1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
declare(strict_types = 1); |
4
|
|
|
|
5
|
|
|
namespace App\Repository; |
6
|
|
|
|
7
|
|
|
use App\Model\PageAssessments; |
8
|
|
|
use App\Model\Project; |
9
|
|
|
use Doctrine\DBAL\Connection; |
10
|
|
|
use Doctrine\Persistence\ManagerRegistry; |
11
|
|
|
use Exception; |
12
|
|
|
use GuzzleHttp\Client; |
13
|
|
|
use Psr\Cache\CacheItemPoolInterface; |
14
|
|
|
use Psr\Log\LoggerInterface; |
15
|
|
|
use Symfony\Component\DependencyInjection\ParameterBag\ParameterBagInterface; |
16
|
|
|
|
17
|
|
|
/** |
18
|
|
|
* This class provides data to the Project class. |
19
|
|
|
* @codeCoverageIgnore |
20
|
|
|
*/ |
21
|
|
|
class ProjectRepository extends Repository |
22
|
|
|
{ |
23
|
|
|
protected PageAssessmentsRepository $assessmentsRepo; |
24
|
|
|
|
25
|
|
|
/** @var string[] Basic metadata if XTools is in single-wiki mode. */ |
26
|
|
|
protected array $singleBasicInfo; |
27
|
|
|
|
28
|
|
|
/** @var string The cache key for the 'all project' metadata. */ |
29
|
|
|
protected string $cacheKeyAllProjects = 'allprojects'; |
30
|
|
|
|
31
|
|
|
/** @var string The configured default project. */ |
32
|
|
|
protected string $defaultProject; |
33
|
|
|
|
34
|
|
|
/** @var bool Whether XTools is configured to run on a single wiki or not. */ |
35
|
|
|
protected bool $singleWiki; |
36
|
|
|
|
37
|
|
|
/** @var array Projects that have opted into showing restricted stats to everyone. */ |
38
|
|
|
protected array $optedIn; |
39
|
|
|
|
40
|
|
|
/** @var string The project's API path. */ |
41
|
|
|
protected string $apiPath; |
42
|
|
|
|
43
|
|
|
/** |
44
|
|
|
* @param ManagerRegistry $managerRegistry |
45
|
|
|
* @param CacheItemPoolInterface $cache |
46
|
|
|
* @param Client $guzzle |
47
|
|
|
* @param LoggerInterface $logger |
48
|
|
|
* @param ParameterBagInterface $parameterBag |
49
|
|
|
* @param bool $isWMF |
50
|
|
|
* @param int $queryTimeout |
51
|
|
|
* @param PageAssessmentsRepository $assessmentsRepo |
52
|
|
|
* @param string $defaultProject |
53
|
|
|
* @param bool $singleWiki |
54
|
|
|
* @param array $optedIn |
55
|
|
|
* @param string $apiPath |
56
|
|
|
*/ |
57
|
|
|
public function __construct( |
58
|
|
|
ManagerRegistry $managerRegistry, |
59
|
|
|
CacheItemPoolInterface $cache, |
60
|
|
|
Client $guzzle, |
61
|
|
|
LoggerInterface $logger, |
62
|
|
|
ParameterBagInterface $parameterBag, |
63
|
|
|
bool $isWMF, |
64
|
|
|
int $queryTimeout, |
65
|
|
|
PageAssessmentsRepository $assessmentsRepo, |
66
|
|
|
string $defaultProject, |
67
|
|
|
bool $singleWiki, |
68
|
|
|
array $optedIn, |
69
|
|
|
string $apiPath |
70
|
|
|
) { |
71
|
|
|
$this->assessmentsRepo = $assessmentsRepo; |
72
|
|
|
$this->defaultProject = $defaultProject; |
73
|
|
|
$this->singleWiki = $singleWiki; |
74
|
|
|
$this->optedIn = $optedIn; |
75
|
|
|
$this->apiPath = $apiPath; |
76
|
|
|
