refactor: differentiate reference marker handling between stream and non-stream modes

- Stream: strip both and [reference:N] markers to prevent
  leaking partial link metadata during incremental output
- Non-stream: convert citation/reference markers to Markdown links for
  Claude Messages, Gemini generateContent, and OpenAI Chat/Responses
- Remove StripReferenceMarkers option from call sites; behavior is now
  determined automatically by stream vs non-stream context
- Extend JS runtime stripReferenceMarkersText() to also match [citation:N]
- Add tests for streaming marker stripping and non-stream link conversion

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
CJACK
2026-05-03 17:53:49 +08:00
parent c099a6f7bf
commit 112bedb05d
14 changed files with 99 additions and 52 deletions

2
API.md
View File

@@ -42,7 +42,7 @@
- Tool Calling 的解析策略在 Go 与 Node Runtime 间保持一致:推荐模型输出 DSML 外壳 `<|DSML|tool_calls>``<|DSML|invoke name="...">``<|DSML|parameter name="...">`;兼容层也接受 DSML wrapper 别名 `<dsml|tool_calls>``<|tool_calls>``<tool_calls>`、常见 DSML 分隔符漏写形态(如 `<|DSML tool_calls>`)、`DSML` 与工具标签名黏连的常见 typo`<DSMLtool_calls>`),以及旧式 canonical XML `<tool_calls>``<invoke name="...">``<parameter name="...">`。实现上采用窄容错结构扫描:只有 `tool_calls` wrapper 或可修复的缺失 opening wrapper 会进入工具路径,裸 `<invoke>` 不计为已支持语法;流式场景继续执行防泄漏筛分。若参数体本身是合法 JSON 字面量(如 `123``true``null`、数组或对象),会按结构化值输出,不再一律当作字符串;若 CDATA 偶发漏闭合,则会在最终 parse / flush 恢复阶段做窄修复,尽量保住已完整包裹的外层工具调用。
- `Admin API` 将配置与运行时策略分开:`/admin/config*` 管静态配置,`/admin/settings*` 管运行时行为。
- 当上游返回 thinking-only 响应(模型输出了推理链但无可见文本)时,非流式补全会自动重试一次:以多轮对话 follow-up 方式追加 prompt 后缀 `"Previous reply had no visible output. Please regenerate the visible final answer or tool call now."` 并设置 `parent_message_id` 在同一 DeepSeek session 内让模型重新输出;重试最大 1 次。
- 引用标记剥离strip reference markers当前为固定开启的运行时行为所有协议适配层统一生效
- 引用标记处理边界:流式输出默认隐藏 `[citation:N]` / `[reference:N]` 这类上游内部占位符;非流式输出默认把 DeepSeek 搜索引用标记转换为 Markdown 引用链接
---