1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
namespace Queryr\Replicator\Cli\Command; |
4
|
|
|
|
5
|
|
|
use BatchingIterator\BatchingIterator; |
6
|
|
|
use Queryr\Replicator\Cli\Import\PagesImporterCli; |
7
|
|
|
use Queryr\Replicator\EntityIdListNormalizer; |
8
|
|
|
use Queryr\Replicator\EntitySource\Api\GetEntitiesClient; |
9
|
|
|
use Queryr\Replicator\EntitySource\Api\Http; |
10
|
|
|
use Queryr\Replicator\EntitySource\BatchingEntityPageFetcher; |
11
|
|
|
use Queryr\Replicator\EntitySource\ReferencedEntityPageIterator; |
12
|
|
|
use Symfony\Component\Console\Input\InputArgument; |
13
|
|
|
use Symfony\Component\Console\Input\InputInterface; |
14
|
|
|
use Symfony\Component\Console\Input\InputOption; |
15
|
|
|
use Symfony\Component\Console\Output\OutputInterface; |
16
|
|
|
use Wikibase\DataModel\Entity\BasicEntityIdParser; |
17
|
|
|
|
18
|
|
|
/** |
19
|
|
|
* @licence GNU GPL v2+ |
20
|
|
|
* @author Jeroen De Dauw < [email protected] > |
21
|
|
|
*/ |
22
|
|
|
class ApiImportCommand extends ImportCommandBase { |
23
|
|
|
|
24
|
1 |
|
protected function configure() { |
25
|
1 |
|
$this->setName( 'import:api' ); |
26
|
1 |
|
$this->setDescription( 'Imports entities via a Wikibase Repo web API' ); |
27
|
|
|
|
28
|
1 |
|
$this->addArgument( |
29
|
1 |
|
'entities', |
30
|
1 |
|
InputArgument::IS_ARRAY, |
31
|
1 |
|
'The IDs of the entities to import, separated by spaces. ID ranges can be specified: Q1-Q100' |
32
|
|
|
); |
33
|
|
|
|
34
|
|
|
// $this->addOption( |
35
|
|
|
// 'url', |
36
|
|
|
// null, |
37
|
|
|
// InputOption::VALUE_OPTIONAL, |
38
|
|
|
// 'The full url of the Wikibase Repo web API to use', |
39
|
|
|
// 'https://www.wikidata.org/w/api.php' |
40
|
|
|
// ); |
41
|
|
|
|
42
|
1 |
|
$this->addOption( |
43
|
1 |
|
'batchsize', |
44
|
1 |
|
'b', |
45
|
1 |
|
InputOption::VALUE_OPTIONAL, |
46
|
1 |
|
'The number of API requests to bundle together', |
47
|
1 |
|
10 |
48
|
|
|
); |
49
|
|
|
|
50
|
1 |
|
$this->addOption( |
51
|
1 |
|
'include-references', |
52
|
1 |
|
'r', |
53
|
1 |
|
InputOption::VALUE_NONE, |
54
|
1 |
|
'If referenced entities should also be imported' |
55
|
|
|
); |
56
|
1 |
|
} |
57
|
|
|
|
58
|
|
|
/** |
59
|
|
|
* @var Http|null |
60
|
|
|
*/ |
61
|
|
|
private $http; |
62
|
|
|
|
63
|
1 |
|
public function setHttp( Http $http ) { |
64
|
1 |
|
$this->http = $http; |
65
|
1 |
|
} |
66
|
|
|
|
67
|
1 |
|
protected function executeCommand( InputInterface $input, OutputInterface $output ) { |
68
|
1 |
|
$importer = new PagesImporterCli( $input, $output, $this->factory ); |
|
|
|
|
69
|
|
|
|
70
|
1 |
|
$importer->runImport( $this->getEntityPageIterator( $input ) ); |
71
|
1 |
|
} |
72
|
|
|
|
73
|
1 |
|
private function getEntityPageIterator( InputInterface $input ): \Iterator { |
74
|
1 |
|
$http = $this->http === null ? new Http() : $this->http; |
75
|
|
|
|
76
|
1 |
|
$idListNormalizer = new EntityIdListNormalizer( new BasicEntityIdParser() ); |
77
|
1 |
|
$ids = []; |
78
|
|
|
|
79
|
1 |
|
foreach ( $idListNormalizer->getNormalized( $input->getArgument( 'entities' ) ) as $id ) { |
|
|
|
|
80
|
1 |
|
$ids[] = $id->getSerialization(); |
81
|
|
|
} |
82
|
|
|
|
83
|
1 |
|
$batchingFetcher = new BatchingEntityPageFetcher( |
84
|
1 |
|
new GetEntitiesClient( $http ), |
85
|
1 |
|
$ids |
86
|
|
|
); |
87
|
|
|
|
88
|
1 |
|
$iterator = new BatchingIterator( $batchingFetcher ); |
89
|
1 |
|
$iterator->setMaxBatchSize( (int)$input->getOption( 'batchsize' ) ); |
90
|
|
|
|
91
|
1 |
|
if ( $input->getOption( 'include-references' ) ) { |
92
|
|
|
return new ReferencedEntityPageIterator( |
93
|
|
|
$iterator, |
94
|
|
|
$batchingFetcher, |
95
|
|
|
$this->factory->newLegacyEntityDeserializer() |
96
|
|
|
); |
97
|
|
|
} |
98
|
|
|
|
99
|
1 |
|
return $iterator; |
100
|
|
|
} |
101
|
|
|
|
102
|
|
|
} |
103
|
|
|
|
Unless you are absolutely sure that the expression can never be null because of other conditions, we strongly recommend to add an additional type check to your code: