mdbq 3.4.0__py3-none-any.whl → 3.4.2__py3-none-any.whl

mdbq/aggregation/datashow.py

@@ -1,4 +1,5 @@
 # -*- coding: UTF-8 –*-
+import decimal
 import os
 import re
 import socket
@@ -91,7 +92,9 @@ else:
 
 class DataShow:
     def __init__(self):
-        self.path = '/Users/xigua/Downloads'
+        self.path = '/Users/xigua/Downloads/html文件'
+        if not os.path.isdir(self.path):
+            os.makedirs(self.path)
         root = tk.Tk()
         self.screen_width = root.winfo_screenwidth()
         self.screen_height = root.winfo_screenheight()
@@ -119,100 +122,171 @@ class DataShow:
         df = pd.concat(__res, ignore_index=True)
         return df
 
-    def dpll_bak(self, db_name='聚合数据', table_name='店铺流量来源构成', pro_list=None, filename='店铺流量来源'):
-        if not pro_list:
-            pro_list = ['日期', '三级来源', '访客数']
-        df = self.getdata(db_name=db_name, table_name=table_name, pro_list=pro_list, start_date='2024-11-01', end_date=self.end_date)
-        if len(df) == 0:
-            print(f'数据不能为空: {table_name}')
-            return
-        df = df[df['三级来源'] != '汇总']
-        df['日期'] = pd.to_datetime(df['日期'])
-        today = datetime.date.today()
-
-        def st_date(num=1):
-            return pd.to_datetime(today - datetime.timedelta(days=num))
-        max_date = df['日期'].max().strftime('%Y-%m-%d')
-        df1 = df[df['日期'] >= st_date(1)]
-        df2 = df[df['日期'] >= st_date(7)]
-        df3 = df[df['日期'] >= st_date(30)]
-        df2 = df2.groupby(
-            ['三级来源'],
-            as_index=False).agg(
-            **{
-                '访客数': ('访客数', np.sum),
-            }
+    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):
+        """
+        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
+        pov_set = self.getdata(
+            db_name='属性设置3',
+            table_name=f'城市等级',
+            pro_list=[],
+            start_date=start_date,
+            end_date=end_date
+        )
+        # print(pov_set)
+        # 城市
+        pro_list = ['日期', '店铺名称', '城市', '销售额', '退款额']
+        year = datetime.datetime.today().year
+        df_city = self.getdata(
+            db_name=db_name,
+            table_name=f'地域分析_城市_{year}',
+            pro_list=pro_list,
+            start_date=start_date,
+            end_date=end_date
+        )
+        df_city = df_city[df_city['店铺名称'] == '万里马官方旗舰店']
+        df_city = df_city.groupby(['店铺名称', '城市'], as_index=False).agg(
+            **{'销售额': ('销售额', np.sum), '退款额': ('退款额', np.sum)})
+        df_city = df_city[df_city['销售额'] > 0]
+
+        # 省份
+        pro_list = ['日期', '店铺名称', '省份', '销售额', '退款额']
+        year = datetime.datetime.today().year
+        df_pov = self.getdata(
+            db_name=db_name,
+            table_name=f'地域分析_省份_{year}',
+            pro_list=pro_list,
+            start_date=start_date,
+            end_date=end_date
+        )
+        df_pov = df_pov[df_pov['店铺名称'] == '万里马官方旗舰店']
+        # print(df_pov[df_pov['省份'] == '广东'])
+        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_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)]
+
+        # 将城市等级添加到df
+        pov_set = pov_set[['城市等级', '城市']]
+        pov_set.drop_duplicates(subset='城市', keep='last', inplace=True, ignore_index=True)
+        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())]
+        df_city.drop_duplicates(subset='城市', keep='last', inplace=True, ignore_index=True)
+        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
+        t_p2 = []
+        for i in range(3):
+            t_p2.extend([{"type": "bar"}])  # 饼图类型
+        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,
         )
-        df3 = df3.groupby(
-            ['三级来源'],
-            as_index=False).agg(
-            **{
-                '访客数': ('访客数', np.sum),
-            }
+        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),
         )
-        # print(df)
-        labels1 = df1['三级来源'].tolist()
-        values1 = df1['访客数'].tolist()
-        labels2 = df2['三级来源'].tolist()
-        values2 = df2['访客数'].tolist()
-        labels3 = df3['三级来源'].tolist()
-        values3 = df3['访客数'].tolist()
-
-        def make_sub(data_list, num):
-            # 创建一个具有1行2列的网格布局
-            t_p = []
-            for i in range(num):
-                t_p.extend([{"type": "pie"}])
-            fig = make_subplots(rows=1, cols=num, specs=[t_p])
-            pie_title = {1: 1, 2: 7, 3: 30}
-            i = 1
-            for item in data_list:
-                # 计算每个扇区的百分比，并找出哪些扇区应该被保留
-                total = sum(item['值'])
-                # 计算每个扇区的百分比，并找出哪些扇区应该被保留
-                threshold_percentage = 0.1  # 阈值百分比
-                filtered_indices = [i for i, value in enumerate(item['值']) if
-                                    (value / total) * 100 >= threshold_percentage]
-                # 提取被保留的扇区的标签和值
-                filtered_labels = [item['键'][i] for i in filtered_indices]
-                filtered_values = [item['值'][i] for i in filtered_indices]
+        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,
+            )
+            x = 0.14 + 0.355 * (row % 3)
+            y = 1 - 0.575 * (row // 3)
+            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
 
