Parser::scrapeUrl() - Code Metrics - Inspection of "MediaEmbed: removed support for custom schemes" - s9e/TextFormatter - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( 06920f...7e343b )

by Josh

created 2017-12-16 13:36 UTC

Parser::scrapeUrl() B

↳ Parent: Parser

Complexity

Conditions	6
Paths	5

Size

Total Lines	19
Code Lines	7

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	8
CRAP Score	6

Importance

Changes

Metric	Value
dl	0
loc	19
ccs	8
cts	8
cp	1
rs	8.8571
c	0
b	0
f	0
cc	6
eloc	7
nc	5
nop	4
crap	6

<?php

/**
* @package   s9e\TextFormatter
* @copyright Copyright (c) 2010-2017 The s9e Authors
* @license   http://www.opensource.org/licenses/mit-license.php The MIT License
*/
namespace s9e\TextFormatter\Plugins\MediaEmbed;

use s9e\TextFormatter\Parser as TagStack;
use s9e\TextFormatter\Parser\Tag;
use s9e\TextFormatter\Plugins\ParserBase;
use s9e\TextFormatter\Utils\Http;

class Parser extends ParserBase
{
	/**
	* @var \s9e\TextFormatter\Utils\Http\Client Client used to perform HTTP request
	*/
	protected static $client;

	/**
	* {@inheritdoc}
	*/
	public function parse($text, array $matches)
	{
		foreach ($matches as $m)
		{
			$url = $m[0][0];
			$pos = $m[0][1];
			$len = strlen($url);

			// Give that tag priority over other tags such as Autolink's
			$tag = $this->parser->addSelfClosingTag($this->config['tagName'], $pos, $len, -10);
			$tag->setAttribute('url', $url);
		}
	}

	/**
	* Filter a MEDIA tag
	*
	* This will always invalidate the original tag, and possibly replace it with the tag that
	* corresponds to the media site
	*
	* @param  Tag      $tag      The original tag
	* @param  TagStack $tagStack Parser instance, so that we can add the new tag to the stack
	* @param  array    $sites    Map of [host => siteId]
	* @return bool               Unconditionally FALSE
	*/
	public static function filterTag(Tag $tag, TagStack $tagStack, array $sites)
	{
		if ($tag->hasAttribute('site'))
		{
			self::addTagFromMediaId($tag, $tagStack, $sites);
		}
		elseif ($tag->hasAttribute('url'))
		{
			self::addTagFromMediaUrl($tag, $tagStack, $sites);
		}

		return false;
	}

	/**
	* Test whether a given tag has at least one non-default attribute
	*
	* @param  Tag  $tag The original tag
	* @return bool      Whether the tag contains an attribute not named "url"
	*/
	public static function hasNonDefaultAttribute(Tag $tag)
	{
		foreach ($tag->getAttributes() as $attrName => $void)
		{
			if ($attrName !== 'url')
			{
				return true;
			}
		}

		return false;
	}

	/**
	* Scrape the content of an URL to extract some data
	*
	* @param  Tag    $tag          Source tag
	* @param  array  $scrapeConfig Array of scrape directives
	* @param  string $cacheDir     Path to the cache directory
	* @return bool                 Unconditionally TRUE
	*/
	public static function scrape(Tag $tag, array $scrapeConfig, $cacheDir = null)
	{
		if ($tag->hasAttribute('url'))
		{
			// Ensure that the URL actually looks like a URL if we want to use it to scrape
			$url = $tag->getAttribute('url');
			if (preg_match('#^https?://[^<>"\'\\s]+$#Di', $url))
			{
				$url = strtolower(substr($url, 0, 5)) . substr($url, 5);
				foreach ($scrapeConfig as $scrape)
				{
					self::scrapeEntry($url, $tag, $scrape, $cacheDir);
				}
			}
		}

		return true;
	}

	//==============================================================================================
	// Internals
	//==============================================================================================

