mslxdff 0.1.203 → 0.1.208

package/src/responses/translate.js

@@ -1,311 +1,311 @@
-/**
- * Responses ⇄ Chat 翻译层（给 Codex/ChatGPT 桌面端用的 /v1/responses 端点）。
- * 纯函数，无网络、无副作用。无状态：previous_response_id 不支持（stateless 网关）。
- */
-let seq = 0;
-export function newResponseId() {
-  return `resp_${Date.now().toString(36)}${(seq++).toString(36)}`;
-}
-
-// 上游块可能是 Buffer/Uint8Array：String() 会变成 "100,97,..." 数字串，必须解码
-const _decoder = new TextDecoder();
-export function chunkToString(c) {
-  if (c == null) return "";
-  if (typeof c === "string") return c;
-  if (typeof Buffer !== "undefined" && Buffer.isBuffer(c)) return c.toString("utf8");
-  if (c instanceof Uint8Array) return _decoder.decode(c);
-  return String(c);
-}
-
-// content 数组 → 文本（tool_calls 摘要之外的纯文本部分，历史行为）
-function inputTextOf(content) {
-  if (typeof content === "string") return content;
-  if (Array.isArray(content)) {
-    return content.filter((p) => p && (p.type === "input_text" || p.type === "text")).map((p) => p.text || "").join("");
-  }
-  return String(content ?? "");
-}
-
-// content 数组里的图片（responses 规范 input_image part）→ chat 规范 image_url part。
-// 此前被 inputTextOf 静默丢掉 → 下游根本拿不到图，模型"看不到"（2026-09-22 实测）。
-function inputImagesOf(content) {
-  if (!Array.isArray(content)) return [];
-  const out = [];
-  for (const p of content) {
-    if (!p || typeof p !== "object") continue;
-    if (p.type === "input_image" || p.type === "image_url") {
-      const url = typeof p.image_url === "string" ? p.image_url : p.image_url?.url;
-      if (url) out.push({ type: "image_url", image_url: { url } });
-    }
-  }
-  return out;
-}
-
-// POST /v1/responses body → chat completions body（直接喂现有 pipeline）
-export function responsesToChatBody(req = {}) {
-  const model = String(req.model || "").trim();
-  if (!model) throw new Error("responses: 缺少 model");
-  const messages = [];
-  if (req.instructions) messages.push({ role: "system", content: String(req.instructions) });
-  const input = req.input;
-  const items = typeof input === "string" ? [{ type: "message", role: "user", content: input }] : Array.isArray(input) ? input : [];
-  // 加密思考往返：input 里的 reasoning item 挂到下一条 assistant 消息（thinking 跨轮必需，不能丢）。
-  // 上游按 item id 查重：同一 reasoning item 重复出现 → 400 "Duplicate item found"（2026-09-22 实测，
-  // 客户端重试/重放历史时会把同一 item 存两份），故按 id 保首个、丢后续——重复项无合法语义，删了不吃亏。
-  let pendingReasoning = [];
-  const seenReasoningIds = new Set();
-  // 同 call_id 的 function_call_output 重复出现 → 上游 400 "Duplicate function_call_output"
-  // （2026-09-23 实测，客户端重试/重放历史时会把同一工具结果存两份），故同样按 call_id 保首个、丢后续。
-  const seenToolResultIds = new Set();
-  for (const it of items) {
-    if (!it || typeof it !== "object") continue;
-    if (it.type === "reasoning") {
-      if (it.id || it.encrypted_content) {
-        const key = String(it.id || "");
-        if (key && seenReasoningIds.has(key)) continue;
-        if (key) seenReasoningIds.add(key);
-        pendingReasoning.push({ id: it.id, encrypted_content: it.encrypted_content, summary: it.summary });
-      }
-      continue;
-    }
-    // responses 规范：message item 的 type 可省（AI SDK/opencode 就不发）→ 有 role 即按 message 处理
-    if (it.type === "message" || (!it.type && it.role)) {
-      const imgs = inputImagesOf(it.content);
-      const msg = { role: it.role || "user", content: inputTextOf(it.content) };
-      // 有图时 content 用 chat 多模态数组形状（text + image_url），下游 responses 通道才能带图上游
-      if (imgs.length) msg.content = [...(msg.content ? [{ type: "text", text: msg.content }] : []), ...imgs];
-      if (pendingReasoning.length && msg.role === "assistant") { msg.reasoning_items = pendingReasoning; pendingReasoning = []; }
-      messages.push(msg);
-    } else if (it.type === "function_call") {
-      const msg = {
-        role: "assistant", content: "",
-        tool_calls: [{ id: it.call_id || it.id || "", type: "function", function: { name: it.name || "", arguments: it.arguments || "" } }],
