Admin-unlimited Dev Workspaces: an admin-owned workspace is now exempt from the max-workspace-count limit, the max-tunnel-count limit, and the whole idle-stop/idle-warn/retention-delete lifecycle. Resolved once in quota.resolve() as Limits.unlimited (owner uid checked against get_admin_uids()), consumed at the three enforcement points (provision.ensure, provision.publish_tunnel, WorkspaceService._advance_lifecycle). Also hardens get_admin_uids()/get_primary_admin_uid() against a partially-schemaed users table (uid/role column guard), which a fresh test/init_db() path could hit. AI gateway per-model automatic fallback: any gateway_models route (chat/embed/image) can now name a fallback_model, picked on /admin/gateway from a select box of other configured public model names of the same kind only (never an internal upstream model id). When a route fails after its own retries are exhausted, the gateway retries once, automatically, against the fallback's own provider/pricing/key, before any bytes reach the client (including for a streaming response). One hop only, no chains or cycles; self-reference and cross-kind fallbacks are rejected at write time. AI gateway real upstream streaming and thinking-default control: stream:true is now forwarded to the upstream and relayed to the client as real SSE chunks (measured TTFT/inter-token latency) instead of a simulated split response, and every chat/vision call explicitly disables model "thinking" by default (admin-overridable via gateway_thinking), with per-dialect handling for DeepSeek, OpenRouter, and Ollama. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TjdKTWgWpW2SMNW8SFqxz5
104 lines
3.7 KiB
Python
104 lines
3.7 KiB
Python
# retoor <retoor@molodetz.nl>
|
|
|
|
from devplacepy.services.openai_gateway.thinking import (
|
|
apply_thinking,
|
|
client_thinking_enabled,
|
|
thinking_dialect,
|
|
)
|
|
|
|
|
|
def test_dialect_from_upstream_url():
|
|
assert thinking_dialect("https://api.deepseek.com/chat/completions") == "deepseek"
|
|
assert thinking_dialect("https://openrouter.ai/api/v1/chat/completions") == "openrouter"
|
|
assert thinking_dialect("http://127.0.0.1:11434/api/chat") == "ollama"
|
|
assert thinking_dialect("http://ollama.local/v1/chat/completions") == "ollama"
|
|
assert thinking_dialect("https://unknown.example/v1/chat/completions") == "deepseek"
|
|
|
|
|
|
def test_client_intent_unspecified():
|
|
assert client_thinking_enabled({}) is None
|
|
assert client_thinking_enabled({"messages": []}) is None
|
|
assert client_thinking_enabled(None) is None
|
|
|
|
|
|
def test_client_intent_disable():
|
|
assert client_thinking_enabled({"think": False}) is False
|
|
assert client_thinking_enabled({"think": "false"}) is False
|
|
assert client_thinking_enabled({"thinking": {"type": "disabled"}}) is False
|
|
assert client_thinking_enabled({"reasoning": {"effort": "none"}}) is False
|
|
assert client_thinking_enabled({"reasoning": {"enabled": False}}) is False
|
|
assert client_thinking_enabled({"enable_thinking": False}) is False
|
|
assert client_thinking_enabled(
|
|
{"chat_template_kwargs": {"enable_thinking": False}}
|
|
) is False
|
|
|
|
|
|
def test_client_intent_enable():
|
|
assert client_thinking_enabled({"think": True}) is True
|
|
assert client_thinking_enabled({"think": "high"}) is True
|
|
assert client_thinking_enabled({"thinking": {"type": "enabled"}}) is True
|
|
assert client_thinking_enabled({"reasoning": {"effort": "low"}}) is True
|
|
assert client_thinking_enabled({"reasoning_effort": "max"}) is True
|
|
|
|
|
|
def test_default_disables_deepseek_thinking():
|
|
payload = apply_thinking(
|
|
{"messages": [{"role": "user", "content": "hi"}]},
|
|
"https://api.deepseek.com/chat/completions",
|
|
)
|
|
assert payload["thinking"] == {"type": "disabled"}
|
|
assert "think" not in payload
|
|
assert "reasoning" not in payload
|
|
assert "reasoning_effort" not in payload
|
|
|
|
|
|
def test_default_disables_openrouter_and_ollama():
|
|
openrouter = apply_thinking({}, "https://openrouter.ai/api/v1/chat/completions")
|
|
assert openrouter["reasoning"] == {"effort": "none"}
|
|
assert "thinking" not in openrouter
|
|
ollama = apply_thinking({}, "http://127.0.0.1:11434/api/chat")
|
|
assert ollama["think"] is False
|
|
assert "thinking" not in ollama
|
|
|
|
|
|
def test_client_can_enable_thinking():
|
|
payload = apply_thinking(
|
|
{"think": True, "messages": []},
|
|
"https://api.deepseek.com/chat/completions",
|
|
)
|
|
assert payload["thinking"] == {"type": "enabled"}
|
|
assert "think" not in payload
|
|
|
|
|
|
def test_client_disable_beats_admin_default_on():
|
|
payload = apply_thinking(
|
|
{"thinking": {"type": "disabled"}},
|
|
"https://api.deepseek.com/chat/completions",
|
|
default_enabled=True,
|
|
)
|
|
assert payload["thinking"] == {"type": "disabled"}
|
|
|
|
|
|
def test_admin_default_on_when_client_silent():
|
|
payload = apply_thinking(
|
|
{"messages": []},
|
|
"https://api.deepseek.com/chat/completions",
|
|
default_enabled=True,
|
|
)
|
|
assert payload["thinking"] == {"type": "enabled"}
|
|
|
|
|
|
def test_strips_conflicting_client_fields():
|
|
payload = apply_thinking(
|
|
{
|
|
"think": False,
|
|
"reasoning_effort": "high",
|
|
"chat_template_kwargs": {"enable_thinking": True, "other": 1},
|
|
},
|
|
"https://api.deepseek.com/chat/completions",
|
|
)
|
|
assert payload["thinking"] == {"type": "disabled"}
|
|
assert "think" not in payload
|
|
assert "reasoning_effort" not in payload
|
|
assert payload["chat_template_kwargs"] == {"other": 1}
|