PHP MongoDB ORM
- Separated models and connection setup (no collection names in models)
- References
- Default values
- Retrieving native mongoclient objects (just in case)
- Lazy-loading of references
- Lazy-loading of models in result (not fetching all at once, not caching)
- Memory-safe (see lazy-loading)
$autoload = array(
'ru\\olamedia\\kanon\\mongo\\connection' => 'connection.php',
'ru\\olamedia\\kanon\\mongo\\collection' => 'collection.php',
'ru\\olamedia\\kanon\\mongo\\model' => 'model.php',
'ru\\olamedia\\kanon\\mongo\\result' => 'result.php',
'ru\\olamedia\\kanon\\mongo\\reference' => 'reference.php',
'ru\\olamedia\\kanon\\mongo\\referenceSet' => 'referenceSet.php',
'ru\\olamedia\\kanon\\mongo\\typeManager' => 'typeManager.php',
);- string
- int | integer
- float | double (both are double in MongoDB)
- bool | boolean
- timestamp (\MongoTimestamp, MongoTimestamp is used by sharding. If you're not looking to write sharding tools, what you probably want is MongoDate.)
- datetime | date | time (\MongoDate, MongoDB stores dates as milliseconds past the epoch.)
- reference (@see http://php.net/mongodbref )
- references (array of "reference"s)
<?php
namespace my\namespace;
use ru\olamedia\kanon\mongo\model;
class article extends model{
protected static $_properties = [
'parent' => [
'model'=>'my\\namespace\\article',
'type'=>'reference'
],
'name' => [
'default'=>'',
'type'=>'string'
]
];
}use ru\olamedia\kanon\mongo;
mongo\connection::getInstance('default')
->connect([
'connection' => [
'hostnames' => 'localhost',
'database' => 'dbname',
]
])
->registerModel(
'my\\namespace\\article', // class name
'articles' // collection name
);$article1 = new article();
$article1->name = 'New article';
$article1->save();
$article2 = new article(['name' => 'Hello world']);
$article2->parent = $article1; // reference
$article2->save();
var_dump($article1->getId());
echo $article1->getId(); // MongoDB uses _id field with \MongoId objects convertable toString().
var_dump($article1->name);
$article1->delete();$result = article::find($criteria, $fields);
$result = article::findOne($criteria, $fields);
// $result is instance of ru\olamedia\kanon\mongo\result,
// which is a wrapper to \MongoCursor able to return model instances instead of arrays
// to get \MongoCursor back:
$cursor = $result->getCursor();
// creating result from \MongoCursor:
$result = new result('my\\namespace\\article', $cursor);
// using native methods:
$result->asc('name');
$result->desc('name');
$result->sort(['name'=>1]); // @see http://php.net/mongocursor.sort
$result->skip(3); // @see http://php.net/mongocursor.skip
$result->limit(10); // see http://php.net/mongocursor.limit
// using collections:
$collection = collection::forClass('my\\namespace\\article'); // \ru\olamedia\kanon\mongo\collection
$mongoCollection = $collection->select(); // \MongoCollection
$cursor = $mongoCollection->find($criteria, $fields); // \MongoCursor
$result = new result('my\\namespace\\article', $cursor); // \ru\olamedia\kanon\mongo\result
foreach ($result as $article){ // \my\namespace\article
echo $article->name.'<br />';
}Implements \Iterator, \Countable
$articles = article::find();
echo count($articles);
foreach ($articles as $article){
}- PHP >= 5.3
- PECL mongoclient >= 1.3.0
MIT