-      };
-      if (pendingReasoning.length) { msg.reasoning_items = pendingReasoning; pendingReasoning = []; }
-      messages.push(msg);
-    } else if (it.type === "function_call_output") {
-      const _tkey = String(it.call_id || "");
-      if (_tkey) { if (seenToolResultIds.has(_tkey)) continue; seenToolResultIds.add(_tkey); }
-      messages.push({ role: "tool", tool_call_id: it.call_id || "", content: typeof it.output === "string" ? it.output : JSON.stringify(it.output ?? "") });
-    }
-  }
-  // 空会话直接抛错，**绝不伪造一句 "hi" 打上游**：那会把假提问写进用户上下文并白烧一发额度。
-  // 口径与 anthropic/request-body.js:83-84 一致（同一仓库同一类场景不该有两套策略）。
-  // 触发面比"空 input"更广：input 里全是未知 item 类型（如 `{type:"unknown_item"}`）时也走这里
-  // —— 静默发一句 hi 会让客户端拿到一个与它请求毫无关系的回答。
-  if (!messages.length) throw new Error("responses: 缺少用户消息");
-  const body = { model, messages, stream: Boolean(req.stream) };
-  if (Array.isArray(req.tools) && req.tools.length) {
-    body.tools = req.tools.map((t) => (t?.type === "function"
-      ? { type: "function", function: { name: t.name, description: t.description || "", parameters: t.parameters || {} } }
-      : t));
-  }
-  if (req.tool_choice) {
-    const tc = req.tool_choice;
-    body.tool_choice = tc?.type === "function" ? { type: "function", function: { name: tc.name } } : tc;
-  }
-  if (req.max_output_tokens != null) body.max_tokens = Number(req.max_output_tokens);
-  if (req.temperature != null) body.temperature = req.temperature;
-  if (req.top_p != null) body.top_p = req.top_p;
-  if (req.parallel_tool_calls != null) body.parallel_tool_calls = req.parallel_tool_calls;
-  // 思考档位透传：Responses 口径是 reasoning.effort，chat 口径用 reasoning_effort（上游通用）。
-  // 只收非空 trimmed 字符串（空串不算声明，免得顶掉下游默认值）——见 change fix-thinking-effort-passthrough。
-  const eff = req.reasoning && typeof req.reasoning === "object" ? req.reasoning.effort : undefined;
-  if (typeof eff === "string" && eff.trim()) body.reasoning_effort = eff.trim();
-  return body;
-}
-
-// chat 口径 usage → Responses 口径（codex 解 ResponseCompleted 硬要 input_tokens/output_tokens，缺则整轮作废）
-export function toResponsesUsage(u = {}) {
-  const num = (v) => (Number.isFinite(Number(v)) ? Number(v) : 0);
-  const src = u && typeof u === "object" ? u : {};
-  const pt = num(src.prompt_tokens ?? src.input_tokens);
-  const ct = num(src.completion_tokens ?? src.output_tokens);
-  return {
-    input_tokens: pt,
-    input_tokens_details: { cached_tokens: num(src.input_tokens_details?.cached_tokens ?? src.prompt_tokens_details?.cached_tokens) },
-    output_tokens: ct,
-    output_tokens_details: { reasoning_tokens: num(src.output_tokens_details?.reasoning_tokens ?? src.completion_tokens_details?.reasoning_tokens) },
-    total_tokens: num(src.total_tokens ?? pt + ct),
-  };
-}
-
-function messageToOutputItems(message = {}) {
-  const out = [];
-  const text = typeof message.content === "string" ? message.content : "";
-  if (text) {
-    out.push({ type: "message", id: `msg_${newResponseId()}`, status: "completed", role: "assistant", content: [{ type: "output_text", text, annotations: [] }] });
-  }
-  for (const tc of message.tool_calls || []) {
-    out.push({ type: "function_call", id: `fc_${newResponseId()}`, call_id: tc.id || "", name: tc.function?.name || "", arguments: tc.function?.arguments || "", status: "completed" });
-  }
-  return out;
-}
-
-// chat JSON → Response 对象（非流式）
-export function chatJsonToResponse(chatJson = {}, model = "") {
-  const choice = chatJson.choices?.[0] || {};
-  const output = messageToOutputItems(choice.message || {});
-  return {
-    id: chatJson.id && String(chatJson.id).startsWith("resp_") ? chatJson.id : newResponseId(),
-    object: "response",
-    created_at: chatJson.created || Math.floor(Date.now() / 1000),
-    status: choice.finish_reason === "length" ? "incomplete" : "completed",
-    model: model || chatJson.model || "",
-    output,
