mongoose-schema-jsonschema
The module allows to create json schema from Mongoose schema by adding
jsonSchema method to mongoose.Schema, mongoose.Model and mongoose.Query
classes
Contents
Installation
npm install mongoose-schema-jsonschemaSchema Build Configuration
Since v1.4.0 it is able to configure how jsonSchema() works.
To do that package was extended with config function.
const config = require('mongoose-schema-jsonschema/config');
config({
// ... options go here
});Currently there are two options that affects build process:
-
forceRebuild:
boolean-- mongoose-schema-jsonschema caches json schemas built for mongoose schemas. That means we cannot built updated jsonSchema after some updates were made in the mongoose schema that already has jsonSchema. To resolve this issue theforceRebuildwas added (see sample bellow) -
fieldOptionsMapping:
{ [key: string]: string } | Array<string|[string, string]>- allows to specify how to convert some custom options specified in the mongoose field definition.
const mongoose = require('mongoose-schema-jsonschema')();
const config = require('mongoose-schema-jsonschema/config');
const { Schema } = mongoose;
const BookSchema = new Schema({
title: { type: String, required: true, notes: 'Book Title' },
year: Number,
author: { type: String, required: true },
});
const fieldOptionsMapping = {
notes: 'x-notes',
};
config({ fieldOptionsMapping });
console.dir(BookSchema.jsonSchema(), { depth: null });
config({ fieldOptionsMapping: [], forceRebuild: true }); // reseting
console.dir(BookSchema.jsonSchema(), { depth: null });Output:
{
type: 'object',
properties: {
title: { type: 'string', 'x-notes': 'Book Title' },
year: { type: 'number' },
author: { type: 'string' },
_id: { type: 'string', pattern: '^[0-9a-fA-F]{24}$' }
},
required: [ 'title', 'author' ]
}
{
type: 'object',
properties: {
title: { type: 'string' },
year: { type: 'number' },
author: { type: 'string' },
_id: { type: 'string', pattern: '^[0-9a-fA-F]{24}$' }
},
required: [ 'title', 'author' ]
}Samples
Let's build json schema from simple mongoose schema
const mongoose = require('mongoose');
require('mongoose-schema-jsonschema')(mongoose);
const Schema = mongoose.Schema;
const BookSchema = new Schema({
title: { type: String, required: true },
year: Number,
author: { type: String, required: true },
});
const jsonSchema = BookSchema.jsonSchema();
console.dir(jsonSchema, { depth: null });Output:
{
type: 'object',
properties: {
title: { type: 'string' },
year: { type: 'number' },
author: { type: 'string' },
_id: { type: 'string', pattern: '^[0-9a-fA-F]{24}$' }
},
required: [ 'title', 'author' ]
}The mongoose.Model.jsonSchema method allows to build json schema considering the field selection and population
const mongoose = require('mongoose');
require('mongoose-schema-jsonschema')(mongoose);
const Schema = mongoose.Schema;
const BookSchema = new Schema({
title: { type: String, required: true },
year: Number,
author: { type: Schema.Types.ObjectId, required: true, ref: 'Person' }
});
const PersonSchema = new Schema({
firstName: { type: String, required: true },
lastName: { type: String, required: true },
dateOfBirth: Date
});
const Book = mongoose.model('Book', BookSchema);
const Person = mongoose.model('Person', PersonSchema)
console.dir(Book.jsonSchema('title year'), { depth: null });
console.dir(Book.jsonSchema('', 'author'), { depth: null });Output:
{
title: 'Book',
type: 'object',
properties: {
title: { type: 'string' },
year: { type: 'number' },
_id: { type: 'string', pattern: '^[0-9a-fA-F]{24}$' }
}
}
{
title: 'Book',
type: 'object',
properties: {
title: { type: 'string' },
year: { type: 'number' },
author: {
title: 'Person',
type: 'object',
properties: {
firstName: { type: 'string' },
lastName: { type: 'string' },
dateOfBirth: { type: 'string', format: 'date-time' },
_id: { type: 'string', pattern: '^[0-9a-fA-F]{24}$' },
__v: { type: 'number' }
},
required: [ 'firstName', 'lastName' ],
'x-ref': 'Person',
description: 'Refers to Person'
},
