mdbq 3.4.2__py3-none-any.whl → 3.4.3__py3-none-any.whl

mdbq/aggregation/datashow.py

@@ -122,14 +122,14 @@ class DataShow:
         df = pd.concat(__res, ignore_index=True)
         return df
 
-    def pov_city(self, db_name='生意经3', filename='销售地域分布', start_date=None, end_date=None, percentage=None):
+    def pov_city(self, db_name='生意经3', filename='销售地域分布', start_date=None, end_date=None, percent=None):
         """
         生意经  省份城市销售分析
         """
         if not start_date:
             start_date = self.start_date
         if not end_date:
-            end_date = self.end_date
+            end_date = self.today.strftime('%Y-%m-%d')
         pov_set = self.getdata(
             db_name='属性设置3',
             table_name=f'城市等级',
@@ -168,11 +168,27 @@ class DataShow:
         df_pov = df_pov.groupby(['店铺名称', '省份'], as_index=False).agg(
             **{'销售额': ('销售额', np.sum), '退款额': ('退款额', np.sum)})
         df_pov.drop_duplicates(subset='省份', keep='last', inplace=True, ignore_index=True)
+
+        # df_pov2： gmv 的饼图
+        df_pov['gmv销售'] = df_pov.apply(lambda x: x['销售额'] + x['退款额'], axis=1)
+        df_pov.sort_values(['gmv销售'], ascending=[False], ignore_index=True, inplace=True)
+        df_pov2 = df_pov.copy()
+        sales_sum = df_pov2['gmv销售'].sum()
+        df_pov2['省份'] = df_pov2.apply(lambda x: '其他' if (x['gmv销售'] / sales_sum) < percent else x['省份'], axis=1)
+
+        # df_pov3： 销售额的饼图
         df_pov.sort_values(['销售额'], ascending=[False], ignore_index=True, inplace=True)
-        df_pov = df_pov[df_pov['省份'] != '其他']
-        percentages = df_pov['销售额'] / df_pov['销售额'].sum() * 100
-        df_pov1 = df_pov.head(10)
-        data_list = [('销售 top省份', df_pov1['省份'].tolist(), df_pov1['销售额'].tolist(), percentages)]
+        df_pov3 = df_pov.copy()
+        sales_sum = df_pov3['销售额'].sum()
+        df_pov3['省份'] = df_pov3.apply(lambda x: '其他' if (x['销售额'] / sales_sum) < 0.016 else x['省份'], axis=1)
+
+        # df_pov1： 省份 销售额 堆叠柱形图
+        df_pov1 = df_pov.copy()
+        df_pov1 = df_pov1.head(15)
+        pov_sales_sum = df_pov1['销售额'].tolist()
+        pov_refunds = df_pov1['退款额'].tolist()
+        percentages = df_pov1['gmv销售'] / df_pov1['gmv销售'].sum() * 100
+        bar_list = [('省份销售/退款', df_pov1['省份'].tolist(), pov_sales_sum, percentages, pov_refunds)]
 
