Parser::scrapeUrl() - Code Metrics - Inspection of "Parser: removed implicit invalidation in tag filte..." - s9e/TextFormatter - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( ff6da6...21a9fc )

by Josh

created 2017-12-28 17:16 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 void
	*/
	public static function filterTag(Tag $tag, TagStack $tagStack, array $sites)
	{
		$tag->invalidate();
		if ($tag->hasAttribute('site'))
		{
			self::addTagFromMediaId($tag, $tagStack, $sites);
		}
		elseif ($tag->hasAttribute('url'))
		{
			self::addTagFromMediaUrl($tag, $tagStack, $sites);
		}
	}

	/**
	* Invalidate given tag if it doesn't have at least one non-default attribute
	*
	* @param  Tag  $tag The original tag
	* @return void
	*/
	public static function hasNonDefaultAttribute(Tag $tag)
	{
		foreach ($tag->getAttributes() as $attrName => $void)
		{
			if ($attrName !== 'url')
			{
				return;
			}
		}

		$tag->invalidate();
	}

	/**
	* 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 $client;
21
22		/**
23		* {@inheritdoc}
24		*/
25	407	public function parse($text, array $matches)
26		{
27	407	foreach ($matches as $m)
28		{
29	407	$url = $m[0][0];
30	407	$pos = $m[0][1];
31	407	$len = strlen($url);
32
33		// Give that tag priority over other tags such as Autolink's
34	407	$tag = $this->parser->addSelfClosingTag($this->config['tagName'], $pos, $len, -10);
35	407	$tag->setAttribute('url', $url);
36		}
37	407	}
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 void
49		*/
50	407	public static function filterTag(Tag $tag, TagStack $tagStack, array $sites)
51		{
52	407	$tag->invalidate();
53	407	if ($tag->hasAttribute('site'))
54		{
55	5	self::addTagFromMediaId($tag, $tagStack, $sites);
56		}
57	403	elseif ($tag->hasAttribute('url'))
58		{
59	403	self::addTagFromMediaUrl($tag, $tagStack, $sites);
60		}
61	407	}
62
63		/**
64		* Invalidate given tag if it doesn't have at least one non-default attribute
65		*
66		* @param Tag $tag The original tag
67		* @return void
68		*/
69	60	public static function hasNonDefaultAttribute(Tag $tag)
70		{
71	60	foreach ($tag->getAttributes() as $attrName => $void)
72		{
73	60	if ($attrName !== 'url')
74		{
75	60	return;
76		}
77		}
78
79	6	$tag->invalidate();
80	6	}
81
82		/**
83		* Scrape the content of an URL to extract some data
84		*
85		* @param Tag $tag Source tag
86		* @param array $scrapeConfig Array of scrape directives
87		* @param string $cacheDir Path to the cache directory
88		* @return bool Unconditionally TRUE
89		*/
90	412	public static function scrape(Tag $tag, array $scrapeConfig, $cacheDir = null)
91		{
92	412	if ($tag->hasAttribute('url'))
93		{
94		// Ensure that the URL actually looks like a URL if we want to use it to scrape
95	411	$url = $tag->getAttribute('url');
96	411	if (preg_match('#^https?://[^<>"\'\\s]+$#Di', $url))
97		{
98	406	$url = strtolower(substr($url, 0, 5)) . substr($url, 5);
99	406	foreach ($scrapeConfig as $scrape)
100		{
101	143	self::scrapeEntry($url, $tag, $scrape, $cacheDir);
102		}
103		}
104		}
105
106	412	return true;
107		}
108
109		//==============================================================================================
110		// Internals
111		//==============================================================================================
112
113		/**
114		* Add a site tag
115		*
116		* @param Tag $tag The original tag
117		* @param TagStack $tagStack Parser instance, so that we can add the new tag to the stack
118		* @param string $siteId Site ID
119		* @return void
120		*/
121	405	protected static function addSiteTag(Tag $tag, TagStack $tagStack, $siteId)
