111 lines
4.1 KiB
Go
111 lines
4.1 KiB
Go
// direct_extract_test.go — task 6:extractor 模式端到端(本地萃卡→POST rag_ingest_card)。
|
||
package main
|
||
|
||
import (
|
||
"encoding/json"
|
||
"io"
|
||
"net/http"
|
||
"net/http/httptest"
|
||
"os"
|
||
"path/filepath"
|
||
"strings"
|
||
"testing"
|
||
)
|
||
|
||
// 完整鏈(claude stub 版):丟原稿 → 萃卡落地本地 → 只有「卡片」被 POST 到 rag_ingest_card
|
||
// → 原文從未離開本機 → manifest 標 ingested(下一輪不重送)。
|
||
func TestDirectExtractorModeE2E(t *testing.T) {
|
||
root := t.TempDir()
|
||
if err := os.WriteFile(filepath.Join(root, "報銷規則.md"), []byte("# 原稿機密內容 XYZZY"), 0o644); err != nil {
|
||
t.Fatal(err)
|
||
}
|
||
|
||
// 假 cypher:收 rag_ingest_card、驗 payload、記帳
|
||
var posted []map[string]any
|
||
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
||
if !strings.HasSuffix(r.URL.Path, "/webhooks/named/demo/rag_ingest_card/trigger") {
|
||
t.Errorf("打錯端點:%s", r.URL.Path)
|
||
}
|
||
body, _ := io.ReadAll(r.Body)
|
||
var m map[string]any
|
||
_ = json.Unmarshal(body, &m)
|
||
posted = append(posted, m)
|
||
_ = json.NewEncoder(w).Encode(map[string]any{"success": true})
|
||
}))
|
||
defer srv.Close()
|
||
|
||
// stub claude:把原稿萃成卡(模擬 /wiki-capture 行為)
|
||
stubDir := t.TempDir()
|
||
stub := filepath.Join(stubDir, "claude")
|
||
script := "#!/bin/sh\nmkdir -p system-dev/wiki/cards\nprintf '# 報銷規則\\n## 一句話定義\\n測試卡\\n## 關聯\\n- 報銷規則 >> 屬於 >> 財務\\n' > 'system-dev/wiki/cards/報銷規則.md'\n"
|
||
if err := os.WriteFile(stub, []byte(script), 0o755); err != nil {
|
||
t.Fatal(err)
|
||
}
|
||
|
||
cfg := &DirectConfig{
|
||
WatchFolders: []string{root},
|
||
Manifest: filepath.Join(t.TempDir(), "m.json"),
|
||
CypherURL: srv.URL, Namespace: "demo", APIKey: "demo",
|
||
Library: "kb", Extractor: "claude", ClaudeBin: stub,
|
||
CardIngestWF: "rag_ingest_card", MaxRemoved: DefaultMaxRemovedRatio,
|
||
}
|
||
results, exit, _ := RunDirectOnce(cfg, false)
|
||
if exit != 0 {
|
||
t.Fatalf("exit=%d results=%+v", exit, results)
|
||
}
|
||
if len(results) != 1 || results[0].Status != "ingested" {
|
||
t.Fatalf("results=%+v", results)
|
||
}
|
||
// 卡片落地本地(用戶看得到自己的 wiki)
|
||
if _, err := os.Stat(filepath.Join(root, "system-dev", "wiki", "cards", "報銷規則.md")); err != nil {
|
||
t.Fatalf("卡片未落地:%v", err)
|
||
}
|
||
// 上雲的是卡片、不是原文
|
||
if len(posted) != 1 {
|
||
t.Fatalf("應恰好 POST 一張卡,got %d", len(posted))
|
||
}
|
||
cc, _ := posted[0]["card_content"].(string)
|
||
if !strings.Contains(cc, "## 一句話定義") {
|
||
t.Fatalf("card_content 不是卡片:%.80s", cc)
|
||
}
|
||
if strings.Contains(cc, "XYZZY") {
|
||
t.Fatal("原文內容洩上雲=違反四步定稿邊界")
|
||
}
|
||
if p, _ := posted[0]["path"].(string); p != "system-dev/wiki/cards/報銷規則.md" {
|
||
t.Fatalf("path=%q", p)
|
||
}
|
||
// 第二輪:原稿沒變 → 不重萃不重送
|
||
results2, exit2, _ := RunDirectOnce(cfg, false)
|
||
if exit2 != 0 || len(results2) != 0 || len(posted) != 1 {
|
||
t.Fatalf("第二輪應零事件:results=%+v posted=%d", results2, len(posted))
|
||
}
|
||
}
|
||
|
||
// 萃取失敗=該檔標 failed、exit=1、manifest 不標(下輪重試),其他檔不受影響。
|
||
func TestDirectExtractorFailKeepsRetry(t *testing.T) {
|
||
root := t.TempDir()
|
||
if err := os.WriteFile(filepath.Join(root, "a.md"), []byte("x"), 0o644); err != nil {
|
||
t.Fatal(err)
|
||
}
|
||
stubDir := t.TempDir()
|
||
stub := filepath.Join(stubDir, "claude")
|
||
if err := os.WriteFile(stub, []byte("#!/bin/sh\nexit 3\n"), 0o755); err != nil {
|
||
t.Fatal(err)
|
||
}
|
||
cfg := &DirectConfig{
|
||
WatchFolders: []string{root},
|
||
Manifest: filepath.Join(t.TempDir(), "m.json"),
|
||
CypherURL: "https://x.example", Namespace: "demo", APIKey: "demo",
|
||
Extractor: "claude", ClaudeBin: stub, MaxRemoved: DefaultMaxRemovedRatio,
|
||
}
|
||
results, exit, _ := RunDirectOnce(cfg, false)
|
||
if exit != 1 || len(results) != 1 || results[0].Status != "failed" {
|
||
t.Fatalf("exit=%d results=%+v", exit, results)
|
||
}
|
||
// 再跑一輪:仍是同一個事件(manifest 沒標 ingested=會重試)
|
||
results2, _, _ := RunDirectOnce(cfg, false)
|
||
if len(results2) != 1 {
|
||
t.Fatalf("失敗檔應重試:%+v", results2)
|
||
}
|
||
}
|