Merge pull request #11 from fourTheorem/vitest

Use vitest for unit tests

transcript-orchestration/functions/audio-transcoder/app.ts

@@ -65,8 +65,7 @@ export const handleEvent = middify(async (event: TranscodeEvent, context: Contex
     try {
       unlink(tempInputFilePath)
       unlink(tempOutputFilePath)
-    } catch (err) {
-      /* istanbul ignore next */
+    } /* c8 ignore next 3 */ catch (err) {
       logger.warn('Failed to delete temporary files', { err })
     }
   }

transcript-orchestration/functions/audio-transcoder/tests/unit/app.test.ts

@@ -5,31 +5,27 @@ import { resolve } from 'node:path'
 import { mockClient } from 'aws-sdk-client-mock'
 import { sdkStreamMixin } from '@aws-sdk/util-stream-node'
 import { S3Client, PutObjectCommand, GetObjectCommand } from '@aws-sdk/client-s3'
-import tap from 'tap'
-import { AudioTranscodeOptions } from '../../transcode'
 import { mockedContext } from '../../../mock-utils/lambda-context'
-
-const mockS3 = mockClient(S3Client)
-
+import { test, expect, vi } from 'vitest'
 process.env.BUCKET_NAME = 'test-bucket'
+import * as app from '../../app'
+import * as transcode from '../../transcode'
 
-const testState: { options?: AudioTranscodeOptions } = {}
-
-const { handleEvent } = tap.mock('../../app', {
-  '../../transcode': {
-    transcodeAudio: async (options: AudioTranscodeOptions) => {
-      testState.options = options
-      await writeFile(options.outputFilePath, 'mp3 contents')
-    }
-  }
-})
+const mockS3 = mockClient(S3Client)
+vi.mock('../../transcode', () => ({
+  transcodeAudio: vi.fn(async (options) => {
+    writeFile(options.outputFilePath, 'mp3 contents')
+  })
+}))
 
-tap.test('converts audio using FFmpeg', async (t) => {
+test('converts audio using FFmpeg', async () => {
   const stream = createReadStream(resolve(__dirname, '../../../../../sample-audio/sample1.m4a'))
   const sdkStream = sdkStreamMixin(stream)
   mockS3.on(GetObjectCommand).resolves({ Body: sdkStream })
   mockS3.on(PutObjectCommand).resolves({})
-  await handleEvent({ audioInputKey: 'sample1.m4a', audioOutputKey: 'sample1.mp3' }, mockedContext)
-  t.ok(testState.options?.inputFilePath)
-  t.ok(testState.options?.outputFilePath)
+  await app.handleEvent({ audioInputKey: 'sample1.m4a', audioOutputKey: 'sample1.mp3' }, mockedContext)
+  expect(transcode.transcodeAudio).toHaveBeenCalledWith(expect.objectContaining({
+    inputFilePath: expect.any(String),
+    outputFilePath: expect.any(String)
+  }))
 })

transcript-orchestration/functions/lib/tests/envs.test.ts

@@ -1,12 +1,12 @@
 process.env.EXAMPLE_KEY_1 = 'test'
 
 import envs from '../envs'
-import { test } from 'tap'
+import { test, assert } from 'vitest'
 
-test('envs retrieves environment variables if available', async (t) => {
-  t.equal(envs.EXAMPLE_KEY_1, 'test')  
+test('envs retrieves environment variables if available', async () => {
+  assert.equal(envs.EXAMPLE_KEY_1, 'test')  
 })
 
-test('envs throws an error if an environment variable is missing', async (t) => {
-  t.throws(() => envs.EXAMPLE_KEY_2)
+test('envs throws an error if an environment variable is missing', async () => {
+  assert.throws(() => envs.EXAMPLE_KEY_2)
 })

transcript-orchestration/functions/mock-utils/lambda-context.ts

@@ -1,6 +1,5 @@
 import { Context } from 'aws-lambda';
 
