Feature/lmstudio provider (#30)

* chore: just a plan update

* Add LM Studio as local AI provider (OpenAI-compatible, like Ollama)

* Convert WebP thumbnails to JPEG before image analysis for LM Studio compatibility

* Strengthen language enforcement in image analysis prompt for local models

* Use i18n localized prompts for image analysis instead of English instructions

* Add airplane mode (Flugmodus) with status bar toggle and offline model preferences

* Fix flightmode: persist model IDs, skip network when offline, airplane icon

* Auto-fallback to offline models in airplane mode for chat, title, and image analysis

* Auto-select first local model as offline fallback when no explicit offline model configured

* Block git fetch/pull/push and site upload in airplane mode

* fix: thumbnails optimized for AI

* fix: error handling in airplane mode

---------

Co-authored-by: hugo <hugoms@me.com>
This commit is contained in:
Georg Bauer
2026-03-02 13:35:42 +01:00
committed by GitHub
parent 4b4a9c1c8b
commit 5747925503
34 changed files with 2215 additions and 105 deletions

View File

@@ -140,13 +140,15 @@ describe('ProviderRegistry', () => {
});
it('getProviderStatus() reports all providers', () => {
expect(registry.getProviderStatus()).toEqual({ opencode: false, mistral: false, ollama: false });
expect(registry.getProviderStatus()).toEqual({ opencode: false, mistral: false, ollama: false, lmstudio: false, offlineMode: false });
registry.setOpencodeKey('test');
expect(registry.getProviderStatus()).toEqual({ opencode: true, mistral: false, ollama: false });
expect(registry.getProviderStatus()).toEqual({ opencode: true, mistral: false, ollama: false, lmstudio: false, offlineMode: false });
registry.setMistralKey('test2');
expect(registry.getProviderStatus()).toEqual({ opencode: true, mistral: true, ollama: false });
expect(registry.getProviderStatus()).toEqual({ opencode: true, mistral: true, ollama: false, lmstudio: false, offlineMode: false });
registry.setOllamaEnabled(true);
expect(registry.getProviderStatus()).toEqual({ opencode: true, mistral: true, ollama: true });
expect(registry.getProviderStatus()).toEqual({ opencode: true, mistral: true, ollama: true, lmstudio: false, offlineMode: false });
registry.setLmstudioEnabled(true);
expect(registry.getProviderStatus()).toEqual({ opencode: true, mistral: true, ollama: true, lmstudio: true, offlineMode: false });
});
it('isProviderKeySet() checks per-provider', () => {
@@ -444,6 +446,120 @@ describe('OneShotTasks', () => {
expect(result.error).toContain('thumbnail');
});
it('uses pre-generated AI JPEG thumbnail without sharp conversion', async () => {
registry.setOpencodeKey('test-key');
chatEngine.getSetting.mockResolvedValue('claude-sonnet-4');
mediaEngine.getMedia.mockResolvedValue({
id: 'media-1',
mimeType: 'image/jpeg',
filename: 'photo.jpg',
});
// Tiny valid JPEG — simulates the pre-generated 'ai' thumbnail
const jpegBase64 = '/9j/2wBDAAYEBQYFBAYGBQYHBwYIChAKCgkJChQODwwQFxQYGBcUFhYaHSUfGhsjHBYWICwgIyYnKSopGR8tMC0oMCUoKSj/2wBDAQcHBwoIChMKChMoGhYaKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCj/wAARCAACAAIDASIAAhEBAxEB/8QAFAABAAAAAAAAAAAAAAAAAAAAAP/EABQQAQAAAAAAAAAAAAAAAAAAAAD/xAAUAQEAAAAAAAAAAAAAAAAAAAAA/8QAFBEBAAAAAAAAAAAAAAAAAAAAAP/aAAwDAQACEQMRAD8AAA//2Q==';
// Return JPEG for 'ai' size, null for others
mediaEngine.getThumbnailDataUrl.mockImplementation(async (_id: string, size: string) => {
if (size === 'ai') return `data:image/jpeg;base64,${jpegBase64}`;
return null;
});
const originalFetch = globalThis.fetch;
let capturedBody: any = null;
globalThis.fetch = vi.fn().mockImplementation(async (url: string, init: any) => {
if (init?.body) {
capturedBody = JSON.parse(init.body);
}
return new Response(JSON.stringify({
id: 'msg_test',
type: 'message',
role: 'assistant',
content: [{ type: 'text', text: '{"title": "Test", "alt": "Test image", "caption": "A test"}' }],
model: 'claude-sonnet-4',
stop_reason: 'end_turn',
usage: { input_tokens: 100, output_tokens: 30, cache_creation_input_tokens: 0, cache_read_input_tokens: 0 },
}), { status: 200, headers: { 'Content-Type': 'application/json' } });
});
try {
const result = await tasks.analyzeMediaImage('media-1', 'en');
