CustomAdapter::getElementText() - Code Metrics - zrashwani/news-scrapper - Measure and Improve Code Quality continuously with Scrutinizer

CustomAdapter::getElementText() B
last analyzed 2017-08-17 08:42 UTC

↳ Parent: CustomAdapter

Complexity

Conditions	4
Paths	5

Size

Total Lines	23
Code Lines	14

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
dl	0
loc	23
rs	8.7972
c	0
b	0
f	0
nc	5
cc	4
eloc	14
nop	3

<?php

namespace Zrashwani\NewsScrapper\Adapters;

use Symfony\Component\DomCrawler\Crawler;
use Zrashwani\NewsScrapper\Selector;

/**
 * Adapter to extract page data from un-structured HTML document
 * @author Zeid Rashwani <zrashwani.com>
 */
class CustomAdapter extends AbstractAdapter
{

    private $authorSelector;
    private $bodySelector;
    private $descriptionSelector;
    private $imageSelector;
    private $keywordsSelector;
    private $publishDateSelector;
    private $titleSelector;

    /**
     * adapter used to fill in the missing selectors data by default values
     * @var DefaultAdapter $fallbackAdapter
     */
    private $fallbackAdapter;

    public function __construct()
    {
        $this->fallbackAdapter = new DefaultAdapter();
    }

    public function setAuthorSelector($selector)
    {
        $this->authorSelector = $selector;
        return $this;
    }

    public function setBodySelector($selector)
    {
        $this->bodySelector = $selector;
        return $this;
    }

    public function setDescriptionSelector($selector)
    {
        $this->descriptionSelector = $selector;
        return $this;
    }

    public function setImageSelector($selector)
    {
        $this->imageSelector = $selector;
        return $this;
    }

    public function setKeywordsSelector($selector)
    {
        $this->keywordsSelector = $selector;
        return $this;
    }

    public function setPublishDateSelector($selector)
    {
        $this->publishDateSelector = $selector;
        return $this;
    }

    public function setTitleSelector($selector)
    {
        $this->titleSelector = $selector;
        return $this;
    }

    public function extractAuthor(Crawler $crawler)

    {
        $ret = $this->getElementText($crawler, $this->authorSelector);
        if (empty($ret) === true) {
            $ret = $this->fallbackAdapter->extractAuthor($crawler);
        }
        return $ret;
    }

    public function extractBody(Crawler $crawler)
    {
        $ret = $this->getElementText($crawler, $this->bodySelector);
        return $this->normalizeHtml($ret);
    }

    public function extractDescription(Crawler $crawler)

    {
        $ret = $this->getElementText($crawler, $this->descriptionSelector);
        if (empty($ret) === true) {
            $ret = $this->fallbackAdapter->extractDescription($crawler);
        }
        return $ret;
    }

    public function extractImage(Crawler $crawler)
    {

        if (empty($this->imageSelector) === false) {
            $ret = $this->getSrcByImgSelector($crawler, $this->imageSelector);
        }
        if (empty($ret) === true) {
            $ret = $this->fallbackAdapter->extractImage($crawler);
        }

        if (empty($ret) === false) {
            return $this->normalizeLink($ret);
function myFunction($a) {
    switch ($a) {
        case 'foo':
            $x = 1;
            break;

        case 'bar':
            $x = 2;
            break;
    }

    // $x is potentially undefined here.
    echo $x;
}
        } else {
            return null;
        }
    }

    public function extractKeywords(Crawler $crawler)
    {
        $ret = $this->getElementText($crawler, $this->keywordsSelector);
        if (empty($ret) === true) {
            return $this->fallbackAdapter->extractKeywords($crawler);
        } else {
            return $this->normalizeKeywords(explode(',', $ret));
        }
    }

    public function extractPublishDate(Crawler $crawler)

    {
        $ret = $this->getElementText($crawler, $this->publishDateSelector);
        if (empty($ret) === true) {
            $ret = $this->fallbackAdapter->extractPublishDate($crawler);
        }
        return $ret;
    }

    public function extractTitle(Crawler $crawler)