-    usage: toResponsesUsage(chatJson.usage),
-  };
-}
-
-// chat SSE chunk 流 → responses SSE 事件流（逐块翻译，tool_calls 按 index 累积）
-export function createChunkTranslator(model = "") {
-  const id = newResponseId();
-  const createdAt = Math.floor(Date.now() / 1000);
-  let buf = "";
-  let textItemOpen = false;
-  let textLen = 0;
-  let lastFinish = "stop";
-  let lastUsage = null;
-  const tools = new Map(); // index → {id, name, args, announced, outputIndex}
-  // output_index 按 item 出现顺序动态分配（reasoning 可能先于 message/tool 出现）
-  let nextIdx = 0;
-  let textIdx = null;
-  // 加密思考（thinking 跨轮）：sse.js 首帧带 x_reasoning_item（含加密态），translator 据此建 reasoning item
-  let reasoningOpen = false;
-  let reasoningId = null;
-  let reasoningEncrypted = null;
-  let reasoningIdx = null;
-  let reasoningText = "";
-  const ev = (type, extra = {}) => ({ type, ...extra });
-
-  function begin() {
-    return [ev("response.created", { response: { id, object: "response", created_at: createdAt, status: "in_progress", model, output: [] } })];
-  }
-
-  function ensureTextItem() {
-    if (textItemOpen) return [];
-    textItemOpen = true;
-    textIdx = nextIdx++;
-    const item = { type: "message", id: `msg_${id}`, status: "in_progress", role: "assistant", content: [{ type: "output_text", text: "", annotations: [] }] };
-    return [ev("response.output_item.added", { output_index: textIdx, item }), ev("response.content_part.added", { item_id: item.id, output_index: textIdx, content_index: 0, part: { type: "output_text", text: "", annotations: [] } })];
-  }
-
-  let fullText = "";
-  function pushText(delta) {
-    if (!delta) return [];
-    const out = ensureTextItem();
-    textLen += delta.length;
-    fullText += delta;
-    out.push(ev("response.output_text.delta", { item_id: `msg_${id}`, output_index: textIdx, content_index: 0, delta }));
-    return out;
-  }
-
-  function pushReasoning(delta, meta) {
-    const out = ensureReasoningItem(meta);
-    reasoningText += delta;
-    if (delta) out.push(ev("response.reasoning_summary_text.delta", { item_id: reasoningId, output_index: reasoningIdx, summary_index: 0, delta }));
-    return out;
-  }
-
-  function ensureReasoningItem(meta) {
-    if (reasoningOpen) {
-      // 补丁：加密态可能晚到（sse.js 在收尾帧补发无文本的 x_reasoning_item）
-      if (!reasoningEncrypted && typeof meta?.encrypted_content === "string") reasoningEncrypted = meta.encrypted_content;
-      return [];
-    }
-    reasoningOpen = true;
-    reasoningId = meta?.id || `rs_${id}`;
-    reasoningEncrypted = typeof meta?.encrypted_content === "string" ? meta.encrypted_content : null;
-    reasoningIdx = nextIdx++;
-    const item = { type: "reasoning", id: reasoningId, summary: [], ...(reasoningEncrypted ? { encrypted_content: reasoningEncrypted } : {}) };
-    return [
-      ev("response.output_item.added", { output_index: reasoningIdx, item }),
-      ev("response.reasoning_summary_part.added", { item_id: reasoningId, output_index: reasoningIdx, summary_index: 0, part: { type: "summary_text", text: "" } }),
-    ];
-  }
-
-  function pushTool(tc) {
-    const idx = Number(tc.index ?? 0);
-    let t = tools.get(idx);
-    if (!t) { t = { id: "", name: "", args: "", announced: false, outputIndex: null }; tools.set(idx, t); }
-    if (tc.id) t.id = tc.id;
-    if (tc.function?.name) t.name += tc.function.name;
-    const frag = tc.function?.arguments || "";
-    const out = [];
-    if (!t.announced && t.id && t.name) {
-      t.announced = true;
-      t.outputIndex = nextIdx++;
-      out.push(ev("response.output_item.added", { output_index: t.outputIndex, item: { type: "function_call", id: `fc_${id}_${idx}`, call_id: t.id, name: t.name, arguments: "" } }));
-    }
-    if (frag) {
-      if (!t.announced) { t.args += frag; return out; } // id/name 未到先攒着
-      t.args += frag;
-      out.push(ev("response.function_call_arguments.delta", { item_id: `fc_${id}_${idx}`, output_index: t.outputIndex, delta: frag }));
