HtmlConverter::getEnvironment() - Code Metrics - Inspection of "Allow to inject Environment object" - thephpleague/html-to-markdown - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Pull Request — master (#118)

by Filip

created 2016-12-19 14:33 UTC

HtmlConverter::getEnvironment() A

↳ Parent: HtmlConverter

Complexity

Conditions	1
Paths	1

Size

Total Lines	4
Code Lines	2

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	0
CRAP Score	2

Importance

Changes

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

<?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
{
    /**
     * @var Environment
     */
    protected $environment;

    /**
     * Constructor
     *
     * @param array $options Configuration options
     * @param Environment $environment
     */
    public function __construct(array $options = array(), Environment $environment = null)
    {
        $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'      => '**', // Set to '__' if you prefer the underlined style
            'italic_style'    => '_', // Set to '*' if you prefer the asterisk 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`
        );

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

        $this->environment->getConfig()->merge($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 $html
     *
     * @return string The Markdown version of the html
     */
    public function convert($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();

        $markdown = $this->sanitize($markdown);

        return $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));
                }
            }
        }

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

        return $markdown;
    }
}


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
18		{
19		/**
20		* @var Environment
21		*/
22		protected $environment;
23
24		/**
25		* Constructor
26		*
27		* @param array $options Configuration options
28		* @param Environment $environment
29		*/
30	81	public function __construct(array $options = array(), Environment $environment = null)
31		{
32		$defaults = array(
33	81	'header_style' => 'setext', // Set to 'atx' to output H1 and H2 headers as # Header1 and ## Header2
34	54	'suppress_errors' => true, // Set to false to show warnings when loading malformed HTML
35	54	'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.
36	54	'bold_style' => '**', // Set to '__' if you prefer the underlined style
37	54	'italic_style' => '_', // Set to '*' if you prefer the asterisk style
38	54	'remove_nodes' => '', // space-separated list of dom nodes that should be removed. example: 'meta style script'
39	54	'hard_break' => false, // Set to true to turn <br> into `\n` instead of ` \n`
40	54	);
41
42	81	$env = $environment instanceof Environment ? $environment : Environment::createDefaultEnvironment($defaults);
43	81	$this->environment = $env;
44
45	81	$this->environment->getConfig()->merge($options);
46	81	}
47
48		/**
49		* @return Environment
50		*/
51		public function getEnvironment()
52		{
53		return $this->environment;
54		}
55
56		/**
57		* @return Configuration
58		*/
59	78	public function getConfig()
60		{
61	78	return $this->environment->getConfig();
62		}
63
64		/**
65		* Convert
66		*
67		* @see HtmlConverter::convert
68		*
69		* @param string $html
70		*
71		* @return string The Markdown version of the html
72		*/
73	3	public function __invoke($html)
74		{
75	3	return $this->convert($html);
76		}
77
78		/**
79		* Convert
80		*
81		* Loads HTML and passes to getMarkdown()
82		*
83		* @param $html
84		*
85		* @return string The Markdown version of the html
86		*/
87	81	public function convert($html)
88		{
89	81	if (trim($html) === '') {
90	3	return '';
91		}
92
93	78	$document = $this->createDOMDocument($html);
94
95		// Work on the entire DOM tree (including head and body)
96	78	if (!($root = $document->getElementsByTagName('html')->item(0))) {
97		throw new \InvalidArgumentException('Invalid HTML was provided');
98		}
99
100	78	$rootElement = new Element($root);
101	78	$this->convertChildren($rootElement);
102
103		// Store the now-modified DOMDocument as a string
104	78	$markdown = $document->saveHTML();
105
106	78	$markdown = $this->sanitize($markdown);
107
108	78	return $markdown;
109		}
110
111		/**
112		* @param string $html
113		*
114		* @return \DOMDocument
115		*/
116	78	private function createDOMDocument($html)
117		{
118	78	$document = new \DOMDocument();
119
120	78	if ($this->getConfig()->getOption('suppress_errors')) {
