an array; these will be passed
     * through the config and set.
     * The advantage of this is that it can allow for static use (IE in Laravel).
     * An example being:
     *
     * HtmlConverter::setOptions(['strip_tags' => true])->convert('<h1>test</h1>');
     */
    public function setOptions(array $options)
    {
        $config = $this->getConfig();

        foreach ($options as $key => $option) {
            $config->setOption($key, $option);
        }

        return $this;
    }
}


1		<?php
2
3		namespace League\HTMLToMarkdown;
4
5		/**
6		* Class HtmlConverter
7		*
8		* A helper class to convert HTML to Markdown.
9		*
10		* @author Colin O'Dell <[email protected]>
11		* @author Nick Cernis <[email protected]>
12		*
13		* @link https://github.com/thephpleague/html-to-markdown/ Latest version on GitHub.
14		*
15		* @license http://www.opensource.org/licenses/mit-license.php MIT
16		*/
17		class HtmlConverter implements HtmlConverterInterface
18		{
19		/**
20		* @var Environment
21		*/
22		protected $environment;
23
24		/**
25		* Constructor
26		*
27		* @param Environment\|array $options Environment object or configuration options
28		*/
29	90	public function __construct($options = array())
30		{
31	90	if ($options instanceof Environment) {
32	3	$this->environment = $options;
33	89	} elseif (is_array($options)) {
34		$defaults = array(
35	87	'header_style' => 'setext', // Set to 'atx' to output H1 and H2 headers as # Header1 and ## Header2
36	58	'suppress_errors' => true, // Set to false to show warnings when loading malformed HTML
37	58	'strip_tags' => false, // Set to true to strip tags that don't have markdown equivalents. N.B. Strips tags, not their content. Useful to clean MS Word HTML output.
38	58	'bold_style' => '**', // DEPRECATED: Set to '__' if you prefer the underlined style
39	58	'italic_style' => '*', // DEPRECATED: Set to '_' if you prefer the underlined style
40	58	'remove_nodes' => '', // space-separated list of dom nodes that should be removed. example: 'meta style script'
41	58	'hard_break' => false, // Set to true to turn <br> into `\n` instead of ` \n`
42	58	'list_item_style' => '-', // Set the default character for each <li> in a <ul>. Can be '-', '*', or '+'
43	58	'preserve_comments' => false, // Set to true to preserve comments, or set to an array of strings to preserve specific comments
44	58	'use_autolinks' => true, // Set to true to use simple link syntax if possible. Will always use []() if set to false
45	58	);
46
47	87	$this->environment = Environment::createDefaultEnvironment($defaults);
48
49	87	$this->environment->getConfig()->merge($options);
50	58	}
51	90	}
52
53		/**
54		* @return Environment
55		*/
56		public function getEnvironment()
57		{
58		return $this->environment;
59		}
60
61		/**
62		* @return Configuration
63		*/
64	87	public function getConfig()
65		{
66	87	return $this->environment->getConfig();
67		}
68
69		/**
70		* Convert
71		*
72		* @see HtmlConverter::convert
73		*
74		* @param string $html
75		*
76		* @return string The Markdown version of the html
77		*/
78	3	public function __invoke($html)
79		{
80	3	return $this->convert($html);
81		}
82
83		/**
84		* Convert
85		*
86		* Loads HTML and passes to getMarkdown()
87		*
88		* @param string $html
89		*
90		* @throws \InvalidArgumentException
91		*
92		* @return string The Markdown version of the html
93		*/
94	90	public function convert($html)
95		{
96	90	if (trim($html) === '') {
97	3	return '';
98		}
99
100	87	$document = $this->createDOMDocument($html);
101
102		// Work on the entire DOM tree (including head and body)
103	87	if (!($root = $document->getElementsByTagName('html')->item(0))) {
104		throw new \InvalidArgumentException('Invalid HTML was provided');
105		}
106
107	87	$rootElement = new Element($root);
108	87	$this->convertChildren($rootElement);
109
110		// Store the now-modified DOMDocument as a string
111	87	$markdown = $document->saveHTML();
112
113	87	return $this->sanitize($markdown);
114		}
115
116		/**
117		* @param string $html
118		*
119		* @return \DOMDocument
120		*/
121	87	private function createDOMDocument($html)
122		{
123	87	$document = new \DOMDocument();
124
125	87	if ($this->getConfig()->getOption('suppress_errors')) {
126		// Suppress conversion errors (from http://bit.ly/pCCRSX)
127	84	libxml_use_internal_errors(true);
128	56	}
129
130		// Hack to load utf-8 HTML (from http://bit.ly/pVDyCt)
131	87	$document->loadHTML('<?xml encoding="UTF-8">' . $html);
132	87	$document->encoding = 'UTF-8';
133
134	87	if ($this->getConfig()->getOption('suppress_errors')) {
135	84	libxml_clear_errors();
136	56	}
137
138	87	return $document;
139		}
140
141		/**
142		* Convert Children
143		*
144		* Recursive function to drill into the DOM and convert each node into Markdown from the inside out.
145		*
146		* Finds children of each node and convert those to #text nodes containing their Markdown equivalent,
