tl;dr

Yes, it is possible to have multiple documents with a field set to null or not defined, while enforcing unique "actual" values.

requirements:

• MongoDB v3.2+.
• Knowing your concrete value type(s) in advance (e.g, always a string or object when not null).

If you're not interested in the details, feel free to skip to the implementation section.

longer version

To supplement @Nolan's answer, starting with MongoDB v3.2 you can use a partial unique index with a filter expression.

The partial filter expression has limitations. It can only include the following:

• equality expressions (i.e. field: value or using the $eq operator),
• $exists: true expression,
• $gt, $gte, $lt, $lte expressions,
• $type expressions,
• $and operator at the top-level only

This means that the trivial expression {"yourField"{$ne: null}} cannot be used.

However, assuming that your field always uses the same type, you can use a $type expression.

{ field: { $type: <BSON type number> | <String alias> } }

MongoDB v3.6 added support for specifying multiple possible types, which can be passed as an array:

{ field: { $type: [ <BSON type1> , <BSON type2>, ... ] } }

which means that it allows the value to be of any of a number of multiple types when not null.

Therefore, if we want to allow the email field in the example below to accept either string or, say, binary data values, an appropriate $type expression would be:

{email: {$type: ["string", "binData"]}}

implementation

mongoose

You can specify it in a mongoose schema:

const UsersSchema = new Schema({
  name: {type: String, trim: true, index: true, required: true},
  email: {
    type: String, trim: true, index: {
      unique: true,
      partialFilterExpression: {email: {$type: "string"}}
    }
  }
});

or directly add it to the collection (which uses the native node.js driver):

User.collection.createIndex("email", {
  unique: true,
  partialFilterExpression: {
    "email": {
      $type: "string"
    }
  }
});

native mongodb driver

using collection.createIndex

db.collection('users').createIndex({
    "email": 1
  }, {
    unique: true,
    partialFilterExpression: {
      "email": {
        $type: "string"
      }
    }
  },
  function (err, results) {
    // ...
  }
);

mongodb shell

using db.collection.createIndex:

db.users.createIndex({
  "email": 1
}, {
  unique: true, 
  partialFilterExpression: {
    "email": {$type: "string"}
  }
})

This will allow inserting multiple records with a null email, or without an email field at all, but not with the same email string.

Just a quick update to those researching this topic.

The selected answer will work, but you might want to consider using partial indexes instead.

Changed in version 3.2: Starting in MongoDB 3.2, MongoDB provides the option to create partial indexes. Partial indexes offer a superset of the functionality of sparse indexes. If you are using MongoDB 3.2 or later, partial indexes should be preferred over sparse indexes.

More doco on partial indexes: https://docs.mongodb.com/manual/core/index-partial/

Actually, only first document where "email" as field does not exist will get save successfully. Subsequent saves where "email" is not present will fail while giving error ( see code snippet below). For the reason look at MongoDB official documentation with respect to Unique Indexes and Missing Keys here at http://www.mongodb.org/display/DOCS/Indexes#Indexes-UniqueIndexes.

  // NOTE: Code to executed in mongo console.

  db.things.ensureIndex({firstname: 1}, {unique: true});
  db.things.save({lastname: "Smith"});

  // Next operation will fail because of the unique index on firstname.
  db.things.save({lastname: "Jones"});

By definition unique index can only allow one value to be stored only once. If you consider null as one such value it can only be inserted once! You are correct in your approach by ensuring and validating it at application level. That is how it can be done.

You may also like to read this http://www.mongodb.org/display/DOCS/Querying+and+nulls