opc-agent 4.1.0 → 4.1.2

package/tests/voice-call.test.ts

@@ -1,47 +1,47 @@
-import { describe, it, expect } from 'vitest';
-import { VoiceCallManager } from '../src/channels/voice-call';
-
-describe('VoiceCallManager', () => {
-  it('should create with config', () => {
-    const mgr = new VoiceCallManager({ provider: 'twilio', credentials: { sid: 'x', token: 'y' } });
-    expect(mgr).toBeInstanceOf(VoiceCallManager);
-  });
-
-  it('should throw without credentials on startCall', async () => {
-    const mgr = new VoiceCallManager({ provider: 'twilio' });
-    await expect(mgr.startCall('+1234')).rejects.toThrow('credentials');
-  });
-
-  it('should start and end a call', async () => {
-    const mgr = new VoiceCallManager({ provider: 'webrtc', credentials: { key: 'val' } });
-    const callId = await mgr.startCall('+1234');
-    expect(callId).toBeTruthy();
-    expect(mgr.getCallStatus(callId)).toBe('ringing');
-    await mgr.endCall(callId);
-    expect(mgr.getCallStatus(callId)).toBe('ended');
-  });
-
-  it('should handle incoming calls', async () => {
-    const mgr = new VoiceCallManager({ provider: 'sip', credentials: { key: 'val' } });
-    const result = await new Promise<{ callId: string; from: string }>((resolve) => {
-      mgr.onIncoming((callId, from) => resolve({ callId, from }));
-      mgr.simulateIncoming('+5678');
-    });
-    expect(result.callId).toBeTruthy();
-    expect(result.from).toBe('+5678');
-  });
-
-  it('should list active calls', async () => {
-    const mgr = new VoiceCallManager({ provider: 'twilio', credentials: { sid: 'x', token: 'y' } });
-    const id1 = await mgr.startCall('+111');
-    const id2 = await mgr.startCall('+222');
-    expect(mgr.listActiveCalls().length).toBe(2);
-    await mgr.endCall(id1);
-    expect(mgr.listActiveCalls().length).toBe(1);
-  });
-
-  it('should throw on unknown callId', () => {
-    const mgr = new VoiceCallManager({ provider: 'twilio', credentials: { sid: 'x' } });
-    expect(() => mgr.getCallStatus('unknown')).toThrow('not found');
-  });
-});
+import { describe, it, expect } from 'vitest';
+import { VoiceCallManager } from '../src/channels/voice-call';
+
+describe('VoiceCallManager', () => {
+  it('should create with config', () => {
+    const mgr = new VoiceCallManager({ provider: 'twilio', credentials: { sid: 'x', token: 'y' } });
+    expect(mgr).toBeInstanceOf(VoiceCallManager);
+  });
+
+  it('should throw without credentials on startCall', async () => {
+    const mgr = new VoiceCallManager({ provider: 'twilio' });
+    await expect(mgr.startCall('+1234')).rejects.toThrow('credentials');
+  });
+
+  it('should start and end a call', async () => {
+    const mgr = new VoiceCallManager({ provider: 'webrtc', credentials: { key: 'val' } });
+    const callId = await mgr.startCall('+1234');
+    expect(callId).toBeTruthy();
+    expect(mgr.getCallStatus(callId)).toBe('ringing');
+    await mgr.endCall(callId);
+    expect(mgr.getCallStatus(callId)).toBe('ended');
+  });
+
+  it('should handle incoming calls', async () => {
+    const mgr = new VoiceCallManager({ provider: 'sip', credentials: { key: 'val' } });
+    const result = await new Promise<{ callId: string; from: string }>((resolve) => {
+      mgr.onIncoming((callId, from) => resolve({ callId, from }));
+      mgr.simulateIncoming('+5678');
+    });
+    expect(result.callId).toBeTruthy();
+    expect(result.from).toBe('+5678');
+  });
+
+  it('should list active calls', async () => {
+    const mgr = new VoiceCallManager({ provider: 'twilio', credentials: { sid: 'x', token: 'y' } });
+    const id1 = await mgr.startCall('+111');
+    const id2 = await mgr.startCall('+222');
+    expect(mgr.listActiveCalls().length).toBe(2);
+    await mgr.endCall(id1);
+    expect(mgr.listActiveCalls().length).toBe(1);
+  });
+
+  it('should throw on unknown callId', () => {
