From 0b061c5fa133bf930981816d483d563231a63952 Mon Sep 17 00:00:00 2001
From: a1gaoshanlaiyige <65058021+a1gaoshanlaiyige@users.noreply.github.com>
Date: Sun, 6 Sep 2026 16:42:57 +0800
Subject: [PATCH] fix(gateway): use JSON escaping for message cache promotion

Use encoding/json instead of Go string quoting when promoting message content to cached text blocks. Add regression coverage for control characters, Unicode, literal backslashes, text preservation, and idempotence.
---
 .../service/gateway_messages_cache.go         |  6 +-
 .../gateway_messages_cache_json_test.go       | 59 +++++++++++++++++++
 2 files changed, 64 insertions(+), 1 deletion(-)
 create mode 100644 backend/internal/service/gateway_messages_cache_json_test.go

diff --git a/backend/internal/service/gateway_messages_cache.go b/backend/internal/service/gateway_messages_cache.go
index 8f7e6b5b7d..646f8b2177 100644
--- a/backend/internal/service/gateway_messages_cache.go
+++ b/backend/internal/service/gateway_messages_cache.go
@@ -2,6 +2,7 @@ package service
 
 import (
 	"context"
+	"encoding/json"
 	"fmt"
 
 	"github.com/Wei-Shaw/sub2api/internal/pkg/claude"
@@ -157,5 +158,8 @@ func injectCacheControlOnLastContentBlock(body []byte, idx int, msg *gjson.Resul
 // mustJSONString 把一个 Go string 序列化为合法 JSON string（含引号），
 // 用于 sjson.SetRawBytes 场景下手工拼 JSON。
 func mustJSONString(s string) string {
-	return fmt.Sprintf("%q", s)
+	// Go string quoting can emit non-JSON escapes such as \x7f or \a.
+	// Marshaling a string cannot fail.
+	encoded, _ := json.Marshal(s)
+	return string(encoded)
 }
diff --git a/backend/internal/service/gateway_messages_cache_json_test.go b/backend/internal/service/gateway_messages_cache_json_test.go
new file mode 100644
index 0000000000..26759c4d72
--- /dev/null
+++ b/backend/internal/service/gateway_messages_cache_json_test.go
@@ -0,0 +1,59 @@
+//go:build unit
+
+package service
+
+import (
+	"encoding/json"
+	"strings"
+	"testing"
+
+	"github.com/stretchr/testify/require"
+	"github.com/tidwall/gjson"
+)
+
+func TestAddMessageCacheBreakpoints_JSONStringEscaping(t *testing.T) {
+	var controls strings.Builder
+	for r := rune(0); r <= 0x9f; r++ {
+		controls.WriteRune(r)
+	}
+	tests := []struct{ name, text string }{
+		{"delete_character", "before\x7fafter"},
+		{"bell_and_vertical_tab", "before\a\vafter"},
+		{"nul_and_escape", "before\x00\x1bafter"},
+		{"all_ascii_and_c1_controls", controls.String()},
+		{"supplementary_nonprinting_rune", "before\U000e0067after"},
+		{"unicode_and_json_metacharacters", "中文 😀 <>& \"quote\" \\slash\n\r\t\b\f"},
+		{"literal_backslash_sequences", `literal \x7f \a \v \U000e0067`},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			// Both the final message and the second-to-last user turn are promoted
+			// from strings to text blocks by the cache rewrite.
+			body, err := json.Marshal(map[string]any{
+				"messages": []map[string]any{
+					{"role": "user", "content": tt.text},
+					{"role": "assistant", "content": "ack"},
+					{"role": "user", "content": "next"},
+					{"role": "assistant", "content": tt.text},
+				},
+			})
+			require.NoError(t, err)
+			out := addMessageCacheBreakpoints(body)
+			// gjson accepts some malformed input, so validate with encoding/json
+			// before inspecting the rewritten content and cache metadata.
+			var decoded any
+			require.NoError(t, json.Unmarshal(out, &decoded))
+			for _, path := range []string{"messages.0.content", "messages.3.content"} {
+				content := gjson.GetBytes(out, path)
+				require.True(t, content.IsArray())
+				require.Len(t, content.Array(), 1)
+				block := content.Array()[0]
+				require.Equal(t, "text", block.Get("type").String())
+				require.Equal(t, tt.text, block.Get("text").String())
+				require.Equal(t, "ephemeral", block.Get("cache_control.type").String())
+				require.Equal(t, "5m", block.Get("cache_control.ttl").String())
+			}
+			require.JSONEq(t, string(out), string(addMessageCacheBreakpoints(out)))
+		})
+	}
+}