         # 将城市等级添加到df
         pov_set = pov_set[['城市等级', '城市']]
@@ -180,72 +196,93 @@ class DataShow:
         df_city = pd.merge(df_city, pov_set, left_on=['城市'], right_on=['城市'], how='left')
         df_level = df_city.groupby(['店铺名称', '城市等级'], as_index=False).agg(
             **{'销售额': ('销售额', np.sum), '退款额': ('退款额', np.sum)})
-        city_level_list = [('按城市等级', df_level['城市等级'].tolist(), df_level['销售额'].tolist())]
+        pie_list = [
+            ('按城市等级', df_level['城市等级'].tolist(), df_level['销售额'].tolist()),
+            ('净销售 top省份', df_pov3['省份'].tolist(), df_pov3['销售额'].tolist()),
+            ('GMV top省份', df_pov2['省份'].tolist(), df_pov2['gmv销售'].tolist())
+        ]
+
+        # df_city1： 城市 销售额 堆叠柱形图
         df_city.drop_duplicates(subset='城市', keep='last', inplace=True, ignore_index=True)
+        df_city['gmv销售'] = df_city.apply(lambda x: x['销售额'] + x['退款额'], axis=1)
         df_city.sort_values(['销售额'], ascending=[False], ignore_index=True, inplace=True)
         df_city = df_city[df_city['城市'] != '其他']
-        percentages = df_city['销售额'] / df_city['销售额'].sum() * 100
-        df_city1 = df_city.head(10)
-        data_list += [('销售 top城市', df_city1['城市'].tolist(), df_city1['销售额'].tolist(), percentages)]
-
-        # 退款 top 城市
-        df_city.sort_values(['退款额'], ascending=[False], ignore_index=True, inplace=True)
-        percentages = df_city['退款额'] / df_city['退款额'].sum() * 100
-        df_city2 = df_city.head(10)
-        data_list += [('退款 top城市', df_city2['城市'].tolist(), df_city2['退款额'].tolist(), percentages)]
-
-        t_p1 = [{"type": "pie"}]
-        for i in range(2):
-            t_p1.extend([{"type": "bar"}])  # 折线图类型
+        percentages = df_city['gmv销售'] / df_city['gmv销售'].sum() * 100
+        df_city1 = df_city.head(15)
+        city_sales_sum = df_city1['销售额'].tolist()
+        city_refunds = df_city1['退款额'].tolist()
+        bar_list += [('城市销售/退款', df_city1['城市'].tolist(), city_sales_sum, percentages, city_refunds)]
+
+        t_p1 = []
+        for i in range(3):
+            t_p1.extend([{"type": "pie"}])
         t_p2 = []
         for i in range(3):
-            t_p2.extend([{"type": "bar"}])  # 饼图类型
+            t_p2.extend([{"type": "bar"}])
         specs = [t_p1, t_p2]
         fig = make_subplots(rows=2, cols=3, specs=specs)
-        title, labels, values = city_level_list[0]
-        # 添加饼图
-        fig.add_trace(
-            go.Pie(
-                labels=labels,
-                values=values,
-                name=title,
-                textinfo='label+percent'
-            ),
-            row=1,
-            col=1,
-        )
-        x = 0.14
-        y = 1
-        fig.add_annotation(
-            text=title,
-            x=x,
-            y=y,
-            xref='paper',  # # 相对于整个图表区域
-            yref='paper',
-            showarrow=True,  # 显示箭头
-            align="left",  # 文本对齐方式
-            font=dict(size=14),
-        )
-        row = 1
-        col = 1
-        for item in data_list:
-            title, labels, values, percentages = item
-            bar = go.Bar(
-                x=labels,
-                y=values,
-                name=title,
-                orientation='v',  # 垂直柱形图
-                text=percentages.map('{:.2f}%'.format),  # 设置要显示的文本（百分比）
-                textposition = 'outside',  # 设置文本位置在柱形图外部
-                width=0.55  # 调整柱子最大宽度
-            )
-            fig.add_trace(
-                bar,
-                row=row // 3 + 1,
-                col=col % 3 + 1,
-            )
+
+        row = 0
+        col = 0
+        for i in range(6):
+            if row // 3 == 0:
+                try:
+                    title, labels, values = pie_list[col % 3]
+                except:
+                    row += 1
+                    col += 1
+                    continue
+                # 添加饼图
+                fig.add_trace(
+                    go.Pie(
+                        labels=labels,
+                        values=values,
+                        name=title,
+                        textinfo='label+percent'
+                    ),
+                    row=row//3 + 1,
+                    col=col % 3 + 1,
+                )
+            else:
+                try:
+                    title, labels, values, percentages, refunds = bar_list[col % 3]
+                except:
+                    row += 1
+                    col += 1
+                    continue
+                bar = go.Bar(
+                    x=labels,
+                    y=values,
+                    name='销售额',
+                    orientation='v',  # 垂直柱形图
+                    # text=percentages.map('{:.1f}%'.format),  # 设置要显示的文本（百分比）
+                    # textposition = 'outside',  # 设置文本位置在柱形图外部
+                    width=0.55,  # 调整柱子最大宽度
+                    # marker_color='blue',
+                )
+                fig.add_trace(
+                    bar,
+                    row=row // 3 + 1,
+                    col=col % 3 + 1,
+                )
+                bar = go.Bar(
+                    x=labels,
+                    y=refunds,
+                    name='退款额',
+                    orientation='v',  # 垂直柱形图
+                    text=percentages.map('{:.1f}%'.format),  # 设置要显示的文本（百分比）
+                    textposition='outside',  # 设置文本位置在柱形图外部
+                    width=0.55,  # 调整柱子最大宽度
+                    # marker_color = 'red',
+                )
+                fig.add_trace(
+                    bar,
+                    row=row // 3 + 1,
+                    col=col % 3 + 1,
+                )
+
             x = 0.14 + 0.355 * (row % 3)
-            y = 1 - 0.575 * (row // 3)
+            y = 0.99 - 0.58 * (row // 3)
             fig.add_annotation(
                 text=title,
                 x=x,
@@ -254,7 +291,7 @@ class DataShow:
                 yref='paper',
                 showarrow=True,  # 显示箭头
                 align="left",  # 文本对齐方式
-                font=dict(size=14),
+                font=dict(size=14)
             )
             row += 1
             col += 1
@@ -264,20 +301,21 @@ class DataShow:
             margin=dict(
                 l=100,  # 左边距
                 r=100,
-                t=100,  # 上边距
-                b=100,
+                t=80,  # 上边距
+                b=80,
             ),
             legend=dict(
                 orientation='v',  # 图例方向（'h' 表示水平，'v' 表示垂直）
                 font=dict(
                     size=12  # 图例字体大小
                 )
-            )
+            ),
+            barmode='stack',  # stack(堆叠)、group(并列)、overlay(覆盖)、relative(相对)
         )
         fig.add_annotation(
-            text=f'统计时间周期: {start_date}~{end_date}',
+            text=f'统计时间周期: {start_date}~{end_date}     tips: 饼图剔除了销售<{f"{percent * 100}%"}的数据',
             x=0.5,
-            y=-0.1,
+            y=-0.09,
             xref='paper',  # # 相对于整个图表区域
             yref='paper',
             showarrow=False,  # 显示箭头
@@ -621,8 +659,8 @@ class DataShow:
                 y=values,
                 name=table_name,
                 orientation='v',  # 垂直柱形图
-                text=percentages.map('{:.2f}%'.format),  # 设置要显示的文本（百分比）
-                # textposition = 'outside',  # 设置文本位置在柱形图外部
+                text=percentages.map('{:.1f}%'.format),  # 设置要显示的文本（百分比）
+                textposition = 'outside',  # 设置文本位置在柱形图外部
                 width=0.55  # 调整柱子最大宽度
             )
             row = count // 3 + 1
@@ -753,8 +791,8 @@ class DataShow:
                 y=values,
                 name=table_name,
                 orientation='v',  # 垂直柱形图
-                text=percentages.map('{:.2f}%'.format),  # 设置要显示的文本（百分比）
-                # textposition = 'outside',  # 设置文本位置在柱形图外部
+                text=percentages.map('{:.1f}%'.format),  # 设置要显示的文本（百分比）
+                textposition = 'outside',  # 设置文本位置在柱形图外部
                 width=0.55  # 调整柱子最大宽度
             )
             row = count // 3 + 1
@@ -829,44 +867,44 @@ class DataShow:
 def main():
     ds = DataShow()
 
