Crawler::clearString() - Code Metrics - juniorb2ss/zServices - Measure and Improve Code Quality continuously with Scrutinizer

Crawler::clearString() A
last analyzed 2017-08-07 13:13 UTC

↳ Parent: Crawler

Complexity

Conditions	1
Paths	1

Size

Total Lines	4
Code Lines	2

Duplication

Lines	0
Ratio	0 %

Importance

Changes

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

<?php namespace zServices\ReceitaFederal\Services\Portais\AN;

use zServices\Miscellany\Interfaces\CrawlerInterface;
use Symfony\Component\DomCrawler\Crawler as BaseCrawler;
use zServices\Miscellany\Exceptions\NoSelectorsConfigured;
use zServices\Miscellany\Exceptions\InvalidCaptcha;
use zServices\Miscellany\Exceptions\ErrorFoundData;

/**
* 
*/
class Crawler extends BaseCrawler implements CrawlerInterface
{

	/**
	 * Selectors to crawler
	 * @var array
	 */
	private $selectors = [];

	/**
	 * Get selectors and initialize crawler in HTML
	 * @param string $html      
	 * @param array  $selectors
	 * @return void

	 */
	public function __construct($html, $selectors)
	{
		$this->selectors = $selectors;

		parent::__construct($html);
	}

	/**
	 * Verifica antes de fazer o crawler se possui erros
	 * na requisição
	 * @return boolean|null 
	 */
	public function hasError()
	{
		if (count($this->selectors) == 0) {
			throw new NoSelectorsConfigured("NoSelectorsConfigured", 1);
		}

		// verifica se a página seguida na requisição 
		// é página de erro da receita federal
		$node = $this->filter($this->selectors['error']);

		if ($node->count()) {
			 throw new ErrorFoundData($this->clearString($node->text()), 1);
		}
        
		// CNPJ informado é válido?
		if ($this->filter('#imgCaptcha')->count()) {
			throw new InvalidCaptcha('Captcha inválido', 99);
		}
	}

	/**
	 * Extrai informações do HTML através do DOM
	 *
	 * @return array
	 */
	public function scraping()

	{
		$scrapped = [];

		$this->hasError();

		foreach ($this->selectors as $name => $selector) {
			if (is_string($selector)) {
				$node = $this->scrap($selector);

				if ($node->count()) {
					$scrapped[$name] = $this->clearString($node->text());
				}
			} elseif (is_array($selector)) {
				foreach ($selector as $selector => $repeat) {
$collection = json_decode($data, true);
if ( ! is_array($collection)) {
    throw new \RuntimeException('$collection must be an array.');
}

foreach ($collection as $item) { /** ... */ }
					$node = $this->scrap($selector);
					if ($node->count()) {
						foreach ($node->filter($repeat) as $loop)
						{
							$scrapped[$name][] = $this->clearString($loop->nodeValue);
						}
					}
				}
			}
		}

		return $scrapped;
	}

	/**
	 * Limpa o valor repassado
	 * @param  string $string
	 * @return string
	 */
	public function clearString($string)
	{
		return trim(preg_replace(['/[\s]+/mu'], ' ', $string));
	}

	/**
	 * Filtra selector no crawler
	 */
	public function scrap($selector)
	{
		$node = $this->filter($selector);
		return $node;
	}
}

