feat: add user_id index to profiles table for faster lookups

The profiles table previously lacked an index on the user_id column, causing full table scans during user lookups. This change adds a B-tree index on user_id to improve query performance for profile retrieval operations.

tests/e2e/openai/v1/embeddings.py

@@ -0,0 +1,80 @@
+# retoor <retoor@molodetz.nl>
+
+import requests
+from tests.conftest import BASE_URL
+from devplacepy.database import get_table
+
+
+def _config(page, **fields):
+    page.request.post(f"{BASE_URL}/admin/services/openai/config", form=fields)
+
+
+def test_embeddings_disabled_returns_503(alice):
+    page, _ = alice
+    page.request.post(f"{BASE_URL}/admin/services/openai/stop")
+    try:
+        r = requests.post(
+            f"{BASE_URL}/openai/v1/embeddings", json={"input": "hello"}
+        )
+        assert r.status_code == 503
+    finally:
+        page.request.post(f"{BASE_URL}/admin/services/openai/start")
+
+
+def test_embeddings_requires_auth(alice):
+    page, _ = alice
+    page.request.post(f"{BASE_URL}/admin/services/openai/start")
+    _config(page, gateway_access_key="embedkey")
+    try:
+        no_creds = requests.post(
+            f"{BASE_URL}/openai/v1/embeddings", json={"input": "hello"}
+        )
+        assert no_creds.status_code == 401
+    finally:
+        _config(page, gateway_access_key="")
+        page.request.post(f"{BASE_URL}/admin/services/openai/start")
+
+
+def test_embeddings_auth_passes_upstream_unreachable(alice):
+    page, user = alice
+    page.request.post(f"{BASE_URL}/admin/services/openai/start")
+    _config(
+        page,
+        gateway_embed_enabled="1",
+        gateway_embed_url="http://127.0.0.1:9/embeddings",
+        gateway_access_key="embedkey",
+    )
+    try:
+        with_key = requests.post(
+            f"{BASE_URL}/openai/v1/embeddings",
+            headers={"X-API-KEY": "embedkey"},
+            json={"model": "molodetz~embed", "input": "hello"},
+        )
+        assert with_key.status_code == 502
+
+        admin_key = get_table("users").find_one(username=user["username"])["api_key"]
+        with_admin = requests.post(
+            f"{BASE_URL}/openai/v1/embeddings",
+            headers={"Authorization": f"Bearer {admin_key}"},
+            json={"model": "molodetz~embed", "input": "hello"},
+        )
+        assert with_admin.status_code == 502
+    finally:
+        _config(page, gateway_access_key="")
+        page.request.post(f"{BASE_URL}/admin/services/openai/start")
+
+
+def test_embeddings_disabled_via_config_returns_503(alice):
+    page, _ = alice
+    page.request.post(f"{BASE_URL}/admin/services/openai/start")
+    _config(page, gateway_embed_enabled="0", gateway_access_key="embedkey")
+    try:
+        r = requests.post(
+            f"{BASE_URL}/openai/v1/embeddings",
+            headers={"X-API-KEY": "embedkey"},
+            json={"input": "hello"},
+        )
+        assert r.status_code == 503
+    finally:
+        _config(page, gateway_embed_enabled="1", gateway_access_key="")
+        page.request.post(f"{BASE_URL}/admin/services/openai/start")

tests/unit/services/openai_gateway/gateway.py

@@ -202,3 +202,85 @@ def test_authorize_require_auth_off_is_open(local_db):
         assert svc.authorize(_make_request_openai_gateway()) is True
     finally:
         set_setting("gateway_require_auth", "1")
+
+
+class FakeEmbedClient_openai_gateway:
+    def __init__(self, *a, **k):
+        self.calls = []
+
+    def build_request(self, method, url, headers=None, json=None, content=None):
+        return FakeRequest(method, url, json)
+
+    async def send(self, request):
+        self.calls.append((request.url, request.json_body))
+        body = request.json_body or {}
+        return FakeResp_openai_gateway(
+            payload={
+                "object": "list",
+                "model": body.get("model"),
+                "data": [{"object": "embedding", "index": 0, "embedding": [0.1, 0.2]}],
+                "usage": {"prompt_tokens": 5, "total_tokens": 5},
+            }
+        )
+
+    async def aclose(self):
+        pass
+
+
+def test_embeddings_remaps_alias_to_configured_model(local_db, monkeypatch):
+    monkeypatch.setattr(gwmod.httpx, "AsyncClient", FakeEmbedClient_openai_gateway)
+    svc = GatewayService()
+    cfg = svc.effective_config()
+    cfg["gateway_force_model"] = False
+    cfg["gateway_embed_enabled"] = True
+    cfg["gateway_embed_model"] = "qwen/qwen3-embedding-8b"
+    rt = svc.runtime()
+    run_async(
+        rt.handle_embeddings(
+            {"model": "molodetz~embed", "input": "hello"},
+            cfg,
+            ("guest", "test"),
+            "test",
+        )
+    )
+    assert rt._client.calls[-1][1]["model"] == "qwen/qwen3-embedding-8b"
+
+
+def test_embeddings_disabled_returns_503(local_db, monkeypatch):
+    monkeypatch.setattr(gwmod.httpx, "AsyncClient", FakeEmbedClient_openai_gateway)
+    svc = GatewayService()
+    cfg = svc.effective_config()
+    cfg["gateway_embed_enabled"] = False
+    rt = svc.runtime()
+    resp = run_async(
+        rt.handle_embeddings(
+            {"input": "hello"}, cfg, ("guest", "test"), "test"
+        )
+    )
+    assert resp.status_code == 503
+
+
+def test_compute_cost_embed_branch():
+    from devplacepy.services.openai_gateway.usage import (
+        Pricing,
+        compute_cost,
+        normalize_usage,
+    )
+
+    pricing = Pricing(
+        chat_cache_hit_per_m=0.0,
+        chat_cache_miss_per_m=0.0,
+        chat_output_per_m=0.0,
+        vision_input_per_m=0.0,
+        vision_output_per_m=0.0,
+        embed_input_per_m=0.01,
+    )
+    usage = {"prompt_tokens": 1_000_000, "total_tokens": 1_000_000}
+    norm = normalize_usage(usage)
+    total, input_cost, output_cost, native = compute_cost(
+        usage, norm, pricing, "embed"
+    )
+    assert native is False
+    assert output_cost == 0.0
+    assert abs(total - 0.01) < 1e-9
+    assert abs(input_cost - 0.01) < 1e-9