    {
        $ret = $this->getElementText($crawler, $this->titleSelector);
        if (empty($ret) === true) {
            $ret = $this->fallbackAdapter->extractTitle($crawler);
        }
        return $ret;
    }

    /**
     * getting text of element by selector (css selector or xpath )
     * @param Crawler $crawler
     * @param string $selector
     * @param \Closure $extractClosure callback function to be used for extraction
     * @return string
     */
    protected function getElementText(Crawler $crawler, $selector, $extractClosure = null)
    {

        if (empty($selector) === true) {
            return null;
        }

        $ret = null;
        if ($extractClosure === null) {
            $extractClosure = function(Crawler $node) use (&$ret) {
                $ret = $node->html();
            };
        }
        if (Selector::isCSS($selector)) {
            $crawler->filter($selector)
                    ->each($extractClosure);
        } else {
            $crawler->filterXPath($selector)
                    ->each($extractClosure);
        }

        return $ret;
    }
}


1		<?php
2
3		namespace Zrashwani\NewsScrapper\Adapters;
4
5		use Symfony\Component\DomCrawler\Crawler;
6		use Zrashwani\NewsScrapper\Selector;
7
8		/**
9		* Adapter to extract page data from un-structured HTML document
10		* @author Zeid Rashwani <zrashwani.com>
11		*/
12		class CustomAdapter extends AbstractAdapter
13		{
14
15		private $authorSelector;
16		private $bodySelector;
17		private $descriptionSelector;
18		private $imageSelector;
19		private $keywordsSelector;
20		private $publishDateSelector;
21		private $titleSelector;
22
23		/**
24		* adapter used to fill in the missing selectors data by default values
25		* @var DefaultAdapter $fallbackAdapter
26		*/
27		private $fallbackAdapter;
28
29		public function __construct()
30		{
31		$this->fallbackAdapter = new DefaultAdapter();
32		}
33
34		public function setAuthorSelector($selector)
35		{
36		$this->authorSelector = $selector;
37		return $this;
38		}
39
40		public function setBodySelector($selector)
41		{
42		$this->bodySelector = $selector;
43		return $this;
44		}
45
46		public function setDescriptionSelector($selector)
47		{
48		$this->descriptionSelector = $selector;
49		return $this;
50		}
51
52		public function setImageSelector($selector)
53		{
54		$this->imageSelector = $selector;
55		return $this;
56		}
57
58		public function setKeywordsSelector($selector)
59		{
60		$this->keywordsSelector = $selector;
61		return $this;
62		}
63
64		public function setPublishDateSelector($selector)
65		{
66		$this->publishDateSelector = $selector;
67		return $this;
68		}
69
70		public function setTitleSelector($selector)
71		{
72		$this->titleSelector = $selector;
73		return $this;
74		}
75
76	View Code Duplication	public function extractAuthor(Crawler $crawler)
		0 ignored issues – show Duplication introduced 2017-08-17 08:40 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...
77		{
78		$ret = $this->getElementText($crawler, $this->authorSelector);
79		if (empty($ret) === true) {
80		$ret = $this->fallbackAdapter->extractAuthor($crawler);
81		}
82		return $ret;
83		}
84
85		public function extractBody(Crawler $crawler)
86		{
87		$ret = $this->getElementText($crawler, $this->bodySelector);
88		return $this->normalizeHtml($ret);
89		}
90
91	View Code Duplication	public function extractDescription(Crawler $crawler)
		0 ignored issues – show Duplication introduced 2017-08-17 08:40 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...
92		{
93		$ret = $this->getElementText($crawler, $this->descriptionSelector);
94		if (empty($ret) === true) {
95		$ret = $this->fallbackAdapter->extractDescription($crawler);
96		}
97		return $ret;
98		}
99
100		public function extractImage(Crawler $crawler)
101		{
102
103		if (empty($this->imageSelector) === false) {
