Update desktop app to android app capabilities.

tests/test_tts_stream.py

@@ -3,13 +3,16 @@ no network, no audio device, no ONNX model."""
 
 import sys
 from pathlib import Path
+from unittest.mock import MagicMock, patch
 
 import numpy as np
+import pytest
 
 sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
 
 from bolt_pet import config as tts_config
-from bolt_pet.audio.tts import chunks_to_int16, model_for, voice_for
+from bolt_pet.audio import tts
+from bolt_pet.audio.tts import TtsError, chunks_to_int16, voice_for
 from bolt_pet.audio.wake_word import NearMissLog
 
 
@@ -83,7 +86,7 @@ def test_clear_resets_peak_and_entries():
     assert log.entries() == [] and log.peak == 0.0
 
 
-# ── voice / model selection (server speak_as) ───────────────────────────────
+# ── voice selection (server speak_as) ───────────────────────────────────────
 
 def test_the_override_voice_wins_over_the_configured_one(monkeypatch):
     monkeypatch.setattr(tts_config, "ELEVENLABS_VOICE_ID", "DEFAULT")
@@ -92,16 +95,81 @@ def test_the_override_voice_wins_over_the_configured_one(monkeypatch):
     assert voice_for(None) == "DEFAULT"
 
 
-def test_english_replies_in_the_default_voice_use_the_default_model(monkeypatch):
-    monkeypatch.setattr(tts_config, "ELEVENLABS_MODEL_ID", "eleven_flash_v2")
-    monkeypatch.setattr(tts_config, "ELEVENLABS_MULTILINGUAL_MODEL_ID", "eleven_flash_v2_5")
-    assert model_for("all good here", None) == "eleven_flash_v2"
+# ── server-hosted synthesis (/desk/tts) ─────────────────────────────────────
+# No local ElevenLabs account: both the whole-clip and streaming paths post
+# to this pet's own server, same as server_client.py's other endpoints.
+
+@pytest.fixture(autouse=True)
+def _configured(monkeypatch):
+    monkeypatch.setattr(tts_config, "SERVER_URL", "http://test-server:5002")
+    monkeypatch.setattr(tts_config, "API_KEY", "test-key")
+    monkeypatch.setattr(tts_config, "SESSION_ID", "pet-test")
+    monkeypatch.setattr(tts_config, "ELEVENLABS_VOICE_ID", "default-voice")
 
 
-def test_a_picked_voice_or_non_english_text_uses_the_multilingual_model(monkeypatch):
-    # eleven_flash_v2 is English-only: it would read either of these as
-    # mangled phonetic English rather than failing outright.
-    monkeypatch.setattr(tts_config, "ELEVENLABS_MODEL_ID", "eleven_flash_v2")
-    monkeypatch.setattr(tts_config, "ELEVENLABS_MULTILINGUAL_MODEL_ID", "eleven_flash_v2_5")
-    assert model_for("all good here", "VOICE1") == "eleven_flash_v2_5"
-    assert model_for("こんにちは", None) == "eleven_flash_v2_5"
+def _mock_response(content=b"", ok=True):
+    resp = MagicMock()
+    resp.content = content
+    resp.raise_for_status = MagicMock() if ok else MagicMock(side_effect=Exception("boom"))
+    return resp
+
+
+def test_synthesize_pcm_posts_to_the_servers_tts_endpoint():
+    pcm_bytes = np.array([1, 2, 3], dtype=np.int16).tobytes()
+    with patch.object(tts.requests, "post") as post:
+        post.return_value = _mock_response(pcm_bytes)
+        pcm, rate = tts.synthesize_pcm("hello there")
+    post.assert_called_once()
+    args, kwargs = post.call_args
+    assert args[0] == "http://test-server:5002/desk/tts"
+    assert kwargs["headers"] == {"X-Desk-Api-Key": "test-key"}
+    assert kwargs["json"] == {
+        "session_id": "pet-test", "text": "hello there", "voice_id": "default-voice",
+    }
+    assert pcm.tolist() == [1, 2, 3]
+    assert rate == tts_config.TTS_SAMPLE_RATE
+
+
+def test_synthesize_pcm_uses_the_override_voice():
+    with patch.object(tts.requests, "post") as post:
+        post.return_value = _mock_response(b"\x01\x00")
+        tts.synthesize_pcm("hi", voice_id="picked-voice")
+    assert post.call_args.kwargs["json"]["voice_id"] == "picked-voice"
+
+
+def test_synthesize_pcm_raises_when_not_configured(monkeypatch):
+    monkeypatch.setattr(tts_config, "API_KEY", "")
+    with pytest.raises(TtsError, match="BOLT_SERVER_URL"):
+        tts.synthesize_pcm("hi")
+
+
+def test_synthesize_pcm_raises_on_empty_audio():
+    with patch.object(tts.requests, "post") as post:
+        post.return_value = _mock_response(b"")
+        with pytest.raises(TtsError, match="no audio"):
+            tts.synthesize_pcm("hi")
+
+
+def test_synthesize_pcm_raises_when_the_request_fails():
+    with patch.object(tts.requests, "post") as post:
+        post.return_value = _mock_response(b"", ok=False)
+        with pytest.raises(TtsError, match="server tts request failed"):
+            tts.synthesize_pcm("hi")
+
+
+def test_stream_pcm_posts_to_the_same_endpoint_with_stream_true():
+    with patch.object(tts.requests, "post") as post:
+        response = MagicMock()
+        response.raise_for_status = MagicMock()
+        response.iter_content.return_value = [np.array([4, 5], dtype=np.int16).tobytes()]
+        post.return_value = response
+        chunks = list(tts.stream_pcm("hi"))
+    assert post.call_args.kwargs["stream"] is True
+    assert post.call_args[0][0] == "http://test-server:5002/desk/tts"
+    assert np.concatenate(chunks).tolist() == [4, 5]
+
+
+def test_stream_pcm_raises_when_not_configured(monkeypatch):
+    monkeypatch.setattr(tts_config, "SERVER_URL", "")
+    with pytest.raises(TtsError, match="BOLT_SERVER_URL"):
+        list(tts.stream_pcm("hi"))