-    # 店铺流量来源
-    ds.dpll()
-    # 多店聚合推广数据
-    ds.tg(
-        days=15,
-        # start_date='2024-11-01',
-        # end_date='2024-11-30',
-    )
-
-    # 商品人群画像
-    item_id_list = [
-        839148235697,
-    ]
-    for item_id in item_id_list:
-        ds.item_crowd(
-            item_id=item_id,
-            lab='全部渠道',
-            option='商详浏览',
-            last_date=None,
-            d_str='近30天',
-        )
-
-    # 达摩盘人群画像
-    crowid_list = [
-        40457166,
-    ]
-    for crowid in crowid_list:
-        ds.crowd(
-            crowd_id=crowid,
-            last_date=None,
-        )
+    # # 店铺流量来源
+    # ds.dpll()
+    # # 多店聚合推广数据
+    # ds.tg(
+    #     days=15,
+    #     # start_date='2024-11-01',
+    #     # end_date='2024-11-30',
+    # )
+    #
+    # # 商品人群画像
+    # item_id_list = [
+    #     839148235697,
+    # ]
+    # for item_id in item_id_list:
+    #     ds.item_crowd(
+    #         item_id=item_id,
+    #         lab='全部渠道',
+    #         option='商详浏览',
+    #         last_date=None,
+    #         d_str='近30天',
+    #     )
+
+    # # 达摩盘人群画像
+    # crowid_list = [
+    #     40457166,
+    # ]
+    # for crowid in crowid_list:
+    #     ds.crowd(
+    #         crowd_id=crowid,
+    #         last_date=None,
+    #     )
 
     ds.pov_city(
         db_name='生意经3',
         filename='销售地域分布',
         start_date='2024-12-01',
-        end_date='2024-12-11',
-        percentage=0.02,
+        end_date=None,
+        percent=0.015,
     )
 
 if __name__ == '__main__':

