From 6bc0f1600636a7e474df7eb84766a881c25ec75d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E6=9D=8E=E6=B8=85?= <liqing@xiaotiai.com>
Date: Thu, 20 Aug 2026 17:07:55 +0800
Subject: [PATCH] feat: expose acceptance Link AI aggregate data

---
 acceptance_service.py | 59 +++++++++++++++++++++++++++++++++++++++++++
 app.py                |  8 ++++++
 crawler_client.py     |  9 +++++++
 3 files changed, 76 insertions(+)

diff --git a/acceptance_service.py b/acceptance_service.py
index 807fa39..552326f 100644
--- a/acceptance_service.py
+++ b/acceptance_service.py
@@ -22,6 +22,7 @@ from crawler_client import (
     collect_keyword_sources,
     fetch_content_detail,
     get_wen_resources,
+    match_feishu_notes,
     query_collection,
     schedule_xapi_screenshots,
 )
@@ -965,3 +966,61 @@ def list_acceptance_notes(public_id: str) -> dict[str, Any]:
             item["parsed_data"] = json.loads(item.pop("raw_json")) if item.get("raw_json") else None
             items.append(item)
     return {"items": items, "total": len(items)}
+
+
+def get_acceptance_link_ai_data(public_id: str) -> dict[str, Any]:
+    """返回 Link AI 所需的笔记、AI 采集结果和逐关键词比对结果。"""
+    with connection() as conn:
+        task = conn.execute(
+            "SELECT * FROM acceptance_tasks WHERE public_task_id=?", (public_id,),
+        ).fetchone()
+        if not task:
+            raise TaskError("ACCEPTANCE_TASK_NOT_FOUND", "验收任务不存在", 404)
+        task_data = _task_dict(conn, task)
+        rows = conn.execute(
+            """SELECT a.keyword,a.credential_id,a.attempt_number,a.crawler_task_id,
+                      a.result_json,a.completed_at
+               FROM acceptance_cookie_attempts a
+               WHERE a.task_id=? AND a.status='completed' AND a.result_json IS NOT NULL
+               ORDER BY a.keyword,a.id""",
+            (task["id"],),
+        ).fetchall()
+
+    ai_results = []
+    for row in rows:
+        try:
+            result = json.loads(row["result_json"])
+        except (TypeError, json.JSONDecodeError):
+            continue
+        ai_results.append({
+            "keyword": row["keyword"],
+            "credential_id": row["credential_id"],
+            "attempt_number": row["attempt_number"],
+            "crawler_task_id": row["crawler_task_id"],
+            "completed_at": row["completed_at"],
+            "result": result,
+        })
+
+    comparisons = []
+    for keyword in task_data["keywords"]:
+        try:
+            comparisons.append({
+                "keyword": keyword,
+                "status": "completed",
+                "data": match_feishu_notes(keyword),
+            })
+        except Exception as exc:
+            comparisons.append({
+                "keyword": keyword,
+                "status": "failed",
+                "error": str(exc),
+                "data": None,
+            })
+
+    return {
+        "task": task_data,
+        "notes": list_acceptance_notes(public_id),
+        "ai_results": {"items": ai_results, "total": len(ai_results)},
+        "note_matches": {"items": comparisons, "total": len(comparisons)},
+        "data_scope": "仅包含当前已完成并已入库的数据",
+    }
diff --git a/app.py b/app.py
index 5587ab5..e95448a 100644
--- a/app.py
+++ b/app.py
@@ -18,6 +18,7 @@ from database import init_database
 from acceptance_service import (
     create_acceptance_task,
     get_acceptance_resources,
+    get_acceptance_link_ai_data,
     get_acceptance_task,
     list_acceptance_notes,
     list_acceptance_tasks,
@@ -286,6 +287,13 @@ def create_app() -> Flask:
     def get_project_acceptance_notes(task_id: str):
         return jsonify({"data": list_acceptance_notes(task_id), "request_id": g.request_id})
 
+    @app.route("/api/v1/acceptance-tasks/<task_id>/link-ai-data", methods=["GET"])
+    def get_project_acceptance_link_ai_data(task_id: str):
+        return jsonify({
+            "data": get_acceptance_link_ai_data(task_id),
+            "request_id": g.request_id,
+        })
+
     @app.route("/api/v1/acceptance-tasks/<task_id>/resume", methods=["POST"])
     def resume_project_acceptance_task(task_id: str):
         return jsonify({"data": resume_acceptance_task(task_id), "request_id": g.request_id})
diff --git a/crawler_client.py b/crawler_client.py
index 33a092e..8366a2b 100644
--- a/crawler_client.py
+++ b/crawler_client.py
@@ -338,6 +338,15 @@ def collect_keyword_sources(keyword: str, credential_id: str | None = None) -> d
     )
 
 
+def match_feishu_notes(keyword: str, task_limit: int = 100) -> dict[str, Any]:
+    """调用爬虫服务，比对飞书笔记和问一问采集来源。"""
+    return _post(
+        "/api/v1/xhs/feishu-notes/matches",
+        {"keyword": keyword, "task_limit": task_limit},
+        retry_transient=True,
+    )
+
+
 def resume_collection(crawler_task_id: str) -> dict[str, Any]:
     """沿用原爬虫任务 ID，从 framework 保存的检查点继续采集。"""