crawler-map.cls.php — LiteSpeed Cache 7.8.1

litespeed-cache / src / crawler-map.cls.php

litespeed-cache / src Last commit date

crawler-map.cls.php

668 lines

1	<?php
2	/**
3	* The Crawler Sitemap Class.
4	*
5	* @package LiteSpeed
6	* @since 1.1.0
7	*/
8
9	namespace LiteSpeed;
10
11	defined( 'WPINC' ) \|\| exit();
12
13	/**
14	* Class Crawler_Map
15	*
16	* Maintains and persists crawler sitemap/blacklist state, parses custom sitemaps,
17	* and exposes helpers to query & mutate crawler results.
18	*/
19	class Crawler_Map extends Root {
20
21	const LOG_TAG = '🐞🗺️';
22
23	const BM_MISS = 1;
24	const BM_HIT = 2;
25	const BM_BLACKLIST = 4;
26
27	/**
28	* Site URL used to simplify URLs.
29	*
30	* @var string
31	*/
32	private $_site_url;
33
34	/**
35	* Main crawler table name.
36	*
37	* @var string
38	*/
39	private $_tb;
40
41	/**
42	* Crawler blacklist table name.
43	*
44	* @var string
45	*/
46	private $_tb_blacklist;
47
48	/**
49	* Data service instance.
50	*
51	* @var \LiteSpeed\Data
52	*/
53	private $__data;
54
55	/**
56	* Timeout (seconds) when fetching sitemaps.
57	*
58	* @var int
59	*/
60	private $_conf_map_timeout;
61
62	/**
63	* Collected URLs from parsed sitemaps.
64	*
65	* @var array<int,string>
66	*/
67	private $_urls = [];
68
69	/**
70	* Instantiate the class.
71	*
72	* @since 1.1.0
73	*/
74	public function __construct() {
75	$this->_site_url = get_site_url();
76	$this->__data = Data::cls();
77	$this->_tb = $this->__data->tb( 'crawler' );
78	$this->_tb_blacklist = $this->__data->tb( 'crawler_blacklist' );
79	// Specify the timeout while parsing the sitemap.
80	$this->_conf_map_timeout = defined( 'LITESPEED_CRAWLER_MAP_TIMEOUT' ) ? constant( 'LITESPEED_CRAWLER_MAP_TIMEOUT' ) : 180;
81	}
82
83	/**
84	* Save URLs crawl status into DB.
85	*
86	* @since 3.0
87	* @access public
88	*
89	* @param array<int,array<int,array{url:string,code:int}>> $items Map of bit => [ id => [url, code] ].
90	* @param int $curr_crawler Current crawler index (0-based).
91	* @return array<int,array>
92	*/
93	public function save_map_status( $items, $curr_crawler ) {
94	global $wpdb;
95	Utility::compatibility();
96
97	$total_crawler = count( Crawler::cls()->list_crawlers() );
98	$total_crawler_pos = $total_crawler - 1;
99
100	// Replace current crawler's position.
101	$curr_crawler = (int) $curr_crawler;
102	foreach ( $items as $bit => $ids ) {
103	// $ids = [ id => [ url, code ], ... ].