{mdbq-3.4.2.dist-info → mdbq-3.4.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mdbq
-Version: 3.4.2
+Version: 3.4.3
 Home-page: https://pypi.org/project/mdbq
 Author: xigua, 
 Author-email: 2587125111@qq.com

{mdbq-3.4.2.dist-info → mdbq-3.4.3.dist-info}/RECORD

@@ -2,7 +2,7 @@ mdbq/__init__.py,sha256=Il5Q9ATdX8yXqVxtP_nYqUhExzxPC_qk_WXQ_4h0exg,16
 mdbq/__version__.py,sha256=y9Mp_8x0BCZSHsdLT_q5tX9wZwd5QgqrSIENLrb6vXA,62
 mdbq/aggregation/__init__.py,sha256=EeDqX2Aml6SPx8363J-v1lz0EcZtgwIBYyCJV6CcEDU,40
 mdbq/aggregation/aggregation.py,sha256=-yzApnlqSN2L0E1YMu5ml-W827qpKQvWPCOI7jj2kzY,80264
-mdbq/aggregation/datashow.py,sha256=k4gUYldnmi_iZJrM7wNtjeenXJl82hUoYcPu6iIL3PU,35864
+mdbq/aggregation/datashow.py,sha256=_pyv7ZmKpBp04bdE_N_RTtNbyBXHcOA-TAQ1vFxl8p8,37881
 mdbq/aggregation/optimize_data.py,sha256=RXIv7cACCgYyehAxMjUYi_S7rVyjIwXKWMaM3nduGtA,3068
 mdbq/aggregation/query_data.py,sha256=9NALeHTP9tblOEPyntLBRtdroLG_qN9qWi34Hg4rXFM,178891
 mdbq/bdup/__init__.py,sha256=AkhsGk81SkG1c8FqDH5tRq-8MZmFobVbN60DTyukYTY,28
@@ -34,7 +34,7 @@ mdbq/pbix/refresh_all.py,sha256=OBT9EewSZ0aRS9vL_FflVn74d4l2G00wzHiikCC4TC0,5926
 mdbq/pbix/refresh_all_old.py,sha256=_pq3WSQ728GPtEG5pfsZI2uTJhU8D6ra-htIk1JXYzw,7192
 mdbq/spider/__init__.py,sha256=RBMFXGy_jd1HXZhngB2T2XTvJqki8P_Fr-pBcwijnew,18
 mdbq/spider/aikucun.py,sha256=v7VO5gtEXR6_4Q6ujbTyu1FHu7TXHcwSQ6hIO249YH0,22208
-mdbq-3.4.2.dist-info/METADATA,sha256=I2lVjMi-WsvegW9ZCQcR4UV8wg4g1A9-mzgVFQ_H7x4,243
-mdbq-3.4.2.dist-info/WHEEL,sha256=eOLhNAGa2EW3wWl_TU484h7q1UNgy0JXjjoqKoxAAQc,92
-mdbq-3.4.2.dist-info/top_level.txt,sha256=2FQ-uLnCSB-OwFiWntzmwosW3X2Xqsg0ewh1axsaylA,5
-mdbq-3.4.2.dist-info/RECORD,,
+mdbq-3.4.3.dist-info/METADATA,sha256=O8d3FiyMrxJoF1-_OcK-YsIQ01m3hQaGxvn8Pk2UoYY,243
+mdbq-3.4.3.dist-info/WHEEL,sha256=eOLhNAGa2EW3wWl_TU484h7q1UNgy0JXjjoqKoxAAQc,92
+mdbq-3.4.3.dist-info/top_level.txt,sha256=2FQ-uLnCSB-OwFiWntzmwosW3X2Xqsg0ewh1axsaylA,5
+mdbq-3.4.3.dist-info/RECORD,,