mocha-chai-rdf

Helpers to simplify testing code which uses RDF/JS data models.

chai matchers plugin

Setup

Register the matcher plugin with chai for example in a mocha test setup file:

import { use } from 'chai'
import rdfMatchers from 'mocha-chai-rdf/matchers.js'

use(rdfMatchers)

Term equality

The plugin overrides the eq assertion to compare RDF terms and clownface pointers.

import { expect } from 'chai'
import type { Term } from '@rdfjs/types'
import type { AnyPointer } from 'clownface'
import rdf from '@zazuko/env'

let actual: Term | AnyPointer
expect(actual).to.eq(rdf.ns.schema.Person)

Additionally, deep equality is also configured (chai v5+).

import { expect } from 'chai'
import type { Term } from '@rdfjs/types'
import type { AnyPointer } from 'clownface'
import rdf from '@zazuko/env'
  
let value: Record<{ type: Term }> // for example, results from SPARQL SELECT
expect(value).to.deep.eq([
  { type: rdf.ns.schema.Person }
])

Dataset snapshot testing

Usage

import type { DatasetCore } from '@rdfjs/types'
import { use } from 'chai'
import snapshots from 'mocha-chai-rdf/snapshots.js'

use(snapshots)

let dataset: DatasetCore
expect(dataset).canonical.toMatchSnapshot()

In-memory dataset in mocha tests

Basic example

In the example below, the call to createStore will load the RDF file test/my-test.spec.ts.ttl and populate an in-memory store with its contents. The store and other properties are then available in the test suite context under .rdf key.

// test/my-test.spec.ts
import { createStore } from 'mocha-chai-rdf/store.js'

describe('my test suite', () => {
  beforeEach(createStore(import.meta.url))
  
  context('foo', () => {
    context('bar', () => {
      it('should do something', function () {
        // you can access the dataset, graph and store
        // loaded from test fixture RDF file
        let { 
          dataset,       // RDF/JS dataset
          graph,         // clownface graph pointer
          store,         // oxigraph in-memory store
          parsingClient, // sparql-http-client-compatible client which returns parsed results
          streamClient   // sparql-http-client-compatible client which stream results
        } = this.rdf
      })
    })
  })
})

Load graphs from trig dataset files

To load graphs from a trig file, pass the trig option:

beforeEach(createStore(import.meta.url, { format: 'trig' }))

This will load a graph named <foo/bar> from the trig file. As you see, by default the topmost describe block name and the test name itself are omitted.

The titles of describe/context/it calls are encoded with to create valid URIs.

You can customize the options:

• include the default graph by adding includeDefaultGraph: true
• load all graphs by adding loadAll: true
• control the graph name by adding sliceTestPath: [number, number]. The default is [1, -1] which, as described above, omits the topmost describe block name and the test name itself, hence the first and last part are sliced from the array of block titles.