feat(tools/io): 内置 IO/网络工具（file(路径安全)/http(SSRF防护)/web-search/web-fetch/knowledge）+ 大小截断

plugins/tools/io/web_fetch_test.go

@@ -0,0 +1,108 @@
+package io
+
+import (
+	"context"
+	"errors"
+	"io"
+	"net/http"
+	"net/http/httptest"
+	"strings"
+	"testing"
+
+	"git.yeij.top/AskaEth/Cyrene_v2/internal/config"
+)
+
+// webFetchToolForTest builds a tools set with opts and returns the web-fetch
+// tool.
+func webFetchToolForTest(t *testing.T, opts ...Option) *webFetchTool {
+	t.Helper()
+	return findTool(t, buildTools(t, &config.Config{}, opts...), "web-fetch").(*webFetchTool)
+}
+
+func TestWebFetch_TextExtraction(t *testing.T) {
+	t.Parallel()
+	doc := `<!doctype html><html><head><title>T</title>` +
+		`<style>.x{color:red}</style><script>alert(1)</script></head>` +
+		`<body><h1>Hello</h1><p>World &amp; more</p><!-- hidden --></body></html>`
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "text/html; charset=utf-8")
+		_, _ = io.WriteString(w, doc)
+	}))
+	defer srv.Close()
+	wf := webFetchToolForTest(t, WithAllowedHosts(serverHost(srv)), WithHTTPClient(srv.Client()))
+
+	got, err := wf.Execute(context.Background(), map[string]any{"url": srv.URL})
+	if err != nil {
+		t.Fatalf("Execute: %v", err)
+	}
+	m := got.(map[string]any)
+	text := m["text"].(string)
+	if !strings.Contains(text, "Hello") || !strings.Contains(text, "World & more") {
+		t.Fatalf("text = %q", text)
+	}
+	if strings.Contains(text, "<") || strings.Contains(text, "alert") {
+		t.Fatalf("text still has tags/script: %q", text)
+	}
+	if m["content_type"] != "text/html" {
+		t.Fatalf("content_type = %q", m["content_type"])
+	}
+}
+
+func TestWebFetch_RejectsBinary(t *testing.T) {
+	t.Parallel()
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "image/png")
+		_, _ = io.WriteString(w, "\x89PNG")
+	}))
+	defer srv.Close()
+	wf := webFetchToolForTest(t, WithAllowedHosts(serverHost(srv)), WithHTTPClient(srv.Client()))
+
+	_, err := wf.Execute(context.Background(), map[string]any{"url": srv.URL})
+	if err == nil || !strings.Contains(err.Error(), "unexpected content type") {
+		t.Fatalf("expected content-type error, got %v", err)
+	}
+}
+
+func TestWebFetch_SSRFDenied(t *testing.T) {
+	t.Parallel()
+	wf := webFetchToolForTest(t)
+	_, err := wf.Execute(context.Background(), map[string]any{"url": "http://localhost:9/"})
+	if !errors.Is(err, ErrSSRF) {
+		t.Fatalf("expected ErrSSRF, got %v", err)
+	}
+}
+
+func TestWebFetch_SizeLimit(t *testing.T) {
+	t.Parallel()
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "text/plain")
+		_, _ = io.WriteString(w, strings.Repeat("z", 200))
+	}))
+	defer srv.Close()
+	wf := webFetchToolForTest(t, WithAllowedHosts(serverHost(srv)), WithHTTPClient(srv.Client()))
+
+	got, err := wf.Execute(context.Background(), map[string]any{"url": srv.URL, "max_bytes": 10})
+	if err != nil {
+		t.Fatalf("Execute: %v", err)
+	}
+	m := got.(map[string]any)
+	if !m["truncated"].(bool) {
+		t.Fatal("expected truncated=true")
+	}
+	if len(m["text"].(string)) != 10 {
+		t.Fatalf("text len = %d", len(m["text"].(string)))
+	}
+}
+
+func TestWebFetch_Non200Status(t *testing.T) {
+	t.Parallel()
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		http.Error(w, "bad", http.StatusNotFound)
+	}))
+	defer srv.Close()
+	wf := webFetchToolForTest(t, WithAllowedHosts(serverHost(srv)), WithHTTPClient(srv.Client()))
+	_, err := wf.Execute(context.Background(), map[string]any{"url": srv.URL})
+	if err == nil || !strings.Contains(err.Error(), "status 404") {
+		t.Fatalf("expected status-404 error, got %v", err)
+	}
+}