PdoCrawlQueue::hasAlreadyBeenProcessed() - Code Metrics - Inspection of "Add change crawl queue support" - spatie/crawler - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Pull Request — master (#87)

by Peter

created 2017-12-08 22:08 UTC

PdoCrawlQueue::hasAlreadyBeenProcessed() A

↳ Parent: PdoCrawlQueue

Complexity

Conditions	1
Paths	1

Size

Total Lines	14
Code Lines	9

Duplication

Lines	13
Ratio	92.86 %

Importance

Changes

Metric	Value
dl	13
loc	14
rs	9.4285
c	0
b	0
f	0
cc	1
eloc	9
nc	1
nop	1

<?php

namespace Spatie\Crawler;

use Spatie\Crawler\Exception\UrlNotFoundByIndex;

class PdoCrawlQueue implements CrawlQueue
{
    const STATUS_PENDING = 1;
    const STATUS_PROCESSED = 2;

    /** @var \Pdo */
    private $pdo;

    /** @var string */
    private $table;

    public function __construct(\Pdo $pdo, string $table)
    {
        $this->pdo = $pdo;
        $this->table = $table;
    }

    public function add(CrawlUrl $url): CrawlQueue
    {
        if ($this->has($url)) {
            return $this;
        }

        $data = serialize($url);

        $this->pdo->beginTransaction();

        try {
            $statement = $this->pdo->prepare(
                'insert into '.$this->table.' (url, status, object) values (:url, :status, :object)'
            );
            $statement->execute([
                'url' => $url->url,
                'object' => $data,
                'status' => self::STATUS_PENDING,
            ]);
            $statement->closeCursor();
            $url->setId($this->pdo->lastInsertId());

            $statement = $this->pdo->prepare(
                'update '.$this->table.' set object = :object where id = :id'
            );

            $data = serialize($url);

            $statement->execute([
                'object' => $data,
                'id' => $url->getId(),
            ]);
            $statement->closeCursor();
            $this->pdo->commit();
        } catch (\Exception $exception) {
            $this->pdo->rollBack();
            throw $exception;
        }

        return $this;
    }

    public function hasPendingUrls(): bool
    {
        $statement = $this->pdo->prepare(
            'select url from '.$this->table.' where status = :status limit 1'
        );
        $statement->execute([
            'status' => self::STATUS_PENDING,
        ]);
        $result = $statement->fetch();
        $statement->closeCursor();

        return (bool) $result;
    }

    /**
     * @param int $id
     *
     * @return \Spatie\Crawler\CrawlUrl|null
     */
    public function getUrlById(int $id): CrawlUrl
    {
        $statement = $this->pdo->prepare(
            'select object from '.$this->table.' where id = :id'
        );
        $statement->execute([
            'id' => $id,
        ]);
        $result = $statement->fetch();
        $statement->closeCursor();

        if (! $result) {
            throw new UrlNotFoundByIndex(sprintf('#%d crawl url not found in database', $id));
        }

        return unserialize($result['object']);
    }

    public function hasAlreadyBeenProcessed(CrawlUrl $url)

    {
        $statement = $this->pdo->prepare(
            'select id from '.$this->table.' where url = :url and status = :status'
        );
        $statement->execute([
            'url' => $url->url,
            'status' => self::STATUS_PROCESSED,
        ]);
        $result = $statement->fetch();
        $statement->closeCursor();

        return (bool) $result;
    }

    public function markAsProcessed(CrawlUrl $crawlUrl)
    {
        $statement = $this->pdo->prepare(
            'update  '.$this->table.' set status = :status where url = :url'
        );
        $statement->execute([
            'url' => $crawlUrl->url,
            'status' => self::STATUS_PROCESSED,
        ]);
        $statement->closeCursor();
    }

    /**
     * @param CrawlUrl|Url $crawlUrl
     *
     * @return bool
     */
    public function has($crawlUrl): bool
    {
        if ($crawlUrl instanceof Url) {
            $crawlUrl = CrawlUrl::create($crawlUrl);
        }

        $statement = $this->pdo->prepare(
            'select id from '.$this->table.' where url = :url'
        );

        $statement->execute([
            'url' => $crawlUrl->url,
        ]);
        $result = $statement->fetch();

        return (bool) $result;
    }

    /** @return \Spatie\Crawler\CrawlUrl|null */
    public function getPendingUrl()