_id: { type: 'string', pattern: '^[0-9a-fA-F]{24}$' },
__v: { type: 'number' }
},
required: [ 'title', 'author' ]
}const mongoose = require('mongoose');
const extendMongooose = require('mongoose-schema-jsonschema');
extendMongooose(mongoose);
const { Schema } = mongoose;
const BookSchema = new Schema({
title: { type: String, required: true },
year: Number,
author: { type: Schema.Types.ObjectId, required: true, ref: 'Person' }
});
const Book = mongoose.model('Book', BookSchema);
const Q = Book.find().select('title').limit(5);
console.dir(Q.jsonSchema(), { depth: null });Output:
{
title: 'List of books',
type: 'array',
items: {
type: 'object',
properties: {
title: { type: 'string' },
_id: { type: 'string', pattern: '^[0-9a-fA-F]{24}$' }
}
},
maxItems: 5
}Validation tools
Created by mongoose-schema-jsonschema json-schema's could be used for document validation with:
Specifications
mongoose.Schema.prototype.jsonSchema
Builds the json schema based on the Mongooose schema. if schema has been already built the method returns new deep copy
Method considers the schema.options.toJSON.virtuals to included
the virtual paths (without detailed description)
Declaration:
function schema_jsonSchema(name) { ... }Parameters:
- name:
String- Name of the object - Returns
Object- json schema
mongoose.Model.jsonSchema
Builds json schema for model considering the selection and population
if fields specified the method removes required constraints
Declaration:
function model_jsonSchema(fields, populate) { ... }Parameters:
- fields:
String|Array|Object- mongoose selection object - populate:
String|Object- mongoose population options - Returns
Object- json schema
mongoose.Query.prototype.jsonSchema
Builds json schema considering the query type and query options.
The method returns the schema for array if query type is find and
the schema for single document if query type is findOne or findOneAnd*.
In case when the method returns schema for array the collection name is used to
form title of the resulting schema. In findOne* case the title is the name
of the appropriate model.
Declaration:
function query_jsonSchema() { ... }Parameters:
- Returns
Object- json schema
Custom Schema Types Support
If you use custom Schema Types you should define the jsonSchema method for your type-class(es).
The base functionality is accessible from your code by calling base-class methods:
newSchemaType.prototype.jsonSchema = function() {
// Simple types (strings, numbers, bools):
const jsonSchema = mongoose.SchemaType.prototype.jsonSchema.call(this);
// Date:
const jsonSchema = Types.Date.prototype.jsonSchema.call(this);
// ObjectId
const jsonSchema = Types.ObjectId.prototype.jsonSchema.call(this);
// for Array (or DocumentArray)
const jsonSchema = Types.Array.prototype.jsonSchema.call(this);
// for Embedded documents
const jsonSchema = Types.Embedded.prototype.jsonSchema.call(this);
// for Mixed documents:
const jsonSchema = Types.Mixed.prototype.jsonSchema.call(this);
/*
*
* Place your code instead of this comment
*
*/
return jsonSchema;
}Releases
- version 1.0 - Basic functionality
- version 1.1 - Mongoose.Query support implemented
- version 1.1.5 - uuid issue fixed, ajv compliance verified
- version 1.1.8 - Schema.Types.Mixed issue fixed
- version 1.1.9 - readonly settings support added
- version 1.1.11 - required issue fixed issue#2
- version 1.1.12 - mixed-type fields description and title support added (fix for issue: issue#3)
- version 1.1.15 - support for [email protected] ensured issue#8
- version 1.3.0
- version 1.3.1 - support
minlenghtandmaxlengthissue#21 - version 1.4.0 - broken - schema build configurations (
forceRebuildandfieldOptionsMapping) - version 1.4.2 - fix for broken version 1.4.0 issue#22
- version 1.4.4 - fix for field constaints issue#25
- version 2.0.0 - Support for [email protected]. Node v8.x.x, v9.x.x are no longer supported (use v1.4.7 of the lib)
- version 2.1.0 - Support for [email protected] and Node v14.x, v16.x, v18.x
Supported versions
- node.js: 14.x, 16.x, 18.x
- mongoose: 5.x, 6.x, 7.x