mongoose-unique-validator is a plugin which adds pre-save validation for unique fields within a Mongoose schema.
This makes error handling much easier, since you will get a Mongoose validation error when you attempt to violate a unique constraint, rather than an E11000 error from MongoDB.
Yarn: yarn add mongoose-unique-validator
NPM: npm install --save mongoose-unique-validator
Then, apply the plugin to your schema:
var mongoose = require('mongoose');
var uniqueValidator = require('mongoose-unique-validator');
var mySchema = mongoose.Schema(/* put your schema definition here */);
mySchema.plugin(uniqueValidator);
Let’s say you have a user schema. You can easily add validation for the unique constraints in this schema by applying
the uniqueValidator
plugin to your user schema:
var mongoose = require('mongoose');
var uniqueValidator = require('mongoose-unique-validator');
// Define your schema as normal.
var userSchema = mongoose.Schema({
username: { type: String, required: true, unique: true },
email: { type: String, index: true, unique: true, required: true },
password: { type: String, required: true }
});
// Apply the uniqueValidator plugin to userSchema.
userSchema.plugin(uniqueValidator);
Now when you try to save a user, the unique validator will check for duplicate database entries and report them just like any other validation error:
var user = new User({ username: 'JohnSmith', email: 'john.smith@gmail.com', password: 'j0hnNYb0i' });
user.save(function (err) {
console.log(err);
});
{
message: 'Validation failed',
name: 'ValidationError',
errors: {
username: {
message: 'Error, expected `username` to be unique. Value: `JohnSmith`',
name: 'ValidatorError',
kind: 'unique',
path: 'username',
value: 'JohnSmith'
}
}
}
When using findOneAndUpdate
and related methods, mongoose doesn't automatically run validation. To trigger this,
you need to pass a configuration object. For technical reasons, this plugin requires that you also set the context
option to query
.
{ runValidators: true, context: 'query' }
A full example:
User.findOneAndUpdate(
{ email: 'old-email@example.com' },
{ email: 'new-email@example.com' },
{ runValidators: true, context: 'query' },
function(err) {
// ...
}
)
You can pass through a custom error type as part of the optional options
argument:
userSchema.plugin(uniqueValidator, { type: 'mongoose-unique-validator' });
After running the above example the output will be:
{
message: 'Validation failed',
name: 'ValidationError',
errors: {
username: {
message: 'Error, expected `username` to be unique. Value: `JohnSmith`',
name: 'ValidatorError',
kind: 'mongoose-unique-validator',
path: 'username',
value: 'JohnSmith'
}
}
}
You can also specify a default custom error type by overriding the plugin defaults.type
variable:
uniqueValidator.defaults.type = 'mongoose-unique-validator'
You can pass through a custom error message as part of the optional options
argument:
userSchema.plugin(uniqueValidator, { message: 'Error, expected {PATH} to be unique.' });
You have access to all of the standard Mongoose error message templating:
{PATH}
{VALUE}
{TYPE}
You can also specify a default custom error message by overriding the plugin defaults.message
variable:
uniqueValidator.defaults.message = 'Error, expected {PATH} to be unique.'
For case-insensitive matches, include the uniqueCaseInsensitive
option in your schema. Queries will treat john.smith@gmail.com
and John.Smith@gmail.com
as duplicates.
var userSchema = mongoose.Schema({
username: { type: String, required: true, unique: true },
email: { type: String, index: true, unique: true, required: true, uniqueCaseInsensitive: true },
password: { type: String, required: true }
});
For additional unique-constraint conditions (ex: only enforce unique constraint on non soft-deleted records), the MongoDB option partialFilterExpression
can be used.
Note: the option index
must be passed as an object containing unique: true
, or else partialFilterExpression
will be ignored.
var userSchema = mongoose.Schema({
username: { type: String, required: true, unique: true },
email: {
type: String,
required: true,
index: {
unique: true,
partialFilterExpression: { deleted: false }
}
},
password: { type: String, required: true }
});
Because we rely on async operations to verify whether a document exists in the database, it's possible for two queries to execute at the same time, both get 0 back, and then both insert into MongoDB.
Outside of automatically locking the collection or forcing a single connection, there's no real solution.
For most of our users this won't be a problem, but is an edge case to be aware of.