-    }
-    return out;
-  }
-
-  const dbg = { chunks: 0, textChars: 0, toolDeltas: 0, skippedLines: 0, jsonFails: 0, reasoningChars: 0 };
-  function push(data) {
-    buf += chunkToString(data);
-    const out = [];
-    let nl;
-    while ((nl = buf.indexOf("\n")) >= 0) {
-      const line = buf.slice(0, nl).trim();
-      buf = buf.slice(nl + 1);
-      if (!line || line.startsWith(":")) continue;
-      if (!line.startsWith("data:")) { dbg.skippedLines++; continue; }
-      const payload = line.slice(5).trim();
-      if (payload === "[DONE]") continue;
-      let chunk;
-      try { chunk = JSON.parse(payload); } catch { dbg.jsonFails++; continue; }
-      dbg.chunks++;
-      const delta = chunk.choices?.[0]?.delta || {};
-      const fr = chunk.choices?.[0]?.finish_reason;
-      if (fr) lastFinish = fr;
-      if (chunk.usage) lastUsage = chunk.usage;
-      if (typeof delta.content === "string" && delta.content) { dbg.textChars += delta.content.length; out.push(...pushText(delta.content)); }
-      for (const tc of delta.tool_calls || []) { dbg.toolDeltas++; out.push(...pushTool(tc)); }
-      const rc = typeof delta.reasoning_content === "string" ? delta.reasoning_content : "";
-      if (rc || chunk.x_reasoning_item) { if (rc) dbg.reasoningChars += rc.length; out.push(...pushReasoning(rc, chunk.x_reasoning_item)); }
-    }
-    return out;
-  }
-
-  function end({ finish = "stop", usage = null } = {}) {
-    const out = [];
-    if (reasoningOpen) {
-      out.push(ev("response.reasoning_summary_text.done", { item_id: reasoningId, output_index: reasoningIdx, summary_index: 0, text: reasoningText }));
-      out.push(ev("response.reasoning_summary_part.done", { item_id: reasoningId, output_index: reasoningIdx, summary_index: 0, part: { type: "summary_text", text: reasoningText } }));
-      out.push(ev("response.output_item.done", { output_index: reasoningIdx, item: { type: "reasoning", id: reasoningId, summary: [{ type: "summary_text", text: reasoningText }], ...(reasoningEncrypted ? { encrypted_content: reasoningEncrypted } : {}) } }));
-    }
-    if (textItemOpen) {
-      out.push(ev("response.output_text.done", { item_id: `msg_${id}`, output_index: textIdx, content_index: 0, text: fullText }));
-      out.push(ev("response.content_part.done", { item_id: `msg_${id}`, output_index: textIdx, content_index: 0, part: { type: "output_text", text: fullText, annotations: [] } }));
-      out.push(ev("response.output_item.done", { output_index: textIdx, item: { type: "message", id: `msg_${id}`, status: "completed", role: "assistant", content: [{ type: "output_text", text: fullText, annotations: [] }] } }));
-    }
-    for (const [idx, t] of tools) {
-      if (!t.id && !t.name && !t.args) continue;
-      if (!t.announced) {
-        t.announced = true;
-        t.outputIndex = nextIdx++;
-        out.push(ev("response.output_item.added", { output_index: t.outputIndex, item: { type: "function_call", id: `fc_${id}_${idx}`, call_id: t.id, name: t.name, arguments: "" } }));
-      }
-      if (t.args) out.push(ev("response.function_call_arguments.done", { item_id: `fc_${id}_${idx}`, output_index: t.outputIndex, arguments: t.args }));
-      out.push(ev("response.output_item.done", { output_index: t.outputIndex, item: { type: "function_call", id: `fc_${id}_${idx}`, call_id: t.id, name: t.name, arguments: t.args, status: "completed" } }));
-    }
-    void textLen;
-    out.push(ev("response.completed", { response: { id, object: "response", created_at: createdAt, status: finish === "length" ? "incomplete" : "completed", model, output: [], usage: toResponsesUsage(usage) } }));
-    return out;
-  }
-
-  function getFinal() {
-    return { finish: lastFinish, usage: lastUsage };
-  }
-
-  function stats() {
-    return { ...dbg, outEvents: null, textItemOpen, toolCalls: tools.size };
-  }
-
