Crawler::updateInboundLinksCounter() - Code Metrics - Inspection of "Difference inbound nofollow and follow" - PiedWeb/SeoPocketCrawler - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( ff6fd6...9fafdc )

by Dev

created 2019-01-25 08:38 UTC

Crawler::updateInboundLinksCounter() A

↳ Parent: Crawler

Complexity

Conditions	5
Paths	4

Size

Total Lines	14
Code Lines	12

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	10
CRAP Score	5.1158

Importance

Changes

Metric	Value
eloc	12
dl	0
loc	14
ccs	10
cts	12
cp	0.8333
rs	9.5555
c	0
b	0
f	0
cc	5
nc	4
nop	3
crap	5.1158

<?php

namespace PiedWeb\SeoPocketCrawler;

use PiedWeb\UrlHarvester\Harvest;
use PiedWeb\UrlHarvester\Indexable;
use PiedWeb\UrlHarvester\Link;
use Spatie\Robots\RobotsTxt;

class Crawler
{
    const FOLLOW = 1;
    const NOFOLLOW = 2;

    /**
     * @var string contain the user agent used during the crawl
     */
    protected $userAgent;

    /**
     * @var string crawl id
     */
    protected $id;

    /**
     * @var RobotsTxt page to ignore during the crawl
     */
    protected $ignore;

    /**
     * @var int depth max where to crawl
     */
    protected $limit;

    /**
     * @var string contain https://domain.tdl from start url
     */
    protected $base;

    /**
     * @var bool
     */
    protected $fromCache;

    protected $recorder;
    protected $robotsTxt;
    protected $request;
    protected $wait = 0;

    protected $currentClick = 0;

    protected $counter = 0;

    protected $urls = [];

    public function __construct(
        string $startUrl,
        string $ignore,
        int $limit,
        string $userAgent,
        int $cacheMethod = Recorder::CACHE_ID,
        int $waitInMicroSeconds = 100000
    ) {
        $startUrl = $this->setBaseAndReturnNormalizedStartUrl($startUrl);
        $this->urls[$startUrl] = null;
        $this->id = date('ymdHi').'-'.parse_url($this->base, PHP_URL_HOST);
        $this->ignore = new RobotsTxt($ignore);
        $this->userAgent = $userAgent;
        $this->limit = $limit;
        $this->wait = $waitInMicroSeconds;

        $this->recorder = new Recorder($this->getDataFolder(), $cacheMethod);

        file_put_contents($this->getDataFolder().'/config.json', json_encode([
            'startUrl' => $startUrl,
            'base' => $this->base,
            'ignore' => $ignore,
            'limit' => $limit,
            'userAgent' => $userAgent,
            'cacheMethod' => $cacheMethod,
            'wait' => $waitInMicroSeconds,
        ]));
    }

    public function getId()
    {
        return $this->id;
    }

    protected function setBaseAndReturnNormalizedStartUrl(string $url): string
    {
        if (!filter_var($url, FILTER_VALIDATE_URL)) {
            throw new \Exception('start is not a valid URL `'.$url.'`');
        }

        $this->base = preg_match('@^(http://|https://)?[^/\?#]+@', $url, $match) ? $match[0] : $url;
        $url = substr($url, strlen($this->base));

        return ('/' != $url[0] ? '/' : '').$url;
    }

    public function getDataFolder()
    {
        return __DIR__.'/../data/'.$this->id;
    }

    public function crawl(bool $debug = false)
    {
        $nothingUpdated = true;

        if ($debug) {
            echo PHP_EOL.PHP_EOL.'// -----'.PHP_EOL.'// '.$this->counter.' crawled / '
                        .count($this->urls).' found '.PHP_EOL.'// -----'.PHP_EOL;
        }

        foreach ($this->urls as $urlToParse => $url) {
            if (null !== $url && (false === $url->can_be_crawled || true === $url->can_be_crawled)) { // déjà crawlé
                continue;
            } elseif ($this->currentClick > $this->limit) {
                continue;
            }

            if ($debug) {
                echo $this->counter.'/'.count($this->urls).'    '.$this->base.$urlToParse.PHP_EOL;
            }