122		{
123	405	$endTag = $tag->getEndTag();
124	405	if ($endTag)
125		{
126	12	$startPos = $tag->getPos();
127	12	$startLen = $tag->getLen();
128	12	$endPos = $endTag->getPos();
129	12	$endLen = $endTag->getLen();
130		}
131		else
132		{
133	401	$startPos = $tag->getPos();
134	401	$startLen = 0;
135	401	$endPos = $tag->getPos() + $tag->getLen();
136	401	$endLen = 0;
137		}
138
139		// Create a new tag and copy this tag's attributes and priority
140	405	$tagStack->addTagPair(strtoupper($siteId), $startPos, $startLen, $endPos, $endLen, $tag->getSortPriority())->setAttributes($tag->getAttributes());
141	405	}
142
143		/**
144		* Add a media site tag based on the attributes of a MEDIA tag
145		*
146		* @param Tag $tag The original tag
147		* @param TagStack $tagStack Parser instance
148		* @param array $sites Map of [host => siteId]
149		* @return void
150		*/
151	5	protected static function addTagFromMediaId(Tag $tag, TagStack $tagStack, array $sites)
152		{
153	5	$siteId = strtolower($tag->getAttribute('site'));
154	5	if (in_array($siteId, $sites, true))
155		{
156	4	self::addSiteTag($tag, $tagStack, $siteId);
157		}
158	5	}
159
160		/**
161		* Add a media site tag based on the url attribute of a MEDIA tag
162		*
163		* @param Tag $tag The original tag
164		* @param TagStack $tagStack Parser instance
165		* @param array $sites Map of [host => siteId]
166		* @return void
167		*/
168	403	protected static function addTagFromMediaUrl(Tag $tag, TagStack $tagStack, array $sites)
169		{
170		// Capture the scheme and (if applicable) host of the URL
171	403	$p = parse_url($tag->getAttribute('url'));
172	403	if (isset($p['scheme']) && isset($sites[$p['scheme'] . ':']))
173		{
174		$siteId = $sites[$p['scheme'] . ':'];
175		}
176	403	elseif (isset($p['host']))
177		{
178	403	$siteId = self::findSiteIdByHost($p['host'], $sites);
179		}
180
181	403	if (!empty($siteId))
182		{
183	401	self::addSiteTag($tag, $tagStack, $siteId);
184		}
185	403	}
186
187		/**
188		* Match a given host to a site ID
189		*
190		* @param string $host Host
191		* @param array $sites Map of [host => siteId]
192		* @return string\|bool Site ID or FALSE
193		*/
194	403	protected static function findSiteIdByHost($host, array $sites)
195		{
196		// Start with the full host then pop domain labels off the start until we get a match
197		do
198		{
199	403	if (isset($sites[$host]))
200		{
201	401	return $sites[$host];
202		}
203
204	264	$pos = strpos($host, '.');
205	264	if ($pos === false)
206		{
207	4	break;
208		}
209
210	264	$host = substr($host, 1 + $pos);
211		}
212	264	while ($host > '');
213
214	4	return false;
215		}
216
217		/**
218		* Return a cached instance of the HTTP client
219		*
220		* @return \s9e\TextFormatter\Utils\Http\Client
221		*/
222	2	protected static function getHttpClient()
223		{
224	2	if (!isset(self::$client))
225		{
226	1	self::$client = Http::getClient();
227		}
228	2	self::$client->timeout = 10;
229
230	2	return self::$client;
231		}
232
233		/**
234		* Replace {@var} tokens in given URL
235		*
236		* @param string $url Original URL
237		* @param string[] $vars Replacements
238		* @return string Modified URL
239		*/
240	20	protected static function replaceTokens($url, array $vars)
241		{
242	20	return preg_replace_callback(
243	20	'#\\{@(\\w+)\\}#',
244	20	function ($m) use ($vars)
245		{
246	20	return (isset($vars[$m[1]])) ? $vars[$m[1]] : '';
247	20	},
248	20	$url
249		);
250		}
251
252		/**
253		* Scrape the content of an URL to extract some data
254		*
255		* @param string $url Original URL
256		* @param Tag $tag Source tag