parent::__construct($managerRegistry, $cache, $guzzle, $logger, $parameterBag, $isWMF, $queryTimeout); |
77
|
|
|
} |
78
|
|
|
|
79
|
|
|
/** |
80
|
|
|
* Convenience method to get a new Project object based on a given identification string. |
81
|
|
|
* @param string $projectIdent The domain name, database name, or URL of a project. |
82
|
|
|
* @return Project |
83
|
|
|
*/ |
84
|
|
|
public function getProject(string $projectIdent): Project |
85
|
|
|
{ |
86
|
|
|
$project = new Project($projectIdent); |
87
|
|
|
$project->setRepository($this); |
88
|
|
|
$project->setPageAssessments(new PageAssessments($this->assessmentsRepo, $project)); |
89
|
|
|
|
90
|
|
|
if ($this->singleWiki) { |
91
|
|
|
$this->setSingleBasicInfo([ |
92
|
|
|
'url' => $this->parameterBag->get('wiki_url'), |
93
|
|
|
'dbName' => '', // Just so this will pass in CI. |
94
|
|
|
// TODO: this will need to be restored for third party support; KEYWORD: isWMF |
95
|
|
|
// 'dbName' => $this->parameterBag->('database_replica_name'), |
96
|
|
|
]); |
97
|
|
|
} |
98
|
|
|
|
99
|
|
|
return $project; |
100
|
|
|
} |
101
|
|
|
|
102
|
|
|
/** |
103
|
|
|
* Get the XTools default project. |
104
|
|
|
* @return Project |
105
|
|
|
*/ |
106
|
|
|
public function getDefaultProject(): Project |
107
|
|
|
{ |
108
|
|
|
return $this->getProject($this->defaultProject); |
109
|
|
|
} |
110
|
|
|
|
111
|
|
|
/** |
112
|
|
|
* Get the global 'meta' project, which is either Meta (if this is Labs) or the default project. |
113
|
|
|
* @return Project |
114
|
|
|
*/ |
115
|
|
|
public function getGlobalProject(): Project |
116
|
|
|
{ |
117
|
|
|
if ($this->isWMF) { |
118
|
|
|
return $this->getProject('metawiki'); |
119
|
|
|
} else { |
120
|
|
|
return $this->getDefaultProject(); |
121
|
|
|
} |
122
|
|
|
} |
123
|
|
|
|
124
|
|
|
/** |
125
|
|
|
* For single-wiki installations, you must manually set the wiki URL and database name |
126
|
|
|
* (because there's no meta.wiki database to query). |
127
|
|
|
* @param array $metadata |
128
|
|
|
* @throws Exception |
129
|
|
|
*/ |
130
|
|
|
public function setSingleBasicInfo(array $metadata): void |
131
|
|
|
{ |
132
|
|
|
if (!array_key_exists('url', $metadata) || !array_key_exists('dbName', $metadata)) { |
133
|
|
|
$error = "Single-wiki metadata should contain 'url', 'dbName' and 'lang' keys."; |
134
|
|
|
throw new Exception($error); |
135
|
|
|
} |
136
|
|
|
$this->singleBasicInfo = array_intersect_key($metadata, [ |
137
|
|
|
'url' => '', |
138
|
|
|
'dbName' => '', |
139
|
|
|
'lang' => '', |
140
|
|
|
]); |
141
|
|
|
} |
142
|
|
|
|
143
|
|
|
/** |
144
|
|
|
* Get the 'dbName', 'url' and 'lang' of all projects. |
145
|
|
|
* @return string[][] Each item has 'dbName', 'url' and 'lang' keys. |
146
|
|
|
*/ |
147
|
|
|
public function getAll(): array |
148
|
|
|
{ |
149
|
|
|
$this->logger->debug(__METHOD__." Getting all projects' metadata"); |
150
|
|
|
// Single wiki mode? |