104	if ( ! $ids ) {
105	continue;
106	}
107	self::debug( 'Update map [crawler] ' . $curr_crawler . ' [bit] ' . $bit . ' [count] ' . count( $ids ) );
108
109	// Update res first, then reason
110	$right_pos = $total_crawler_pos - $curr_crawler;
111	$id_all = implode(',', array_map('intval', array_keys($ids)));
112
113	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.InterpolatedNotPrepared
114	$wpdb->query("UPDATE `$this->_tb` SET res = CONCAT( LEFT( res, $curr_crawler ), '$bit', RIGHT( res, $right_pos ) ) WHERE id IN ( $id_all )");
115
116	// Add blacklist
117	if (Crawler::STATUS_BLACKLIST === $bit \|\| Crawler::STATUS_NOCACHE === $bit) {
118	$q = "SELECT a.id, a.url FROM `$this->_tb_blacklist` a LEFT JOIN `$this->_tb` b ON b.url=a.url WHERE b.id IN ( $id_all )";
119	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
120	$existing = $wpdb->get_results($q, ARRAY_A);
121	// Update current crawler status tag in existing blacklist
122	if ($existing) {
123	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.InterpolatedNotPrepared, WordPress.DB.PreparedSQL.NotPrepared
124	$count = $wpdb->query("UPDATE `$this->_tb_blacklist` SET res = CONCAT( LEFT( res, $curr_crawler ), '$bit', RIGHT( res, $right_pos ) ) WHERE id IN ( " . implode(',', array_column($existing, 'id')) . ' )');
125	self::debug('Update blacklist [count] ' . $count);
126	}
127
128	// Append new blacklist
129	if (count($ids) > count($existing)) {
130	$new_urls = array_diff(array_column($ids, 'url'), array_column($existing, 'url'));
131
132	self::debug('Insert into blacklist [count] ' . count($new_urls));
133
134	$q = "INSERT INTO `$this->_tb_blacklist` ( url, res, reason ) VALUES " . implode(',', array_fill(0, count($new_urls), '( %s, %s, %s )'));
135	$data = [];
136	$res = array_fill(0, $total_crawler, '-');
137	$res[$curr_crawler] = $bit;
138	$res = implode('', $res);
139	$default_reason = $total_crawler > 1 ? str_repeat(',', $total_crawler - 1) : ''; // Pre-populate default reason value first, update later
140	foreach ($new_urls as $url) {
141	$data[] = $url;
142	$data[] = $res;
143	$data[] = $default_reason;
144	}
145	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
146	$wpdb->query($wpdb->prepare($q, $data));
147	}
148	}
149
150	// Update sitemap reason w/ HTTP code.
151	$reason_array = [];
152	foreach ( $ids as $row_id => $row ) {
153	$code = (int) $row['code'];
154	if ( empty( $reason_array[ $code ] ) ) {
155	$reason_array[ $code ] = [];
156	}
157	$reason_array[ $code ][] = (int) $row_id;
158	}
159
160	foreach ($reason_array as $code => $v2) {
161	// Complement comma
162	if ($curr_crawler) {
163	$code = ',' . $code;
164	}
165	if ($curr_crawler < $total_crawler_pos) {
166	$code .= ',';
167	}
168
169	// phpcs:ignore WordPress.DB
170	$count = $wpdb->query( "UPDATE `$this->_tb` SET reason=CONCAT(SUBSTRING_INDEX(reason, ',', $curr_crawler), '$code', SUBSTRING_INDEX(reason, ',', -$right_pos)) WHERE id IN (" . implode(',', $v2) . ')' );
171
172	self::debug("Update map reason [code] $code [pos] left $curr_crawler right -$right_pos [count] $count");
173
174	// Update blacklist reason
175	if (Crawler::STATUS_BLACKLIST === $bit \|\| Crawler::STATUS_NOCACHE === $bit) {
176	// phpcs:ignore WordPress.DB
177	$count = $wpdb->query( "UPDATE `$this->_tb_blacklist` a LEFT JOIN `$this->_tb` b ON b.url = a.url SET a.reason=CONCAT(SUBSTRING_INDEX(a.reason, ',', $curr_crawler), '$code', SUBSTRING_INDEX(a.reason, ',', -$right_pos)) WHERE b.id IN (" . implode(',', $v2) . ')' );
178
179	self::debug("Update blacklist [code] $code [pos] left $curr_crawler right -$right_pos [count] $count");
180	}
181	}
182
183	// Reset list.
184	$items[ $bit ] = [];
185	}
186
187	return $items;
188	}
189
190	/**
191	* Add one record to blacklist.
192	* NOTE: $id is sitemap table ID.
193	*
194	* @since 3.0
195	* @access public
196	*
197	* @param int $id Sitemap row ID.
198	* @return void
199	*/
200	public function blacklist_add( $id ) {
201	global $wpdb;
202
203	$id = (int) $id;
204
205	// Build res&reason.