257		* @param array $scrape Array of scrape directives
258		* @param string $cacheDir Path to the cache directory
259		* @return void
260		*/
261	143	protected static function scrapeEntry($url, Tag $tag, array $scrape, $cacheDir)
262		{
263	143	list($matchRegexps, $extractRegexps, $attrNames) = $scrape;
264	143	if (!self::tagIsMissingAnyAttribute($tag, $attrNames))
265		{
266	74	return;
267		}
268
269		// Test whether this URL matches any regexp
270	76	$vars = [];
271	76	$matched = false;
272	76	foreach ((array) $matchRegexps as $matchRegexp)
273		{
274	76	if (preg_match($matchRegexp, $url, $m))
275		{
276	63	$vars += $m;
277	76	$matched = true;
278		}
279		}
280	76	if (!$matched)
281		{
282	20	return;
283		}
284
285		// Add the tag's attributes to the named captures from the "match" regexp
286	63	$vars += $tag->getAttributes();
287
288	63	$scrapeUrl = (isset($scrape[3])) ? self::replaceTokens($scrape[3], $vars) : $url;
289	63	self::scrapeUrl($scrapeUrl, $tag, (array) $extractRegexps, $cacheDir);
290	63	}
291
292		/**
293		* Scrape a URL to help fill a tag's attributes
294		*
295		* @param string $url URL to scrape
296		* @param Tag $tag Tag to fill
297		* @param string[] $regexps Regexps used to extract content from the page
298		* @param string\|null $cacheDir Path to the cache directory
299		* @return void
300		*/
301	63	protected static function scrapeUrl($url, Tag $tag, array $regexps, $cacheDir)
302		{
303	63	$content = self::wget($url, $cacheDir);
304
305		// Execute the extract regexps and fill any missing attribute
306	63	foreach ($regexps as $regexp)
307		{
308	63	if (preg_match($regexp, $content, $m))
309		{
310	61	foreach ($m as $k => $v)
311		{
312	61	if (!is_numeric($k) && !$tag->hasAttribute($k))
313		{
314	63	$tag->setAttribute($k, $v);
315		}
316		}
317		}
318		}
319	63	}
320
321		/**
322		* Test whether a tag is missing any of given attributes
323		*
324		* @param Tag $tag
325		* @param string[] $attrNames
326		* @return bool
327		*/
328	143	protected static function tagIsMissingAnyAttribute(Tag $tag, array $attrNames)
329		{
330	143	foreach ($attrNames as $attrName)
331		{
332	143	if (!$tag->hasAttribute($attrName))
333		{
334	143	return true;
335		}
336		}
337
338	74	return false;
339		}
340
341		/**
342		* Retrieve external content (possibly from the cache)
343		*
344		* If the cache directory exists, the external content will be saved into it. Cached content is
345		* never pruned
346		*
347		* @param string $url URL
348		* @param string $cacheDir Path to the cache directory
349		* @return string External content
350		*/
351	63	protected static function wget($url, $cacheDir = null)
352		{
353	63	$prefix = '';
354	63	$url = preg_replace('(#.*)s', '', $url);
355
356		// Return the content from the cache if applicable
357	63	if (isset($cacheDir) && file_exists($cacheDir))
358		{
359	63	$cacheFile = $cacheDir . '/http.' . crc32($url);
360	63	if (extension_loaded('zlib'))
361		{
362	63	$prefix = 'compress.zlib://';
363	63	$cacheFile .= '.gz';
364		}
365	63	if (file_exists($cacheFile))
366		{
367	61	return file_get_contents($prefix . $cacheFile);
368		}
369		}
370
371		// Retrieve the external content from the source
372	2	$content = @self::getHttpClient()->get($url, ['User-Agent: PHP (not Mozilla)']);
373
374		// Save to the cache if applicable
375	2	if (isset($cacheFile) && !empty($content))
376		{
377	1	file_put_contents($prefix . $cacheFile, $content);
378		}
379
380	2	return $content;
381		}
382		}

s9e / TextFormatter

Push — master ( ff6da6...21a9fc )

Parser::scrapeUrl() B

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like