151
|
|
|
if (!empty($this->singleBasicInfo)) { |
152
|
|
|
return [$this->getOne('')]; |
153
|
|
|
} |
154
|
|
|
|
155
|
|
|
// Maybe we've already fetched it. |
156
|
|
|
if ($this->cache->hasItem($this->cacheKeyAllProjects)) { |
157
|
|
|
return $this->cache->getItem($this->cacheKeyAllProjects)->get(); |
158
|
|
|
} |
159
|
|
|
|
160
|
|
|
if ($this->parameterBag->has("database_meta_table")) { |
161
|
|
|
$table = $this->parameterBag->get('database_meta_name') . '.' . |
162
|
|
|
$this->parameterBag->get('database_meta_table'); |
163
|
|
|
} else { |
164
|
|
|
$table = "meta_p.wiki"; |
165
|
|
|
} |
166
|
|
|
|
167
|
|
|
// Otherwise, fetch all from the database. |
168
|
|
|
$sql = "SELECT dbname AS dbName, url, lang FROM $table"; |
169
|
|
|
$projects = $this->executeProjectsQuery('meta', $sql) |
170
|
|
|
->fetchAllAssociative(); |
171
|
|
|
$projectsMetadata = []; |
172
|
|
|
foreach ($projects as $project) { |
173
|
|
|
$projectsMetadata[$project['dbName']] = $project; |
174
|
|
|
} |
175
|
|
|
|
176
|
|
|
// Cache for one day and return. |
177
|
|
|
return $this->setCache( |
178
|
|
|
$this->cacheKeyAllProjects, |
179
|
|
|
$projectsMetadata, |
180
|
|
|
'P1D' |
181
|
|
|
); |
182
|
|
|
} |
183
|
|
|
|
184
|
|
|
/** |
185
|
|
|
* Get the 'dbName', 'url' and 'lang' of a project. This is all you need to make database queries. |
186
|
|
|
* More comprehensive metadata can be fetched with getMetadata() at the expense of an API call. |
187
|
|
|
* @param string $project A project URL, domain name, or database name. |
188
|
|
|
* @return string[]|null With 'dbName', 'url' and 'lang' keys; or null if not found. |
189
|
|
|
*/ |
190
|
|
|
public function getOne(string $project): ?array |
191
|
|
|
{ |
192
|
|
|
$this->logger->debug(__METHOD__." Getting metadata about $project"); |
193
|
|
|
// For single-wiki setups, every project is the same. |
194
|
|
|
if (isset($this->singleBasicInfo)) { |
195
|
|
|
return $this->singleBasicInfo; |
196
|
|
|
} |
197
|
|
|
|
198
|
|
|
// Remove _p suffix. |
199
|
|
|
$project = rtrim($project, '_p'); |
200
|
|
|
|
201
|
|
|
// For multi-wiki setups, first check the cache. |
202
|
|
|
// First the all-projects cache, then the individual one. |
203
|
|
|
if ($this->cache->hasItem($this->cacheKeyAllProjects)) { |
204
|
|
|
foreach ($this->cache->getItem($this->cacheKeyAllProjects)->get() as $projMetadata) { |
205
|
|
|
if ($projMetadata['dbName'] == "$project" |
206
|
|
|
|| $projMetadata['url'] == "$project" |
207
|
|
|
|| $projMetadata['url'] == "https://$project" |
208
|
|
|
|| $projMetadata['url'] == "https://$project.org" |
209
|
|
|
|| $projMetadata['url'] == "https://www.$project") { |
210
|
|
|
$this->logger->debug(__METHOD__ . " Using cached data for $project"); |
211
|
|
|
return $projMetadata; |
212
|
|
|
} |
213
|
|
|
} |
214
|
|
|
} |
215
|
|
|
$cacheKey = $this->getCacheKey($project, 'project'); |
