SearchResultAdCrawler::setUrlParser() - Code Metrics - Inspection of "Merge pull request #16 from rfussien/develop" - rfussien/leboncoin-crawler - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( 00c88c...56f5c3 )

by Rémi

created 2017-04-06 19:54 UTC

SearchResultAdCrawler::setUrlParser() A

↳ Parent: SearchResultAdCrawler

Complexity

Conditions	1
Paths	1

Size

Total Lines	4
Code Lines	2

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	3
CRAP Score	1

Importance

Changes

Metric	Value
c	0
b	0
f	0
dl	0
loc	4
ccs	3
cts	3
cp	1
rs	10
cc	1
eloc	2
nc	1
nop	1
crap	1

<?php

namespace Lbc\Crawler;

use Lbc\Filter\DefaultSanitizer;
use Lbc\Filter\PriceSanitizer;
use Lbc\Parser\AdUrlParser;
use Lbc\Parser\SearchResultUrlParser;
use League\Uri\Schemes\Http;
use Symfony\Component\DomCrawler\Crawler;

/**
 * Class SearchResultAdCrawler
 * @package Lbc\Crawler
 */
class SearchResultAdCrawler extends CrawlerAbstract
{
    /**
     * @param $url
     * @return SearchResultUrlParser
     */
    protected function setUrlParser($url)
    {
        $this->url = new AdUrlParser($url);
    }

    /**
     * Return the Ad's ID
     *
     * @return string
     */
    public function getId()
    {
        return $this->url->getId();
class A
{
    public function foo() { }
}

class B extends A
{
    public function bar() { }
}

/**
 * @param A|B $x
 */
function someFunction($x)
{
    $x->foo(); // This call is fine as the method exists in A and B.
    $x->bar(); // This method only exists in B and might cause an error.
}
    }

    /**
     * Return the title
     *
     * @return mixed
     */
    public function getTitle()
    {
        return DefaultSanitizer::clean($this->node->filter('h2')->text());
    }

    /**
     * Return the price
     *
     * @return int
     */
    public function getPrice()
    {
        if ($this->node->filter('*[itemprop=price]')->count()) {
            return PriceSanitizer::clean(
                $this->node->filter('*[itemprop=price]')->text()
            );
        }
        return 0;
    }

    /**
     * Return the Ad's URL
     *
     * @return string
     */
    public function getUrl()
    {
        return (string)Http::createFromString($this->url)->withScheme('https');
function acceptsInteger($int) { }

$x = '123'; // string "123"

// Instead of
acceptsInteger($x);

// we recommend to use
acceptsInteger((integer) $x);
    }

    /**
     * Return the data and time the ad was created
     *
     * @return string
     */
    public function getCreatedAt()
    {
        $node = $this->node
            ->filter('*[itemprop=availabilityStarts]')
            ->first();

        $date = $node->attr('content');

        $time = $this->getFieldValue($node, 0, function ($value) {
            $value = trim($value);

            return substr($value, strpos($value, ',') + 2);
        });

        return $date . ' ' . $time;
    }

    /**
     * Return the thumb picture url
     *
     * @return null|string
     */
    public function getThumb()
    {
        $image = $this->node
            ->filter('.item_imagePic .lazyload[data-imgsrc]')
            ->first();

        if (0 === $image->count()) {
            return null;
        }

        $src = $image
            ->attr('data-imgsrc');

        return (string)Http::createFromString($src)->withScheme('https');
    }

    /**
     * Return the number of picture of the ad
     *
     * @return int
     */
    public function getNbImage()
    {
        $node = $this->node->filter('.item_imageNumber');

        return $this->getFieldValue($node, 0, function ($value) {
            return (int)trim($value);
        });
    }

    /**
     * @return mixed
     */
    public function getPlacement()

    {
        $node = $this->node->filter('*[itemprop=availableAtOrFrom]');

        return $this->getFieldValue($node, '', function ($value) {
            return preg_replace('/\s+/', ' ', trim($value));
        });
    }

    /**
     * @return mixed
     */
    public function getType()