+    const mgr = new VoiceCallManager({ provider: 'twilio', credentials: { sid: 'x' } });
+    expect(() => mgr.getCallStatus('unknown')).toThrow('not found');
+  });
+});

package/tests/voice-enhanced.test.ts

@@ -1,169 +1,169 @@
-import { describe, it, expect, vi } from 'vitest';
-import { VoiceChannel, createVoiceProviders, EdgeTTSProvider, WhisperSTTProvider, OpenAITTSProvider } from '../src/channels/voice';
-import type { STTProvider, TTSProvider, VoiceConfig } from '../src/channels/voice';
-
-describe('VoiceChannel Enhanced', () => {
-  // ── VoiceConfig Parsing ───────────────────────────────────
-
-  it('should parse VoiceConfig with edge-tts', () => {
-    const config: VoiceConfig = {
-      sttProvider: 'whisper',
-      ttsProvider: 'edge-tts',
-      voice: 'en-US-GuyNeural',
-      language: 'en',
-    };
-    const { stt, tts } = createVoiceProviders(config);
-    expect(stt).toBeUndefined(); // no API key
-    expect(tts).toBeDefined();
-    expect(tts!.name).toBe('edge-tts');
-  });
-
-  it('should parse VoiceConfig with whisper + openai-tts', () => {
-    const config: VoiceConfig = {
-      sttProvider: 'whisper',
-      ttsProvider: 'openai-tts',
-      sttApiKey: 'sk-test',
-      ttsApiKey: 'sk-test',
-      voice: 'nova',
-    };
-    const { stt, tts } = createVoiceProviders(config);
-    expect(stt).toBeDefined();
-    expect(stt!.name).toBe('whisper');
-    expect(tts).toBeDefined();
-    expect(tts!.name).toBe('openai-tts');
-  });
-
-  it('should return undefined for web-speech (node-only)', () => {
-    const { stt } = createVoiceProviders({
-      sttProvider: 'web-speech',
-      ttsProvider: 'edge-tts',
-    });
-    expect(stt).toBeUndefined();
-  });
-
-  // ── Transcribe (mock) ─────────────────────────────────────
-
-  it('should transcribe audio with mock STT provider', async () => {
-    const mockSTT: STTProvider = {
-      name: 'mock-stt',
-      transcribe: vi.fn().mockResolvedValue('Hello world'),
-    };
-    const channel = new VoiceChannel({ sttProvider: mockSTT });
-    const text = await channel.transcribe(Buffer.from('audio-data'));
-    expect(text).toBe('Hello world');
-    expect(mockSTT.transcribe).toHaveBeenCalled();
-  });
-
-  it('should throw if no STT provider for transcribe', async () => {
-    const channel = new VoiceChannel();
-    await expect(channel.transcribe(Buffer.from('audio'))).rejects.toThrow('No STT provider');
-  });
-
-  // ── Synthesize (mock) ─────────────────────────────────────
-
-  it('should synthesize text with mock TTS provider', async () => {
-    const mockTTS: TTSProvider = {
-      name: 'mock-tts',
-      synthesize: vi.fn().mockResolvedValue(Buffer.from('audio-output')),
-    };
-    const channel = new VoiceChannel({ ttsProvider: mockTTS });
-    const audio = await channel.synthesize('Hello world');
-    expect(audio).toEqual(Buffer.from('audio-output'));
-    expect(mockTTS.synthesize).toHaveBeenCalledWith('Hello world', { voice: undefined });
-  });
-
-  it('should throw if no TTS provider for synthesize', async () => {
-    const channel = new VoiceChannel();
-    await expect(channel.synthesize('Hello')).rejects.toThrow('No TTS provider');
-  });
-
-  // ── Conversation Mode ─────────────────────────────────────
-
-  it('should start and stop conversation mode', async () => {
-    const channel = new VoiceChannel();
-    await channel.startConversation(async (text) => `Echo: ${text}`);
-    expect(channel.isConversationActive()).toBe(true);
-    channel.stopConversation();
-    expect(channel.isConversationActive()).toBe(false);
-  });
-
-  // ── processConversationTurn ───────────────────────────────
-
-  it('should process a conversation turn', async () => {
-    const mockSTT: STTProvider = {
-      name: 'mock-stt',
-      transcribe: vi.fn().mockResolvedValue('What time is it?'),
-    };