-/* istanbul ignore next */
 export const mockedContext: Context = {
   callbackWaitsForEmptyEventLoop: false,
   functionName: 'mockFunction',

transcript-orchestration/functions/s3-keys/tests/unit/app.test.ts

@@ -1,15 +1,13 @@
-import { sdkStreamMixin } from '@aws-sdk/util-stream-node'
-import tap from 'tap'
-import { mockedContext } from '../../../mock-utils/lambda-context'
+import { test, expect } from 'vitest'
 import { handleEvent, S3KeysEvent } from '../../app'
 
 process.env.BUCKET_NAME = 'test-bucket'
 
 const testState: { options?: S3KeysEvent } = {}
 
-tap.test('constructs keys from audio input key', async (t) => {
+test('constructs keys from audio input key', async (t) => {
   const result = await handleEvent({ audioInputKey: 'audio/50.m4a' })
-  t.same(result, {
+  expect(result).toEqual({
     mp3Key: 'audio/50.mp3',
     whisperPrefix: 'whisper-batch-output',
     whisperOutputKey: 'whisper-batch-output/50.json',

transcript-orchestration/functions/transcript-processor/tests/unit/app.test.ts

@@ -5,7 +5,7 @@ import { resolve } from 'node:path'
 import { mockClient } from 'aws-sdk-client-mock'
 import { sdkStreamMixin } from '@aws-sdk/util-stream-node'
 import { S3Client, PutObjectCommand, GetObjectCommand } from '@aws-sdk/client-s3'
-import tap from 'tap'
+import { test, assert } from 'vitest'
 
 import { VocabularySubstitutions } from '../../vocabulary'
 import { MergedTranscript } from '../../types'
@@ -27,7 +27,7 @@ const substitutions: VocabularySubstitutions = [
 ]
 
 for (const vocabularySubstitutions of [undefined, substitutions]) {
-  tap.test(`transcript processor generates a merged transcript ${vocabularySubstitutions ? 'with': 'without'} substitutions`, async (t) => {
+  test(`transcript processor generates a merged transcript ${vocabularySubstitutions ? 'with': 'without'} substitutions`, async (t) => {
     const transcribeOutputStream = createReadStream(resolve(__dirname, './resources/1.transcribe'))
     const whisperOutputStream = createReadStream(resolve(__dirname, './resources/1.whisper.out.json'))
 
@@ -48,17 +48,17 @@ for (const vocabularySubstitutions of [undefined, substitutions]) {
     })
 
     mockS3.on(PutObjectCommand).callsFake((input) => {
-      t.equal(input.Bucket, process.env.BUCKET_NAME)
-      t.equal(input.Key, processedTranscriptKey)
+      assert.equal(input.Bucket, process.env.BUCKET_NAME)
+      assert.equal(input.Key, processedTranscriptKey)
       const transcript = JSON.parse(input.Body) as any as MergedTranscript
-      t.equal(transcript.segments.length, 295) 
+      assert.equal(transcript.segments.length, 295) 
       let prevEnd = 0
       for (const segment of transcript.segments) {
         const distance = Number(segment.start.toFixed(2)) - Number(prevEnd.toFixed(2))
-        t.ok(distance >= 0, `${JSON.stringify(segment)} >= ${prevEnd} (${distance})`)
-        t.ok(segment.end > segment.start, JSON.stringify(segment))
-        t.ok(segment.text.length > 0, JSON.stringify(segment))
-        t.ok(segment.speakerLabel.startsWith('spk_'), JSON.stringify(segment))
+        assert.ok(distance >= 0, `${JSON.stringify(segment)} >= ${prevEnd} (${distance})`)
+        assert.ok(segment.end > segment.start, JSON.stringify(segment))
+        assert.ok(segment.text.length > 0, JSON.stringify(segment))
+        assert.ok(segment.speakerLabel.startsWith('spk_'), JSON.stringify(segment))
         prevEnd = segment.end
       }
     })

transcript-orchestration/functions/transcript-processor/tests/unit/process-transcript.test.ts

@@ -1,23 +1,23 @@
-import { closestSpeakerChange, merge } from '../../process-transcripts'
+import { test, assert } from 'vitest'
 
-import tap from 'tap'
+import { closestSpeakerChange, merge } from '../../process-transcripts'
 import { TranscribeSpeakerSegment } from '../../types'
 
-tap.test('it finds the closes speaker change', async (t) => {
+test('it finds the closes speaker change', () => {
   const speakerChangeIndex = [
     { speakerLabel: 'spk_0', start: 0 },
     { speakerLabel: 'spk_1', start: 3 }
   ]
 