    {
        $node = $this->node->filter('*[itemprop=category]');

        return $this->getFieldValue($node, false, function ($value) {
            if ('pro' === preg_replace('/[\s()]+/', '', $value)) {
                return 'pro';
            }

            return 'part';
        });
    }

    /**
     * @return object
     */
    public function getAll()
    {
        return (object)[
            'id'         => $this->getId(),
            'title'      => $this->getTitle(),
            'price'      => $this->getPrice(),
            'url'        => $this->getUrl(),
            'created_at' => $this->getCreatedAt(),
            'thumb'      => $this->getThumb(),
            'nb_image'   => $this->getNbImage(),
            'placement'  => $this->getPlacement(),
            'type'       => $this->getType(),
        ];
    }

    /**
     * Return the field's value
     *
     * @param $node
     * @param $defaultValue
     * @param $callback
     * @param string $funcName
     * @param string $funcParam
     *
     * @return mixed
     */
    private function getFieldValue(
        Crawler $node,
        $defaultValue,
        $callback,
        $funcName = 'text',
        $funcParam = ''
    ) {
        if ($node->count()) {
            return $callback($node->$funcName($funcParam));
        }

        return $defaultValue;
    }
}


1			<?php
2
3			namespace Lbc\Crawler;
4
5			use Lbc\Filter\DefaultSanitizer;
6			use Lbc\Filter\PriceSanitizer;
7			use Lbc\Parser\AdUrlParser;
8			use Lbc\Parser\SearchResultUrlParser;
9			use League\Uri\Schemes\Http;
10			use Symfony\Component\DomCrawler\Crawler;
11
12			/**
13			* Class SearchResultAdCrawler
14			* @package Lbc\Crawler
15			*/
16			class SearchResultAdCrawler extends CrawlerAbstract
17			{
18			/**
19			* @param $url
20			* @return SearchResultUrlParser
21			*/
22	10		protected function setUrlParser($url)
23			{
24	10		$this->url = new AdUrlParser($url);
25	10		}
26
27			/**
28			* Return the Ad's ID
29			*
30			* @return string
31			*/
32	10		public function getId()
33			{
34	10		return $this->url->getId();
			0 ignored issues – show Bug introduced 2017-04-06 19:53 UTC by Report Bug Copy Issue Report The method `getId` does only exist in `Lbc\Parser\AdUrlParser`, but not in `Lbc\Parser\SearchResultUrlParser`. It seems like the method you are trying to call exists only in some of the possible types. Let’s take a look at an example: class A { public function foo() { } } class B extends A { public function bar() { } } /** * @param A\|B $x / function someFunction($x) { $x->foo(); // This call is fine as the method exists in A and B. $x->bar(); // This method only exists in B and might cause an error. } Available Fixes Add an additional type-check: /* * @param A\|B $x / function someFunction($x) { $x->foo(); if ($x instanceof B) { $x->bar(); } } Only allow a single type to be passed if the variable comes from a parameter: function someFunction(B $x) { /* ... */ } Loading history...
35			}
36
37			/**
38			* Return the title
39			*
40			* @return mixed
41			*/
42	10		public function getTitle()
43			{
44	10		return DefaultSanitizer::clean($this->node->filter('h2')->text());
45			}
46
47			/**
48			* Return the price
49			*
50			* @return int
51			*/
52	10		public function getPrice()
53			{
54	10		if ($this->node->filter('*[itemprop=price]')->count()) {
55	10		return PriceSanitizer::clean(
56	10		$this->node->filter('*[itemprop=price]')->text()
57	10		);
58			}
59			return 0;
60			}
61
62			/**
63			* Return the Ad's URL
64			*
65			* @return string
66			*/
67	10		public function getUrl()
68			{
69	10		return (string)Http::createFromString($this->url)->withScheme('https');