206	$total_crawler = count( Crawler::cls()->list_crawlers() );
207	$res = str_repeat(Crawler::STATUS_BLACKLIST, $total_crawler);
208	$reason = implode(',', array_fill(0, $total_crawler, 'Man'));
209
210	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.InterpolatedNotPrepared
211	$row = $wpdb->get_row("SELECT a.url, b.id FROM `$this->_tb` a LEFT JOIN `$this->_tb_blacklist` b ON b.url = a.url WHERE a.id = '$id'", ARRAY_A);
212	if (!$row) {
213	self::debug('blacklist failed to add [id] ' . $id);
214	return;
215	}
216
217	self::debug('Add to blacklist [url] ' . $row['url']);
218
219	$q = "UPDATE `$this->_tb` SET res = %s, reason = %s WHERE id = %d";
220	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
221	$wpdb->query($wpdb->prepare($q, [ $res, $reason, $id ]));
222
223	if ($row['id']) {
224	$q = "UPDATE `$this->_tb_blacklist` SET res = %s, reason = %s WHERE id = %d";
225	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
226	$wpdb->query($wpdb->prepare($q, [ $res, $reason, $row['id'] ]));
227	} else {
228	$q = "INSERT INTO `$this->_tb_blacklist` (url, res, reason) VALUES (%s, %s, %s)";
229	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
230	$wpdb->query($wpdb->prepare($q, [ $row['url'], $res, $reason ]));
231	}
232	}
233
234	/**
235	* Delete one record from blacklist.
236	*
237	* @since 3.0
238	* @access public
239	*
240	* @param int $id Blacklist row ID.
241	* @return void
242	*/
243	public function blacklist_del( $id ) {
244	global $wpdb;
245	if ( ! $this->__data->tb_exist( 'crawler_blacklist' ) ) {
246	return;
247	}
248
249	$id = (int) $id;
250	self::debug('blacklist delete [id] ' . $id);
251
252	$sql = sprintf(
253	"UPDATE `%s` SET res=REPLACE(REPLACE(res, '%s', '-'), '%s', '-') WHERE url=(SELECT url FROM `%s` WHERE id=%d)",
254	$this->_tb,
255	Crawler::STATUS_NOCACHE,
256	Crawler::STATUS_BLACKLIST,
257	$this->_tb_blacklist,
258	$id
259	);
260	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
261	$wpdb->query($sql);
262	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.InterpolatedNotPrepared
263	$wpdb->query("DELETE FROM `$this->_tb_blacklist` WHERE id='$id'");
264	}
265
266	/**
267	* Empty blacklist.
268	*
269	* @since 3.0
270	* @access public
271	* @return void
272	*/
273	public function blacklist_empty() {
274	global $wpdb;
275
276	if ( ! $this->__data->tb_exist( 'crawler_blacklist' ) ) {
277	return;
278	}
279
280	self::debug('Truncate blacklist');
281	$sql = sprintf("UPDATE `%s` SET res=REPLACE(REPLACE(res, '%s', '-'), '%s', '-')", $this->_tb, Crawler::STATUS_NOCACHE, Crawler::STATUS_BLACKLIST);
282	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
283	$wpdb->query($sql);
284	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.InterpolatedNotPrepared
285	$wpdb->query("TRUNCATE `$this->_tb_blacklist`");
286	}
287
288	/**
289	* List blacklist.
290	*
291	* @since 3.0
292	* @access public
293	*
294	* @param int\|false $limit Number of rows to fetch, or false for all.
295	* @param int\|false $offset Offset for pagination, or false to auto-calc.
296	* @return array<int,array<string,mixed>>
297	*/
298	public function list_blacklist( $limit = false, $offset = false ) {
299	global $wpdb;
300
301	if ( ! $this->__data->tb_exist( 'crawler_blacklist' ) ) {
302	return [];
303	}
304
305	$q = "SELECT * FROM `$this->_tb_blacklist` ORDER BY id DESC";
306
307	if ( false !== $limit ) {
308	if ( false === $offset ) {
309	$total = $this->count_blacklist();
310	$offset = Utility::pagination($total, $limit, true);
311	}
312	$q .= ' LIMIT %d, %d';
313	// phpcs:ignore WordPress.DB.PreparedSQL.NotPrepared
314	$q = $wpdb->prepare($q, $offset, $limit);
315	}
316	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
317	return $wpdb->get_results($q, ARRAY_A);
318	}
319
320	/**
321	* Count blacklist.
