Files
Enterprise-Knowledge-Base/cli/internal/sse/accumulator_test.go
T
nullkey e236be1ced fix(cli): correct KB id detection, SSE terminal-frame, and CI test isolation
Three defects surfaced during end-to-end RAG verification — the first two
block real chat usage, the third makes Linux CI flaky:

1. KB id detection — `IsKBID` was checking
   `strings.HasPrefix(s, "kb_")`, but WeKnora generates KB ids as bare
   UUIDs (internal/types/knowledge_base.go: `uuid.New().String()` stored
   in a `varchar(36)` column). Real ids therefore fell through to the
   name-resolution path:

     $ weknora chat ... --kb a32a63ff-fb36-4874-bcaa-30f48570a694
     Error: knowledge base not found: a32a63ff-...

   Switched the discriminator to a UUID regex
   (`^[0-9a-fA-F]{8}-…-[0-9a-fA-F]{12}$`). KB names are arbitrary
   user-supplied strings, so the canonical 8-4-4-4-12 form is an
   unambiguous signal. Mirrors gcloud `--project`'s id-vs-name detection.

2. SSE terminal-frame — the accumulator's `Append` was gating
   finalization on `r.Done`, but the server's KnowledgeQAStream protocol
   emits a leading `agent_query` frame with `done=true` to deliver
   session + message metadata *before* the answer fragments arrive:

     event: message
     data: {"response_type":"agent_query","content":"","done":true,…}

     event: message
     data: {"response_type":"answer","content":"你好","done":false}
     …
     event: message
     data: {"response_type":"complete","content":"","done":true}

   The accumulator therefore flipped to `finished=true` on frame #1 and
   discarded every subsequent answer fragment — `weknora chat … --json`
   returned `answer: ""` even though the LLM reported completion_tokens
   > 0. Fixed: terminate only on `response_type == complete`.
   References still captured opportunistically (they may arrive on a
   dedicated `references` event before the terminator).

3. doctor credential_storage CI isolation — the check probes the real
   OS keyring via `secrets.NewBestEffortStore()`: present on macOS dev
   machines → StatusOK; absent on Linux CI runners without libsecret /
   Gnome-Keyring → StatusWarn ("falling back to file store"). That
   host-dependence was leaking into two test classes that assumed
   StatusOK:

     * cmd/doctor/doctor_test.go: TestDoctor_AllOK and
       TestDoctor_NoConfig_StillRunsCredentialStorage already had a
       withCredStoreFactory seam but didn't use it. Added the pin.

     * acceptance/contract/envelope_test.go: doctor.success_offline
       and doctor.error_network golden cases. The contract test runs
       through the cobra tree in-process and shares cmd/doctor's
       package-level credStoreFactory var — but couldn't reach it
       because the existing seam was unexported.

   Fix: export `doctor.SetCredStoreFactoryForTest(fn) (restore func())`
   for out-of-package tests; acceptance/contract/helpers_test.go adds
   a TestMain that pins the factory to a MemStore-returning closure
   for the whole suite (MemStore is neither *FileStore nor a real
   keyring, so doctor's type-switch hits StatusOK). Production stays
   at secrets.NewBestEffortStore — only the test hook is now reachable
   from across packages.

Test fixtures and goldens that used the old `kb_xxx` literals or
`Done: true` terminators were rewritten to use real UUIDs and
`ResponseType: ResponseTypeComplete` respectively. Per-command --help
text and Long descriptions / Examples now show a UUID rather than
`kb_…` so users see the correct shape from the start. New
TestAccumulator_IgnoresAgentQueryDone pins the SSE terminator bug so
it can't regress.

Tests: 24 cli packages green on macOS dev + Linux/macOS/Windows CI
matrix. Verified end-to-end against a live WeKnora server: `weknora
chat "..." --kb <UUID> --no-stream --json` returns the full LLM answer
in the envelope, live token streaming in TTY mode works, and the
credential_storage check renders deterministic envelopes across hosts.
2026-05-12 13:20:42 +08:00