            $nothingUpdated = false;
            ++$this->counter;

            $harvest = $this->harvest($urlToParse);
            $this->urls[$urlToParse]->setDiscovered(count($this->urls));

            $this->cacheRobotsTxt($harvest);

            $this->cacheRequest($harvest);

            usleep($this->wait);

            if ($this->counter / 500 == round($this->counter / 500)) {
                echo $debug ? '    --- auto-save'.PHP_EOL : '';
                $this->recorder->record($this->urls);
            }
        }

        ++$this->currentClick;

        // Record after each Level:
        $this->recorder->record($this->urls);

        $record = $nothingUpdated || $this->currentClick >= $this->limit;

        return $record ? null : $this->crawl($debug);
    }

    protected function cacheRobotsTxt($harvest)
    {
        if (null === $this->robotsTxt && $harvest instanceof Harvest) {
            $this->robotsTxt = $harvest->getRobotsTxt();
        }

        return $this;
    }

    protected function cacheRequest($harvest)
    {
        if ($harvest instanceof Harvest && null !== $harvest->getResponse()->getRequest()) {
            $this->request = $harvest->getResponse()->getRequest();
        }

        return $this;
    }

    protected function loadRobotsTxt(Harvest $harvest)
    {
        if (null !== $this->robotsTxt) {
            $harvest->setRobotsTxt($this->robotsTxt);
        }

        return $this;
    }

    protected function getHarvest(Url $url)
    {
        return Harvest::fromUrl(
            $this->base.$url->uri,
            $this->userAgent,
            'en,en-US;q=0.5',
            $this->request
        );
    }

    protected function harvest(string $urlToParse)
    {
        $this->urls[$urlToParse] = $this->urls[$urlToParse] ?? new Url($this->base.$urlToParse, $this->currentClick);
        $url = $this->urls[$urlToParse];

        $url->can_be_crawled = $this->ignore->allows($this->base.$urlToParse, $this->userAgent);

        if (false === $url->can_be_crawled) {
            return;
        }

        $harvest = $this->getHarvest($url);

        if (!$harvest instanceof Harvest) {
            $url->indexable = Indexable::NOT_INDEXABLE_NETWORK_ERROR;

            return;
        }

        $this->loadRobotsTxt($harvest);

        $url->indexable = $harvest->isIndexable(); // slow ~30%

        if (Indexable::NOT_INDEXABLE_3XX === $url->indexable) {
            $redir = $harvest->getRedirection();
            if (false !== $redir) {
                $links = Harvest::LINK_INTERNAL === $harvest->getType($redir) ? [new Link($redir)] : [];
            }
        } else {
            $this->recorder->cache($harvest, $url);

            $mimeType = $harvest->getResponse()->getMimeType();
            $url->mime_type = 'text/html' == $mimeType ? 1 : $mimeType;

            $this->recorder->recordOutboundLink($url, $harvest->getLinks()); // ~10%
            $url->links = count($harvest->getLinks());
            $url->links_duplicate = $harvest->getNbrDuplicateLinks();
            $url->links_internal = count($harvest->getLinks(Harvest::LINK_INTERNAL));
            $url->links_self = count($harvest->getLinks(Harvest::LINK_SELF));
            $url->links_sub = count($harvest->getLinks(Harvest::LINK_SUB));
            $url->links_external = count($harvest->getLinks(Harvest::LINK_EXTERNAL));
            $links = $harvest->getLinks(Harvest::LINK_INTERNAL);

            //$url->ratio_text_code = $harvest->getRatioTxtCode(); // Slow ~30%
            $url->words_count = $harvest->getTextAnalysis()->getWordNumber();
            $url->load_time = $harvest->getResponse()->getInfo('total_time');
            $url->size = $harvest->getResponse()->getInfo('size_download');