322	*
323	* @return int\|false
324	*/
325	public function count_blacklist() {
326	global $wpdb;
327
328	if ( ! $this->__data->tb_exist( 'crawler_blacklist' ) ) {
329	return false;
330	}
331
332	$q = "SELECT COUNT(*) FROM `$this->_tb_blacklist`";
333	// phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
334	return $wpdb->get_var($q);
335	}
336
337	/**
338	* Empty sitemap.
339	*
340	* @since 3.0
341	* @access public
342	* @return void
343	*/
344	public function empty_map() {
345	Data::cls()->tb_del( 'crawler' );
346
347	$msg = __( 'Sitemap cleaned successfully', 'litespeed-cache' );
348	Admin_Display::success( $msg );
349	}
350
351	/**
352	* List generated sitemap.
353	*
354	* @since 3.0
355	* @access public
356	*
357	* @param int $limit Number of rows per page.
358	* @param int\|bool $offset Offset for pagination, or false to auto-calc.
359	* @return array<int,array<string,mixed>>
360	*/
361	public function list_map( $limit, $offset = false ) {
362	global $wpdb;
363
364	if ( ! $this->__data->tb_exist( 'crawler' ) ) {
365	return [];
366	}
367
368	if ( false === $offset ) {
369	$total = $this->count_map();
370	$offset = Utility::pagination($total, $limit, true);
371	}
372
373	$type = Router::verify_type();
374
375	$req_uri_like = '';
376	// phpcs:ignore WordPress.Security.NonceVerification.Missing
377	if ( ! empty( $_POST['kw'] ) ) {
378	// phpcs:ignore WordPress.Security.NonceVerification.Missing
379	$kw = sanitize_text_field( wp_unslash( $_POST['kw'] ) );
380	$q = "SELECT * FROM `$this->_tb` WHERE url LIKE %s";
381	if ( 'hit' === $type ) {
382	$q .= " AND res LIKE '%" . Crawler::STATUS_HIT . "%'";
383	}
384	if ( 'miss' === $type ) {
385	$q .= " AND res LIKE '%" . Crawler::STATUS_MISS . "%'";
386	}
387	if ( 'blacklisted' === $type ) {
388	$q .= " AND res LIKE '%" . Crawler::STATUS_BLACKLIST . "%'";
389	}
390	$q .= ' ORDER BY id LIMIT %d, %d';
391	$req_uri_like = '%' . $wpdb->esc_like( $kw ) . '%';
392	return $wpdb->get_results( $wpdb->prepare( $q, $req_uri_like, $offset, $limit ), ARRAY_A ); // phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
393	}
394
395	$q = "SELECT * FROM `$this->_tb`";
396	if ( 'hit' === $type ) {
397	$q .= " WHERE res LIKE '%" . Crawler::STATUS_HIT . "%'";
398	}
399	if ( 'miss' === $type ) {
400	$q .= " WHERE res LIKE '%" . Crawler::STATUS_MISS . "%'";
401	}
402	if ( 'blacklisted' === $type ) {
403	$q .= " WHERE res LIKE '%" . Crawler::STATUS_BLACKLIST . "%'";
404	}
405	$q .= ' ORDER BY id LIMIT %d, %d';
406
407	return $wpdb->get_results( $wpdb->prepare( $q, $offset, $limit ), ARRAY_A ); // phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
408	}
409
410	/**
411	* Count sitemap.
412	*
413	* @return int\|false
414	*/
415	public function count_map() {
416	global $wpdb;
417
418	if ( ! $this->__data->tb_exist( 'crawler' ) ) {
419	return false;
420	}
421
422	$q = "SELECT COUNT(*) FROM `$this->_tb`";
423
424	$type = Router::verify_type();
425	if ( 'hit' === $type ) {
426	$q .= " WHERE res LIKE '%" . Crawler::STATUS_HIT . "%'";
427	}
428	if ( 'miss' === $type ) {
429	$q .= " WHERE res LIKE '%" . Crawler::STATUS_MISS . "%'";
430	}
431	if ( 'blacklisted' === $type ) {
432	$q .= " WHERE res LIKE '%" . Crawler::STATUS_BLACKLIST . "%'";
433	}
434
435	return $wpdb->get_var( $q ); // phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
436	}
437
438	/**
439	* Generate sitemap.