-    const mockTTS: TTSProvider = {
-      name: 'mock-tts',
-      synthesize: vi.fn().mockResolvedValue(Buffer.from('audio-response')),
-    };
-    const channel = new VoiceChannel({ sttProvider: mockSTT, ttsProvider: mockTTS });
-
-    const result = await channel.processConversationTurn(
-      Buffer.from('audio'),
-      async (text) => `It is 3pm, you said: ${text}`,
-    );
-
-    expect(result.text).toBe('What time is it?');
-    expect(result.response).toContain('3pm');
-    expect(result.audioResponse).toBeDefined();
-  });
-
-  // ── processAudio with handler ─────────────────────────────
-
-  it('should process audio end-to-end', async () => {
-    const mockSTT: STTProvider = {
-      name: 'mock-stt',
-      transcribe: vi.fn().mockResolvedValue('Hello agent'),
-    };
-    const mockTTS: TTSProvider = {
-      name: 'mock-tts',
-      synthesize: vi.fn().mockResolvedValue(Buffer.from('tts-audio')),
-    };
-    const channel = new VoiceChannel({ sttProvider: mockSTT, ttsProvider: mockTTS });
-    channel.onMessage(async (msg) => ({
-      id: 'r1',
-      role: 'assistant' as const,
-      content: `Reply to: ${msg.content}`,
-      timestamp: Date.now(),
-    }));
-
-    const result = await channel.processAudio(Buffer.from('raw-audio'));
-    expect(result.text).toBe('Hello agent');
-    expect(result.response).toBe('Reply to: Hello agent');
-    expect(result.audioResponse).toBeDefined();
-  });
-
-  // ── Provider instances ────────────────────────────────────
-
-  it('should create EdgeTTSProvider with default voice', () => {
-    const provider = new EdgeTTSProvider();
-    expect(provider.name).toBe('edge-tts');
-  });
-
-  it('should create EdgeTTSProvider with custom voice', () => {
-    const provider = new EdgeTTSProvider('zh-CN-XiaoxiaoNeural');
-    expect(provider.name).toBe('edge-tts');
-  });
-
-  // ── Channel lifecycle ─────────────────────────────────────
-
-  it('should start and stop channel', async () => {
-    const channel = new VoiceChannel();
-    expect(channel.isRunning()).toBe(false);
-    await channel.start();
-    expect(channel.isRunning()).toBe(true);
-    await channel.stop();
-    expect(channel.isRunning()).toBe(false);
-  });
-
-  it('should set STT/TTS providers dynamically', () => {
-    const channel = new VoiceChannel();
-    const mockSTT: STTProvider = { name: 'test-stt', transcribe: vi.fn() };
-    const mockTTS: TTSProvider = { name: 'test-tts', synthesize: vi.fn() };
-    channel.setSTTProvider(mockSTT);
-    channel.setTTSProvider(mockTTS);
-    // No error means success
-  });
-});
+import { describe, it, expect, vi } from 'vitest';
+import { VoiceChannel, createVoiceProviders, EdgeTTSProvider, WhisperSTTProvider, OpenAITTSProvider } from '../src/channels/voice';
+import type { STTProvider, TTSProvider, VoiceConfig } from '../src/channels/voice';
+
+describe('VoiceChannel Enhanced', () => {
+  // ── VoiceConfig Parsing ───────────────────────────────────
+
+  it('should parse VoiceConfig with edge-tts', () => {
+    const config: VoiceConfig = {
+      sttProvider: 'whisper',
+      ttsProvider: 'edge-tts',
+      voice: 'en-US-GuyNeural',
+      language: 'en',
+    };
+    const { stt, tts } = createVoiceProviders(config);
+    expect(stt).toBeUndefined(); // no API key
+    expect(tts).toBeDefined();
+    expect(tts!.name).toBe('edge-tts');
+  });
+
+  it('should parse VoiceConfig with whisper + openai-tts', () => {
+    const config: VoiceConfig = {
+      sttProvider: 'whisper',
+      ttsProvider: 'openai-tts',
+      sttApiKey: 'sk-test',
+      ttsApiKey: 'sk-test',
+      voice: 'nova',
+    };
+    const { stt, tts } = createVoiceProviders(config);
+    expect(stt).toBeDefined();
+    expect(stt!.name).toBe('whisper');
+    expect(tts).toBeDefined();
+    expect(tts!.name).toBe('openai-tts');
+  });
+
+  it('should return undefined for web-speech (node-only)', () => {
