package api import ( "bytes" "context" "crypto/tls" "encoding/json" "io" "log/slog" "net/http" "net/http/httptest" "net/netip" "os" "path/filepath" "strings" "sync" "testing" "time" "git.wntrmute.dev/kyle/sift/internal/auth" "git.wntrmute.dev/kyle/sift/internal/digest" "git.wntrmute.dev/kyle/sift/internal/fetch" "git.wntrmute.dev/kyle/sift/internal/guard" ) func setupAPI(t *testing.T, g *guard.Guard) (*Server, string) { t.Helper() dir := t.TempDir() keyOut := filepath.Join(dir, "k") keys := filepath.Join(dir, "keys.toml") if err := auth.Add(keys, "test", "hermes-fleet", keyOut); err != nil { t.Fatal(err) } kb, err := os.ReadFile(keyOut) if err != nil { t.Fatal(err) } store, err := auth.Load(keys) if err != nil { t.Fatal(err) } if g == nil { g = &guard.Guard{} } f := &fetch.Fetcher{Guard: g, UA: "sift-test"} s := New("test", g, f, store, slog.New(slog.NewTextHandler(io.Discard, nil))) return s, strings.TrimSpace(string(kb)) } func postJSON(t *testing.T, h http.Handler, path, key string, body any) *httptest.ResponseRecorder { t.Helper() b, _ := json.Marshal(body) req := httptest.NewRequest(http.MethodPost, path, bytes.NewReader(b)) req.Header.Set("Content-Type", "application/json") if key != "" { req.Header.Set("Authorization", "Bearer "+key) } rec := httptest.NewRecorder() h.ServeHTTP(rec, req) return rec } func TestAuthMissingUnknown(t *testing.T) { s, key := setupAPI(t, nil) rec := postJSON(t, s, "/v2/scrape", "", map[string]string{"url": "https://example.com/"}) if rec.Code != 401 { t.Fatalf("missing key: %d %s", rec.Code, rec.Body.String()) } rec = postJSON(t, s, "/v2/scrape", "nope", map[string]string{"url": "https://example.com/"}) if rec.Code != 401 { t.Fatalf("bad key: %d", rec.Code) } rec = postJSON(t, s, "/healthz", "", nil) _ = key } func TestHealthzNoAuth(t *testing.T) { s, _ := setupAPI(t, nil) req := httptest.NewRequest(http.MethodGet, "/healthz", nil) rec := httptest.NewRecorder() s.ServeHTTP(rec, req) if rec.Code != 200 { t.Fatal(rec.Code) } } func TestDenyPrefixThroughScrape(t *testing.T) { g := &guard.Guard{DenyPrefixes: []netip.Prefix{netip.MustParsePrefix("8.8.8.8/32")}} s, key := setupAPI(t, g) rec := postJSON(t, s, "/v2/scrape", key, map[string]string{"url": "http://8.8.8.8/"}) if rec.Code != 403 { t.Fatalf("got %d %s", rec.Code, rec.Body.String()) } if !strings.Contains(rec.Body.String(), "blocked:") { t.Fatal(rec.Body.String()) } if strings.Contains(rec.Body.String(), `"success": true`) { t.Fatal("success true") } } func TestGuardLoopbackAndTailnet(t *testing.T) { s, key := setupAPI(t, &guard.Guard{}) for _, u := range []string{ "http://127.0.0.1/", "http://100.95.252.120:8000", } { rec := postJSON(t, s, "/v2/scrape", key, map[string]string{"url": u}) if rec.Code != 403 { t.Fatalf("%s: %d %s", u, rec.Code, rec.Body.String()) } } } func TestRedirectToPrivate(t *testing.T) { g := &guard.Guard{AllowPrefixes: []netip.Prefix{netip.MustParsePrefix("127.0.0.0/8")}} ts := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { http.Redirect(w, r, "http://10.0.0.5/secret", http.StatusFound) })) defer ts.Close() s, key := setupAPI(t, g) rec := postJSON(t, s, "/v2/scrape", key, map[string]string{"url": ts.URL}) if rec.Code != 403 { t.Fatalf("got %d %s", rec.Code, rec.Body.String()) } } func TestSkipTLSVerificationIgnored(t *testing.T) { g := &guard.Guard{AllowPrefixes: []netip.Prefix{netip.MustParsePrefix("127.0.0.0/8")}} ts := httptest.NewTLSServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { w.Header().Set("Content-Type", "text/plain") _, _ = w.Write([]byte("hi")) })) defer ts.Close() s, key := setupAPI(t, g) rec := postJSON(t, s, "/v2/scrape", key, map[string]any{ "url": ts.URL, "skipTlsVerification": true, }) if rec.Code == 200 { t.Fatalf("bad cert succeeded: %s", rec.Body.String()) } } func TestDigestFallbackWhenUnset(t *testing.T) { g := &guard.Guard{AllowPrefixes: []netip.Prefix{netip.MustParsePrefix("127.0.0.0/8")}} ts := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { w.Header().Set("Content-Type", "text/html") _, _ = w.Write([]byte("