	/**
	* Add a site tag
	*
	* @param  Tag      $tag      The original tag
	* @param  TagStack $tagStack Parser instance, so that we can add the new tag to the stack
	* @param  string   $siteId   Site ID
	* @return void
	*/
	protected static function addSiteTag(Tag $tag, TagStack $tagStack, $siteId)
	{
		$endTag = $tag->getEndTag();
		if ($endTag)
		{
			$startPos = $tag->getPos();
			$startLen = $tag->getLen();
			$endPos   = $endTag->getPos();
			$endLen   = $endTag->getLen();
		}
		else
		{
			$startPos = $tag->getPos();
			$startLen = 0;
			$endPos   = $tag->getPos() + $tag->getLen();
			$endLen   = 0;
		}

		// Create a new tag and copy this tag's attributes and priority
		$tagStack->addTagPair(strtoupper($siteId), $startPos, $startLen, $endPos, $endLen, $tag->getSortPriority())->setAttributes($tag->getAttributes());
	}

	/**
	* Add a media site tag based on the attributes of a MEDIA tag
	*
	* @param  Tag      $tag      The original tag
	* @param  TagStack $tagStack Parser instance
	* @param  array    $sites    Map of [host => siteId]
	* @return void
	*/
	protected static function addTagFromMediaId(Tag $tag, TagStack $tagStack, array $sites)
	{
		$siteId = strtolower($tag->getAttribute('site'));
		if (in_array($siteId, $sites, true))
		{
			self::addSiteTag($tag, $tagStack, $siteId);
		}
	}

	/**
	* Add a media site tag based on the url attribute of a MEDIA tag
	*
	* @param  Tag      $tag      The original tag
	* @param  TagStack $tagStack Parser instance
	* @param  array    $sites    Map of [host => siteId]
	* @return void
	*/
	protected static function addTagFromMediaUrl(Tag $tag, TagStack $tagStack, array $sites)
	{
		// Capture the scheme and (if applicable) host of the URL
		$p = parse_url($tag->getAttribute('url'));
		if (isset($p['scheme']) && isset($sites[$p['scheme'] . ':']))
		{
			$siteId = $sites[$p['scheme'] . ':'];
		}
		elseif (isset($p['host']))
		{
			$siteId = self::findSiteIdByHost($p['host'], $sites);
		}

		if (!empty($siteId))
		{
			self::addSiteTag($tag, $tagStack, $siteId);
		}
	}

	/**
	* Match a given host to a site ID
	*
	* @param  string      $host  Host
	* @param  array       $sites Map of [host => siteId]
	* @return string|bool        Site ID or FALSE
	*/
	protected static function findSiteIdByHost($host, array $sites)
	{
		// Start with the full host then pop domain labels off the start until we get a match
		do
		{
			if (isset($sites[$host]))
			{
				return $sites[$host];
			}

			$pos = strpos($host, '.');
			if ($pos === false)
			{
				break;
			}

			$host = substr($host, 1 + $pos);
		}
		while ($host > '');

		return false;
	}

	/**
	* Return a cached instance of the HTTP client
	*
	* @return \s9e\TextFormatter\Utils\Http\Client
	*/
	protected static function getHttpClient()
	{
		if (!isset(self::$client))
		{
			self::$client = Http::getClient();
		}
		self::$client->timeout = 10;

		return self::$client;
	}

	/**
	* Replace {@var} tokens in given URL
	*
	* @param  string   $url  Original URL
	* @param  string[] $vars Replacements
	* @return string         Modified URL
	*/
	protected static function replaceTokens($url, array $vars)
	{
		return preg_replace_callback(
			'#\\{@(\\w+)\\}#',
			function ($m) use ($vars)
			{
				return (isset($vars[$m[1]])) ? $vars[$m[1]] : '';
			},
			$url
		);
	}