    {
        $statement = $this->pdo->prepare(
            'select object from '.$this->table.' where status = :status limit 1'
        );
        $statement->execute([
            'status' => self::STATUS_PENDING,
        ]);
        $result = $statement->fetch();
        $statement->closeCursor();

        if (! $result) {
            return;
        }

        return unserialize($result['object']);
    }
}


1		<?php
2
3		namespace Spatie\Crawler;
4
5		use Spatie\Crawler\Exception\UrlNotFoundByIndex;
6
7		class PdoCrawlQueue implements CrawlQueue
8		{
9		const STATUS_PENDING = 1;
10		const STATUS_PROCESSED = 2;
11
12		/** @var \Pdo */
13		private $pdo;
14
15		/** @var string */
16		private $table;
17
18		public function __construct(\Pdo $pdo, string $table)
19		{
20		$this->pdo = $pdo;
21		$this->table = $table;
22		}
23
24		public function add(CrawlUrl $url): CrawlQueue
25		{
26		if ($this->has($url)) {
27		return $this;
28		}
29
30		$data = serialize($url);
31
32		$this->pdo->beginTransaction();
33
34		try {
35		$statement = $this->pdo->prepare(
36		'insert into '.$this->table.' (url, status, object) values (:url, :status, :object)'
37		);
38		$statement->execute([
39		'url' => $url->url,
40		'object' => $data,
41		'status' => self::STATUS_PENDING,
42		]);
43		$statement->closeCursor();
44		$url->setId($this->pdo->lastInsertId());
45
46		$statement = $this->pdo->prepare(
47		'update '.$this->table.' set object = :object where id = :id'
48		);
49
50		$data = serialize($url);
51
52		$statement->execute([
53		'object' => $data,
54		'id' => $url->getId(),
55		]);
56		$statement->closeCursor();
57		$this->pdo->commit();
58		} catch (\Exception $exception) {
59		$this->pdo->rollBack();
60		throw $exception;
61		}
62
63		return $this;
64		}
65
66		public function hasPendingUrls(): bool
67		{
68		$statement = $this->pdo->prepare(
69		'select url from '.$this->table.' where status = :status limit 1'
70		);
71		$statement->execute([
72		'status' => self::STATUS_PENDING,
73		]);
74		$result = $statement->fetch();
75		$statement->closeCursor();
76
77		return (bool) $result;
78		}
79
80		/**
81		* @param int $id
82		*
83		* @return \Spatie\Crawler\CrawlUrl\|null
84		*/
85		public function getUrlById(int $id): CrawlUrl
86		{
87		$statement = $this->pdo->prepare(
88		'select object from '.$this->table.' where id = :id'
89		);
90		$statement->execute([
91		'id' => $id,
92		]);
93		$result = $statement->fetch();
94		$statement->closeCursor();
95
96		if (! $result) {
97		throw new UrlNotFoundByIndex(sprintf('#%d crawl url not found in database', $id));
98		}
99
100		return unserialize($result['object']);
101		}
102
103	View Code Duplication	public function hasAlreadyBeenProcessed(CrawlUrl $url)
		0 ignored issues – show Duplication introduced 2017-12-08 22:05 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...
104		{
105		$statement = $this->pdo->prepare(
106		'select id from '.$this->table.' where url = :url and status = :status'
107		);
108		$statement->execute([
109		'url' => $url->url,
110		'status' => self::STATUS_PROCESSED,
111		]);
112		$result = $statement->fetch();
113		$statement->closeCursor();
114
115		return (bool) $result;
116		}
117
118		public function markAsProcessed(CrawlUrl $crawlUrl)
119		{
120		$statement = $this->pdo->prepare(
121		'update '.$this->table.' set status = :status where url = :url'
122		);
123		$statement->execute([
124		'url' => $crawlUrl->url,
125		'status' => self::STATUS_PROCESSED,
126		]);
127		$statement->closeCursor();
128		}
129
130		/**
131		* @param CrawlUrl\|Url $crawlUrl
132		*
133		* @return bool
134		*/
135		public function has($crawlUrl): bool
136		{
137		if ($crawlUrl instanceof Url) {
138		$crawlUrl = CrawlUrl::create($crawlUrl);
139		}
140
141		$statement = $this->pdo->prepare(
142		'select id from '.$this->table.' where url = :url'
143		);
144
145		$statement->execute([
146		'url' => $crawlUrl->url,
147		]);
148		$result = $statement->fetch();
149
150		return (bool) $result;
151		}
152
153		/** @return \Spatie\Crawler\CrawlUrl\|null */
154	View Code Duplication	public function getPendingUrl()
		0 ignored issues – show Duplication introduced 2017-12-08 22:05 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...
155		{
156		$statement = $this->pdo->prepare(
157		'select object from '.$this->table.' where status = :status limit 1'
158		);
159		$statement->execute([
160		'status' => self::STATUS_PENDING,
161		]);
162		$result = $statement->fetch();
163		$statement->closeCursor();
164
165		if (! $result) {
166		return;
167		}
168
169		return unserialize($result['object']);
170		}
171		}
172

spatie / crawler

Pull Request — master (#87)

PdoCrawlQueue::hasAlreadyBeenProcessed() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like