Changeset 0.17.1 (#76)
This commit is contained in:
@@ -11,38 +11,48 @@ import (
|
||||
"git.gobha.me/xcaliber/chat-switchboard/database"
|
||||
"git.gobha.me/xcaliber/chat-switchboard/models"
|
||||
"git.gobha.me/xcaliber/chat-switchboard/roles"
|
||||
"git.gobha.me/xcaliber/chat-switchboard/store"
|
||||
postgres "git.gobha.me/xcaliber/chat-switchboard/store/postgres"
|
||||
sqlite "git.gobha.me/xcaliber/chat-switchboard/store/sqlite"
|
||||
"git.gobha.me/xcaliber/chat-switchboard/treepath"
|
||||
)
|
||||
|
||||
// ═══════════════════════════════════════════
|
||||
// Live Compaction Tests
|
||||
// ═══════════════════════════════════════════
|
||||
// Requires: TEST_DATABASE_URL + VENICE_API_KEY
|
||||
// Requires: TEST_DATABASE_URL + PROVIDER_KEY (or VENICE_API_KEY)
|
||||
//
|
||||
// Tests the full Compact() pipeline end-to-end:
|
||||
// seed messages → call utility model → verify summary tree node
|
||||
// ═══════════════════════════════════════════
|
||||
|
||||
// dialectStores returns the correct store bundle for the active dialect.
|
||||
func dialectStores() store.Stores {
|
||||
if database.IsSQLite() {
|
||||
return sqlite.NewStores(database.TestDB)
|
||||
}
|
||||
return postgres.NewStores(database.TestDB)
|
||||
}
|
||||
|
||||
func TestLive_CompactionFullPipeline(t *testing.T) {
|
||||
h := setupHarness(t)
|
||||
veniceKey := requireVeniceKey(t)
|
||||
pc := requireLiveProvider(t)
|
||||
userID, adminToken := h.createAdminUser("compactadmin", "compact@test.com")
|
||||
|
||||
// Set up Venice provider + enable qwen3-4b
|
||||
configID, _ := setupVeniceWithModel(t, h, adminToken, veniceKey, veniceTestModel)
|
||||
// Set up provider + enable first model
|
||||
configID, modelID := setupProviderWithModel(t, h, adminToken, pc)
|
||||
|
||||
// Configure utility role → qwen3-4b
|
||||
// Configure utility role
|
||||
w := h.request("PUT", "/api/v1/admin/roles/utility", adminToken, map[string]interface{}{
|
||||
"primary": map[string]interface{}{
|
||||
"provider_config_id": configID,
|
||||
"model_id": veniceTestModel,
|
||||
"model_id": modelID,
|
||||
},
|
||||
})
|
||||
if w.Code != http.StatusOK {
|
||||
t.Fatalf("configure utility role: %d: %s", w.Code, w.Body.String())
|
||||
}
|
||||
t.Log(" ✓ Utility role configured with", veniceTestModel)
|
||||
t.Log(" ✓ Utility role configured with", modelID)
|
||||
|
||||
// Create a channel with enough messages to summarize
|
||||
w = h.request("POST", "/api/v1/channels", adminToken, map[string]interface{}{
|
||||
@@ -69,33 +79,30 @@ func TestLive_CompactionFullPipeline(t *testing.T) {
|
||||
{"assistant", "Early morning is best for temples. Fushimi Inari at sunrise (5-6am) is magical and nearly empty. Kinkaku-ji opens at 9am — arrive right at opening. Arashiyama bamboo grove is best before 8am. For Senso-ji in Tokyo, go at dawn for beautiful photos. Weekdays are always less crowded than weekends."},
|
||||
}
|
||||
|
||||
ph := database.PH
|
||||
var lastMsgID string
|
||||
for _, m := range msgs {
|
||||
var parentPtr *string
|
||||
if lastMsgID != "" {
|
||||
parentPtr = &lastMsgID
|
||||
}
|
||||
err := database.TestDB.QueryRow(`
|
||||
INSERT INTO messages (channel_id, parent_id, role, content, sibling_index)
|
||||
VALUES ($1, $2, $3, $4, 0)
|
||||
RETURNING id
|
||||