+    const { stt } = createVoiceProviders({
+      sttProvider: 'web-speech',
+      ttsProvider: 'edge-tts',
+    });
+    expect(stt).toBeUndefined();
+  });
+
+  // ── Transcribe (mock) ─────────────────────────────────────
+
+  it('should transcribe audio with mock STT provider', async () => {
+    const mockSTT: STTProvider = {
+      name: 'mock-stt',
+      transcribe: vi.fn().mockResolvedValue('Hello world'),
+    };
+    const channel = new VoiceChannel({ sttProvider: mockSTT });
+    const text = await channel.transcribe(Buffer.from('audio-data'));
+    expect(text).toBe('Hello world');
+    expect(mockSTT.transcribe).toHaveBeenCalled();
+  });
+
+  it('should throw if no STT provider for transcribe', async () => {
+    const channel = new VoiceChannel();
+    await expect(channel.transcribe(Buffer.from('audio'))).rejects.toThrow('No STT provider');
+  });
+
+  // ── Synthesize (mock) ─────────────────────────────────────
+
+  it('should synthesize text with mock TTS provider', async () => {
+    const mockTTS: TTSProvider = {
+      name: 'mock-tts',
+      synthesize: vi.fn().mockResolvedValue(Buffer.from('audio-output')),
+    };
+    const channel = new VoiceChannel({ ttsProvider: mockTTS });
+    const audio = await channel.synthesize('Hello world');
+    expect(audio).toEqual(Buffer.from('audio-output'));
+    expect(mockTTS.synthesize).toHaveBeenCalledWith('Hello world', { voice: undefined });
+  });
+
+  it('should throw if no TTS provider for synthesize', async () => {
+    const channel = new VoiceChannel();
+    await expect(channel.synthesize('Hello')).rejects.toThrow('No TTS provider');
+  });
+
+  // ── Conversation Mode ─────────────────────────────────────
+
+  it('should start and stop conversation mode', async () => {
+    const channel = new VoiceChannel();
+    await channel.startConversation(async (text) => `Echo: ${text}`);
+    expect(channel.isConversationActive()).toBe(true);
+    channel.stopConversation();
+    expect(channel.isConversationActive()).toBe(false);
+  });
+
+  // ── processConversationTurn ───────────────────────────────
+
+  it('should process a conversation turn', async () => {
+    const mockSTT: STTProvider = {
+      name: 'mock-stt',
+      transcribe: vi.fn().mockResolvedValue('What time is it?'),
+    };
+    const mockTTS: TTSProvider = {
+      name: 'mock-tts',
+      synthesize: vi.fn().mockResolvedValue(Buffer.from('audio-response')),
+    };
+    const channel = new VoiceChannel({ sttProvider: mockSTT, ttsProvider: mockTTS });
+
+    const result = await channel.processConversationTurn(
+      Buffer.from('audio'),
+      async (text) => `It is 3pm, you said: ${text}`,
+    );
+
+    expect(result.text).toBe('What time is it?');
+    expect(result.response).toContain('3pm');
+    expect(result.audioResponse).toBeDefined();
+  });
+
+  // ── processAudio with handler ─────────────────────────────
+
+  it('should process audio end-to-end', async () => {
+    const mockSTT: STTProvider = {
+      name: 'mock-stt',
+      transcribe: vi.fn().mockResolvedValue('Hello agent'),
+    };
+    const mockTTS: TTSProvider = {
+      name: 'mock-tts',
+      synthesize: vi.fn().mockResolvedValue(Buffer.from('tts-audio')),
+    };
+    const channel = new VoiceChannel({ sttProvider: mockSTT, ttsProvider: mockTTS });
+    channel.onMessage(async (msg) => ({
+      id: 'r1',
+      role: 'assistant' as const,
+      content: `Reply to: ${msg.content}`,
+      timestamp: Date.now(),
+    }));
+
+    const result = await channel.processAudio(Buffer.from('raw-audio'));
+    expect(result.text).toBe('Hello agent');
+    expect(result.response).toBe('Reply to: Hello agent');
+    expect(result.audioResponse).toBeDefined();
+  });
+
+  // ── Provider instances ────────────────────────────────────
+
