update

api/api_v1/endpoints/query.py

@@ -1,4 +1,5 @@
 import pandas as pd
+import numpy as np
 from fastapi import APIRouter, Depends, Request
 from motor.motor_asyncio import AsyncIOMotorDatabase
 
@@ -275,5 +276,69 @@ async def funnel_model(
             _ = date_data.setdefault(key[0].strftime('%Y-%m-%d'), {})
             _[key[1]] = tmp
 
-    resp = {'list': data_list, 'date_data': date_data, 'title': groupby + cond_level}
+    resp = {'list': data_list, 'date_data': date_data, 'title': groupby + cond_level, 'level': cond_level
+            }
     return schemas.Msg(code=0, msg='ok', data=resp)
+
+
+@router.post("/scatter_model_sql")
+async def scatter_model_sql(
+        request: Request,
+        game: str,
+        analysis: BehaviorAnalysis = Depends(BehaviorAnalysis),
+        current_user: schemas.UserDB = Depends(deps.get_current_user)
+) -> schemas.Msg:
+    """分布分析 sql"""
+    await analysis.init()
+    data = analysis.scatter_model_sql()
+    return schemas.Msg(code=0, msg='ok', data=[data])
+
+
+@router.post("/scatter_model")
+async def scatter_model_sql(
+        request: Request,
+        game: str,
+        ckdb: CKDrive = Depends(get_ck_db),
+        db: AsyncIOMotorDatabase = Depends(get_database),
+        analysis: BehaviorAnalysis = Depends(BehaviorAnalysis),
+        current_user: schemas.UserDB = Depends(deps.get_current_user)
+) -> schemas.Msg:
+    """分布分析 模型"""
+    await analysis.init()
+    res = analysis.scatter_model_sql()
+    sql = res['sql']
+    df = await ckdb.query_dataframe(sql)
+    interval_type = res['interval_type']
+    analysis = res['analysis']
+    quota_interval_arr = res['quota_interval_arr']
+    if analysis != 'number_of_days':
+        max_v = int(df['values'].max())
+        min_v = int(df['values'].min())
+        interval = (max_v - min_v) // 10
+        resp = {}
+        if not quota_interval_arr:
+            resp['label'] = [f'[{i},{i + interval})' for i in range(min_v, max_v, interval)]
+            bins = [i for i in range(min_v, max_v + interval, interval)]
+        else:
+            quota_interval_arr = [-float('inf')] + quota_interval_arr+[float('inf')]
+            resp['label'] = []
+            bins = [quota_interval_arr[0]]
+            for i, v in enumerate(quota_interval_arr[1:]):
+                resp['label'].append(f'[{quota_interval_arr[i]},{v})')
+                bins.append(v)
+
+        for key, tmp_df in df.groupby('date'):
+            bins_s = pd.cut(tmp_df['values'], bins=bins,
+                            right=False).value_counts()
+            bins_s.sort_index(inplace=True)
+            total = int(bins_s.sum())
+            resp[key.strftime('%Y-%m-%d')] = {'n': bins_s.to_list(), 'total': total,
+                                              'p': round(bins_s * 100 / total, 2).to_list()}
+        return schemas.Msg(code=0, msg='ok', data=resp)
+
+    if interval_type == 'def' and analysis == 'number_of_days':
+        resp = {}
+        for key, tmp_df in df.groupby('date'):
+            total = int(tmp_df['values'].sum())
+            resp[key.strftime('%Y-%m-%d')] = {'n': total, 'total': total, 'p': 100}
+        return schemas.Msg(code=0, msg='ok', data=resp)

api/api_v1/endpoints/test.py

@@ -20,7 +20,7 @@ async def test(
         current_user: schemas.UserDB = Depends(deps.get_current_user),
         analysis: BehaviorAnalysis = Depends(BehaviorAnalysis)) -> schemas.Msg:
     await analysis.init()
-    query = analysis.funnel_model_sql()
+    query = analysis.scatter_model_sql()
     data = {
         'game': game,
         'analysis': analysis.game,

core/config.py

@@ -11,12 +11,8 @@ class Settings(BaseSettings):
 
     BACKEND_CORS_ORIGINS: List[str] = ['*']
 
-
-
     CASBIN_COLL: str = 'casbin_rule'
 
-
-
     SUPERUSER_EMAIL: str = '15392746632@qq.com'
     SUPERUSER_PASSWORD: str = '123456'
     SUPERUSER_NAME: str = 'root'
@@ -74,6 +70,15 @@ class Settings(BaseSettings):
                     "Nullable(Float)": 'float',
                     "Float": 'float', }
 
+    CK_FUNC = {
+        'sum': lambda x: func.sum(x),
+        'avg': lambda x: func.round(func.avg(x), 2),
+        'median': lambda x: func.median(x),
+        'max': lambda x: func.max(x),
+        'min': lambda x: func.min(x),
+        'distinct_count': lambda x: func.count(func.distinct_count(x)),
+    }
+
     CK_OPERATOR = {
         'int': [{
             'id': 'sum',

models/behavior_analysis.py

@@ -280,3 +280,49 @@ ORDER BY level
                 'date_range': self.date_range,
                 'cond_level': cond_level
                 }
+
+    def scatter_model_sql(self):
+        event = self.events[0]
+        event_name = event['eventName']
+        analysis = event['analysis']
+        e_account_id_col = getattr(self.event_tbl.c, '#account_id')
+        event_name_col = getattr(self.event_tbl.c, '#event_name')
+        event_time_col = getattr(self.event_tbl.c, '#event_time').label('date')
+        event_date_col = settings.TIME_GRAIN_EXPRESSIONS[self.time_particle](event_time_col, self.zone_time)
+
+        quota_interval_arr = event.get('quotaIntervalArr')
+
+        where = [
+            event_date_col >= self.start_date,
+            event_date_col <= self.end_date,
+            event_name_col == event_name
+
+        ]
+        if analysis in ['times', 'number_of_days', 'number_of_hours']:
+
+            qry = sa.select(event_date_col, func.count().label('values')) \
+                .where(and_(*where)) \
+                .group_by(event_date_col, e_account_id_col)
+
+            sql = str(qry.compile(compile_kwargs={"literal_binds": True}))
+            print(sql)
+            return {
+                'sql': sql,
+                'interval_type': event['intervalType'],
+                'analysis': analysis,
+                'quota_interval_arr': quota_interval_arr
+            }
+        elif event.get('quota'):
+            event_attr_col = getattr(self.event_tbl.c, event['quota'])
+
+            qry = sa.select(event_date_col, settings.CK_FUNC[analysis](event_attr_col).label('values')) \
+                .where(and_(*where)) \
+                .group_by(event_date_col, e_account_id_col)
+            sql = str(qry.compile(compile_kwargs={"literal_binds": True}))
+            print(sql)
+            return {
+                'sql': sql,
+                'interval_type': event['intervalType'],
+                'analysis': analysis,
+                'quota_interval_arr': quota_interval_arr
+            }