	/**
	* Scrape the content of an URL to extract some data
	*
	* @param  string $url      Original URL
	* @param  Tag    $tag      Source tag
	* @param  array  $scrape   Array of scrape directives
	* @param  string $cacheDir Path to the cache directory
	* @return void
	*/
	protected static function scrapeEntry($url, Tag $tag, array $scrape, $cacheDir)
	{
		list($matchRegexps, $extractRegexps, $attrNames) = $scrape;
		if (!self::tagIsMissingAnyAttribute($tag, $attrNames))
		{
			return;
		}

		// Test whether this URL matches any regexp
		$vars    = [];
		$matched = false;
		foreach ((array) $matchRegexps as $matchRegexp)
		{
			if (preg_match($matchRegexp, $url, $m))
			{
				$vars   += $m;
				$matched = true;
			}
		}
		if (!$matched)
		{
			return;
		}

		// Add the tag's attributes to the named captures from the "match" regexp
		$vars += $tag->getAttributes();

		$scrapeUrl = (isset($scrape[3])) ? self::replaceTokens($scrape[3], $vars) : $url;
		self::scrapeUrl($scrapeUrl, $tag, (array) $extractRegexps, $cacheDir);
	}

	/**
	* Scrape a URL to help fill a tag's attributes
	*
	* @param  string      $url      URL to scrape
	* @param  Tag         $tag      Tag to fill
	* @param  string[]    $regexps  Regexps used to extract content from the page
	* @param  string|null $cacheDir Path to the cache directory
	* @return void
	*/
	protected static function scrapeUrl($url, Tag $tag, array $regexps, $cacheDir)
	{
		$content = self::wget($url, $cacheDir);

		// Execute the extract regexps and fill any missing attribute
		foreach ($regexps as $regexp)
		{
			if (preg_match($regexp, $content, $m))
			{
				foreach ($m as $k => $v)
				{
					if (!is_numeric($k) && !$tag->hasAttribute($k))
					{
						$tag->setAttribute($k, $v);
					}
				}
			}
		}
	}

	/**
	* Test whether a tag is missing any of given attributes
	*
	* @param  Tag      $tag
	* @param  string[] $attrNames
	* @return bool
	*/
	protected static function tagIsMissingAnyAttribute(Tag $tag, array $attrNames)
	{
		foreach ($attrNames as $attrName)
		{
			if (!$tag->hasAttribute($attrName))
			{
				return true;
			}
		}

		return false;
	}

	/**
	* Retrieve external content (possibly from the cache)
	*
	* If the cache directory exists, the external content will be saved into it. Cached content is
	* never pruned
	*
	* @param  string $url      URL
	* @param  string $cacheDir Path to the cache directory
	* @return string           External content
	*/
	protected static function wget($url, $cacheDir = null)
	{
		$prefix = '';
		$url    = preg_replace('(#.*)s', '', $url);

		// Return the content from the cache if applicable
		if (isset($cacheDir) && file_exists($cacheDir))
		{
			$cacheFile = $cacheDir . '/http.' . crc32($url);
			if (extension_loaded('zlib'))
			{
				$prefix     = 'compress.zlib://';
				$cacheFile .= '.gz';
			}
			if (file_exists($cacheFile))
			{
				return file_get_contents($prefix . $cacheFile);
			}
		}

		// Retrieve the external content from the source
		$content = @self::getHttpClient()->get($url, ['User-Agent: PHP (not Mozilla)']);

		// Save to the cache if applicable
		if (isset($cacheFile) && !empty($content))
		{
			file_put_contents($prefix . $cacheFile, $content);
		}