-  t.same(closestSpeakerChange(speakerChangeIndex, 0)?.speakerLabel, 'spk_0')
-  t.same(closestSpeakerChange(speakerChangeIndex, 1)?.speakerLabel, 'spk_0')
-  t.same(closestSpeakerChange(speakerChangeIndex, 2)?.speakerLabel, 'spk_0')
-  t.same(closestSpeakerChange(speakerChangeIndex, 3)?.speakerLabel, 'spk_1')
-  t.same(closestSpeakerChange(speakerChangeIndex, 4)?.speakerLabel, 'spk_1')
-  t.same(closestSpeakerChange(speakerChangeIndex, 100)?.speakerLabel, 'spk_1')
+  assert.equal(closestSpeakerChange(speakerChangeIndex, 0)?.speakerLabel, 'spk_0')
+  assert.equal(closestSpeakerChange(speakerChangeIndex, 1)?.speakerLabel, 'spk_0')
+  assert.equal(closestSpeakerChange(speakerChangeIndex, 2)?.speakerLabel, 'spk_0')
+  assert.equal(closestSpeakerChange(speakerChangeIndex, 3)?.speakerLabel, 'spk_1')
+  assert.equal(closestSpeakerChange(speakerChangeIndex, 4)?.speakerLabel, 'spk_1')
+  assert.equal(closestSpeakerChange(speakerChangeIndex, 100)?.speakerLabel, 'spk_1')
 })
 
-tap.test('it merges a simple set of files', async (t) => {
+test('it merges a simple set of files', () => {
   const whisperSegments = [{
     start: 0,
     end: 2,
@@ -74,10 +74,10 @@ tap.test('it merges a simple set of files', async (t) => {
     ]
   }
 
-  t.same(result, expectedResult)
+  assert.deepEqual(result, expectedResult)
 })
 
-tap.test('it merges segments where the first transcribe segment starts after the intiial audio silence', async (t) => {
+test('it merges segments where the first transcribe segment starts after the intiial audio silence', async () => {
   const whisperSegments = [
     { start: 0, end: 7.04, text: ' Node.js is considered by many a game changer, possibly the biggest innovation of the decade' },
     { start: 7.04, end: 36.64, text: ' in web development.' }
@@ -105,23 +105,23 @@ tap.test('it merges segments where the first transcribe segment starts after the
       }
     ]
   }
-  t.same(result, expectedResult)
+  assert.deepEqual(result, expectedResult)
 })
 
-tap.test('it identifies speaker as unknown if there is no speaker data', async (t) => {
+test('it identifies speaker as unknown if there is no speaker data', async () => {
   const whisperSegments = [
     { start: 0, end: 1, text: 'Hello' },
     { start: 1, end: 2, text: 'Goodbye' }
   ]
   const transcribeSegments: TranscribeSpeakerSegment[] = []
   const result = merge(whisperSegments, transcribeSegments)
-  t.same(result.segments.length, 2)
+  assert.equal(result.segments.length, 2)
   for (const segment of result.segments) {
-    t.equal(segment.speakerLabel, 'unknown')
+    assert.equal(segment.speakerLabel, 'unknown')
   }
 })
 
-tap.test('it splits a segment if the speaker changes mid-sentence', async (t) => {
+test('it splits a segment if the speaker changes mid-sentence', () => {
   const whisperSegments = [
     { start: 0, end: 1, text: 'Hello. My name is Bob and I am here with' },
     { start: 1, end: 2, text: ' Alice. How are you today, Alice? I am good actually' },
@@ -159,5 +159,5 @@ tap.test('it splits a segment if the speaker changes mid-sentence', async (t) =>
     ]
   }
   const result = merge(whisperSegments, transcribeSegments)
-  t.same(result, expectedResult)
+  assert.deepEqual(result, expectedResult)
 })

transcript-orchestration/functions/transcript-processor/tests/unit/vocabulary.test.ts

@@ -1,9 +1,9 @@
 import { substituteVocabulary, VocabularySubstitutions } from '../../vocabulary'
 
-import tap from 'tap'
+import { test, assert } from 'vitest'
 import { MergedTranscript } from '../../types'
 
-tap.test('it substitutes with literal and regex searches', async (t) => {
+test('it substitutes with literal and regex searches', async () => {
   const transcript: MergedTranscript = {
     speakers: {
       'spk_0': 'a',
@@ -47,6 +47,6 @@ tap.test('it substitutes with literal and regex searches', async (t) => {
     }
   ]
   substituteVocabulary(transcript, vocab)
-  t.equal(transcript.segments[0].text, 'Hello my name is Eoin and this is AWS Bites.')
-  t.equal(transcript.segments[1].text, ' Hi, my name is Luciano and we are going to talk about Lambda!')
+  assert.equal(transcript.segments[0].text, 'Hello my name is Eoin and this is AWS Bites.')
+  assert.equal(transcript.segments[1].text, ' Hi, my name is Luciano and we are going to talk about Lambda!')
 })