-  return { id, begin, push, end, getFinal, stats };
-}
+/**
+ * Responses ⇄ Chat 翻译层（给 Codex/ChatGPT 桌面端用的 /v1/responses 端点）。
+ * 纯函数，无网络、无副作用。无状态：previous_response_id 不支持（stateless 网关）。
+ */
+let seq = 0;
+export function newResponseId() {
+  return `resp_${Date.now().toString(36)}${(seq++).toString(36)}`;
+}
+
+// 上游块可能是 Buffer/Uint8Array：String() 会变成 "100,97,..." 数字串，必须解码
+const _decoder = new TextDecoder();
+export function chunkToString(c) {
+  if (c == null) return "";
+  if (typeof c === "string") return c;
+  if (typeof Buffer !== "undefined" && Buffer.isBuffer(c)) return c.toString("utf8");
+  if (c instanceof Uint8Array) return _decoder.decode(c);
+  return String(c);
+}
+
+// content 数组 → 文本（tool_calls 摘要之外的纯文本部分，历史行为）
+function inputTextOf(content) {
+  if (typeof content === "string") return content;
+  if (Array.isArray(content)) {
+    return content.filter((p) => p && (p.type === "input_text" || p.type === "text")).map((p) => p.text || "").join("");
+  }
+  return String(content ?? "");
+}
+
+// content 数组里的图片（responses 规范 input_image part）→ chat 规范 image_url part。
+// 此前被 inputTextOf 静默丢掉 → 下游根本拿不到图，模型"看不到"（2026-09-22 实测）。
+function inputImagesOf(content) {
+  if (!Array.isArray(content)) return [];
+  const out = [];
+  for (const p of content) {
+    if (!p || typeof p !== "object") continue;
+    if (p.type === "input_image" || p.type === "image_url") {
+      const url = typeof p.image_url === "string" ? p.image_url : p.image_url?.url;
+      if (url) out.push({ type: "image_url", image_url: { url } });
+    }
+  }
+  return out;
+}
+
+// POST /v1/responses body → chat completions body（直接喂现有 pipeline）
+export function responsesToChatBody(req = {}) {
+  const model = String(req.model || "").trim();
+  if (!model) throw new Error("responses: 缺少 model");
+  const messages = [];
+  if (req.instructions) messages.push({ role: "system", content: String(req.instructions) });
+  const input = req.input;
+  const items = typeof input === "string" ? [{ type: "message", role: "user", content: input }] : Array.isArray(input) ? input : [];
+  // 加密思考往返：input 里的 reasoning item 挂到下一条 assistant 消息（thinking 跨轮必需，不能丢）。
+  // 上游按 item id 查重：同一 reasoning item 重复出现 → 400 "Duplicate item found"（2026-09-22 实测，
+  // 客户端重试/重放历史时会把同一 item 存两份），故按 id 保首个、丢后续——重复项无合法语义，删了不吃亏。
+  let pendingReasoning = [];
+  const seenReasoningIds = new Set();
+  // 同 call_id 的 function_call_output 重复出现 → 上游 400 "Duplicate function_call_output"
+  // （2026-09-23 实测，客户端重试/重放历史时会把同一工具结果存两份），故同样按 call_id 保首个、丢后续。
+  const seenToolResultIds = new Set();
+  for (const it of items) {
+    if (!it || typeof it !== "object") continue;
+    if (it.type === "reasoning") {
+      if (it.id || it.encrypted_content) {
+        const key = String(it.id || "");
+        if (key && seenReasoningIds.has(key)) continue;
+        if (key) seenReasoningIds.add(key);
+        pendingReasoning.push({ id: it.id, encrypted_content: it.encrypted_content, summary: it.summary });
+      }
+      continue;
+    }
+    // responses 规范：message item 的 type 可省（AI SDK/opencode 就不发）→ 有 role 即按 message 处理
+    if (it.type === "message" || (!it.type && it.role)) {
+      const imgs = inputImagesOf(it.content);
+      const msg = { role: it.role || "user", content: inputTextOf(it.content) };
+      // 有图时 content 用 chat 多模态数组形状（text + image_url），下游 responses 通道才能带图上游
+      if (imgs.length) msg.content = [...(msg.content ? [{ type: "text", text: msg.content }] : []), ...imgs];
+      if (pendingReasoning.length && msg.role === "assistant") { msg.reasoning_items = pendingReasoning; pendingReasoning = []; }
+      messages.push(msg);
+    } else if (it.type === "function_call") {
+      const msg = {
+        role: "assistant", content: "",
+        tool_calls: [{ id: it.call_id || it.id || "", type: "function", function: { name: it.name || "", arguments: it.arguments || "" } }],
+      };