// Check the image was sent as JPEG, not WebP
if (capturedBody?.messages) {
const userMsg = capturedBody.messages.find((m: any) => m.role === 'user');
if (userMsg?.content) {
const imagePart = userMsg.content.find((p: any) => p.type === 'image_url');
if (imagePart?.image_url?.url) {
expect(imagePart.image_url.url).toMatch(/^data:image\/jpeg;base64,/);
expect(imagePart.image_url.url).not.toMatch(/^data:image\/webp;base64,/);
}
}
}
// Also verify it succeeded (may fail on response parsing but the format check is key)
if (result.success) {
expect(result.title).toBe('Test');
}
} finally {
globalThis.fetch = originalFetch;
}
});
it('sends localized prompts based on project language', async () => {
registry.setOpencodeKey('test-key');
chatEngine.getSetting.mockResolvedValue('claude-sonnet-4');
mediaEngine.getMedia.mockResolvedValue({
id: 'media-1',
mimeType: 'image/jpeg',
filename: 'photo.jpg',
});
// Tiny valid JPEG — simulates the pre-generated 'ai' thumbnail
const jpegBase64 = '/9j/2wBDAAYEBQYFBAYGBQYHBwYIChAKCgkJChQODwwQFxQYGBcUFhYaHSUfGhsjHBYWICwgIyYnKSopGR8tMC0oMCUoKSj/2wBDAQcHBwoIChMKChMoGhYaKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCj/wAARCAACAAIDASIAAhEBAxEB/8QAFAABAAAAAAAAAAAAAAAAAAAAAP/EABQQAQAAAAAAAAAAAAAAAAAAAAD/xAAUAQEAAAAAAAAAAAAAAAAAAAAA/8QAFBEBAAAAAAAAAAAAAAAAAAAAAP/aAAwDAQACEQMRAD8AAA//2Q==';
mediaEngine.getThumbnailDataUrl.mockImplementation(async (_id: string, size: string) => {
if (size === 'ai') return `data:image/jpeg;base64,${jpegBase64}`;
return null;
});
const originalFetch = globalThis.fetch;
let capturedBody: any = null;
globalThis.fetch = vi.fn().mockImplementation(async (_url: string, init: any) => {
if (init?.body) {
capturedBody = JSON.parse(init.body);
}
return new Response(JSON.stringify({
id: 'msg_test',
type: 'message',
role: 'assistant',
content: [{ type: 'text', text: '{"title": "Testbild", "alt": "Rotes Quadrat", "caption": "Ein Test"}' }],
model: 'claude-sonnet-4',
stop_reason: 'end_turn',
usage: { input_tokens: 100, output_tokens: 30, cache_creation_input_tokens: 0, cache_read_input_tokens: 0 },
}), { status: 200, headers: { 'Content-Type': 'application/json' } });
});
try {
await tasks.analyzeMediaImage('media-1', 'de');
// System prompt should be in German (from i18n), not contain English instructions
if (capturedBody) {
const systemMsg = capturedBody.messages?.find((m: any) => m.role === 'system')
?? capturedBody.system;
const systemText = typeof systemMsg === 'string' ? systemMsg
: Array.isArray(systemMsg) ? systemMsg.map((p: any) => p.text).join('')
: systemMsg?.content ?? '';
expect(systemText).toContain('Deutsch');
expect(systemText).not.toContain('English');
// User message should also be in German
const userMsg = capturedBody.messages?.find((m: any) => m.role === 'user');
if (userMsg?.content) {
const textPart = Array.isArray(userMsg.content)
? userMsg.content.find((p: any) => p.type === 'text')
: null;
if (textPart?.text) {
expect(textPart.text).toContain('Deutsch');
}
}
}
} finally {
globalThis.fetch = originalFetch;
}
});
it('falls back to claude-sonnet-4-5 when no image analysis model is configured', async () => {
registry.setOpencodeKey('test-key');
chatEngine.getSetting.mockResolvedValue(null);
@@ -452,7 +568,12 @@ describe('OneShotTasks', () => {
mimeType: 'image/jpeg',
filename: 'photo.jpg',
});
mediaEngine.getThumbnailDataUrl.mockResolvedValue('data:image/webp;base64,abc123');
// Tiny valid JPEG — simulates the pre-generated 'ai' thumbnail
const jpegBase64 = '/9j/2wBDAAYEBQYFBAYGBQYHBwYIChAKCgkJChQODwwQFxQYGBcUFhYaHSUfGhsjHBYWICwgIyYnKSopGR8tMC0oMCUoKSj/2wBDAQcHBwoIChMKChMoGhYaKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCj/wAARCAACAAIDASIAAhEBAxEB/8QAFAABAAAAAAAAAAAAAAAAAAAAAP/EABQQAQAAAAAAAAAAAAAAAAAAAAD/xAAUAQEAAAAAAAAAAAAAAAAAAAAA/8QAFBEBAAAAAAAAAAAAAAAAAAAAAP/aAAwDAQACEQMRAD8AAA//2Q==';
mediaEngine.getThumbnailDataUrl.mockImplementation(async (_id: string, size: string) => {
if (size === 'ai') return `data:image/jpeg;base64,${jpegBase64}`;
return null;
});
// Verify the method selects the right model by checking it attempts
// to call the resolver (which hits the network). We mock fetch to