hello

")) })) defer ts.Close() s, key := setupAPI(t, g) rec := postJSON(t, s, "/v1/fetch", key, map[string]any{"url": ts.URL, "mode": "digest", "question": "what"}) if rec.Code != 200 { t.Fatalf("%d %s", rec.Code, rec.Body.String()) } if !strings.Contains(rec.Body.String(), "fallback") { t.Fatalf("want fallback: %s", rec.Body.String()) } } func TestCapDigestFallback(t *testing.T) { big := strings.Repeat("word ", 50000) out, warn := capDigestFallback(big, nil) if digest.EstTokens(out) > digest.SmallDoc { t.Fatalf("tokens %d", digest.EstTokens(out)) } joined := strings.Join(warn, " ") if !strings.Contains(joined, "digest budget") { t.Fatalf("warn %v", warn) } } func TestNever502(t *testing.T) { s, key := setupAPI(t, nil) rec := postJSON(t, s, "/v2/scrape", key, map[string]string{"url": "http://127.0.0.1/"}) if rec.Code == 502 { t.Fatal("returned 502") } _ = key } func TestCleanHTMLThroughScrape(t *testing.T) { g := &guard.Guard{AllowPrefixes: []netip.Prefix{netip.MustParsePrefix("127.0.0.0/8")}} ts := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { w.Header().Set("Content-Type", "text/html; charset=utf-8") _, _ = w.Write([]byte(`Hello

Hi

World

`)) })) defer ts.Close() s, key := setupAPI(t, g) rec := postJSON(t, s, "/v2/scrape", key, map[string]any{ "url": ts.URL, "formats": []string{"markdown"}, }) if rec.Code != 200 { t.Fatalf("%d %s", rec.Code, rec.Body.String()) } var env map[string]any _ = json.Unmarshal(rec.Body.Bytes(), &env) if env["success"] != true { t.Fatal(env) } } func TestJSONJQ(t *testing.T) { g := &guard.Guard{AllowPrefixes: []netip.Prefix{netip.MustParsePrefix("127.0.0.0/8")}} ts := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { w.Header().Set("Content-Type", "application/json") _, _ = w.Write([]byte(`{"a":1,"b":2}`)) })) defer ts.Close() s, key := setupAPI(t, g) rec := postJSON(t, s, "/v1/fetch", key, map[string]any{"url": ts.URL, "jq": ".a", "mode": "clean"}) if rec.Code != 200 { t.Fatalf("%d %s", rec.Code, rec.Body.String()) } if !strings.Contains(rec.Body.String(), "1") { t.Fatal(rec.Body.String()) } } func TestUnsupportedType(t *testing.T) { g := &guard.Guard{AllowPrefixes: []netip.Prefix{netip.MustParsePrefix("127.0.0.0/8")}} ts := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { // P5a: a PDF is supported now; a zip is still not. w.Header().Set("Content-Type", "application/zip") _, _ = w.Write([]byte("PK\x03\x04")) })) defer ts.Close() s, key := setupAPI(t, g) rec := postJSON(t, s, "/v2/scrape", key, map[string]string{"url": ts.URL}) if rec.Code != 415 { t.Fatalf("%d %s", rec.Code, rec.Body.String()) } } func TestFetcherUsesSameGuard(t *testing.T) { g := &guard.Guard{DenyPrefixes: []netip.Prefix{netip.MustParsePrefix("1.2.3.4/32")}} s, key := setupAPI(t, g) if s.Fetch.Guard != g { t.Fatal("fetcher not using the same Guard") } _ = key _ = tls.VersionTLS13 } func TestJQTooLarge(t *testing.T) { g := &guard.Guard{AllowPrefixes: []netip.Prefix{netip.MustParsePrefix("127.0.0.0/8")}} ts := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { w.Header().Set("Content-Type", "application/json") _, _ = w.Write([]byte(`0`)) })) defer ts.Close() s, key := setupAPI(t, g) rec := postJSON(t, s, "/v1/fetch", key, map[string]any{"url": ts.URL, "jq": "range(20000)", "mode": "clean"}) if rec.Code != 400 || !strings.Contains(rec.Body.String(), "jq output too large") { t.Fatalf("%d %s", rec.Code, rec.Body.String()) } } func TestHTTPServerWriteTimeout(t *testing.T) { s := New("test", &guard.Guard{}, nil, nil, slog.New(slog.NewTextHandler(io.Discard, nil))) hs := s.HTTPServer("127.0.0.1:0") if hs.WriteTimeout != 240*time.Second { // P5a: 120 s download + 20 s queue + 60 s extraction fit inside t.Fatalf("WriteTimeout %s", hs.WriteTimeout) } } type recHandler struct { mu sync.Mutex recs []slog.Record } func (h *recHandler) Enabled(context.Context, slog.Level) bool { return true } func (h *recHandler) Handle(_ context.Context, r slog.Record) error { h.mu.Lock() h.recs = append(h.recs, r.Clone()) h.mu.Unlock() return nil } func (h *recHandler) WithAttrs([]slog.Attr) slog.Handler { return h } func (h *recHandler) WithGroup(string) slog.Handler { return h } func recHost(r slog.Record) string { var host string r.Attrs(func(a slog.Attr) bool { if a.Key == "host" { host = a.Value.String() } return true }) return host } func recMsg(r slog.Record) string { return r.Message } func recDump(r slog.Record) string { var b strings.Builder b.WriteString(r.Message) r.Attrs(func(a slog.Attr) bool { b.WriteByte(' ') b.WriteString(a.Key) b.WriteByte('=') b.WriteString(a.Value.String()) return true }) return b.String() } func TestRestrictedCallerLogsBlankHost(t *testing.T) { dir := t.TempDir() keys := filepath.Join(dir, "keys.toml") fleetOut := filepath.Join(dir, "fleet") gsOut := filepath.Join(dir, "gs") if err := auth.Add(keys, "test", "hermes-fleet", fleetOut); err != nil { t.Fatal(err) } if err := auth.Add(keys, "straylight-gs", "guardian-star", gsOut); err != nil { t.Fatal(err) } fleetKey, err := os.ReadFile(fleetOut) if err != nil { t.Fatal(err) } gsKey, err := os.ReadFile(gsOut) if err != nil { t.Fatal(err) } store, err := auth.Load(keys) if err != nil { t.Fatal(err) } g := &guard.Guard{AllowPrefixes: []netip.Prefix{netip.MustParsePrefix("127.0.0.0/8")}} h := &recHandler{} s := New("test", g, &fetch.Fetcher{Guard: g, UA: "sift-test"}, store, slog.New(h)) s.PermitRestricted = []string{"straylight-gs"} ts := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { w.Header().Set("Content-Type", "text/html; charset=utf-8") _, _ = w.Write([]byte(`Hi