216
|
|
|
if ($this->cache->hasItem($cacheKey)) { |
217
|
|
|
return $this->cache->getItem($cacheKey)->get(); |
218
|
|
|
} |
219
|
|
|
|
220
|
|
|
// TODO: make this configurable if XTools is to work on 3rd party wiki farms |
221
|
|
|
$table = "meta_p.wiki"; |
222
|
|
|
|
223
|
|
|
// Otherwise, fetch the project's metadata from the meta.wiki table. |
224
|
|
|
$sql = "SELECT dbname AS dbName, url, lang |
225
|
|
|
FROM $table |
226
|
|
|
WHERE dbname = :project |
227
|
|
|
OR url LIKE :projectUrl |
228
|
|
|
OR url LIKE :projectUrl2 |
229
|
|
|
OR url LIKE :projectUrl3 |
230
|
|
|
OR url LIKE :projectUrl4"; |
231
|
|
|
$basicInfo = $this->executeProjectsQuery('meta', $sql, [ |
232
|
|
|
'project' => $project, |
233
|
|
|
'projectUrl' => "https://$project", |
234
|
|
|
'projectUrl2' => "https://$project.org", |
235
|
|
|
'projectUrl3' => "https://www.$project", |
236
|
|
|
'projectUrl4' => "https://www.$project.org", |
237
|
|
|
])->fetchAssociative(); |
238
|
|
|
$basicInfo = false === $basicInfo ? null : $basicInfo; |
239
|
|
|
|
240
|
|
|
// Cache for one hour and return. |
241
|
|
|
return $this->setCache($cacheKey, $basicInfo, 'PT1H'); |
242
|
|
|
} |
243
|
|
|
|
244
|
|
|
/** |
245
|
|
|
* Get metadata about a project, including the 'dbName', 'url' and 'lang' |
246
|
|
|
* |
247
|
|
|
* @param string $projectUrl The project's URL. |
248
|
|
|
* @return array|null With 'dbName', 'url', 'lang', 'general' and 'namespaces' keys. |
249
|
|
|
* 'general' contains: 'wikiName', 'articlePath', 'scriptPath', 'script', |
250
|
|
|
* 'timezone', and 'timezoneOffset'; 'namespaces' contains all namespace |
251
|
|
|
* names, keyed by their IDs. If this function returns null, the API call |
252
|
|
|
* failed. |
253
|
|
|
*/ |
254
|
|
|
public function getMetadata(string $projectUrl): ?array |
255
|
|
|
{ |
256
|
|
|
$cacheKey = $this->getCacheKey(func_get_args(), "project_metadata"); |
257
|
|
|
if ($this->cache->hasItem($cacheKey)) { |
258
|
|
|
return $this->cache->getItem($cacheKey)->get(); |
259
|
|
|
} |
260
|
|
|
|
261
|
|
|
try { |
262
|
|
|
$res = json_decode($this->guzzle->request('GET', $projectUrl.$this->getApiPath(), [ |
263
|
|
|
'query' => [ |
264
|
|
|
'action' => 'query', |
265
|
|
|
'meta' => 'siteinfo', |
266
|
|
|
'siprop' => 'general|namespaces', |
267
|
|
|
'format' => 'json', |
268
|
|
|
], |
269
|
|
|
])->getBody()->getContents(), true); |
270
|
|
|
} catch (Exception $e) { |
271
|
|
|
return null; |
272
|
|
|
} |
273
|
|
|
|
274
|
|
|
$metadata = [ |
275
|
|
|
'general' => [], |
276
|
|
|
'namespaces' => [], |
277
|
|
|
]; |
278
|
|
|
|
279
|
|
|
if (isset($res['query']['general'])) { |
280
|
|
|
$info = $res['query']['general']; |
281
|
|
|
|
282
|
|
|
$metadata['dbName'] = $info['wikiid']; |
283
|
|
|
$metadata['url'] = $info['server']; |
284
|
|
|
$metadata['lang'] = $info['lang']; |
285
|
|
|
|
286
|
|
|
$metadata['general'] = [ |
