feat: initial release v0.3.0

tests/unit/api-config/provider-card-protocol-probe.test.ts

@@ -0,0 +1,83 @@
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import type { CustomModel } from '@/app/[locale]/profile/components/api-config/types'
+import {
+  probeModelLlmProtocolViaApi,
+  shouldProbeModelLlmProtocol,
+  shouldReprobeModelLlmProtocol,
+} from '@/app/[locale]/profile/components/api-config/provider-card/hooks/useProviderCardState'
+
+describe('api-config provider-card protocol probe helpers', () => {
+  beforeEach(() => {
+    vi.clearAllMocks()
+  })
+
+  it('only probes openai-compatible llm models', () => {
+    expect(shouldProbeModelLlmProtocol({ providerId: 'openai-compatible:oa-1', modelType: 'llm' })).toBe(true)
+    expect(shouldProbeModelLlmProtocol({ providerId: 'openai-compatible:oa-1', modelType: 'image' })).toBe(false)
+    expect(shouldProbeModelLlmProtocol({ providerId: 'gemini-compatible:gm-1', modelType: 'llm' })).toBe(false)
+  })
+
+  it('re-probes only when modelId/provider changed on openai-compatible llm', () => {
+    const originalModel: CustomModel = {
+      modelId: 'gpt-4.1-mini',
+      modelKey: 'openai-compatible:oa-1::gpt-4.1-mini',
+      name: 'GPT 4.1 Mini',
+      type: 'llm',
+      provider: 'openai-compatible:oa-1',
+      llmProtocol: 'chat-completions',
+      llmProtocolCheckedAt: '2026-01-01T00:00:00.000Z',
+      price: 0,
+      enabled: true,
+    }
+
+    expect(shouldReprobeModelLlmProtocol({
+      providerId: 'openai-compatible:oa-1',
+      originalModel,
+      nextModelId: 'gpt-4.1-mini',
+    })).toBe(false)
+
+    expect(shouldReprobeModelLlmProtocol({
+      providerId: 'openai-compatible:oa-1',
+      originalModel,
+      nextModelId: 'gpt-4.1',
+    })).toBe(true)
+
+    expect(shouldReprobeModelLlmProtocol({
+      providerId: 'gemini-compatible:gm-1',
+      originalModel,
+      nextModelId: 'gpt-4.1',
+    })).toBe(false)
+  })
+
+  it('parses successful probe response payload', async () => {
+    const fetchMock = vi.fn(async () => new Response(JSON.stringify({
+      success: true,
+      protocol: 'responses',
+      checkedAt: '2026-03-05T10:00:00.000Z',
+    }), { status: 200 }))
+    vi.stubGlobal('fetch', fetchMock)
+
+    const result = await probeModelLlmProtocolViaApi({
+      providerId: 'openai-compatible:oa-1',
+      modelId: 'gpt-4.1-mini',
+    })
+
+    expect(result).toEqual({
+      llmProtocol: 'responses',
+      llmProtocolCheckedAt: '2026-03-05T10:00:00.000Z',
+    })
+  })
+
+  it('throws probe failure code on unsuccessful probe response', async () => {
+    const fetchMock = vi.fn(async () => new Response(JSON.stringify({
+      success: false,
+      code: 'PROBE_INCONCLUSIVE',
+    }), { status: 200 }))
+    vi.stubGlobal('fetch', fetchMock)
+
+    await expect(probeModelLlmProtocolViaApi({
+      providerId: 'openai-compatible:oa-1',
+      modelId: 'gpt-4.1-mini',
+    })).rejects.toThrow('PROBE_INCONCLUSIVE')
+  })
+})