Skip to content

Latest commit

 

History

History
70 lines (56 loc) · 2.29 KB

README.md

File metadata and controls

70 lines (56 loc) · 2.29 KB

mongo-schemer

Provides clear Mongo schema validation error messages by running all validation errors through ajv. Does this by wrapping your Mongo client db in order to catch validation errors and compare against your schema.

Requirements

Installation

npm install mongo-schemer

Usage

Wrap your db with MongoSchemer.explainSchemaErrors():

const { MongoClient } = require('mongodb');
const MongoSchemer = require('mongo-schemer');

MongoClient.connect(dbUrl).then((client) => {
  const db = client.db(dbName);
  if (process.env.DEV) {
    db = MongoSchemer.explainSchemaErrors(db, {
      onError: (errors) => { console.log(errors); },
      includeValidationInError: true,
    });
  }
});

Options
onError allows the enhanced validation errors to be dealt with within its callback. Optional.

includeValidationInError includes the enhanced validation errors within the original MongoDB error under the key validationErrors. This allows the error to be handled at any point in its callstack. Optional.

Example Error
If using the onError callback only the validationErrors array is provided:

{
  message: "Document failed validation",
  ..., // original error details
  validationErrors: [
    {
      message: 'should be date got this should be a date instead',
      keyword: 'bsonType',
      params: { bsonType: 'date' },
      dataPath: '.created',
      schemaPath: '#/properties/created/bsonType'
    },
    {
      message: 'should NOT have additional properties',
      keyword: 'additionalProperties',
      params: { additionalProperty: 'extraFieldThatDoesNotExist' },
      dataPath: '.items[0]',
      schemaPath: '#/properties/items/items/additionalProperties'
    }
  ]
}

See tests for more details.

Production Usage
Running in prod is not recommended due to the overhead of validating documents against the schema.

Caveats

MongoDB does not currently support definitions in the JSON Schema. However, there is an easy workaround using json-schema-ref-parser. See https://brianschiller.com/blog/2018/01/30/pokemon-mongo-schemer for more information.