+      if (pendingReasoning.length) { msg.reasoning_items = pendingReasoning; pendingReasoning = []; }
+      messages.push(msg);
+    } else if (it.type === "function_call_output") {
+      const _tkey = String(it.call_id || "");
+      if (_tkey) { if (seenToolResultIds.has(_tkey)) continue; seenToolResultIds.add(_tkey); }
+      messages.push({ role: "tool", tool_call_id: it.call_id || "", content: typeof it.output === "string" ? it.output : JSON.stringify(it.output ?? "") });
+    }
+  }
+  // 空会话直接抛错，**绝不伪造一句 "hi" 打上游**：那会把假提问写进用户上下文并白烧一发额度。
+  // 口径与 anthropic/request-body.js:83-84 一致（同一仓库同一类场景不该有两套策略）。
+  // 触发面比"空 input"更广：input 里全是未知 item 类型（如 `{type:"unknown_item"}`）时也走这里
+  // —— 静默发一句 hi 会让客户端拿到一个与它请求毫无关系的回答。
+  if (!messages.length) throw new Error("responses: 缺少用户消息");
+  const body = { model, messages, stream: Boolean(req.stream) };
+  if (Array.isArray(req.tools) && req.tools.length) {
+    body.tools = req.tools.map((t) => (t?.type === "function"
+      ? { type: "function", function: { name: t.name, description: t.description || "", parameters: t.parameters || {} } }
+      : t));
+  }
+  if (req.tool_choice) {
+    const tc = req.tool_choice;
+    body.tool_choice = tc?.type === "function" ? { type: "function", function: { name: tc.name } } : tc;
+  }
+  if (req.max_output_tokens != null) body.max_tokens = Number(req.max_output_tokens);
+  if (req.temperature != null) body.temperature = req.temperature;
+  if (req.top_p != null) body.top_p = req.top_p;
+  if (req.parallel_tool_calls != null) body.parallel_tool_calls = req.parallel_tool_calls;
+  // 思考档位透传：Responses 口径是 reasoning.effort，chat 口径用 reasoning_effort（上游通用）。
+  // 只收非空 trimmed 字符串（空串不算声明，免得顶掉下游默认值）——见 change fix-thinking-effort-passthrough。
+  const eff = req.reasoning && typeof req.reasoning === "object" ? req.reasoning.effort : undefined;
+  if (typeof eff === "string" && eff.trim()) body.reasoning_effort = eff.trim();
+  return body;
+}
+
+// chat 口径 usage → Responses 口径（codex 解 ResponseCompleted 硬要 input_tokens/output_tokens，缺则整轮作废）
+export function toResponsesUsage(u = {}) {
+  const num = (v) => (Number.isFinite(Number(v)) ? Number(v) : 0);
+  const src = u && typeof u === "object" ? u : {};
+  const pt = num(src.prompt_tokens ?? src.input_tokens);
+  const ct = num(src.completion_tokens ?? src.output_tokens);
+  return {
+    input_tokens: pt,
+    input_tokens_details: { cached_tokens: num(src.input_tokens_details?.cached_tokens ?? src.prompt_tokens_details?.cached_tokens) },
+    output_tokens: ct,
+    output_tokens_details: { reasoning_tokens: num(src.output_tokens_details?.reasoning_tokens ?? src.completion_tokens_details?.reasoning_tokens) },
+    total_tokens: num(src.total_tokens ?? pt + ct),
+  };
+}
+
+function messageToOutputItems(message = {}) {
+  const out = [];
+  const text = typeof message.content === "string" ? message.content : "";
+  if (text) {
+    out.push({ type: "message", id: `msg_${newResponseId()}`, status: "completed", role: "assistant", content: [{ type: "output_text", text, annotations: [] }] });
+  }
+  for (const tc of message.tool_calls || []) {
+    out.push({ type: "function_call", id: `fc_${newResponseId()}`, call_id: tc.id || "", name: tc.function?.name || "", arguments: tc.function?.arguments || "", status: "completed" });
+  }
+  return out;
+}
+
+// chat JSON → Response 对象（非流式）
+export function chatJsonToResponse(chatJson = {}, model = "") {
+  const choice = chatJson.choices?.[0] || {};
+  const output = messageToOutputItems(choice.message || {});
+  return {
+    id: chatJson.id && String(chatJson.id).startsWith("resp_") ? chatJson.id : newResponseId(),
+    object: "response",
+    created_at: chatJson.created || Math.floor(Date.now() / 1000),
+    status: choice.finish_reason === "length" ? "incomplete" : "completed",
+    model: model || chatJson.model || "",
+    output,
+    usage: toResponsesUsage(chatJson.usage),
+  };
+}
+