287
|
|
|
'wikiName' => $info['sitename'], |
288
|
|
|
'articlePath' => $info['articlepath'], |
289
|
|
|
'scriptPath' => $info['scriptpath'], |
290
|
|
|
'script' => $info['script'], |
291
|
|
|
'timezone' => $info['timezone'], |
292
|
|
|
'timeOffset' => $info['timeoffset'], |
293
|
|
|
'mainpage' => $info['mainpage'], |
294
|
|
|
]; |
295
|
|
|
} |
296
|
|
|
|
297
|
|
|
$this->setNamespaces($res, $metadata); |
298
|
|
|
|
299
|
|
|
// Cache for one hour and return. |
300
|
|
|
return $this->setCache($cacheKey, $metadata, 'PT1H'); |
301
|
|
|
} |
302
|
|
|
|
303
|
|
|
/** |
304
|
|
|
* Set the namespaces on the given $metadata. |
305
|
|
|
* @param array $res As produced by meta=siteinfo API. |
306
|
|
|
* @param array &$metadata The metadata array to modify. |
307
|
|
|
*/ |
308
|
|
|
private function setNamespaces(array $res, array &$metadata): void |
309
|
|
|
{ |
310
|
|
|
if (!isset($res['query']['namespaces'])) { |
311
|
|
|
return; |
312
|
|
|
} |
313
|
|
|
|
314
|
|
|
foreach ($res['query']['namespaces'] as $namespace) { |
315
|
|
|
if ($namespace['id'] < 0) { |
316
|
|
|
continue; |
317
|
|
|
} |
318
|
|
|
|
319
|
|
|
if (isset($namespace['name'])) { |
320
|
|
|
$name = $namespace['name']; |
321
|
|
|
} elseif (isset($namespace['*'])) { |
322
|
|
|
$name = $namespace['*']; |
323
|
|
|
} else { |
324
|
|
|
continue; |
325
|
|
|
} |
326
|
|
|
|
327
|
|
|
$metadata['namespaces'][$namespace['id']] = $name; |
328
|
|
|
} |
329
|
|
|
} |
330
|
|
|
|
331
|
|
|
/** |
332
|
|
|
* Get a list of projects that have opted in to having all their users' restricted statistics available to anyone. |
333
|
|
|
* @return string[] |
334
|
|
|
*/ |
335
|
|
|
public function optedIn(): array |
336
|
|
|
{ |
337
|
|
|
return $this->optedIn; |
338
|
|
|
} |
339
|
|
|
|
340
|
|
|
/** |
341
|
|
|
* The path to api.php. |
342
|
|
|
* @return string |
343
|
|
|
*/ |
344
|
|
|
public function getApiPath(): string |
345
|
|
|
{ |
346
|
|
|
return $this->apiPath; |
347
|
|
|
} |
348
|
|
|
|
349
|
|
|
/** |
350
|
|
|
* Check to see if a page exists on this project and has some content. |
351
|
|
|
* @param Project $project The project. |
352
|
|
|
* @param int $namespaceId The page namespace ID. |
353
|
|
|
* @param string $pageTitle The page title, without namespace. |
354
|
|
|
* @return bool |
355
|
|
|
*/ |
356
|
|
|
public function pageHasContent(Project $project, int $namespaceId, string $pageTitle): bool |
357
|
|
|
{ |
358
|
|
|
$pageTable = $this->getTableName($project->getDatabaseName(), 'page'); |
359
|
|
|
$query = "SELECT page_id " |
360
|
|
|
. " FROM $pageTable " |
361
|
|
|
. " WHERE page_namespace = :ns AND page_title = :title AND page_len > 0 " |
362
|
|
|
. " LIMIT 1"; |
363
|
|
|
$params = [ |
364
|
|
|
'ns' => $namespaceId, |
365
|
|
|
'title' => str_replace(' ', '_', $pageTitle), |
366
|
|
|
]; |
367
|
|
|
$pages = $this->executeProjectsQuery($project, $query, $params) |