147		* starting with the innermost element and working up to the outermost element.
148		*
149		* @param ElementInterface $element
150		*/
151	87	private function convertChildren(ElementInterface $element)
152		{
153		// Don't convert HTML code inside <code> and <pre> blocks to Markdown - that should stay as HTML
154		// except if the current node is a code tag, which needs to be converted by the CodeConverter.
155	87	if ($element->isDescendantOf(array('pre', 'code')) && $element->getTagName() !== 'code') {
156	18	return;
157		}
158
159		// If the node has children, convert those to Markdown first
160	87	if ($element->hasChildren()) {
161	87	foreach ($element->getChildren() as $child) {
162	87	$this->convertChildren($child);
163	58	}
164	58	}
165
166		// Now that child nodes have been converted, convert the original node
167	87	$markdown = $this->convertToMarkdown($element);
168
169		// Create a DOM text node containing the Markdown equivalent of the original node
170
171		// Replace the old $node e.g. '<h3>Title</h3>' with the new $markdown_node e.g. '### Title'
172	87	$element->setFinalMarkdown($markdown);
		0 ignored issues – show Security Bug introduced 2016-02-10 20:51 UTC by Report Bug Copy Issue Report It seems like `$markdown` defined by `$this->convertToMarkdown($element)` on line `167` can also be of type `false`; however, `League\HTMLToMarkdown\El...ace::setFinalMarkdown()` does only seem to accept `string`, did you maybe forget to handle an error condition? This check looks for type mismatches where the missing type is `false`. This is usually indicative of an error condtion. Consider the follow example <?php function getDate($date) { if ($date !== null) { return new DateTime($date); } return false; } This function either returns a new `DateTime` object or false, if there was an error. This is a typical pattern in PHP programming to show that an error has occurred without raising an exception. The calling code should check for this returned `false` before passing on the value to another function or method that may not be able to handle a `false`. Loading history...
173	87	}
174
175		/**
176		* Convert to Markdown
177		*
178		* Converts an individual node into a #text node containing a string of its Markdown equivalent.
179		*
180		* Example: An <h3> node with text content of 'Title' becomes a text node with content of '### Title'
181		*
182		* @param ElementInterface $element
183		*
184		* @return string The converted HTML as Markdown
185		*/
186	87	protected function convertToMarkdown(ElementInterface $element)
187		{
188	87	$tag = $element->getTagName();
189
190		// Strip nodes named in remove_nodes
191	87	$tags_to_remove = explode(' ', $this->getConfig()->getOption('remove_nodes'));
192	87	if (in_array($tag, $tags_to_remove)) {
193	3	return false;
194		}
195
196	87	$converter = $this->environment->getConverterByTag($tag);
197
198	87	return $converter->convert($element);
199		}
200
201		/**
202		* @param string $markdown
203		*
204		* @return string
205		*/
206	87	protected function sanitize($markdown)
207		{
208	87	$markdown = html_entity_decode($markdown, ENT_QUOTES, 'UTF-8');
209	87	$markdown = preg_replace('/<!DOCTYPE [^>]+>/', '', $markdown); // Strip doctype declaration
210	87	$markdown = trim($markdown); // Remove blank spaces at the beggining of the html
211
212		/*
213		* Removing unwanted tags. Tags should be added to the array in the order they are expected.
214		* XML, html and body opening tags should be in that order. Same case with closing tags
215		*/
216	87	$unwanted = array('<?xml encoding="UTF-8">', '<html>', '</html>', '<body>', '</body>', '<head>', '</head>', ' ');
217
218	87	foreach ($unwanted as $tag) {
219	87	if (strpos($tag, '/') === false) {
220		// Opening tags
221	87	if (strpos($markdown, $tag) === 0) {
222	87	$markdown = substr($markdown, strlen($tag));
223	58	}
224	58	} else {
225		// Closing tags
226	87	if (strpos($markdown, $tag) === strlen($markdown) - strlen($tag)) {
227	81	$markdown = substr($markdown, 0, -strlen($tag));
228	52	}
229		}
230	58	}
231
232	87	return trim($markdown, "\n\r\0\x0B");
233		}
234
235		/**
236		* Pass a series of key-value pairs in an array; these will be passed
237		* through the config and set.
238		* The advantage of this is that it can allow for static use (IE in Laravel).
239		* An example being:
240		*
241		* HtmlConverter::setOptions(['strip_tags' => true])->convert('<h1>test</h1>');
242		*/
243		public function setOptions(array $options)
244		{
245		$config = $this->getConfig();
246
247		foreach ($options as $key => $option) {
248		$config->setOption($key, $option);
249		}
250
251		return $this;
252		}
253		}
254

thephpleague / html-to-markdown

Push — master ( a1ddbf...2742fe )

HtmlConverter::convert() A

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like