		return $content;
	}
}

1		<?php
2
3		/**
4		* @package s9e\TextFormatter
5		* @copyright Copyright (c) 2010-2017 The s9e Authors
6		* @license http://www.opensource.org/licenses/mit-license.php The MIT License
7		*/
8		namespace s9e\TextFormatter\Plugins\MediaEmbed;
9
10		use s9e\TextFormatter\Parser as TagStack;
11		use s9e\TextFormatter\Parser\Tag;
12		use s9e\TextFormatter\Plugins\ParserBase;
13		use s9e\TextFormatter\Utils\Http;
14
15		class Parser extends ParserBase
16		{
17		/**
18		* @var \s9e\TextFormatter\Utils\Http\Client Client used to perform HTTP request
19		*/
20		protected static $client;
21
22		/**
23		* {@inheritdoc}
24		*/
25	404	public function parse($text, array $matches)
26		{
27	404	foreach ($matches as $m)
28		{
29	404	$url = $m[0][0];
30	404	$pos = $m[0][1];
31	404	$len = strlen($url);
32
33		// Give that tag priority over other tags such as Autolink's
34	404	$tag = $this->parser->addSelfClosingTag($this->config['tagName'], $pos, $len, -10);
35	404	$tag->setAttribute('url', $url);
36		}
37	404	}
38
39		/**
40		* Filter a MEDIA tag
41		*
42		* This will always invalidate the original tag, and possibly replace it with the tag that
43		* corresponds to the media site
44		*
45		* @param Tag $tag The original tag
46		* @param TagStack $tagStack Parser instance, so that we can add the new tag to the stack
47		* @param array $sites Map of [host => siteId]
48		* @return bool Unconditionally FALSE
49		*/
50	404	public static function filterTag(Tag $tag, TagStack $tagStack, array $sites)
51		{
52	404	if ($tag->hasAttribute('site'))
53		{
54	5	self::addTagFromMediaId($tag, $tagStack, $sites);
55		}
56	400	elseif ($tag->hasAttribute('url'))
57		{
58	400	self::addTagFromMediaUrl($tag, $tagStack, $sites);
59		}
60
61	404	return false;
62		}
63
64		/**
65		* Test whether a given tag has at least one non-default attribute
66		*
67		* @param Tag $tag The original tag
68		* @return bool Whether the tag contains an attribute not named "url"
69		*/
70	65	public static function hasNonDefaultAttribute(Tag $tag)
71		{
72	65	foreach ($tag->getAttributes() as $attrName => $void)
73		{
74	65	if ($attrName !== 'url')
75		{
76	65	return true;
77		}
78		}
79
80	6	return false;
81		}
82
83		/**
84		* Scrape the content of an URL to extract some data
85		*
86		* @param Tag $tag Source tag
87		* @param array $scrapeConfig Array of scrape directives
88		* @param string $cacheDir Path to the cache directory
89		* @return bool Unconditionally TRUE
90		*/
91	409	public static function scrape(Tag $tag, array $scrapeConfig, $cacheDir = null)
92		{
93	409	if ($tag->hasAttribute('url'))
94		{
95		// Ensure that the URL actually looks like a URL if we want to use it to scrape
96	408	$url = $tag->getAttribute('url');
97	408	if (preg_match('#^https?://[^<>"\'\\s]+$#Di', $url))
98		{
99	403	$url = strtolower(substr($url, 0, 5)) . substr($url, 5);
100	403	foreach ($scrapeConfig as $scrape)
101		{
102	143	self::scrapeEntry($url, $tag, $scrape, $cacheDir);
103		}
104		}
105		}
106
107	409	return true;
108		}
109
110		//==============================================================================================
111		// Internals
112		//==============================================================================================
113
114		/**
115		* Add a site tag
116		*
117		* @param Tag $tag The original tag
118		* @param TagStack $tagStack Parser instance, so that we can add the new tag to the stack
119		* @param string $siteId Site ID
120		* @return void
121		*/
122	402	protected static function addSiteTag(Tag $tag, TagStack $tagStack, $siteId)
123		{