            $breadcrumb = $harvest->getBreadCrumb();
            if (is_array($breadcrumb)) {
                $url->breadcrumb_level = count($breadcrumb);
                $url->breadcrumb_first = isset($breadcrumb[1]) ? $breadcrumb[1]->getCleanName() : '';
                $url->breadcrumb_text = $harvest->getBreadCrumb('//');
            }

            $url->title = $harvest->getUniqueTag('head title') ?? '';
            $url->kws = ','.implode(',', array_keys($harvest->getKws())).','; // Slow ~20%
            $url->h1 = $harvest->getUniqueTag('h1') ?? '';
            $url->h1 = $url->title == $url->h1 ? '=' : $url->h1;
        }

        if (isset($links)) {
            $this->updateInboundLinksCounter($url, $links, $harvest);
        }

        return $harvest;
    }

    public function updateInboundLinksCounter(Url $url, array $links, Harvest $harvest)
    {
        $everAdd = [];
        foreach ($links as $link) {
            $newUri = substr($link->getPageUrl(), strlen($this->base));
            $this->urls[$newUri] = $this->urls[$newUri] ?? new Url($link->getPageUrl(), ($this->currentClick + 1));
            if (!isset($everAdd[$newUri])) {
                $everAdd[$newUri] = 1;
                if (!$link->mayFollow() || !$harvest->mayFollow()) {
                    ++$this->urls[$newUri]->inboundlinks_nofollow;
                    $this->recorder->recordInboundLink($url, $this->urls[$newUri], self::NOFOLLOW);
                } else {
                    ++$this->urls[$newUri]->inboundlinks;
                    $this->recorder->recordInboundLink($url, $this->urls[$newUri], self::FOLLOW);
                }
            }
        }
    }
}


