数据可视化上线

elastic/es_connect.py

@@ -8,6 +8,8 @@ from .documents import AchievementDocument, UserDocument, GlobalDocument
 from .indexes import ACHIEVEMENT_INDEX_NAME, USER_INDEX_NAME, GLOBAL_INDEX_NAME
 import hashlib
 import time
+from datetime import datetime, timezone, timedelta
+import threading
 
 # 使用Django的ES连接配置
 connections.create_connection(hosts=['localhost:9200'])
@@ -132,11 +134,11 @@ def insert_data(data):
         bool: 插入成功返回True，失败返回False
     """
     try:
-        # 使用Django-elasticsearch-dsl的方式插入数据
         achievement = AchievementDocument(
             writer_id=data.get('writer_id', ''),
             data=data.get('data', ''),
-            image=data.get('image', '')
+            image=data.get('image', ''),
+            time=datetime.now(timezone.utc).isoformat()
         )
         achievement.save()
         print(f"文档写入成功，内容: {data}")
@@ -303,6 +305,68 @@ def search_by_any_field(keyword):
         print(f"模糊搜索失败: {str(e)}")
         return []
 
+ANALYTICS_CACHE = {"data": None, "ts": 0}
+
+def _compute_hist(range_gte: str, interval: str, fmt: str):
+    from elasticsearch_dsl import Search
+    s = AchievementDocument.search()
+    s = s.filter('range', time={'gte': range_gte, 'lte': 'now'})
+    s = s.extra(size=0)
+    s.aggs.bucket('b', 'date_histogram', field='time', calendar_interval=interval, format=fmt, min_doc_count=0)
+    resp = s.execute()
+    buckets = getattr(resp.aggs, 'b').buckets
+    return [{"label": b.key_as_string, "count": b.doc_count} for b in buckets]
+
+def _compute_type_counts(range_gte: str, types: list):
+    counts = []
+    for t in types:
+        s = AchievementDocument.search()
+        s = s.filter('range', time={'gte': range_gte, 'lte': 'now'})
+        s = s.query('match_phrase', data=str(t))
+        total = s.count()
+        counts.append({"type": str(t), "count": int(total)})
+    return counts
+
+def compute_analytics():
+    types = get_type_list()
+    days = _compute_hist('now-10d/d', 'day', 'yyyy-MM-dd')
+    weeks = _compute_hist('now-10w/w', 'week', 'yyyy-ww')
+    months = _compute_hist('now-10M/M', 'month', 'yyyy-MM')
+    pie_1m = _compute_type_counts('now-1M/M', types)
+    pie_12m = _compute_type_counts('now-12M/M', types)
+    return {
+        "last_10_days": days[-10:],
+        "last_10_weeks": weeks[-10:],
+        "last_10_months": months[-10:],
+        "type_pie_1m": pie_1m,
+        "type_pie_12m": pie_12m,
+    }
+
+def get_analytics_overview(force: bool = False):
+    now_ts = time.time()
+    if force or ANALYTICS_CACHE["data"] is None or (now_ts - ANALYTICS_CACHE["ts"]) > 3600:
+        ANALYTICS_CACHE["data"] = compute_analytics()
+        ANALYTICS_CACHE["ts"] = now_ts
+    return ANALYTICS_CACHE["data"]
+
+def _seconds_until_hour(h: int):
+    now = datetime.now()
+    tgt = now.replace(hour=h, minute=0, second=0, microsecond=0)
+    if tgt <= now:
+        tgt = tgt + timedelta(days=1)
+    return max(0, int((tgt - now).total_seconds()))
+
+def start_daily_analytics_scheduler():
+    def _run_and_reschedule():
+        try:
+            get_analytics_overview(force=True)
+        except Exception as e:
+            print(f"分析任务失败: {e}")
+        finally:
+            threading.Timer(24 * 3600, _run_and_reschedule).start()
+    delay = _seconds_until_hour(3)
+    threading.Timer(delay, _run_and_reschedule).start()
+
 def write_user_data(user_data):
     """
     写入用户数据到 ES
@@ -375,6 +439,23 @@ def get_all_users():
         print(f"获取所有用户失败: {str(e)}")
         return []
 
+def get_user_by_id(user_id):
+    try:
+        search = UserDocument.search()
+        search = search.query("term", user_id=int(user_id))
+        response = search.execute()
+        if response.hits:
+            hit = response.hits[0]
+            return {
+                "user_id": hit.user_id,
+                "username": hit.username,
+                "permission": hit.permission,
+            }
+        return None
+    except Exception as e:
+        print(f"获取用户数据失败: {str(e)}")
+        return None
+
 def delete_user_by_username(username):
     """
     根据用户名删除用户