			0 ignored issues – show Documentation introduced 2017-04-06 19:53 UTC by Report Bug Copy Issue Report `$this->url` is of type `object<Lbc\Parser\AdUrlP...\SearchResultUrlParser>`, but the function expects a `string`. It seems like the type of the argument is not accepted by the function/method which you are calling. In some cases, in particular if PHP’s automatic type-juggling kicks in this might be fine. In other cases, however this might be a bug. We suggest to add an explicit type cast like in the following example: function acceptsInteger($int) { } $x = '123'; // string "123" // Instead of acceptsInteger($x); // we recommend to use acceptsInteger((integer) $x); Loading history...
70			}
71
72			/**
73			* Return the data and time the ad was created
74			*
75			* @return string
76			*/
77	10		public function getCreatedAt()
78			{
79	10		$node = $this->node
80	10		->filter('*[itemprop=availabilityStarts]')
81	10		->first();
82
83	10		$date = $node->attr('content');
84
85			$time = $this->getFieldValue($node, 0, function ($value) {
86	10		$value = trim($value);
87
88	10		return substr($value, strpos($value, ',') + 2);
89	10		});
90
91	10		return $date . ' ' . $time;
92			}
93
94			/**
95			* Return the thumb picture url
96			*
97			* @return null\|string
98			*/
99	10		public function getThumb()
100			{
101	10		$image = $this->node
102	10		->filter('.item_imagePic .lazyload[data-imgsrc]')
103	10		->first();
104
105	10		if (0 === $image->count()) {
106			return null;
107			}
108
109			$src = $image
110	10		->attr('data-imgsrc');
111
112	10		return (string)Http::createFromString($src)->withScheme('https');
113			}
114
115			/**
116			* Return the number of picture of the ad
117			*
118			* @return int
119			*/
120	10		public function getNbImage()
121			{
122	10		$node = $this->node->filter('.item_imageNumber');
123
124			return $this->getFieldValue($node, 0, function ($value) {
125	10		return (int)trim($value);
126	10		});
127			}
128
129			/**
130			* @return mixed
131			*/
132	10	View Code Duplication	public function getPlacement()
			0 ignored issues – show Duplication introduced 2015-12-21 14:14 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...
133			{
134	10		$node = $this->node->filter('*[itemprop=availableAtOrFrom]');
135
136			return $this->getFieldValue($node, '', function ($value) {
137	10		return preg_replace('/\s+/', ' ', trim($value));
138	10		});
139			}
140
141			/**
142			* @return mixed
143			*/
144	10	View Code Duplication	public function getType()
			0 ignored issues – show Duplication introduced 2015-12-21 14:14 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...
145			{
146	10		$node = $this->node->filter('*[itemprop=category]');
147
148	10		return $this->getFieldValue($node, false, function ($value) {
149	10		if ('pro' === preg_replace('/[\s()]+/', '', $value)) {
150	8		return 'pro';
151			}
152
153	10		return 'part';
154	10		});
155			}
156
157			/**
158			* @return object
159			*/
160	10		public function getAll()
161			{
162			return (object)[
163	10		'id' => $this->getId(),
164	10		'title' => $this->getTitle(),
165	10		'price' => $this->getPrice(),
166	10		'url' => $this->getUrl(),
167	10		'created_at' => $this->getCreatedAt(),
168	10		'thumb' => $this->getThumb(),
169	10		'nb_image' => $this->getNbImage(),
170	10		'placement' => $this->getPlacement(),
171	10		'type' => $this->getType(),
172	10		];
173			}
174
175			/**
176			* Return the field's value
177			*
178			* @param $node
179			* @param $defaultValue
180			* @param $callback
181			* @param string $funcName
182			* @param string $funcParam
183			*
184			* @return mixed
185			*/
186	10		private function getFieldValue(
187			Crawler $node,
188			$defaultValue,
189			$callback,
190			$funcName = 'text',
191			$funcParam = ''
192			) {
193	10		if ($node->count()) {
194	10		return $callback($node->$funcName($funcParam));
195			}
196
197			return $defaultValue;
198			}
199			}
200

rfussien / leboncoin-crawler

Push — master ( 00c88c...56f5c3 )

SearchResultAdCrawler::setUrlParser() A

Complexity

Size

Duplication

Code Coverage

Importance

Available Fixes

Duplication Side-by-Side

Filter issues like