From f6788756d30e6394b43d5135048ac74eedc09db6 Mon Sep 17 00:00:00 2001
From: Jerry Yan <792602257@qq.com>
Date: Sat, 31 Jan 2026 16:45:48 +0800
Subject: [PATCH] =?UTF-8?q?fix(annotation):=20=E4=BF=AE=E5=A4=8D=E5=88=86?=
 =?UTF-8?q?=E6=AE=B5=E6=A0=87=E6=B3=A8=E6=95=B0=E6=8D=AE=E7=BB=93=E6=9E=84?=
 =?UTF-8?q?=E5=85=BC=E5=AE=B9=E6=80=A7=E9=97=AE=E9=A2=98?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- 添加分段标注合并异常时的日志记录和警告
- 增加分段标注保存时的详细状态日志
- 修复分段数据结构类型检查逻辑，支持dict和list格式统一转换
- 避免SQLAlchemy变更检测失效的原地修改问题
- 添加旧版list结构向新dict结构的数据迁移兼容处理
---
 .../app/module/annotation/service/editor.py   | 30 ++++++++++++++++++-
 1 file changed, 29 insertions(+), 1 deletion(-)

diff --git a/runtime/datamate-python/app/module/annotation/service/editor.py b/runtime/datamate-python/app/module/annotation/service/editor.py
index 878c389..a5f9789 100644
--- a/runtime/datamate-python/app/module/annotation/service/editor.py
+++ b/runtime/datamate-python/app/module/annotation/service/editor.py
@@ -1018,6 +1018,14 @@ class AnnotationEditorService:
                 request.segment_index,
                 annotation_payload,
             )
+            segment_entries = self._extract_segment_annotations(final_payload)
+            if str(request.segment_index) not in segment_entries:
+                logger.warning(
+                    "分段标注合并异常：未找到当前段落 key，project_id=%s file_id=%s segment_index=%s",
+                    project_id,
+                    file_id,
+                    request.segment_index,
+                )
         else:
             # 非分段模式：直接使用传入的 annotation
             annotation_payload["task"] = ls_task_id
@@ -1057,6 +1065,18 @@ class AnnotationEditorService:
                 else:
                     raise HTTPException(status_code=400, detail="未发现标注内容，请确认无标注/不适用后再保存")
 
+        if request.segment_index is not None:
+            segment_entries = self._extract_segment_annotations(final_payload)
+            logger.info(
+                "分段标注保存：project_id=%s file_id=%s segment_index=%s segments=%s total=%s status=%s",
+                project_id,
+                file_id,
+                request.segment_index,
+                len(segment_entries),
+                segment_total,
+                final_status,
+            )
+
         if existing:
             if request.expected_updated_at and existing.updated_at:
                 if existing.updated_at != request.expected_updated_at.replace(tzinfo=None):
@@ -1126,7 +1146,15 @@ class AnnotationEditorService:
         if not base.get(SEGMENTED_KEY):
             base[SEGMENTED_KEY] = True
         segments = base.get(SEGMENTS_KEY)
-        if not isinstance(segments, dict):
+        if isinstance(segments, dict):
+            # 拷贝一份，避免原地修改导致 SQLAlchemy 变更检测失效
+            segments = dict(segments)
+            base[SEGMENTS_KEY] = segments
+        elif isinstance(segments, list):
+            # 兼容旧的 list 结构，归一化为 dict 结构
+            segments = self._extract_segment_annotations(base)
+            base[SEGMENTS_KEY] = segments
+        else:
             segments = {}
             base[SEGMENTS_KEY] = segments