Skip to main content


Ottoman documents represent a one-to-one mapping to documents as stored in Couchbase Server. Each document is an instance of its Model.

Documents vs Models

Document and Model are distinct classes in Ottoman. The Model class is a subclass of the Document class. When you use the Model constructor, you create a new document.

const MyModel = ottoman.model('Test', new Schema({ name: String }));
const doc = new MyModel();

doc instanceof MyModel; // true
doc instanceof ottoman.Model; // true
doc instanceof ottoman.Document; // true

In Ottoman, a document means an instance of a model. No need to create an instance of the Document class.


When you load documents from Couchbase Server using model functions like findById(), you get an Ottoman document back.

const doc = await MyModel.findOne();

doc instanceof MyModel; // true
doc instanceof ottoman.Model; // true
doc instanceof ottoman.Document; // true


Ottoman documents track changes. You can modify a document using vanilla JavaScript assignments and Ottoman will convert it into Couchbase update operations. = 'foo';

// Ottoman sends an `updateById(, { name: 'foo' })`
// to Couchbase Server.

If the document with the corresponding id is not found, Ottoman will report a DocumentNotFoundError:

const doc = await MyModel.findOne();

// Delete document so `save()` will throw. Ottoman cannot save.
await MyModel.removeById(doc._id); = 'foo';
await; // Throws `DocumentNotFoundError` on `save()`.

The save() function is generally the right way to update a document with Ottoman. With save(), you get full validation and middleware.


Documents are casted and validated before saved. Ottoman casts values to the specified type and then validates them. Internally, Ottoman calls the document's _validate() method before saving.

const schema = new Schema({ name: String, age: { type: Number, min: 0 } });
const Person = ottoman.model('Person', schema);

let p = new Person({ name: 'foo', age: 'bar' });
// Cast to Number failed for value "bar" at path "age"
await p._validate();

let p2 = new Person({ name: 'foo', age: -1 });
// Path `age` (-1) is less than minimum allowed value (0).
await p2._validate();


Population is the process of automatically replacing the specified paths in a document using document(s) from other collection(s). We may populate a single document, multiple documents, a plain object, multiple plain objects, or all objects returned from a query.


import {Schema, model} from 'ottoman';

const personSchema = new Schema({
name: String,
age: Number,
stories: [{ type: String, ref: 'Story' }]

const storySchema = new Schema({
author: { type: String, ref: 'Person' },
title: String,
fans: [{ type: String, ref: 'Person' }]

const Story = model('Story', storySchema);
const Person = model('Person', personSchema);

So far we've created two Models. Our Person model has its stories field set to an array of ids. The ref option is what tells Ottoman which model to use during population, in our case the Story model. All ids we store here must be document ids from the Story model.

Saving Refs

Saving refs to other documents works the same way you normally save properties, just assign the id value:

const author = new Person({
name: 'Ian Fleming',
age: 50

const story1 = new Story({
title: 'Casino Royale',
author: // assign the id from the person


Using Population

So far we haven't done anything much different. We've merely created a Person and a Story. Now let's take a look at populating our Story's author:

// Populate using document
const story = await Story.findById('storyId')
await story._populate('author');
console.log('The author is %s',;
// prints "The author is Ian Fleming"

// Populate using Model
const stories = await Story.find({ title: 'Casino Royale' }, {populate: 'author'})
const story = stories[0];
console.log('The author is %s',;
// prints "The author is Ian Fleming"

Populated paths are no longer set to their original id, their value is replaced with the Ottoman document returned from the database by performing a separate query before returning the results.

Arrays of refs work the same way. Just call the _populate method on the query, and an array of documents will be returned in place of the original ids.

You can see more examples of pupulate here.

Checking Whether a Field is Populated

You can call the _populated() function to check whether a field is populated. If _populated() returns a truthy value, you can assume the field is populated.

story._populated('author'); // truthy

story._depopulate('author'); // Make `author` not populated anymore
story._populated('author'); // undefined

A common reason for checking whether a path is populated is getting the

story._populated('author'); // truthy;

story._depopulate('author'); // Make `author` not populated anymore
story._populated('author'); // false

Advanced Population

Use * as a wildcard to populate all references in the current model. In this example all properties types references will be populated:

await story._populate('*');

Also, you can automatically populate child document references by passing a second integer value to _populate function, this value will tell to Ottoman how many levels deep you want to populate.

// Using on a document
await story._populate('*', 2);

// Using on a Model
const stories = await Story.find({ title: 'Casino Royale' }, { populate: 'author', populateMaxDeep: 2 })

In the above example Ottoman will populate all references on story and story's children. It doesn't matter if they are single or array references.


Beware from setting a large integer value on populate deep argument, it could affect the query performance.

Next Up

Now that we've covered Documents, let's take a look at Query Builder.