1		<?php
2
3		namespace PiedWeb\SeoPocketCrawler;
4
5		use PiedWeb\UrlHarvester\Harvest;
6		use PiedWeb\UrlHarvester\Indexable;
7		use PiedWeb\UrlHarvester\Link;
8		use Spatie\Robots\RobotsTxt;
9
10		class Crawler
11		{
12		const FOLLOW = 1;
13		const NOFOLLOW = 2;
14
15		/**
16		* @var string contain the user agent used during the crawl
17		*/
18		protected $userAgent;
19
20		/**
21		* @var string crawl id
22		*/
23		protected $id;
24
25		/**
26		* @var RobotsTxt page to ignore during the crawl
27		*/
28		protected $ignore;
29
30		/**
31		* @var int depth max where to crawl
32		*/
33		protected $limit;
34
35		/**
36		* @var string contain https://domain.tdl from start url
37		*/
38		protected $base;
39
40		/**
41		* @var bool
42		*/
43		protected $fromCache;
44
45		protected $recorder;
46		protected $robotsTxt;
47		protected $request;
48		protected $wait = 0;
49
50		protected $currentClick = 0;
51
52		protected $counter = 0;
53
54		protected $urls = [];
55
56	6	public function __construct(
57		string $startUrl,
58		string $ignore,
59		int $limit,
60		string $userAgent,
61		int $cacheMethod = Recorder::CACHE_ID,
62		int $waitInMicroSeconds = 100000
63		) {
64	6	$startUrl = $this->setBaseAndReturnNormalizedStartUrl($startUrl);
65	6	$this->urls[$startUrl] = null;
66	6	$this->id = date('ymdHi').'-'.parse_url($this->base, PHP_URL_HOST);
67	6	$this->ignore = new RobotsTxt($ignore);
68	6	$this->userAgent = $userAgent;
69	6	$this->limit = $limit;
70	6	$this->wait = $waitInMicroSeconds;
71
72	6	$this->recorder = new Recorder($this->getDataFolder(), $cacheMethod);
73
74	6	file_put_contents($this->getDataFolder().'/config.json', json_encode([
75	6	'startUrl' => $startUrl,
76	6	'base' => $this->base,
77	6	'ignore' => $ignore,
78	6	'limit' => $limit,
79	6	'userAgent' => $userAgent,
80	6	'cacheMethod' => $cacheMethod,
81	6	'wait' => $waitInMicroSeconds,
82		]));
83	6	}
84
85	3	public function getId()
86		{
87	3	return $this->id;
88		}
89
90	6	protected function setBaseAndReturnNormalizedStartUrl(string $url): string
91		{
92	6	if (!filter_var($url, FILTER_VALIDATE_URL)) {
93		throw new \Exception('start is not a valid URL `'.$url.'`');
94		}
95
96	6	$this->base = preg_match('@^(http://\|https://)?[^/\?#]+@', $url, $match) ? $match[0] : $url;
97	6	$url = substr($url, strlen($this->base));
98
99	6	return ('/' != $url[0] ? '/' : '').$url;
100		}
101
102	6	public function getDataFolder()
103		{
104	6	return __DIR__.'/../data/'.$this->id;
105		}
106
107	6	public function crawl(bool $debug = false)
108		{
109	6	$nothingUpdated = true;
110
111	6	if ($debug) {
112	6	echo PHP_EOL.PHP_EOL.'// -----'.PHP_EOL.'// '.$this->counter.' crawled / '
113	6	.count($this->urls).' found '.PHP_EOL.'// -----'.PHP_EOL;
114		}
115
116	6	foreach ($this->urls as $urlToParse => $url) {
117	6	if (null !== $url && (false === $url->can_be_crawled \|\| true === $url->can_be_crawled)) { // déjà crawlé
118	3	continue;
119	6	} elseif ($this->currentClick > $this->limit) {
120	2	continue;
121		}
122
123	6	if ($debug) {
124	6	echo $this->counter.'/'.count($this->urls).' '.$this->base.$urlToParse.PHP_EOL;
125		}
126
127	6	$nothingUpdated = false;
128	6	++$this->counter;
129
130	6	$harvest = $this->harvest($urlToParse);
131	6	$this->urls[$urlToParse]->setDiscovered(count($this->urls));
132
133	6	$this->cacheRobotsTxt($harvest);
134
135	6	$this->cacheRequest($harvest);
136
137	6	usleep($this->wait);
138
139	6	if ($this->counter / 500 == round($this->counter / 500)) {
140		echo $debug ? ' --- auto-save'.PHP_EOL : '';
141	4	$this->recorder->record($this->urls);
142		}
143		}
144
145	6	++$this->currentClick;
146
147		// Record after each Level:
148	6	$this->recorder->record($this->urls);
149
150	6	$record = $nothingUpdated \|\| $this->currentClick >= $this->limit;
151
152	6	return $record ? null : $this->crawl($debug);
153		}
154
155	6	protected function cacheRobotsTxt($harvest)
156		{
157	6	if (null === $this->robotsTxt && $harvest instanceof Harvest) {
158	4	$this->robotsTxt = $harvest->getRobotsTxt();
159		}
160
161	6	return $this;
162		}
163
164	6	protected function cacheRequest($harvest)
165		{
