json-schema-diff

What is json-schema-diff?

The json-schema-diff npm package is a tool for comparing JSON schemas. It helps identify differences between two JSON schemas, which can be useful for versioning, migration, and ensuring compatibility between different versions of APIs or data structures.

What are json-schema-diff's main functionalities?

Compare JSON Schemas

This feature allows you to compare two JSON schemas and identify the differences between them. The code sample demonstrates how to use the json-schema-diff package to compare two schemas and print the differences.

const jsonSchemaDiff = require('json-schema-diff');

const schema1 = {
  type: 'object',
  properties: {
    name: { type: 'string' },
    age: { type: 'number' }
  },
  required: ['name']
};

const schema2 = {
  type: 'object',
  properties: {
    name: { type: 'string' },
    age: { type: 'number' },
    email: { type: 'string' }
  },
  required: ['name', 'email']
};

jsonSchemaDiff.diffSchemas({
  sourceSchema: schema1,
  destinationSchema: schema2
}).then((diff) => {
  console.log(JSON.stringify(diff, null, 2));
}).catch((error) => {
  console.error(error);
});

Other packages similar to json-schema-diff

json-diff

The json-diff package provides a way to compare two JSON objects and highlight the differences. While it is not specifically designed for JSON schemas, it can be used to compare any JSON data structures. It is more general-purpose compared to json-schema-diff, which is specialized for JSON schema comparison.

deep-diff

The deep-diff package allows for deep comparison of JavaScript objects, including JSON objects. It can be used to find differences between two objects at any depth. Similar to json-diff, it is not specialized for JSON schemas but can be used for a wide range of JSON data comparisons.

diff

The diff package is a general-purpose text comparison tool that can be used to compare JSON strings. It provides a variety of diff algorithms and output formats. While it is not specifically designed for JSON schemas, it can be used to compare JSON data by converting it to strings.

README

Json Schema Diff

A language agnostic CLI tool and nodejs api to identify differences between two json schema files.

Requirements

• nodejs 10.x or higher (tested using 10.x, 12.x, 14.x and 16.x)
• npm 6.x or higher (tested using 6.x and 7.x)

Installation

Install the tool using npm and add it to the package.json

npm install json-schema-diff --save-dev

Description

This tool identifies what has changed between two json schema files. These changes are classified into two groups, added and removed. Using an approach based on set theory this tool is able to calculate these differences to a high level of accuracy.

KEYWORDS.md contains the details of what json schema keywords are supported.

A change is considered an addition when the destination schema has become more permissive relative to the source schema. For example {"type": "string"} -> {"type": ["string", "number"]}.

A change is considered a removal when the destination schema has become more restrictive relative to the source schema. For example {"type": ["string", "number"]} -> {"type": "string"}.

The addition and removal changes detected are returned in JsonSchema format. These schemas represent the set of values that have been added or removed.

Example

Source Schema

{
    "properties": {
        "id": {
            "type": "number"
        }
    },
    "type": "object"
}

Destination Schema

{
    "properties": {
        "id": {
            "type": ["string", "number"]
        }
    },
    "type": "object"
}

Schema representing what was added

All objects that contain an id property of type string. The id property is required because both source and destination schemas accept objects without an id property, so we want to exclude those objects from the added result.

{
    "properties": {
        "id": {
            "type": "string"
        }
    },
    "required": ["id"],
    "type": "object"
}

Schema representing what was removed

All values accepted by the source schema are also accepted by the destination schema, so the removed result is a schema that accepts no values.

false

Usage

Usage as a cli tool

Invoke the tool with a file path to the source schema file and the destination schema file. These files should be in JSON format and be valid according to the json schema draft-07 specification.

The tool will return two json schemas as output, one representing the values that were added by the destination schema and the other representing the values that were removed by the destination schema.

The tool will fail if any removed differences are detected.

Example

/path/to/source-schema.json

{
  "type": "string"
}

/path/to/destination-schema.json

{
  "type": ["string", "number"]
}

Invoking the tool

json-schema-diff /path/to/source-schema.json /path/to/destination-schema.json

Output

Non-breaking changes found between the two schemas.

Values described by the following schema were added:
{
    "type": [
        "number"
    ]
}

Values described by the following schema were removed:
false

Usage as a nodejs api

Invoke the library with the source schema and the destination schema. These objects should be simple javascript objects and be valid according to the json schema draft-07 specification.

For full details of the nodejs api please refer to api-types.d.ts

Example

const jsonSchemaDiff = require('json-schema-diff');

const source = {type: 'string'};
const destination = {type: ['string', 'number']};

const result = await jsonSchemaDiff.diffSchemas({
    sourceSchema: source, 
    destinationSchema: destination
});

if (result.removalsFound) {
    console.log('Something was removed!');
}

if (result.additionsFound) {
    console.log('Something was added!');
}

Additional settings

During the process of parsing input schemas and/or diffing, some mathematical functions and heuristics might be applied to improve the performance of calculations. These do not affect the format or content of the results.

Some of these simplifications, while having a positive effect on memory consumption may require a higher CPU usage. If you come across memory consumption issues or process taking too slow to finish you may give it a try at turning these on/off, via environment variables

Apply law of absorption in cartesian product

• Description: The intersection of two relatively large groups of sets can produce a large number of intermediate sets, which may lead to high memory consumption ( n×m results given inputs of size n and m). Some of these intermediate results can be discarded immediately if it can be proved to be subsets of previously calculated results ( A ∪ (A ∩ B) ⇔ A ).
• Environment variable: JSON_SCHEMA_DIFF_APPLY_ABSORPTION_IN_CARTESIAN_PRODUCT
• Default value: true
• When enabled: Performs additional calculations (requires CPU) to reduce memory usage
• When disabled: Avoid additional computation, with the risk of high memory usage

Changelog

See CHANGELOG.md

Contributing

See CONTRIBUTING.md

License

See LICENSE.txt