124	402	$endTag = $tag->getEndTag();
125	402	if ($endTag)
126		{
127	12	$startPos = $tag->getPos();
128	12	$startLen = $tag->getLen();
129	12	$endPos = $endTag->getPos();
130	12	$endLen = $endTag->getLen();
131		}
132		else
133		{
134	398	$startPos = $tag->getPos();
135	398	$startLen = 0;
136	398	$endPos = $tag->getPos() + $tag->getLen();
137	398	$endLen = 0;
138		}
139
140		// Create a new tag and copy this tag's attributes and priority
141	402	$tagStack->addTagPair(strtoupper($siteId), $startPos, $startLen, $endPos, $endLen, $tag->getSortPriority())->setAttributes($tag->getAttributes());
142	402	}
143
144		/**
145		* Add a media site tag based on the attributes of a MEDIA tag
146		*
147		* @param Tag $tag The original tag
148		* @param TagStack $tagStack Parser instance
149		* @param array $sites Map of [host => siteId]
150		* @return void
151		*/
152	5	protected static function addTagFromMediaId(Tag $tag, TagStack $tagStack, array $sites)
153		{
154	5	$siteId = strtolower($tag->getAttribute('site'));
155	5	if (in_array($siteId, $sites, true))
156		{
157	4	self::addSiteTag($tag, $tagStack, $siteId);
158		}
159	5	}
160
161		/**
162		* Add a media site tag based on the url attribute of a MEDIA tag
163		*
164		* @param Tag $tag The original tag
165		* @param TagStack $tagStack Parser instance
166		* @param array $sites Map of [host => siteId]
167		* @return void
168		*/
169	400	protected static function addTagFromMediaUrl(Tag $tag, TagStack $tagStack, array $sites)
170		{
171		// Capture the scheme and (if applicable) host of the URL
172	400	$p = parse_url($tag->getAttribute('url'));
173	400	if (isset($p['scheme']) && isset($sites[$p['scheme'] . ':']))
174		{
175		$siteId = $sites[$p['scheme'] . ':'];
176		}
177	400	elseif (isset($p['host']))
178		{
179	400	$siteId = self::findSiteIdByHost($p['host'], $sites);
180		}
181
182	400	if (!empty($siteId))
183		{
184	398	self::addSiteTag($tag, $tagStack, $siteId);
185		}
186	400	}
187
188		/**
189		* Match a given host to a site ID
190		*
191		* @param string $host Host
192		* @param array $sites Map of [host => siteId]
193		* @return string\|bool Site ID or FALSE
194		*/
195	400	protected static function findSiteIdByHost($host, array $sites)
196		{
197		// Start with the full host then pop domain labels off the start until we get a match
198		do
199		{
200	400	if (isset($sites[$host]))
201		{
202	398	return $sites[$host];
203		}
204
205	263	$pos = strpos($host, '.');
206	263	if ($pos === false)
207		{
208	4	break;
209		}
210
211	263	$host = substr($host, 1 + $pos);
212		}
213	263	while ($host > '');
214
215	4	return false;
216		}
217
218		/**
219		* Return a cached instance of the HTTP client
220		*
221		* @return \s9e\TextFormatter\Utils\Http\Client
222		*/
223	2	protected static function getHttpClient()
224		{
225	2	if (!isset(self::$client))
226		{
227	1	self::$client = Http::getClient();
228		}
229	2	self::$client->timeout = 10;
230
231	2	return self::$client;
232		}
233
234		/**
235		* Replace {@var} tokens in given URL
236		*
237		* @param string $url Original URL
238		* @param string[] $vars Replacements
239		* @return string Modified URL
240		*/
241	20	protected static function replaceTokens($url, array $vars)
242		{
243	20	return preg_replace_callback(
244	20	'#\\{@(\\w+)\\}#',
245	20	function ($m) use ($vars)
246		{
247	20	return (isset($vars[$m[1]])) ? $vars[$m[1]] : '';
248	20	},
249	20	$url
250		);
251		}
252
253		/**
254		* Scrape the content of an URL to extract some data
255		*
256		* @param string $url Original URL
257		* @param Tag $tag Source tag