ok

`)) })) defer ts.Close() rec := postJSON(t, s, "/v1/fetch", strings.TrimSpace(string(gsKey)), map[string]any{"url": ts.URL, "mode": "clean"}) if rec.Code != 200 { t.Fatalf("gs fetch: %d %s", rec.Code, rec.Body.String()) } var env map[string]any if err := json.Unmarshal(rec.Body.Bytes(), &env); err != nil { t.Fatal(err) } src, _ := env["source"].(map[string]any) final, _ := src["final_url"].(string) if !strings.Contains(final, hostOf(ts.URL)) { t.Fatalf("final_url missing from body, got %q", final) } rec = postJSON(t, s, "/v1/fetch", strings.TrimSpace(string(gsKey)), map[string]any{"url": "http://10.0.0.5/", "mode": "clean"}) if rec.Code != 403 { t.Fatalf("gs refused: %d %s", rec.Code, rec.Body.String()) } const badHost = "this-host-does-not-exist-sift-p25.invalid" rec = postJSON(t, s, "/v1/fetch", strings.TrimSpace(string(gsKey)), map[string]any{"url": "http://" + badHost + "/", "mode": "clean"}) if rec.Code != 403 { t.Fatalf("gs nxdomain: %d %s", rec.Code, rec.Body.String()) } rec = postJSON(t, s, "/v1/fetch", strings.TrimSpace(string(fleetKey)), map[string]any{"url": ts.URL, "mode": "clean"}) if rec.Code != 200 { t.Fatalf("fleet fetch: %d %s", rec.Code, rec.Body.String()) } var gsFetch, gsRefused, fleetFetch string h.mu.Lock() for _, r := range h.recs { switch recMsg(r) { case "fetch": if attrCaller(r) == "straylight-gs" { gsFetch = recHost(r) } if attrCaller(r) == "test" { fleetFetch = recHost(r) } case "refused": if attrCaller(r) == "straylight-gs" { gsRefused = recHost(r) } } } h.mu.Unlock() if gsFetch != "-" { t.Fatalf("gs fetch host=%q want -", gsFetch) } if gsRefused != "-" { t.Fatalf("gs refused host=%q want -", gsRefused) } wantHost := hostOf(ts.URL) if fleetFetch != wantHost { t.Fatalf("fleet fetch host=%q want %q", fleetFetch, wantHost) } h.mu.Lock() for _, r := range h.recs { if attrCaller(r) != "straylight-gs" { continue } dump := recDump(r) if strings.Contains(dump, badHost) || strings.Contains(dump, "10.0.0.5") { t.Fatalf("restricted log leaked host: %s", dump) } } h.mu.Unlock() } func attrCaller(r slog.Record) string { var name string r.Attrs(func(a slog.Attr) bool { if a.Key == "caller" { name = a.Value.String() } return true }) return name }