121		// Suppress conversion errors (from http://bit.ly/pCCRSX)
122	75	libxml_use_internal_errors(true);
123	50	}
124
125		// Hack to load utf-8 HTML (from http://bit.ly/pVDyCt)
126	78	$document->loadHTML('<?xml encoding="UTF-8">' . $html);
127	78	$document->encoding = 'UTF-8';
128
129	78	if ($this->getConfig()->getOption('suppress_errors')) {
130	75	libxml_clear_errors();
131	50	}
132
133	78	return $document;
134		}
135
136		/**
137		* Convert Children
138		*
139		* Recursive function to drill into the DOM and convert each node into Markdown from the inside out.
140		*
141		* Finds children of each node and convert those to #text nodes containing their Markdown equivalent,
142		* starting with the innermost element and working up to the outermost element.
143		*
144		* @param ElementInterface $element
145		*/
146	78	private function convertChildren(ElementInterface $element)
147		{
148		// Don't convert HTML code inside <code> and <pre> blocks to Markdown - that should stay as HTML
149		// except if the current node is a code tag, which needs to be converted by the CodeConverter.
150	78	if ($element->isDescendantOf(array('pre', 'code')) && $element->getTagName() !== 'code') {
151	15	return;
152		}
153
154		// If the node has children, convert those to Markdown first
155	78	if ($element->hasChildren()) {
156	78	foreach ($element->getChildren() as $child) {
157	78	$this->convertChildren($child);
158	52	}
159	52	}
160
161		// Now that child nodes have been converted, convert the original node
162	78	$markdown = $this->convertToMarkdown($element);
163
164		// Create a DOM text node containing the Markdown equivalent of the original node
165
166		// Replace the old $node e.g. '<h3>Title</h3>' with the new $markdown_node e.g. '### Title'
167	78	$element->setFinalMarkdown($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($element)` on line `162` can also be of type `false`; however, `League\HTMLToMarkdown\El...ace::setFinalMarkdown()` does only seem to accept `string`, did you maybe forget to handle an error condition? This check looks for type mismatches where the missing type is `false`. This is usually indicative of an error condtion. Consider the follow example <?php function getDate($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...
168	78	}
169
170		/**
171		* Convert to Markdown
172		*
173		* Converts an individual node into a #text node containing a string of its Markdown equivalent.
174		*
175		* Example: An <h3> node with text content of 'Title' becomes a text node with content of '### Title'
176		*
177		* @param ElementInterface $element
178		*
179		* @return string The converted HTML as Markdown
180		*/
181	78	protected function convertToMarkdown(ElementInterface $element)
182		{
183	78	$tag = $element->getTagName();
184
185		// Strip nodes named in remove_nodes
186	78	$tags_to_remove = explode(' ', $this->getConfig()->getOption('remove_nodes'));
187	78	if (in_array($tag, $tags_to_remove)) {
188	3	return false;
189		}
190
191	78	$converter = $this->environment->getConverterByTag($tag);
192
193	78	return $converter->convert($element);
194		}
195
196		/**
197		* @param string $markdown
198		*
199		* @return string
200		*/
201	78	protected function sanitize($markdown)
202		{
203	78	$markdown = html_entity_decode($markdown, ENT_QUOTES, 'UTF-8');
204	78	$markdown = preg_replace('/<!DOCTYPE [^>]+>/', '', $markdown); // Strip doctype declaration
205	78	$markdown = trim($markdown); // Remove blank spaces at the beggining of the html
206
207		/*
208		* Removing unwanted tags. Tags should be added to the array in the order they are expected.
209		* XML, html and body opening tags should be in that order. Same case with closing tags
210		*/
211	78	$unwanted = array('<?xml encoding="UTF-8">', '<html>', '</html>', '<body>', '</body>', '<head>', '</head>', ' ');
212
213	78	foreach ($unwanted as $tag) {
214	78	if (strpos($tag, '/') === false) {
215		// Opening tags
216	78	if (strpos($markdown, $tag) === 0) {
217	78	$markdown = substr($markdown, strlen($tag));
218	52	}
219	52	} else {
220		// Closing tags
221	78	if (strpos($markdown, $tag) === strlen($markdown) - strlen($tag)) {
222	72	$markdown = substr($markdown, 0, -strlen($tag));
223	46	}
224		}
225	52	}
226
227	78	$markdown = trim($markdown, "\n\r\0\x0B");
228
229	78	return $markdown;
230		}
231		}
232

thephpleague / html-to-markdown

Pull Request — master (#118)

HtmlConverter::getEnvironment() A

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like