124 lines
4.2 KiB
Go

package sse_test
import (
"testing"
"github.com/Tencent/WeKnora/cli/internal/sse"
sdk "github.com/Tencent/WeKnora/client"
)
// terminator is the canonical "stream is done" event the accumulator
// recognises. The server emits response_type=complete after the answer
// fragments and any references; per-event done=true on agent_query /
// answer frames is NOT terminal.
var terminator = &sdk.StreamResponse{ResponseType: sdk.ResponseTypeComplete, Done: true}
func TestAccumulator_AppendsContent(t *testing.T) {
a := &sse.Accumulator{}
a.Append(&sdk.StreamResponse{ResponseType: sdk.ResponseTypeAnswer, Content: "Hello "})
a.Append(&sdk.StreamResponse{ResponseType: sdk.ResponseTypeAnswer, Content: "world"})
if got := a.Result(); got != "Hello world" {
t.Errorf("got %q, want %q", got, "Hello world")
}
if a.Done() {
t.Error("expected Done=false before terminal event")
}
}
func TestAccumulator_FinalizesOnComplete(t *testing.T) {
a := &sse.Accumulator{}
a.Append(&sdk.StreamResponse{ResponseType: sdk.ResponseTypeAnswer, Content: "answer"})
// References may arrive on a dedicated event before the terminator.
a.Append(&sdk.StreamResponse{
ResponseType: sdk.ResponseTypeReferences,
KnowledgeReferences: []*sdk.SearchResult{{KnowledgeID: "k1"}},
})
a.Append(terminator)
if !a.Done() {
t.Error("expected Done=true after response_type=complete")
}
if len(a.References) != 1 {
t.Errorf("expected 1 reference, got %d", len(a.References))
}
if a.References[0].KnowledgeID != "k1" {
t.Errorf("references payload not preserved: %+v", a.References[0])
}
}
func TestAccumulator_IgnoresAgentQueryDone(t *testing.T) {
// Server emits a leading agent_query frame with done=true to deliver
// session metadata; the accumulator must NOT treat that as terminal —
// otherwise the answer fragments that follow would be discarded.
a := &sse.Accumulator{}
a.Append(&sdk.StreamResponse{
ResponseType: sdk.ResponseTypeAgentQuery,
Done: true,
SessionID: "sess_123",
AssistantMessageID: "msg_456",
})
a.Append(&sdk.StreamResponse{ResponseType: sdk.ResponseTypeAnswer, Content: "real answer"})
a.Append(terminator)
if !a.Done() {
t.Error("expected Done=true after response_type=complete")
}
if got := a.Result(); got != "real answer" {
t.Errorf("agent_query done=true should not terminate; got %q", got)
}
if a.SessionID != "sess_123" {
t.Errorf("session metadata from agent_query frame lost: got %q", a.SessionID)
}
}
func TestAccumulator_IgnoresPostComplete(t *testing.T) {
a := &sse.Accumulator{}
a.Append(&sdk.StreamResponse{ResponseType: sdk.ResponseTypeAnswer, Content: "first"})
a.Append(terminator)
a.Append(&sdk.StreamResponse{ResponseType: sdk.ResponseTypeAnswer, Content: "after"})
if got := a.Result(); got != "first" {
t.Errorf("post-complete append should be no-op, got %q", got)
}
}
func TestAccumulator_NilSafe(t *testing.T) {
a := &sse.Accumulator{}
a.Append(nil)
if got := a.Result(); got != "" {
t.Errorf("expected empty result for nil append, got %q", got)
}
if a.Done() {
t.Error("nil append must not finalize")
}
}
func TestAccumulator_CapturesSessionMetadata(t *testing.T) {
a := &sse.Accumulator{}
a.Append(&sdk.StreamResponse{
ResponseType: sdk.ResponseTypeAnswer,
SessionID: "sess_123",
AssistantMessageID: "msg_456",
Content: "hi",
})
a.Append(terminator)
if a.SessionID != "sess_123" {
t.Errorf("SessionID: got %q", a.SessionID)
}
if a.AssistantMessageID != "msg_456" {
t.Errorf("AssistantMessageID: got %q", a.AssistantMessageID)
}
}
func TestAccumulator_FirstSessionMetadataWins(t *testing.T) {
// Subsequent events must not overwrite the first non-empty value — the
// session id is set once at session start and any later override would be
// a server bug we should not silently mask.
a := &sse.Accumulator{}
a.Append(&sdk.StreamResponse{SessionID: "sess_first", AssistantMessageID: "msg_first"})
a.Append(&sdk.StreamResponse{SessionID: "sess_second", AssistantMessageID: "msg_second"})
if a.SessionID != "sess_first" {
t.Errorf("SessionID overwritten: got %q want sess_first", a.SessionID)
}
if a.AssistantMessageID != "msg_first" {
t.Errorf("AssistantMessageID overwritten: got %q want msg_first", a.AssistantMessageID)
}
}