368
|
|
|
->fetchAllAssociative(); |
369
|
|
|
return count($pages) > 0; |
370
|
|
|
} |
371
|
|
|
|
372
|
|
|
/** |
373
|
|
|
* Get a list of the extensions installed on the wiki. |
374
|
|
|
* @param Project $project |
375
|
|
|
* @return string[] |
376
|
|
|
*/ |
377
|
|
|
public function getInstalledExtensions(Project $project): array |
378
|
|
|
{ |
379
|
|
|
$res = json_decode($this->guzzle->request('GET', $project->getApiUrl(), ['query' => [ |
380
|
|
|
'action' => 'query', |
381
|
|
|
'meta' => 'siteinfo', |
382
|
|
|
'siprop' => 'extensions', |
383
|
|
|
'format' => 'json', |
384
|
|
|
]])->getBody()->getContents(), true); |
385
|
|
|
|
386
|
|
|
$extensions = $res['query']['extensions'] ?? []; |
387
|
|
|
return array_map(function ($extension) { |
388
|
|
|
return $extension['name']; |
389
|
|
|
}, $extensions); |
390
|
|
|
} |
391
|
|
|
|
392
|
|
|
/** |
393
|
|
|
* Get a list of users who are in one of the given user groups. |
394
|
|
|
* @param Project $project |
395
|
|
|
* @param string[] $groups List of user groups to look for. |
396
|
|
|
* @param string[] $globalGroups List of global groups to look for. |
397
|
|
|
* @return string[] with keys 'user_name' and 'ug_group' |
398
|
|
|
*/ |
399
|
|
|
public function getUsersInGroups(Project $project, array $groups = [], array $globalGroups = []): array |
400
|
|
|
{ |
401
|
|
|
$cacheKey = $this->getCacheKey(func_get_args(), 'project_useringroups'); |
402
|
|
|
if ($this->cache->hasItem($cacheKey)) { |
403
|
|
|
return $this->cache->getItem($cacheKey)->get(); |
404
|
|
|
} |
405
|
|
|
|
406
|
|
|
$userTable = $project->getTableName('user'); |
407
|
|
|
$userGroupsTable = $project->getTableName('user_groups'); |
408
|
|
|
|
409
|
|
|
$sql = "SELECT user_name, ug_group AS user_group |
410
|
|
|
FROM $userTable |
411
|
|
|
JOIN $userGroupsTable ON ug_user = user_id |
412
|
|
|
WHERE ug_group IN (?) |
413
|
|
|
GROUP BY user_name, ug_group"; |
414
|
|
|
$users = $this->getProjectsConnection($project) |
415
|
|
|
->executeQuery($sql, [$groups], [Connection::PARAM_STR_ARRAY]) |
416
|
|
|
->fetchAllAssociative(); |
417
|
|
|
|
418
|
|
|
if (count($globalGroups) > 0 && $this->isWMF) { |
419
|
|
|
$sql = "SELECT gu_name AS user_name, gug_group AS user_group |
420
|
|
|
FROM centralauth_p.global_user_groups |
421
|
|
|
JOIN centralauth_p.globaluser ON gug_user = gu_id |
422
|
|
|
WHERE gug_group IN (?) |
423
|
|
|
GROUP BY user_name, user_group"; |
424
|
|
|
$globalUsers = $this->getProjectsConnection('centralauth') |
425
|
|
|
->executeQuery($sql, [$globalGroups], [Connection::PARAM_STR_ARRAY]) |
426
|
|
|
->fetchAllAssociative(); |
427
|
|
|
|
428
|
|
|
$users = array_merge($users, $globalUsers); |
429
|
|
|
} |
430
|
|
|
|
431
|
|
|
// Cache for 12 hours and return. |
432
|
|
|
return $this->setCache($cacheKey, $users, 'PT12H'); |
433
|
|
|
} |
434
|
|
|
} |
435
|
|
|
|