You are browsing a version that is no longer maintained.
Maintained
2.10 (upcoming) 2.9.0 (current)
Unmaintained
2.8 2.7 2.6 2.5 2.4 2.3 2.2 2.1 2.0 1.3 1.2 1.1 1.0

Basic Mapping

This chapter explains the basic mapping of objects and properties. Mapping of references and embedded documents will be covered in the next chapter "Reference Mapping".

Mapping Drivers

Doctrine provides several different ways for specifying object document mapping metadata:

  • Docblock Annotations
  • XML
  • Raw PHP Code

If you're wondering which mapping driver gives the best performance, the answer is: None. Once the metadata of a class has been read from the source (annotations or xml) it is stored in an instance of the Doctrine\ODM\MongoDB\Mapping\ClassMetadata class and these instances are stored in the metadata cache. Therefore at the end of the day all drivers perform equally well. If you're not using a metadata cache (not recommended!) then the XML driver might have a slight edge in performance due to the powerful native XML support in PHP.

Introduction to Docblock Annotations

You've probably used docblock annotations in some form already, most likely to provide documentation metadata for a tool like PHPDocumentor (@author, @link, ...). Docblock annotations are a tool to embed metadata inside the documentation section which can then be processed by some tool. Doctrine generalizes the concept of docblock annotations so that they can be used for any kind of metadata and so that it is easy to define new docblock annotations. In order to allow more involved annotation values and to reduce the chances of clashes with other docblock annotations, the Doctrine docblock annotations feature an alternative syntax that is heavily inspired by the Annotation syntax introduced in Java 5.

The implementation of these enhanced docblock annotations is located in the Doctrine\Common\Annotations namespace and therefore part of the Common package. Doctrine docblock annotations support namespaces and nested annotations among other things. The Doctrine MongoDB ODM defines its own set of docblock annotations for supplying object document mapping metadata.

If you're not comfortable with the concept of docblock annotations, don't worry, as mentioned earlier Doctrine provides XML alternative and you could easily implement your own favorite mechanism for defining ODM metadata.

Persistent classes

In order to mark a class for object-relational persistence it needs to be designated as a document. This can be done through the @Document marker annotation.

1<?php

namespace Documents;

/** @Document */
class User
{
}


2


3


4


5


6


7


8

By default, the document will be persisted to a database named doctrine and a collection with the same name as the class name. In order to change that, you can use the db and collection option as follows:

1<?php

namespace Documents;

/** @Document(db="my_db", collection="users") */
class User
{
}


2


3


4


5


6


7


8

Now instances of Documents\User will be persisted into a collection named users in the database my_db.

If you want to omit the db attribute you can configure the default db to use with the setDefaultDB method:

1<?php

$config->setDefaultDB('my_db');


2


3

Doctrine Mapping Types

A Doctrine Mapping Type defines the mapping between a PHP type and a MongoDB type. You can even write your own custom mapping types.

Here is a quick overview of the built-in mapping types:

  • bin
  • bin_bytearray
  • bin_custom
  • bin_func
  • bin_md5
  • bin_uuid
  • boolean
  • collection
  • custom_id
  • date
  • date_immutable
  • decimal128
  • file
  • float
  • hash
  • id
  • int
  • key
  • object_id
  • raw
  • string
  • timestamp

You can read more about the available MongoDB types on php.net.

The Doctrine mapping types are used to convert the local PHP types to the MongoDB types when persisting so that your domain is not bound to MongoDB-specific types. For example a DateTime instance may be converted to MongoDB\BSON\UTCDateTime when you persist your documents, and vice versa during hydration.

Generally, the name of each built-in mapping type hints as to how the value will be converted. This list explains some of the less obvious mapping types:

  • bin: string to MongoDBBSONBinary instance with a "generic" type (default)
  • bin_bytearray: string to MongoDBBSONBinary instance with a "byte array" type
  • bin_custom: string to MongoDBBSONBinary instance with a "custom" type
  • bin_func: string to MongoDBBSONBinary instance with a "function" type
  • bin_md5: string to MongoDBBSONBinary instance with a "md5" type
  • bin_uuid: string to MongoDBBSONBinary instance with a "uuid" type
  • collection: numerically indexed array to MongoDB array
  • date: DateTime to MongoDB\BSON\UTCDateTime
  • date_immutable: DateTimeImmutable to MongoDB\BSON\UTCDateTime
  • decimal128: string to MongoDB\BSON\Decimal128, requires ext-bcmath
  • hash: associative array to MongoDB object
  • id: string to ObjectId by default, but other formats are possible
  • timestamp: string to MongoDB\BSON\Timestamp
  • raw: any type

If you are using the hash type, values within the associative array are passed to MongoDB directly, without being prepared. Only formats suitable for the Mongo driver should be used. If your hash contains values which are not suitable you should either use an embedded document or use formats provided by the MongoDB driver (e.g. \MongoDB\BSON\UTCDateTime instead of \DateTime).

