From c5aac1251d5496ea21f2b1bc09bcf7cee5eea687 Mon Sep 17 00:00:00 2001
From: YanzheL <lee.yanzhe@yanzhe.org>
Date: Thu, 2 Apr 2026 00:11:06 +0800
Subject: [PATCH] fix(gateway): add content-based session hash fallback for
 non-Codex clients

When no explicit session signals (session_id, conversation_id, prompt_cache_key)
are provided, derive a stable session seed from the request body content
(model + tools + system prompt + first user message) to enable sticky routing
and prompt caching for non-Codex clients using the Chat Completions API.

This mirrors the content-based fallback already present in GatewayService.
GenerateSessionHash, adapted for the OpenAI gateway's request formats (both
Chat Completions messages and Responses API input).

JSON fragments are canonicalized via normalizeCompatSeedJSON to ensure
semantically identical requests produce the same seed regardless of
whitespace or key ordering.

Closes #1421
---
 .../service/openai_content_session_seed.go    | 107 ++++++++++++++++++
 .../service/openai_gateway_service.go         |   4 +
 2 files changed, 111 insertions(+)
 create mode 100644 backend/internal/service/openai_content_session_seed.go

diff --git a/backend/internal/service/openai_content_session_seed.go b/backend/internal/service/openai_content_session_seed.go
new file mode 100644
index 00000000..cb8fcb84
--- /dev/null
+++ b/backend/internal/service/openai_content_session_seed.go
@@ -0,0 +1,107 @@
+package service
+
+import (
+	"encoding/json"
+	"strings"
+
+	"github.com/tidwall/gjson"
+)
+
+// contentSessionSeedPrefix prevents collisions between content-derived seeds
+// and explicit session IDs (e.g. "sess-xxx" or "compat_cc_xxx").
+const contentSessionSeedPrefix = "compat_cs_"
+
+// deriveOpenAIContentSessionSeed builds a stable session seed from an
+// OpenAI-format request body. Only fields constant across conversation turns
+// are included: model, tools/functions definitions, system/developer prompts,
+// instructions (Responses API), and the first user message.
+// Supports both Chat Completions (messages) and Responses API (input).
+func deriveOpenAIContentSessionSeed(body []byte) string {
+	if len(body) == 0 {
+		return ""
+	}
+
+	var b strings.Builder
+
+	if model := gjson.GetBytes(body, "model").String(); model != "" {
+		b.WriteString("model=")
+		b.WriteString(model)
+	}
+
+	if tools := gjson.GetBytes(body, "tools"); tools.Exists() && tools.IsArray() && tools.Raw != "[]" {
+		b.WriteString("|tools=")
+		b.WriteString(normalizeCompatSeedJSON(json.RawMessage(tools.Raw)))
+	}
+
+	if funcs := gjson.GetBytes(body, "functions"); funcs.Exists() && funcs.IsArray() && funcs.Raw != "[]" {
+		b.WriteString("|functions=")
+		b.WriteString(normalizeCompatSeedJSON(json.RawMessage(funcs.Raw)))
+	}
+
+	if instr := gjson.GetBytes(body, "instructions").String(); instr != "" {
+		b.WriteString("|instructions=")
+		b.WriteString(instr)
+	}
+
+	firstUserCaptured := false
+
+	msgs := gjson.GetBytes(body, "messages")
+	if msgs.Exists() && msgs.IsArray() {
+		msgs.ForEach(func(_, msg gjson.Result) bool {
+			role := msg.Get("role").String()
+			switch role {
+			case "system", "developer":
+				b.WriteString("|system=")
+				if c := msg.Get("content"); c.Exists() {
+					b.WriteString(normalizeCompatSeedJSON(json.RawMessage(c.Raw)))
+				}
+			case "user":
+				if !firstUserCaptured {
+					b.WriteString("|first_user=")
+					if c := msg.Get("content"); c.Exists() {
+						b.WriteString(normalizeCompatSeedJSON(json.RawMessage(c.Raw)))
+					}
+					firstUserCaptured = true
+				}
+			}
+			return true
+		})
+	} else if inp := gjson.GetBytes(body, "input"); inp.Exists() {
+		if inp.Type == gjson.String {
+			b.WriteString("|input=")
+			b.WriteString(inp.String())
+		} else if inp.IsArray() {
+			inp.ForEach(func(_, item gjson.Result) bool {
+				role := item.Get("role").String()
+				switch role {
+				case "system", "developer":
+					b.WriteString("|system=")
+					if c := item.Get("content"); c.Exists() {
+						b.WriteString(normalizeCompatSeedJSON(json.RawMessage(c.Raw)))
+					}
+				case "user":
+					if !firstUserCaptured {
+						b.WriteString("|first_user=")
+						if c := item.Get("content"); c.Exists() {
+							b.WriteString(normalizeCompatSeedJSON(json.RawMessage(c.Raw)))
+						}
+						firstUserCaptured = true
+					}
+				}
+				if !firstUserCaptured && item.Get("type").String() == "input_text" {
+					b.WriteString("|first_user=")
+					if text := item.Get("text").String(); text != "" {
+						b.WriteString(text)
+					}
+					firstUserCaptured = true
+				}
+				return true
+			})
+		}
+	}
+
+	if b.Len() == 0 {
+		return ""
+	}
+	return contentSessionSeedPrefix + b.String()
+}
diff --git a/backend/internal/service/openai_gateway_service.go b/backend/internal/service/openai_gateway_service.go
index 0a959615..b9f42cd7 100644
--- a/backend/internal/service/openai_gateway_service.go
+++ b/backend/internal/service/openai_gateway_service.go
@@ -1044,6 +1044,7 @@ func (s *OpenAIGatewayService) ExtractSessionID(c *gin.Context, body []byte) str
 //  1. Header: session_id
 //  2. Header: conversation_id
 //  3. Body:   prompt_cache_key (opencode)
+//  4. Body:   content-based fallback (model + system + tools + first user message)
 func (s *OpenAIGatewayService) GenerateSessionHash(c *gin.Context, body []byte) string {
 	if c == nil {
 		return ""
@@ -1056,6 +1057,9 @@ func (s *OpenAIGatewayService) GenerateSessionHash(c *gin.Context, body []byte)
 	if sessionID == "" && len(body) > 0 {
 		sessionID = strings.TrimSpace(gjson.GetBytes(body, "prompt_cache_key").String())
 	}
+	if sessionID == "" && len(body) > 0 {
+		sessionID = deriveOpenAIContentSessionSeed(body)
+	}
 	if sessionID == "" {
 		return ""
 	}