+// chat SSE chunk 流 → responses SSE 事件流（逐块翻译，tool_calls 按 index 累积）
+export function createChunkTranslator(model = "") {
+  const id = newResponseId();
+  const createdAt = Math.floor(Date.now() / 1000);
+  let buf = "";
+  let textItemOpen = false;
+  let textLen = 0;
+  let lastFinish = "stop";
+  let lastUsage = null;
+  const tools = new Map(); // index → {id, name, args, announced, outputIndex}
+  // output_index 按 item 出现顺序动态分配（reasoning 可能先于 message/tool 出现）
+  let nextIdx = 0;
+  let textIdx = null;
+  // 加密思考（thinking 跨轮）：sse.js 首帧带 x_reasoning_item（含加密态），translator 据此建 reasoning item
+  let reasoningOpen = false;
+  let reasoningId = null;
+  let reasoningEncrypted = null;
+  let reasoningIdx = null;
+  let reasoningText = "";
+  const ev = (type, extra = {}) => ({ type, ...extra });
+
+  function begin() {
+    return [ev("response.created", { response: { id, object: "response", created_at: createdAt, status: "in_progress", model, output: [] } })];
+  }
+
+  function ensureTextItem() {
+    if (textItemOpen) return [];
+    textItemOpen = true;
+    textIdx = nextIdx++;
+    const item = { type: "message", id: `msg_${id}`, status: "in_progress", role: "assistant", content: [{ type: "output_text", text: "", annotations: [] }] };
+    return [ev("response.output_item.added", { output_index: textIdx, item }), ev("response.content_part.added", { item_id: item.id, output_index: textIdx, content_index: 0, part: { type: "output_text", text: "", annotations: [] } })];
+  }
+
+  let fullText = "";
+  function pushText(delta) {
+    if (!delta) return [];
+    const out = ensureTextItem();
+    textLen += delta.length;
+    fullText += delta;
+    out.push(ev("response.output_text.delta", { item_id: `msg_${id}`, output_index: textIdx, content_index: 0, delta }));
+    return out;
+  }
+
+  function pushReasoning(delta, meta) {
+    const out = ensureReasoningItem(meta);
+    reasoningText += delta;
+    if (delta) out.push(ev("response.reasoning_summary_text.delta", { item_id: reasoningId, output_index: reasoningIdx, summary_index: 0, delta }));
+    return out;
+  }
+
+  function ensureReasoningItem(meta) {
+    if (reasoningOpen) {
+      // 补丁：加密态可能晚到（sse.js 在收尾帧补发无文本的 x_reasoning_item）
+      if (!reasoningEncrypted && typeof meta?.encrypted_content === "string") reasoningEncrypted = meta.encrypted_content;
+      return [];
+    }
+    reasoningOpen = true;
+    reasoningId = meta?.id || `rs_${id}`;
+    reasoningEncrypted = typeof meta?.encrypted_content === "string" ? meta.encrypted_content : null;
+    reasoningIdx = nextIdx++;
+    const item = { type: "reasoning", id: reasoningId, summary: [], ...(reasoningEncrypted ? { encrypted_content: reasoningEncrypted } : {}) };
+    return [
+      ev("response.output_item.added", { output_index: reasoningIdx, item }),
+      ev("response.reasoning_summary_part.added", { item_id: reasoningId, output_index: reasoningIdx, summary_index: 0, part: { type: "summary_text", text: "" } }),
+    ];
+  }
+
+  function pushTool(tc) {
+    const idx = Number(tc.index ?? 0);
+    let t = tools.get(idx);
+    if (!t) { t = { id: "", name: "", args: "", announced: false, outputIndex: null }; tools.set(idx, t); }
+    if (tc.id) t.id = tc.id;
+    if (tc.function?.name) t.name += tc.function.name;
+    const frag = tc.function?.arguments || "";
+    const out = [];
+    if (!t.announced && t.id && t.name) {
+      t.announced = true;
+      t.outputIndex = nextIdx++;
+      out.push(ev("response.output_item.added", { output_index: t.outputIndex, item: { type: "function_call", id: `fc_${id}_${idx}`, call_id: t.id, name: t.name, arguments: "" } }));
+    }
+    if (frag) {
+      if (!t.announced) { t.args += frag; return out; } // id/name 未到先攒着
+      t.args += frag;
+      out.push(ev("response.function_call_arguments.delta", { item_id: `fc_${id}_${idx}`, output_index: t.outputIndex, delta: frag }));
+    }