258		* @param array $scrape Array of scrape directives
259		* @param string $cacheDir Path to the cache directory
260		* @return void
261		*/
262	143	protected static function scrapeEntry($url, Tag $tag, array $scrape, $cacheDir)
263		{
264	143	list($matchRegexps, $extractRegexps, $attrNames) = $scrape;
265	143	if (!self::tagIsMissingAnyAttribute($tag, $attrNames))
266		{
267	74	return;
268		}
269
270		// Test whether this URL matches any regexp
271	76	$vars = [];
272	76	$matched = false;
273	76	foreach ((array) $matchRegexps as $matchRegexp)
274		{
275	76	if (preg_match($matchRegexp, $url, $m))
276		{
277	63	$vars += $m;
278	76	$matched = true;
279		}
280		}
281	76	if (!$matched)
282		{
283	20	return;
284		}
285
286		// Add the tag's attributes to the named captures from the "match" regexp
287	63	$vars += $tag->getAttributes();
288
289	63	$scrapeUrl = (isset($scrape[3])) ? self::replaceTokens($scrape[3], $vars) : $url;
290	63	self::scrapeUrl($scrapeUrl, $tag, (array) $extractRegexps, $cacheDir);
291	63	}
292
293		/**
294		* Scrape a URL to help fill a tag's attributes
295		*
296		* @param string $url URL to scrape
297		* @param Tag $tag Tag to fill
298		* @param string[] $regexps Regexps used to extract content from the page
299		* @param string\|null $cacheDir Path to the cache directory
300		* @return void
301		*/
302	63	protected static function scrapeUrl($url, Tag $tag, array $regexps, $cacheDir)
303		{
304	63	$content = self::wget($url, $cacheDir);
305
306		// Execute the extract regexps and fill any missing attribute
307	63	foreach ($regexps as $regexp)
308		{
309	63	if (preg_match($regexp, $content, $m))
310		{
311	61	foreach ($m as $k => $v)
312		{
313	61	if (!is_numeric($k) && !$tag->hasAttribute($k))
314		{
315	63	$tag->setAttribute($k, $v);
316		}
317		}
318		}
319		}
320	63	}
321
322		/**
323		* Test whether a tag is missing any of given attributes
324		*
325		* @param Tag $tag
326		* @param string[] $attrNames
327		* @return bool
328		*/
329	143	protected static function tagIsMissingAnyAttribute(Tag $tag, array $attrNames)
330		{
331	143	foreach ($attrNames as $attrName)
332		{
333	143	if (!$tag->hasAttribute($attrName))
334		{
335	143	return true;
336		}
337		}
338
339	74	return false;
340		}
341
342		/**
343		* Retrieve external content (possibly from the cache)
344		*
345		* If the cache directory exists, the external content will be saved into it. Cached content is
346		* never pruned
347		*
348		* @param string $url URL
349		* @param string $cacheDir Path to the cache directory
350		* @return string External content
351		*/
352	63	protected static function wget($url, $cacheDir = null)
353		{
354	63	$prefix = '';
355	63	$url = preg_replace('(#.*)s', '', $url);
356
357		// Return the content from the cache if applicable
358	63	if (isset($cacheDir) && file_exists($cacheDir))
359		{
360	63	$cacheFile = $cacheDir . '/http.' . crc32($url);
361	63	if (extension_loaded('zlib'))
362		{
363	63	$prefix = 'compress.zlib://';
364	63	$cacheFile .= '.gz';
365		}
366	63	if (file_exists($cacheFile))
367		{
368	61	return file_get_contents($prefix . $cacheFile);
369		}
370		}
371
372		// Retrieve the external content from the source
373	2	$content = @self::getHttpClient()->get($url, ['User-Agent: PHP (not Mozilla)']);
374
375		// Save to the cache if applicable
376	2	if (isset($cacheFile) && !empty($content))
377		{
378	1	file_put_contents($prefix . $cacheFile, $content);
379		}
380
381	2	return $content;
382		}
383		}

s9e / TextFormatter

Push — master ( 06920f...7e343b )

Parser::scrapeUrl() B

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like