github/openclaw
1
1
Fork 0
mirror of https://github.com/openclaw/openclaw.git synced 2026-02-19 18:39:20 -05:00
Code Issues Packages Projects Releases Wiki Activity

test(media): dedupe auto-audio fixture wiring

Browse Source
This commit is contained in:
Peter Steinberger
2026-02-18 13:06:21 +00:00
parent d067618600
commit f4db58a5fd
1 changed files with 102 additions and 73 deletions
Download Patch File Download Diff File Expand all files Collapse all files

175
src/media-understanding/runner.auto-audio.test.ts
View File

@@ -11,40 +11,66 @@ import {
runCapability,
} from "./runner.js";
async function withAudioFixture(
run: (params: {
ctx: MsgContext;
media: ReturnType<typeof normalizeMediaAttachments>;
cache: ReturnType<typeof createMediaAttachmentCache>;
}) => Promise<void>,
) {
const originalPath = process.env.PATH;
process.env.PATH = "/usr/bin:/bin";
const tmpPath = path.join(os.tmpdir(), `openclaw-auto-audio-${Date.now()}.wav`);
await fs.writeFile(tmpPath, Buffer.from("RIFF"));
const ctx: MsgContext = { MediaPath: tmpPath, MediaType: "audio/wav" };
const media = normalizeMediaAttachments(ctx);
const cache = createMediaAttachmentCache(media);
try {
await run({ ctx, media, cache });
} finally {
process.env.PATH = originalPath;
await cache.cleanup();
await fs.unlink(tmpPath).catch(() => {});
}
}
function createOpenAiAudioProvider(
transcribeAudio: (req: { model?: string }) => Promise<{ text: string; model: string }>,
) {
return buildProviderRegistry({
openai: {
id: "openai",
capabilities: ["audio"],
transcribeAudio,
},
});
}
function createOpenAiAudioCfg(extra?: Partial<OpenClawConfig>): OpenClawConfig {
return {
models: {
providers: {
openai: {
apiKey: "test-key",
models: [],
},
},
},
...extra,
} as unknown as OpenClawConfig;
}
describe("runCapability auto audio entries", () => {
it("uses provider keys to auto-enable audio transcription", async () => {
const originalPath = process.env.PATH;
process.env.PATH = "/usr/bin:/bin";
const tmpPath = path.join(os.tmpdir(), `openclaw-auto-audio-${Date.now()}.wav`);
await fs.writeFile(tmpPath, Buffer.from("RIFF"));
const ctx: MsgContext = { MediaPath: tmpPath, MediaType: "audio/wav" };
const media = normalizeMediaAttachments(ctx);
const cache = createMediaAttachmentCache(media);
await withAudioFixture(async ({ ctx, media, cache }) => {
let seenModel: string | undefined;
const providerRegistry = createOpenAiAudioProvider(async (req) => {
seenModel = req.model;
return { text: "ok", model: req.model ?? "unknown" };
});
const cfg = createOpenAiAudioCfg();
let seenModel: string | undefined;
const providerRegistry = buildProviderRegistry({
openai: {
id: "openai",
capabilities: ["audio"],
transcribeAudio: async (req) => {
seenModel = req.model;
return { text: "ok", model: req.model };
},
},
});
const cfg = {
models: {
providers: {
openai: {
apiKey: "test-key",
models: [],
},
},
},
} as unknown as OpenClawConfig;
try {
const result = await runCapability({
capability: "audio",
cfg,
@@ -56,49 +82,25 @@ describe("runCapability auto audio entries", () => {
expect(result.outputs[0]?.text).toBe("ok");
expect(seenModel).toBe("gpt-4o-mini-transcribe");
expect(result.decision.outcome).toBe("success");
} finally {
process.env.PATH = originalPath;
await cache.cleanup();
await fs.unlink(tmpPath).catch(() => {});
}
});
});
it("skips auto audio when disabled", async () => {
const originalPath = process.env.PATH;
process.env.PATH = "/usr/bin:/bin";
const tmpPath = path.join(os.tmpdir(), `openclaw-auto-audio-${Date.now()}.wav`);
await fs.writeFile(tmpPath, Buffer.from("RIFF"));
const ctx: MsgContext = { MediaPath: tmpPath, MediaType: "audio/wav" };
const media = normalizeMediaAttachments(ctx);
const cache = createMediaAttachmentCache(media);
const providerRegistry = buildProviderRegistry({
openai: {
id: "openai",
capabilities: ["audio"],
transcribeAudio: async () => ({ text: "ok", model: "whisper-1" }),
},
});
const cfg = {
models: {
providers: {
openai: {
apiKey: "test-key",
models: [],
await withAudioFixture(async ({ ctx, media, cache }) => {
const providerRegistry = createOpenAiAudioProvider(async () => ({
text: "ok",
model: "whisper-1",
}));
const cfg = createOpenAiAudioCfg({
tools: {
media: {
audio: {
enabled: false,
},
},
},
},
tools: {
media: {
audio: {
enabled: false,
},
},
},
} as unknown as OpenClawConfig;
});
try {
const result = await runCapability({
capability: "audio",
cfg,
@@ -109,10 +111,37 @@ describe("runCapability auto audio entries", () => {
});
expect(result.outputs).toHaveLength(0);
expect(result.decision.outcome).toBe("disabled");
} finally {
process.env.PATH = originalPath;
await cache.cleanup();
await fs.unlink(tmpPath).catch(() => {});
}
});
});
it("prefers explicitly configured audio model entries", async () => {
await withAudioFixture(async ({ ctx, media, cache }) => {
let seenModel: string | undefined;
const providerRegistry = createOpenAiAudioProvider(async (req) => {
seenModel = req.model;
return { text: "ok", model: req.model ?? "unknown" };
});
const cfg = createOpenAiAudioCfg({
tools: {
media: {
audio: {
models: [{ provider: "openai", model: "whisper-1" }],
},
},
},
});
const result = await runCapability({
capability: "audio",
cfg,
ctx,
attachments: cache,
media,
providerRegistry,
});
expect(result.outputs[0]?.text).toBe("ok");
expect(seenModel).toBe("whisper-1");
});
});
});