+  it('should create EdgeTTSProvider with default voice', () => {
+    const provider = new EdgeTTSProvider();
+    expect(provider.name).toBe('edge-tts');
+  });
+
+  it('should create EdgeTTSProvider with custom voice', () => {
+    const provider = new EdgeTTSProvider('zh-CN-XiaoxiaoNeural');
+    expect(provider.name).toBe('edge-tts');
+  });
+
+  // ── Channel lifecycle ─────────────────────────────────────
+
+  it('should start and stop channel', async () => {
+    const channel = new VoiceChannel();
+    expect(channel.isRunning()).toBe(false);
+    await channel.start();
+    expect(channel.isRunning()).toBe(true);
+    await channel.stop();
+    expect(channel.isRunning()).toBe(false);
+  });
+
+  it('should set STT/TTS providers dynamically', () => {
+    const channel = new VoiceChannel();
+    const mockSTT: STTProvider = { name: 'test-stt', transcribe: vi.fn() };
+    const mockTTS: TTSProvider = { name: 'test-tts', synthesize: vi.fn() };
+    channel.setSTTProvider(mockSTT);
+    channel.setTTSProvider(mockTTS);
+    // No error means success
+  });
+});

package/tests/voice-interaction.test.ts

@@ -1,38 +1,38 @@
-import { describe, it, expect } from 'vitest';
-
-/**
- * Voice interaction tests — browser-only APIs (Web Speech API).
- * These tests verify the integration logic rather than the browser APIs themselves.
- */
-describe('Voice Interaction', () => {
-  it('should define voice feature flags', () => {
-    // In a browser context, these would be available
-    // Here we just verify our integration assumptions
-    expect(typeof globalThis).toBe('object');
-  });
-
-  it('should handle missing SpeechRecognition gracefully', () => {
-    // Simulate no SpeechRecognition API
-    const SpeechRecognition = (globalThis as any).SpeechRecognition || (globalThis as any).webkitSpeechRecognition;
-    expect(SpeechRecognition).toBeUndefined(); // Not available in Node
-  });
-
-  it('should handle missing speechSynthesis gracefully', () => {
-    const synth = (globalThis as any).speechSynthesis;
-    expect(synth).toBeUndefined(); // Not available in Node
-  });
-});
-
-describe('Voice TTS button injection', () => {
-  it('should create a button element with correct attributes', () => {
-    // Simulate DOM creation logic
-    const btn = {
-      className: 'tts-btn',
-      textContent: '🔊',
-      title: 'Read aloud',
-    };
-    expect(btn.className).toBe('tts-btn');
-    expect(btn.textContent).toBe('🔊');
-    expect(btn.title).toBe('Read aloud');
-  });
-});
+import { describe, it, expect } from 'vitest';
+
+/**
+ * Voice interaction tests — browser-only APIs (Web Speech API).
+ * These tests verify the integration logic rather than the browser APIs themselves.
+ */
+describe('Voice Interaction', () => {
+  it('should define voice feature flags', () => {
+    // In a browser context, these would be available
+    // Here we just verify our integration assumptions
+    expect(typeof globalThis).toBe('object');
+  });
+
+  it('should handle missing SpeechRecognition gracefully', () => {
+    // Simulate no SpeechRecognition API
+    const SpeechRecognition = (globalThis as any).SpeechRecognition || (globalThis as any).webkitSpeechRecognition;
+    expect(SpeechRecognition).toBeUndefined(); // Not available in Node
+  });
+
+  it('should handle missing speechSynthesis gracefully', () => {
+    const synth = (globalThis as any).speechSynthesis;
+    expect(synth).toBeUndefined(); // Not available in Node
+  });
+});
+
+describe('Voice TTS button injection', () => {
+  it('should create a button element with correct attributes', () => {
+    // Simulate DOM creation logic
+    const btn = {
+      className: 'tts-btn',
+      textContent: '🔊',
+      title: 'Read aloud',
+    };
+    expect(btn.className).toBe('tts-btn');
+    expect(btn.textContent).toBe('🔊');
+    expect(btn.title).toBe('Read aloud');
+  });
+});