`, channelID, parentPtr, m.role, m.content).Scan(&lastMsgID)
|
||||
err := database.TestDB.QueryRow(
|
||||
"INSERT INTO messages (channel_id, parent_id, role, content, sibling_index) VALUES ("+ph(1)+", "+ph(2)+", "+ph(3)+", "+ph(4)+", 0) RETURNING id",
|
||||
channelID, parentPtr, m.role, m.content).Scan(&lastMsgID)
|
||||
if err != nil {
|
||||
t.Fatalf("seed message: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
// Set cursor to last message
|
||||
database.TestDB.Exec(`
|
||||
INSERT INTO channel_cursors (channel_id, user_id, active_leaf_id)
|
||||
VALUES ($1, $2, $3)
|
||||
ON CONFLICT (channel_id, user_id) DO UPDATE SET active_leaf_id = $3
|
||||
`, channelID, userID, lastMsgID)
|
||||
database.TestDB.Exec(
|
||||
"INSERT INTO channel_cursors (channel_id, user_id, active_leaf_id) VALUES ("+ph(1)+", "+ph(2)+", "+ph(3)+") ON CONFLICT (channel_id, user_id) DO UPDATE SET active_leaf_id = excluded.active_leaf_id",
|
||||
channelID, userID, lastMsgID)
|
||||
|
||||
t.Logf(" ✓ Seeded %d messages in channel %s", len(msgs), channelID)
|
||||
|
||||
// ── Run compaction ──
|
||||
stores := postgres.NewStores(database.TestDB)
|
||||
stores := dialectStores()
|
||||
resolver := roles.NewResolver(stores, nil)
|
||||
svc := compaction.NewService(stores, resolver)
|
||||
|
||||
@@ -119,16 +126,16 @@ func TestLive_CompactionFullPipeline(t *testing.T) {
|
||||
if result.Content == "" {
|
||||
t.Fatal("summary content should not be empty")
|
||||
}
|
||||
if result.Model != veniceTestModel {
|
||||
t.Errorf("model = %q, want %q", result.Model, veniceTestModel)
|
||||
if result.Model != modelID {
|
||||
t.Errorf("model = %q, want %q", result.Model, modelID)
|
||||
}
|
||||
|
||||
// Verify summary message exists in the tree
|
||||
var summaryContent, summaryRole string
|
||||
var metadataRaw []byte
|
||||
err = database.TestDB.QueryRow(`
|
||||
SELECT role, content, metadata FROM messages WHERE id = $1
|
||||
`, result.SummaryID).Scan(&summaryRole, &summaryContent, &metadataRaw)
|
||||
err = database.TestDB.QueryRow(
|
||||
"SELECT role, content, metadata FROM messages WHERE id = "+ph(1),
|
||||
result.SummaryID).Scan(&summaryRole, &summaryContent, &metadataRaw)
|
||||
if err != nil {
|
||||
t.Fatalf("query summary message: %v", err)
|
||||
}
|
||||
@@ -152,7 +159,6 @@ func TestLive_CompactionFullPipeline(t *testing.T) {
|
||||
t.Fatalf("GetActivePath after compaction: %v", err)
|
||||
}
|
||||
|
||||
// The last message in the path should be the summary
|
||||
if len(path) == 0 {
|
||||
t.Fatal("path should not be empty after compaction")
|
||||
}
|
||||
@@ -164,9 +170,9 @@ func TestLive_CompactionFullPipeline(t *testing.T) {
|
||||
|
||||
// Verify usage was logged
|
||||
var usageCount int
|
||||
database.TestDB.QueryRow(`
|
||||
SELECT COUNT(*) FROM usage_log WHERE channel_id = $1 AND role = 'utility'
|
||||
`, channelID).Scan(&usageCount)
|
||||
database.TestDB.QueryRow(
|
||||
"SELECT COUNT(*) FROM usage_log WHERE channel_id = "+ph(1)+" AND role = 'utility'",
|
||||
channelID).Scan(&usageCount)
|
||||
if usageCount == 0 {
|
||||
t.Error("expected usage_log entry for utility role")
|
||||
}
|
||||
@@ -177,24 +183,31 @@ func TestLive_CompactionFullPipeline(t *testing.T) {
|
||||
// rejects input that exceeds the utility model's context window.