Property Mapping

After a class has been marked as a document it can specify mappings for its instance fields. Here we will only look at simple fields that hold scalar values like strings, numbers, etc. References to other objects and embedded objects are covered in the chapter "Reference Mapping".

Identifiers

Every document class needs an identifier. You designate the field that serves as the identifier with the @Id marker annotation. Here is an example:

1<?php

namespace Documents;

/** @Document */
class User
{
    /** @Id */
    private $id;
}


2


3


4


5


6


7


8


9


10

You can configure custom ID strategies if you don't want to use the default object ID. The available strategies are:

  • AUTO - Uses the native generated ObjectId.
  • ALNUM - Generates an alpha-numeric string (based on an incrementing value).
  • CUSTOM - Defers generation to a AbstractIdGenerator implementation specified in the class option.
  • INCREMENT - Uses another collection to auto increment an integer identifier.
  • UUID - Generates a UUID identifier.
  • NONE - Do not generate any identifier. ID must be manually set.

Here is an example how to manually set a string identifier for your documents:

1<?php

/** Document */
class MyPersistentClass
{
    /** @Id(strategy="NONE", type="string") */
    private $id;

    public function setId(string $id): void
    {
        $this->id = $id;
    }

    //...
}


2


3


4


5


6


7


8


9


10


11


12


13


14


15

When using the NONE strategy you will have to explicitly set an id before persisting the document:

1<?php

//...

$document = new MyPersistentClass();
$document->setId('my_unique_identifier');
$dm->persist($document);
$dm->flush();


2


3


4


5


6


7


8

Now you can retrieve the document later:

1<?php

//...

$document = $dm->find(MyPersistentClass::class, 'my_unique_identifier');


2


3


4


5

You can define your own ID generator by extending the Doctrine\ODM\MongoDB\Id\AbstractIdGenerator class and specifying the class as an option for the CUSTOM strategy:

1<?php

/** Document */
class MyPersistentClass
{
    /** @Id(strategy="CUSTOM", type="string", options={"class"="Vendor\Specific\Generator"}) */
    private $id;

    public function setId(string $id): void
    {
        $this->id = $id;
    }

    //...
}


2


3


4


5


6


7


8


9


10


11


12


13


14


15

Fields

To mark a property for document persistence the @Field docblock annotation can be used. This annotation usually requires at least 1 attribute to be set, the type. The type attribute specifies the Doctrine Mapping Type to use for the field. If the type is not specified, 'string' is used as the default mapping type since it is the most flexible.

Example:

1<?php

namespace Documents;

/** @Document */
class User
{
    // ...

    /** @Field(type="string") */
    private $username;
}


2


3


4


5


6


7


8


9


10


11


12

In that example we mapped the property id to the field id using the mapping type id and the property name is mapped to the field name with the default mapping type string. As you can see, by default the mongo field names are assumed to be the same as the property names. To specify a different name for the field, you can use the name attribute of the Field annotation as follows:

1<?php

/** @Field(name="db_name") */
private $name;


2


3


4

Multiple Document Types in a Collection

You can easily store multiple types of documents in a single collection. This requires specifying the same collection name, discriminatorField, and (optionally) discriminatorMap mapping options for each class that will share the collection. Here is an example:

1<?php

/**
 * @Document(collection="my_documents")
 * @DiscriminatorField("type")
 * @DiscriminatorMap({"article"=Article::class, "album"=Album::class})
 */
class Article
{
    // ...
}

/**
 * @Document(collection="my_documents")
 * @DiscriminatorField("type")
 * @DiscriminatorMap({"article"=Article::class, "album"=Album::class})
 */
class Album
{
    // ...
}


2


3


4


5


6


7


8


9


10


11


12


13


14


15


16


17


18


19


20


21

All instances of Article and Album will be stored in the my_documents collection. You can query for the documents of a particular class just like you normally would and the results will automatically be limited based on the discriminator value for that class.

If you wish to query for multiple types of documents from the collection, you may pass an array of document class names when creating a query builder:

1<?php

$query = $dm->createQuery([Article::class, Album::class]);
$documents = $query->execute();


2


3


4

The above will return a cursor that will allow you to iterate over all Article and Album documents in the collections.