1		<?php namespace zServices\ReceitaFederal\Services\Portais\AN;
2
3		use zServices\Miscellany\Interfaces\CrawlerInterface;
4		use Symfony\Component\DomCrawler\Crawler as BaseCrawler;
5		use zServices\Miscellany\Exceptions\NoSelectorsConfigured;
6		use zServices\Miscellany\Exceptions\InvalidCaptcha;
7		use zServices\Miscellany\Exceptions\ErrorFoundData;
8
9		/**
10		*
11		*/
12		class Crawler extends BaseCrawler implements CrawlerInterface
13		{
14
15		/**
16		* Selectors to crawler
17		* @var array
18		*/
19		private $selectors = [];
20
21		/**
22		* Get selectors and initialize crawler in HTML
23		* @param string $html
24		* @param array $selectors
25		* @return void
		0 ignored issues – show Comprehensibility Best Practice introduced 2016-05-02 17:42 UTC by Report Bug Copy Issue Report Adding a `@return` annotation to constructors is generally not recommended as a constructor does not have a meaningful return value. Adding a `@return` annotation to a constructor is not recommended, since a constructor does not have a meaningful return value. Please refer to the PHP core documentation on constructors. Loading history...
26		*/
27		public function __construct($html, $selectors)
28		{
29		$this->selectors = $selectors;
30
31		parent::__construct($html);
32		}
33
34		/**
35		* Verifica antes de fazer o crawler se possui erros
36		* na requisição
37		* @return boolean\|null
38		*/
39		public function hasError()
40		{
41		if (count($this->selectors) == 0) {
42		throw new NoSelectorsConfigured("NoSelectorsConfigured", 1);
43		}
44
45		// verifica se a página seguida na requisição
46		// é página de erro da receita federal
47		$node = $this->filter($this->selectors['error']);
48
49		if ($node->count()) {
50		throw new ErrorFoundData($this->clearString($node->text()), 1);
51		}
52
53		// CNPJ informado é válido?
54		if ($this->filter('#imgCaptcha')->count()) {
55		throw new InvalidCaptcha('Captcha inválido', 99);
56		}
57		}
58
59		/**
60		* Extrai informações do HTML através do DOM
61		*
62		* @return array
63		*/
64	View Code Duplication	public function scraping()
		0 ignored issues – show Duplication introduced 2017-08-02 16:49 UTC by Report Bug Copy Issue Report This method seems to be duplicated in your project. Duplicated code is one of the most pungent code smells. If you need to duplicate the same code in three or more different places, we strongly encourage you to look into extracting the code into a single class or operation. You can also find more detailed suggestions in the “Code” section of your repository. Loading history...
65		{
66		$scrapped = [];
67
68		$this->hasError();
69
70		foreach ($this->selectors as $name => $selector) {
71		if (is_string($selector)) {
72		$node = $this->scrap($selector);
73
74		if ($node->count()) {
75		$scrapped[$name] = $this->clearString($node->text());
76		}
77		} elseif (is_array($selector)) {
78		foreach ($selector as $selector => $repeat) {
		0 ignored issues – show Bug introduced 2016-05-02 15:20 UTC by Report Bug Copy Issue Report The expression `$selector` of type `integer\|string\|array` is not guaranteed to be traversable. How about adding an additional type check? There are different options of fixing this problem. If you want to be on the safe side, you can add an additional type-check: $collection = json_decode($data, true); if ( ! is_array($collection)) { throw new \RuntimeException('$collection must be an array.'); } foreach ($collection as $item) { /** ... / } If you are sure that the expression is traversable, you might want to add a doc comment cast* to improve IDE auto-completion and static analysis: /** @var array $collection / $collection = json_decode($data, true); foreach ($collection as $item) { /* .. / } Mark the issue as a false-positive*: Just hover the remove button, in the top-right corner of this issue for more options. Loading history...
79		$node = $this->scrap($selector);
80		if ($node->count()) {
81		foreach ($node->filter($repeat) as $loop)
82		{
83		$scrapped[$name][] = $this->clearString($loop->nodeValue);
84		}
85		}
86		}
87		}
88		}
89
90		return $scrapped;
91		}
92
93		/**
94		* Limpa o valor repassado
95		* @param string $string
96		* @return string
97		*/
98		public function clearString($string)
99		{
100		return trim(preg_replace(['/[\s]+/mu'], ' ', $string));
101		}
102
103		/**
104		* Filtra selector no crawler
105		*/
106		public function scrap($selector)
107		{
108		$node = $this->filter($selector);
109		return $node;
110		}
111		}

juniorb2ss / zServices

GitHub Access Token became invalid

Crawler::clearString() A last analyzed 2017-08-07 13:13 UTC

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like

Crawler::clearString() A
last analyzed 2017-08-07 13:13 UTC