166	6	if ($harvest instanceof Harvest && null !== $harvest->getResponse()->getRequest()) {
167	4	$this->request = $harvest->getResponse()->getRequest();
168		}
169
170	6	return $this;
171		}
172
173	4	protected function loadRobotsTxt(Harvest $harvest)
174		{
175	4	if (null !== $this->robotsTxt) {
176		$harvest->setRobotsTxt($this->robotsTxt);
177		}
178
179	4	return $this;
180		}
181
182	6	protected function getHarvest(Url $url)
183		{
184	6	return Harvest::fromUrl(
185	6	$this->base.$url->uri,
186	6	$this->userAgent,
187	6	'en,en-US;q=0.5',
188	6	$this->request
189		);
190		}
191
192	6	protected function harvest(string $urlToParse)
193		{
194	6	$this->urls[$urlToParse] = $this->urls[$urlToParse] ?? new Url($this->base.$urlToParse, $this->currentClick);
195	6	$url = $this->urls[$urlToParse];
196
197	6	$url->can_be_crawled = $this->ignore->allows($this->base.$urlToParse, $this->userAgent);
198
199	6	if (false === $url->can_be_crawled) {
200		return;
201		}
202
203	6	$harvest = $this->getHarvest($url);
204
205	6	if (!$harvest instanceof Harvest) {
206	2	$url->indexable = Indexable::NOT_INDEXABLE_NETWORK_ERROR;
207
208	2	return;
209		}
210
211	4	$this->loadRobotsTxt($harvest);
212
213	4	$url->indexable = $harvest->isIndexable(); // slow ~30%
214
215	4	if (Indexable::NOT_INDEXABLE_3XX === $url->indexable) {
216		$redir = $harvest->getRedirection();
217		if (false !== $redir) {
218		$links = Harvest::LINK_INTERNAL === $harvest->getType($redir) ? [new Link($redir)] : [];
219		}
220		} else {
221	4	$this->recorder->cache($harvest, $url);
222
223	4	$mimeType = $harvest->getResponse()->getMimeType();
224	4	$url->mime_type = 'text/html' == $mimeType ? 1 : $mimeType;
225
226	4	$this->recorder->recordOutboundLink($url, $harvest->getLinks()); // ~10%
227	4	$url->links = count($harvest->getLinks());
228	4	$url->links_duplicate = $harvest->getNbrDuplicateLinks();
229	4	$url->links_internal = count($harvest->getLinks(Harvest::LINK_INTERNAL));
230	4	$url->links_self = count($harvest->getLinks(Harvest::LINK_SELF));
231	4	$url->links_sub = count($harvest->getLinks(Harvest::LINK_SUB));
232	4	$url->links_external = count($harvest->getLinks(Harvest::LINK_EXTERNAL));
233	4	$links = $harvest->getLinks(Harvest::LINK_INTERNAL);
234
235		//$url->ratio_text_code = $harvest->getRatioTxtCode(); // Slow ~30%
236	4	$url->words_count = $harvest->getTextAnalysis()->getWordNumber();
237	4	$url->load_time = $harvest->getResponse()->getInfo('total_time');
238	4	$url->size = $harvest->getResponse()->getInfo('size_download');
239
240	4	$breadcrumb = $harvest->getBreadCrumb();
241	4	if (is_array($breadcrumb)) {
242		$url->breadcrumb_level = count($breadcrumb);
243		$url->breadcrumb_first = isset($breadcrumb[1]) ? $breadcrumb[1]->getCleanName() : '';
244		$url->breadcrumb_text = $harvest->getBreadCrumb('//');
245		}
246
247	4	$url->title = $harvest->getUniqueTag('head title') ?? '';
248	4	$url->kws = ','.implode(',', array_keys($harvest->getKws())).','; // Slow ~20%
249	4	$url->h1 = $harvest->getUniqueTag('h1') ?? '';
250	4	$url->h1 = $url->title == $url->h1 ? '=' : $url->h1;
251		}
252
253	4	if (isset($links)) {
254	4	$this->updateInboundLinksCounter($url, $links, $harvest);
255		}
256
257	4	return $harvest;
258		}
259
260	4	public function updateInboundLinksCounter(Url $url, array $links, Harvest $harvest)
261		{
262	4	$everAdd = [];
263	4	foreach ($links as $link) {
264	4	$newUri = substr($link->getPageUrl(), strlen($this->base));
265	4	$this->urls[$newUri] = $this->urls[$newUri] ?? new Url($link->getPageUrl(), ($this->currentClick + 1));
266	4	if (!isset($everAdd[$newUri])) {
267	4	$everAdd[$newUri] = 1;
268	4	if (!$link->mayFollow() \|\| !$harvest->mayFollow()) {
269		++$this->urls[$newUri]->inboundlinks_nofollow;
270		$this->recorder->recordInboundLink($url, $this->urls[$newUri], self::NOFOLLOW);
271		} else {
272	4	++$this->urls[$newUri]->inboundlinks;
273	4	$this->recorder->recordInboundLink($url, $this->urls[$newUri], self::FOLLOW);
274		}
275		}
276		}
277	4	}
278		}
279

PiedWeb / SeoPocketCrawler

Push — master ( ff6fd6...9fafdc )

Crawler::updateInboundLinksCounter() A

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like