+    return out;
+  }
+
+  const dbg = { chunks: 0, textChars: 0, toolDeltas: 0, skippedLines: 0, jsonFails: 0, reasoningChars: 0 };
+  function push(data) {
+    buf += chunkToString(data);
+    const out = [];
+    let nl;
+    while ((nl = buf.indexOf("\n")) >= 0) {
+      const line = buf.slice(0, nl).trim();
+      buf = buf.slice(nl + 1);
+      if (!line || line.startsWith(":")) continue;
+      if (!line.startsWith("data:")) { dbg.skippedLines++; continue; }
+      const payload = line.slice(5).trim();
+      if (payload === "[DONE]") continue;
+      let chunk;
+      try { chunk = JSON.parse(payload); } catch { dbg.jsonFails++; continue; }
+      dbg.chunks++;
+      const delta = chunk.choices?.[0]?.delta || {};
+      const fr = chunk.choices?.[0]?.finish_reason;
+      if (fr) lastFinish = fr;
+      if (chunk.usage) lastUsage = chunk.usage;
+      if (typeof delta.content === "string" && delta.content) { dbg.textChars += delta.content.length; out.push(...pushText(delta.content)); }
+      for (const tc of delta.tool_calls || []) { dbg.toolDeltas++; out.push(...pushTool(tc)); }
+      const rc = typeof delta.reasoning_content === "string" ? delta.reasoning_content : "";
+      if (rc || chunk.x_reasoning_item) { if (rc) dbg.reasoningChars += rc.length; out.push(...pushReasoning(rc, chunk.x_reasoning_item)); }
+    }
+    return out;
+  }
+
+  function end({ finish = "stop", usage = null } = {}) {
+    const out = [];
+    if (reasoningOpen) {
+      out.push(ev("response.reasoning_summary_text.done", { item_id: reasoningId, output_index: reasoningIdx, summary_index: 0, text: reasoningText }));
+      out.push(ev("response.reasoning_summary_part.done", { item_id: reasoningId, output_index: reasoningIdx, summary_index: 0, part: { type: "summary_text", text: reasoningText } }));
+      out.push(ev("response.output_item.done", { output_index: reasoningIdx, item: { type: "reasoning", id: reasoningId, summary: [{ type: "summary_text", text: reasoningText }], ...(reasoningEncrypted ? { encrypted_content: reasoningEncrypted } : {}) } }));
+    }
+    if (textItemOpen) {
+      out.push(ev("response.output_text.done", { item_id: `msg_${id}`, output_index: textIdx, content_index: 0, text: fullText }));
+      out.push(ev("response.content_part.done", { item_id: `msg_${id}`, output_index: textIdx, content_index: 0, part: { type: "output_text", text: fullText, annotations: [] } }));
+      out.push(ev("response.output_item.done", { output_index: textIdx, item: { type: "message", id: `msg_${id}`, status: "completed", role: "assistant", content: [{ type: "output_text", text: fullText, annotations: [] }] } }));
+    }
+    for (const [idx, t] of tools) {
+      if (!t.id && !t.name && !t.args) continue;
+      if (!t.announced) {
+        t.announced = true;
+        t.outputIndex = nextIdx++;
+        out.push(ev("response.output_item.added", { output_index: t.outputIndex, item: { type: "function_call", id: `fc_${id}_${idx}`, call_id: t.id, name: t.name, arguments: "" } }));
+      }
+      if (t.args) out.push(ev("response.function_call_arguments.done", { item_id: `fc_${id}_${idx}`, output_index: t.outputIndex, arguments: t.args }));
+      out.push(ev("response.output_item.done", { output_index: t.outputIndex, item: { type: "function_call", id: `fc_${id}_${idx}`, call_id: t.id, name: t.name, arguments: t.args, status: "completed" } }));
+    }
+    void textLen;
+    out.push(ev("response.completed", { response: { id, object: "response", created_at: createdAt, status: finish === "length" ? "incomplete" : "completed", model, output: [], usage: toResponsesUsage(usage) } }));
+    return out;
+  }
+
+  function getFinal() {
+    return { finish: lastFinish, usage: lastUsage };
+  }
+
+  function stats() {
+    return { ...dbg, outEvents: null, textItemOpen, toolCalls: tools.size };
+  }
+
+  return { id, begin, push, end, getFinal, stats };
+}