104		$ret = $this->getSrcByImgSelector($crawler, $this->imageSelector);
105		}
106		if (empty($ret) === true) {
107		$ret = $this->fallbackAdapter->extractImage($crawler);
108		}
109
110		if (empty($ret) === false) {
111		return $this->normalizeLink($ret);
		0 ignored issues – show Bug introduced 2015-11-28 14:05 UTC by Report Bug Copy Issue Report The variable `$ret` does not seem to be defined for all execution paths leading up to this point. If you define a variable conditionally, it can happen that it is not defined for all execution paths. Let’s take a look at an example: function myFunction($a) { switch ($a) { case 'foo': $x = 1; break; case 'bar': $x = 2; break; } // $x is potentially undefined here. echo $x; } In the above example, the variable `$x` is defined if you pass “foo” or “bar” as argument for `$a`. However, since the `switch` statement has no default case statement, if you pass any other value, the variable `$x` would be undefined. Available Fixes Check for existence of the variable explicitly: function myFunction($a) { switch ($a) { case 'foo': $x = 1; break; case 'bar': $x = 2; break; } if (isset($x)) { // Make sure it's always set. echo $x; } } Define a default value for the variable: function myFunction($a) { $x = ''; // Set a default which gets overridden for certain paths. switch ($a) { case 'foo': $x = 1; break; case 'bar': $x = 2; break; } echo $x; } Add a value for the missing path: function myFunction($a) { switch ($a) { case 'foo': $x = 1; break; case 'bar': $x = 2; break; // We add support for the missing case. default: $x = ''; break; } echo $x; } Loading history...
112		} else {
113		return null;
114		}
115		}
116
117		public function extractKeywords(Crawler $crawler)
118		{
119		$ret = $this->getElementText($crawler, $this->keywordsSelector);
120		if (empty($ret) === true) {
121		return $this->fallbackAdapter->extractKeywords($crawler);
122		} else {
123		return $this->normalizeKeywords(explode(',', $ret));
124		}
125		}
126
127	View Code Duplication	public function extractPublishDate(Crawler $crawler)
		0 ignored issues – show Duplication introduced 2017-08-17 08:40 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...
128		{
129		$ret = $this->getElementText($crawler, $this->publishDateSelector);
130		if (empty($ret) === true) {
131		$ret = $this->fallbackAdapter->extractPublishDate($crawler);
132		}
133		return $ret;
134		}
135
136	View Code Duplication	public function extractTitle(Crawler $crawler)
		0 ignored issues – show Duplication introduced 2015-11-28 14:05 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...
137		{
138		$ret = $this->getElementText($crawler, $this->titleSelector);
139		if (empty($ret) === true) {
140		$ret = $this->fallbackAdapter->extractTitle($crawler);
141		}
142		return $ret;
143		}
144
145		/**
146		* getting text of element by selector (css selector or xpath )
147		* @param Crawler $crawler
148		* @param string $selector
149		* @param \Closure $extractClosure callback function to be used for extraction
150		* @return string
151		*/
152		protected function getElementText(Crawler $crawler, $selector, $extractClosure = null)
153		{
154
155		if (empty($selector) === true) {
156		return null;
157		}
158
159		$ret = null;
160		if ($extractClosure === null) {
161		$extractClosure = function(Crawler $node) use (&$ret) {
162		$ret = $node->html();
163		};
164		}
165		if (Selector::isCSS($selector)) {
166		$crawler->filter($selector)
167		->each($extractClosure);
168		} else {
169		$crawler->filterXPath($selector)
170		->each($extractClosure);
171		}
172
173		return $ret;
174		}
175		}
176

zrashwani / news-scrapper

CustomAdapter::getElementText() B last analyzed 2017-08-17 08:42 UTC

Complexity

Size

Duplication

Importance

Available Fixes

Duplication Side-by-Side

Filter issues like

CustomAdapter::getElementText() B
last analyzed 2017-08-17 08:42 UTC