|
||||
func TestLive_CompactionContextBudgetGuardRail(t *testing.T) {
|
||||
h := setupHarness(t)
|
||||
veniceKey := requireVeniceKey(t)
|
||||
pc := requireLiveProvider(t)
|
||||
_, adminToken := h.createAdminUser("guardrail_admin", "guardrail@test.com")
|
||||
userID := database.SeedTestUser(t, "guardrail_user", "gruser@test.com")
|
||||
|
||||
// Set up Venice + qwen3-4b (32K context)
|
||||
configID, catalogEntryID := setupVeniceWithModel(t, h, adminToken, veniceKey, veniceTestModel)
|
||||
// Set up provider + enable first model
|
||||
configID, modelID := setupProviderWithModel(t, h, adminToken, pc)
|
||||
|
||||
ph := database.PH
|
||||
|
||||
// Set max_context to 32000 in catalog (in case sync didn't populate it)
|
||||
database.TestDB.Exec(`
|
||||
UPDATE model_catalog SET capabilities = capabilities || '{"max_context": 32000}'::jsonb
|
||||
WHERE id = $1
|
||||
`, catalogEntryID)
|
||||
if database.IsSQLite() {
|
||||
database.TestDB.Exec(
|
||||
"UPDATE model_catalog SET capabilities = json_set(COALESCE(capabilities,'{}'), '$.max_context', 32000) WHERE provider_config_id = "+ph(1)+" AND model_id = "+ph(2),
|
||||
configID, modelID)
|
||||
} else {
|
||||
database.TestDB.Exec(
|
||||
"UPDATE model_catalog SET capabilities = capabilities || '{\"max_context\": 32000}'::jsonb WHERE provider_config_id = "+ph(1)+" AND model_id = "+ph(2),
|
||||
configID, modelID)
|
||||
}
|
||||
|
||||
// Configure utility role
|
||||
h.request("PUT", "/api/v1/admin/roles/utility", adminToken, map[string]interface{}{
|
||||
"primary": map[string]interface{}{
|
||||
"provider_config_id": configID,
|
||||
"model_id": veniceTestModel,
|
||||
"model_id": modelID,
|
||||
},
|
||||
})
|
||||
|
||||
@@ -207,7 +220,6 @@ func TestLive_CompactionContextBudgetGuardRail(t *testing.T) {
|
||||
channelID := ch["id"].(string)
|
||||
|
||||
// Seed 20 messages × 8000 chars each = 160K chars ≈ 40K tokens
|
||||
// This exceeds 32K × 0.80 = 25.6K token ceiling
|
||||
bigContent := make([]byte, 8000)
|
||||
for i := range bigContent {
|
||||
bigContent[i] = 'a'
|
||||
@@ -222,19 +234,16 @@ func TestLive_CompactionContextBudgetGuardRail(t *testing.T) {
|
||||
if i%2 == 1 {
|
||||
role = "assistant"
|
||||
}
|
||||
database.TestDB.QueryRow(`
|
||||
INSERT INTO messages (channel_id, parent_id, role, content, sibling_index)
|
||||
VALUES ($1, $2, $3, $4, 0) RETURNING id
|
||||
`, channelID, parentPtr, role, string(bigContent)).Scan(&lastMsgID)
|
||||
database.TestDB.QueryRow(
|
||||
"INSERT INTO messages (channel_id, parent_id, role, content, sibling_index) VALUES ("+ph(1)+", "+ph(2)+", "+ph(3)+", "+ph(4)+", 0) RETURNING id",
|
||||
channelID, parentPtr, role, string(bigContent)).Scan(&lastMsgID)
|
||||
}
|
||||
database.TestDB.Exec(`
|
||||
INSERT INTO channel_cursors (channel_id, user_id, active_leaf_id)
|
||||
VALUES ($1, $2, $3)
|
||||
ON CONFLICT (channel_id, user_id) DO UPDATE SET active_leaf_id = $3
|
||||
`, channelID, userID, lastMsgID)
|
||||
database.TestDB.Exec(
|
||||
"INSERT INTO channel_cursors (channel_id, user_id, active_leaf_id) VALUES ("+ph(1)+", "+ph(2)+", "+ph(3)+") ON CONFLICT (channel_id, user_id) DO UPDATE SET active_leaf_id = excluded.active_leaf_id",
|
||||
channelID, userID, lastMsgID)
|
||||
|
||||
// Run compaction — should fail with context budget error
|
||||
stores := postgres.NewStores(database.TestDB)
|
||||
stores := dialectStores()
|
||||
resolver := roles.NewResolver(stores, nil)
|
||||
svc := compaction.NewService(stores, resolver)
|
||||
|
||||
@@ -250,7 +259,6 @@ func TestLive_CompactionContextBudgetGuardRail(t *testing.T) {
|
||||
|
||||
t.Logf(" ✓ Guard rail triggered: %v", err)
|
||||
|
||||
// Verify it's specifically a context budget error
|
||||
if !strings.Contains(err.Error(), "context window") {
|
||||
t.Errorf("expected context window error, got: %v", err)
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user