mongoose/docs/populate.jade

Mongoose MongoDB ODM

extends layout

block content
  h2 Query Population
  :markdown
    There are no joins in MongoDB but sometimes we still want references to documents in other collections. This is where [query#populate](./api.html#query_Query-populate) comes in.

    `ObjectIds` can refer to another document in a collection within our database and be `populate()`d when querying:
  :js
    var mongoose = require('mongoose')
      , Schema = mongoose.Schema

    var PersonSchema = new Schema({
      name    : String,
      age     : Number,
      stories : [{ type: Schema.Types.ObjectId, ref: 'Story' }]
    });

    var StorySchema = new Schema({
      _creator : { type: Schema.Types.ObjectId, ref: 'Person' },
      title    : String,
      fans     : [{ type: Schema.Types.ObjectId, ref: 'Person' }]
    });

    var Story  = mongoose.model('Story', StorySchema);
    var Person = mongoose.model('Person', PersonSchema);
  :markdown
    So far we've created two `models`. Our `Person` model has it's `stories` field set to an array of `ObjectId`s. The `ref` option is what tells Mongoose in which model to look, in our case the `Story` model. All `_id`s we store here must be document _ids from the `Story` model. We also added a `_creator` `ObjectId` to our `Story` schema which refers to a single `Person`.
  h3 Saving refs
  :markdown
    Saving refs to other documents works the same way you normally save objectids, just assign an `ObjectId`:
  :js
    var aaron = new Person({ name: 'Aaron', age: 100 });

    aaron.save(function (err) {
      if (err) return handleError(err);

      var story1 = new Story({
        title: "Once upon a timex.",
        _creator: aaron._id    // assign an ObjectId
      });

      story1.save(function (err) {
        if (err) return handleError(err);
        // thats it!
      });
    })
  h3 Population
  :markdown
    So far we haven't done anything special. We've merely created a `Person` and a `Story`. Now let's take a look at populating our story's `_creator`:
  :js
    Story
    .findOne({ title: /timex/ })
    .populate('_creator')
    .exec(function (err, story) {
      if (err) return handleError(err);
      console.log('The creator is %s', story._creator.name); // prints "The creator is Aaron"
    })

  :markdown
    Populated paths are no longer set to their original `ObjectId`s, their value is replaced with the mongoose document returned from the database by performing a separate query before returning the results.

    Arrays of `ObjectId` refs work the same way. Just call the [populate](./api.html#query_Query-populate) method on the query and an array of documents will be returned _in place_ of the `ObjectIds`.

  h3 Field selection
  :markdown
    What if we only want a few specific fields returned for the query? This can be accomplished by passing the usual [field name syntax](./api.html#query_Query-select) as the second argument to the populate method:
  :js
    Story
    .findOne({ title: /timex/i })
    .populate('_creator', 'name') // only return the Persons name
    .exec(function (err, story) {
      if (err) return handleError(err);

      console.log('The creator is %s', story._creator.name);
      // prints "The creator is Aaron"

      console.log('The creators age is %s', story._creator.age);
      // prints "The creators age is null'
    })
  h3 Query conditions for populate
  :markdown
    What if we wanted to populate our fans array based on their age, and return, at most, any 5 of them?
  :js
    Story
    .find(...)
    .populate('fans', null, { age: { $gte: 21 }}, { limit: 5 })
  :markdown
    Done. `Conditions` and `options` for populate queries are passed as the third and fourth arguments respectively.
  h3 Refs to children
  :markdown
    We may find however, if we use the `aaron` object, we are unable to get a list of the stories. This is because no `story` objects were ever 'pushed' on to `aaron.stories`.

    There are two perspectives to this story. First, it's nice to have `aaron` know which are his stories.
  :js
    aaron.stories.push(story1);
    aaron.save();
  :markdown
    This allows us to perform a `find` and `populate` combo:
  :js
    Person
    .findOne({ name: 'Aaron' })
    .populate('stories') // only works if we pushed refs to children
    .exec(function (err, person) {
      if (err) return handleError(err);
      console.log(person);
    })
  :markdown
    However, it is debatable that we really want two sets of pointers as they may get out of sync. So we could instead merely `find()` the documents we are interested in.
  :js
    Story
    .find({ _creator: aaron._id })
    .populate('_creator') // not really necessary
    .exec(function (err, stories) {
      if (err) return handleError(err);
      console.log('The stories are an array: ', stories);
    })

  h3 Updating refs
  :markdown
    Now that we have a `story` we realized that the `_creator` was incorrect. We can update `ObjectId` refs the same as any other property through the magic of Mongooses internal casting:
  :js
    var guille = new Person({ name: 'Guillermo' });
    guille.save(function (err) {
      if (err) return handleError(err);

      story._creator = guille; // or guille._id

      story.save(function (err) {
        if (err) return handleError(err);

        Story
        .findOne({ title: /timex/i })
        .populate('_creator', ['name'])
        .exec(function (err, story) {
          if (err) return handleError(err);

          console.log('The creator is %s', story._creator.name)
          // prints "The creator is Guillermo"
        })
      })
    })
  h4 NOTE:
  :markdown
    The documents returned from calling [populate](./api.html#query_Query-populate) become fully functional, `remove`able, `save`able documents. Do not confuse them with [sub docs](./subdocs.html). Take caution when calling its remove method because you'll be removing it from the database, not just the array.