iop/apps/edge/internal/openai/server_test.go

package openai

import (
	"context"
	"encoding/json"
	"net/http"
	"net/http/httptest"
	"strings"
	"testing"
	"time"

	edgeservice "iop/apps/edge/internal/service"
	"iop/packages/config"
	iop "iop/proto/gen/iop"
)

type fakeRunService struct {
	req    edgeservice.SubmitRunRequest
	events chan *iop.RunEvent
}

func (s *fakeRunService) SubmitRun(_ context.Context, req edgeservice.SubmitRunRequest) (*edgeservice.RunHandle, error) {
	s.req = req
	return &edgeservice.RunHandle{
		RunDispatch: edgeservice.RunDispatch{
			RunID:      "run-test",
			Target:     req.Target,
			TimeoutSec: 5,
		},
		RunStream: edgeservice.RunStream{
			Events:     s.events,
			NodeEvents: make(chan *iop.EdgeNodeEvent),
		},
	}, nil
}

func TestChatCompletionsDispatchesConfiguredOllamaTarget(t *testing.T) {
	fake := &fakeRunService{events: make(chan *iop.RunEvent, 3)}
	fake.events <- &iop.RunEvent{Type: "delta", Delta: "hello"}
	fake.events <- &iop.RunEvent{Type: "delta", Delta: " world"}
	fake.events <- &iop.RunEvent{Type: "complete", Usage: &iop.Usage{InputTokens: 2, OutputTokens: 2}}

	srv := NewServer(config.EdgeOpenAIConf{
		Adapter:    "ollama",
		Target:     "llama-fixed",
		SessionID:  "cline",
		TimeoutSec: 15,
	}, fake, nil)
	req := httptest.NewRequest(http.MethodPost, "/v1/chat/completions", strings.NewReader(`{
		"model":"client-model",
		"messages":[{"role":"system","content":"brief"},{"role":"user","content":"say hello"}]
	}`))
	w := httptest.NewRecorder()

	srv.handleChatCompletions(w, req)

	if w.Code != http.StatusOK {
		t.Fatalf("status: got %d body=%s", w.Code, w.Body.String())
	}
	if fake.req.Adapter != "ollama" || fake.req.Target != "llama-fixed" {
		t.Fatalf("dispatch target mismatch: %+v", fake.req)
	}
	if fake.req.SessionID != "cline" || fake.req.TimeoutSec != 15 {
		t.Fatalf("execution config mismatch: %+v", fake.req)
	}
	if !strings.Contains(fake.req.Prompt, "system: brief") || !strings.Contains(fake.req.Prompt, "user: say hello") {
		t.Fatalf("prompt did not include messages: %q", fake.req.Prompt)
	}

	var resp chatCompletionResponse
	if err := json.Unmarshal(w.Body.Bytes(), &resp); err != nil {
		t.Fatalf("decode response: %v", err)
	}
	if resp.Choices[0].Message.Content != "hello world" {
		t.Fatalf("content: got %q", resp.Choices[0].Message.Content)
	}
	if resp.Usage == nil || resp.Usage.TotalTokens != 4 {
		t.Fatalf("usage: %+v", resp.Usage)
	}
}

func TestChatCompletionsUsesRequestModelWhenNoConfiguredTarget(t *testing.T) {
	fake := &fakeRunService{events: make(chan *iop.RunEvent, 2)}
	fake.events <- &iop.RunEvent{Type: "delta", Delta: "ok"}
	fake.events <- &iop.RunEvent{Type: "complete"}

	srv := NewServer(config.EdgeOpenAIConf{Adapter: "ollama"}, fake, nil)
	req := httptest.NewRequest(http.MethodPost, "/v1/chat/completions", strings.NewReader(`{
		"model":"from-request",
		"messages":[{"role":"user","content":"hi"}]
	}`))
	w := httptest.NewRecorder()

	srv.handleChatCompletions(w, req)

	if w.Code != http.StatusOK {
		t.Fatalf("status: got %d body=%s", w.Code, w.Body.String())
	}
	if fake.req.Target != "from-request" {
		t.Fatalf("target: got %q", fake.req.Target)
	}
}

func TestChatCompletionsStreamsSSE(t *testing.T) {
	fake := &fakeRunService{events: make(chan *iop.RunEvent, 3)}
	fake.events <- &iop.RunEvent{Type: "delta", Delta: "hi"}
	fake.events <- &iop.RunEvent{Type: "complete"}

	srv := NewServer(config.EdgeOpenAIConf{Adapter: "ollama"}, fake, nil)
	req := httptest.NewRequest(http.MethodPost, "/v1/chat/completions", strings.NewReader(`{
		"model":"stream-model",
		"stream":true,
		"messages":[{"role":"user","content":"hi"}]
	}`))
	w := httptest.NewRecorder()

	srv.handleChatCompletions(w, req)

	if w.Code != http.StatusOK {
		t.Fatalf("status: got %d body=%s", w.Code, w.Body.String())
	}
	body := w.Body.String()
	if !strings.Contains(body, `"content":"hi"`) || !strings.Contains(body, "data: [DONE]") {
		t.Fatalf("unexpected SSE body:\n%s", body)
	}
}

func TestModelsUsesConfiguredModelsOrTarget(t *testing.T) {
	srv := NewServer(config.EdgeOpenAIConf{Target: "fallback-model"}, &fakeRunService{}, nil)
	req := httptest.NewRequest(http.MethodGet, "/v1/models", nil)
	w := httptest.NewRecorder()
	srv.handleModels(w, req)
	if w.Code != http.StatusOK {
		t.Fatalf("status: got %d", w.Code)
	}
	if !strings.Contains(w.Body.String(), "fallback-model") {
		t.Fatalf("expected target model, got %s", w.Body.String())
	}

	srv = NewServer(config.EdgeOpenAIConf{Models: []string{"a", "b"}}, &fakeRunService{}, nil)
	w = httptest.NewRecorder()
	srv.handleModels(w, req)
	if !strings.Contains(w.Body.String(), `"id":"a"`) || !strings.Contains(w.Body.String(), `"id":"b"`) {
		t.Fatalf("expected configured models, got %s", w.Body.String())
	}
}

func TestCollectRunTextTimesOut(t *testing.T) {
	handle := &edgeservice.RunHandle{
		RunDispatch: edgeservice.RunDispatch{TimeoutSec: 1},
		RunStream: edgeservice.RunStream{
			Events:     make(chan *iop.RunEvent),
			NodeEvents: make(chan *iop.EdgeNodeEvent),
		},
	}
	ctx, cancel := context.WithTimeout(context.Background(), 1500*time.Millisecond)
	defer cancel()
	_, _, err := collectRunText(ctx, handle)
	if err == nil {
		t.Fatal("expected timeout error")
	}
}