440	*
441	* @since 1.1.0
442	* @access public
443	*
444	* @param bool $manual Whether triggered manually from UI.
445	* @return void
446	*/
447	public function gen( $manual = false ) {
448	$count = $this->_gen();
449
450	if ( ! $count ) {
451	Admin_Display::error( __( 'No valid sitemap parsed for crawler.', 'litespeed-cache' ) );
452	return;
453	}
454
455	if ( ! wp_doing_cron() && $manual ) {
456	$msg = sprintf( __( 'Sitemap created successfully: %d items', 'litespeed-cache' ), $count );
457	Admin_Display::success( $msg );
458	}
459	}
460
461	/**
462	* Generate the sitemap.
463	*
464	* @since 1.1.0
465	* @access private
466	* @return int\|false Number of URLs generated or false on failure.
467	*/
468	private function _gen() {
469	global $wpdb;
470
471	if ( ! $this->__data->tb_exist( 'crawler' ) ) {
472	$this->__data->tb_create( 'crawler' );
473	}
474
475	if ( ! $this->__data->tb_exist( 'crawler_blacklist' ) ) {
476	$this->__data->tb_create( 'crawler_blacklist' );
477	}
478
479	// Use custom sitemap.
480	$sitemap = $this->conf( Base::O_CRAWLER_SITEMAP );
481	if ( ! $sitemap ) {
482	return false;
483	}
484
485	$offset = strlen( $this->_site_url );
486	$sitemap = Utility::sanitize_lines( $sitemap );
487
488	try {
489	foreach ( $sitemap as $this_map ) {
490	$this->_parse( $this_map );
491	}
492	} catch ( \Exception $e ) {
493	self::debug( '❌ failed to parse custom sitemap: ' . $e->getMessage() );
494	}
495
496	if ( is_array( $this->_urls ) && ! empty( $this->_urls ) ) {
497	if ( defined( 'LITESPEED_CRAWLER_DROP_DOMAIN' ) && constant( 'LITESPEED_CRAWLER_DROP_DOMAIN' ) ) {
498	foreach ( $this->_urls as $k => $v ) {
499	if ( 0 !== stripos( $v, $this->_site_url ) ) {
500	unset( $this->_urls[ $k ] );
501	continue;
502	}
503	$this->_urls[ $k ] = substr( $v, $offset );
504	}
505	}
506
507	$this->_urls = array_values( array_unique( $this->_urls ) );
508	}
509
510	self::debug( 'Truncate sitemap' );
511	$wpdb->query( "TRUNCATE `$this->_tb`" ); // phpcs:ignore WordPress.DB.PreparedSQL.InterpolatedNotPrepared, WordPress.DB.DirectDatabaseQuery.DirectQuery
512
513	self::debug( 'Generate sitemap' );
514
515	// Filter URLs in blacklist.
516	$blacklist = $this->list_blacklist();
517
518	$full_blacklisted = [];
519	$partial_blacklisted = [];
520	foreach ( $blacklist as $v ) {
521	if ( false === strpos( $v['res'], '-' ) ) {
522	// Full blacklisted.
523	$full_blacklisted[] = $v['url'];
524	} else {
525	// Replace existing reason.
526	$v['reason'] = explode( ',', $v['reason'] );
527	$v['reason'] = array_map(
528	function ( $element ) {
529	return $element ? 'Existed' : '';
530	},
531	$v['reason']
532	);
533	$v['reason'] = implode( ',', $v['reason'] );
534	$partial_blacklisted[ $v['url'] ] = [
535	'res' => $v['res'],
536	'reason' => $v['reason'],
537	];
538	}
539	}
540
541	// Drop all blacklisted URLs.
542	$this->_urls = array_diff( $this->_urls, $full_blacklisted );
543
544	// Default res & reason.
545	$crawler_count = count( Crawler::cls()->list_crawlers() );
546	$default_res = str_repeat( '-', $crawler_count );
547	$default_reason = $crawler_count > 1 ? str_repeat( ',', $crawler_count - 1 ) : '';
548
549	$data = [];
550	foreach ( $this->_urls as $url ) {
551	$data[] = $url;
552	$data[] = array_key_exists( $url, $partial_blacklisted ) ? $partial_blacklisted[ $url ]['res'] : $default_res;
553	$data[] = array_key_exists( $url, $partial_blacklisted ) ? $partial_blacklisted[ $url ]['reason'] : $default_reason;
554	}
555
556	foreach ( array_chunk( $data, 300 ) as $data2 ) {
557	$this->_save( $data2 );
558	}
559
560	// Reset crawler.
561	Crawler::cls()->reset_pos();
562
563	return count( $this->_urls );
564	}
565
566	/**
567	* Save data to table.
568	*
569	* @since 3.0
570	* @access private
571	*
572	* @param array<int,string> $data Flat array (url,res,reason, url,res,reason, ...).
573	* @param string $fields Fields list for insert (default url,res,reason).
574	* @return void
575	*/
576	private function _save( $data, $fields = 'url,res,reason' ) {
577	global $wpdb;
578
579	if ( empty( $data ) ) {
580	return;
581	}
582
583	$q = "INSERT INTO `$this->_tb` ( {$fields} ) VALUES ";
584
585	// Add placeholder.
586	$q .= Utility::chunk_placeholder( $data, $fields );
587
588	// Store data.
589	$wpdb->query( $wpdb->prepare( $q, $data ) ); // phpcs:ignore WordPress.DB.DirectDatabaseQuery.DirectQuery, WordPress.DB.PreparedSQL.NotPrepared
590	}
591
592	/**
593	* Parse custom sitemap and collect urls.
594	*
595	* @since 1.1.1
596	* @access private
597	*
598	* @param string $sitemap Absolute sitemap URL.
599	* @return void
600	* @throws \Exception If remote read or parsing fails.
601	*/
602	private function _parse( $sitemap ) {
603	/**
604	* Read via wp func to avoid allow_url_fopen = off
605	*
606	* @since 2.2.7
607	*/
608	$response = wp_safe_remote_get(
609	$sitemap,
610	[
611	'timeout' => $this->_conf_map_timeout,
612	'sslverify' => false,
613	]
614	);
615	if ( is_wp_error( $response ) ) {
616	$error_message = $response->get_error_message();
617	self::debug( 'failed to read sitemap: ' . $error_message );
618	throw new \Exception( 'Failed to remote read ' . esc_url( $sitemap ) );
619	}
620
621	$xml_object = simplexml_load_string($response['body'], null, LIBXML_NOCDATA);
622	if (!$xml_object) {
623	if ($this->_urls) {
624	return;
625	}
626	throw new \Exception('Failed to parse xml ' . esc_url( $sitemap ));
627	}
628
629	// start parsing.
630	$xml_array = (array) $xml_object;
631	if ( ! empty( $xml_array['sitemap'] ) ) {
632	// parse sitemap set.
633	if ( is_object( $xml_array['sitemap'] ) ) {
634	$xml_array['sitemap'] = (array) $xml_array['sitemap'];
635	}
636
637	if ( ! empty( $xml_array['sitemap']['loc'] ) ) {
638	// is single sitemap.
639	$this->_parse( (string) $xml_array['sitemap']['loc'] );
640	} else {
641	// parse multiple sitemaps.
642	foreach ( (array) $xml_array['sitemap'] as $val ) {
643	$val = (array) $val;
644	if ( ! empty( $val['loc'] ) ) {
645	$this->_parse( (string) $val['loc'] ); // recursive parse sitemap.
646	}
647	}
648	}
649	} elseif ( ! empty( $xml_array['url'] ) ) {
650	// parse url set.
651	if ( is_object( $xml_array['url'] ) ) {
652	$xml_array['url'] = (array) $xml_array['url'];
653	}
654	// if only 1 element.
655	if ( ! empty( $xml_array['url']['loc'] ) ) {
656	$this->_urls[] = (string) $xml_array['url']['loc'];
657	} else {
658	foreach ( (array) $xml_array['url'] as $val ) {
659	$val = (array) $val;
660	if ( ! empty( $val['loc'] ) ) {
661	$this->_urls[] = (string) $val['loc'];
662	}
663	}
664	}
665	}
666	}
667	}
668