-                # 添加饼图
-                fig.add_trace(
-                    go.Pie(labels=filtered_labels, values=filtered_values, name=f'pie {i}', textinfo='label+percent'),
-                    row=1, col=i)
-                # fig.add_trace(go.Pie(labels=item['键'], values=item['值'], name=f'最近{pie_title[i]}天', textinfo='label+percent'), row=1, col=i)
-                fig.add_annotation(
-                    text=f'最近{pie_title[i]}天',
-                    x=0.15 + 0.35 * (i - 1),
-                    y=0.98,
-                    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=16),
-                )
-                i += 1
-                fig.update_layout(
-                    title_text=f'店铺流量来源   最近一天: {max_date}',
-                    xaxis_title='X Axis',
-                    yaxis_title='Y Axis',
-                    # width=self.screen_width // 1.4,
-                    # height=self.screen_width // 2,
-                    margin=dict(
-                        l=100,   # 左边距
-                        r=300,
-                        t=100,  # 上边距
-                        b=400,
-                    ),
+        fig.update_layout(
+            title_text=f'销售地域分布',
+            margin=dict(
+                l=100,  # 左边距
+                r=100,
+                t=100,  # 上边距
+                b=100,
+            ),
+            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
                 )
-                fig.update_layout(xaxis_showgrid=False, yaxis_showgrid=False, xaxis_visible=False, yaxis_visible=False)
-            return fig
-
-        data_list = [{'键': labels1, '值': values1}, {'键': labels2, '值': values2}, {'键': labels3, '值': values3}]
-        fig = make_sub(data_list=data_list, num=3)
+            )
+        )
+        fig.add_annotation(
+            text=f'统计时间周期: {start_date}~{end_date}',
+            x=0.5,
+            y=-0.1,
+            xref='paper',  # # 相对于整个图表区域
+            yref='paper',
+            showarrow=False,  # 显示箭头
+            align="left",  # 文本对齐方式
+            font=dict(size=12),
+        )
         fig.write_html(os.path.join(self.path, f'{filename}.html'))
 
+
     def dpll(self, db_name='聚合数据', table_name='店铺流量来源构成', pro_list=None, filename='店铺流量来源'):
         if not pro_list:
             pro_list = ['日期', '店铺名称', '类别', '来源构成', '二级来源', '三级来源', '访客数']
@@ -324,7 +398,7 @@ class DataShow:
                 )
                 count2 += 1
         fig.update_layout(
-            title_text=f'店铺流量来源   最近数据: {max_date}',
+            title_text=f'店铺流量来源',
             # xaxis_title='X Axis',
             # yaxis_title='Y Axis',
             # width=self.screen_width // 1.4,
@@ -345,6 +419,16 @@ class DataShow:
                 )
             )
         )
+        fig.add_annotation(
+            text=f'最近数据日期: {max_date}',
+            x=0.5,
+            y=-0.25,
+            xref='paper',  # # 相对于整个图表区域
+            yref='paper',
+            showarrow=False,  # 显示箭头
+            align="left",  # 文本对齐方式
+            font=dict(size=12),
+        )
         fig.write_html(os.path.join(self.path, f'{filename}.html'))
 
     def tg(self, db_name='聚合数据', table_name='多店推广场景_按日聚合', pro_list=None, filename='多店推广场景', days=None, start_date=None, end_date=None):
@@ -600,7 +684,7 @@ class DataShow:
             align="left",  # 文本对齐方式
             font=dict(size=12),
         )
-        fig.write_html(os.path.join(self.path, f'{filename}.html'))
+        fig.write_html(os.path.join(self.path, f'{filename}_{item_id}.html'))
 
     def crowd(self, db_name='人群画像2', table_list=None, pro_list=None, filename='达摩盘人群画像', crowd_id=None, last_date=None):
         # item_ids = [696017020186, 714066010148, 830890472575]
@@ -739,29 +823,51 @@ class DataShow:
             align="left",  # 文本对齐方式
             font=dict(size=12),
         )
-        fig.write_html(os.path.join(self.path, f'{filename}.html'))
+        fig.write_html(os.path.join(self.path, f'{filename}_{crowd_name[:15]}.html'))
 
 
 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',
     )
-    ds.item_crowd(
-        item_id=839148235697,
-        lab='全部渠道',
-        option='商详浏览',
-        last_date=None,
-        d_str='近30天',
-    )
-    ds.crowd(
-        crowd_id=40457166,
-        last_date=None,
-    )
 
+    # 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,
+    )
 
 if __name__ == '__main__':
     main()

mdbq/aggregation/query_data.py

@@ -1193,6 +1193,83 @@ class MysqlDatasQuery:
         )
         return True
 
+    def item_up(self, db_name='聚合数据', table_name='淘宝店铺货品'):
+        start_date, end_date = self.months_data(num=self.months)
+        projection = {}
+        df_set = self.download.data_to_df(
+            db_name='属性设置3',
+            table_name=f'货品年份基准',
+            start_date=start_date,
+            end_date=end_date,
+            projection={'商品id':1, '上市年份':1},
+        )
+        df = self.download.data_to_df(
+            db_name='市场数据3',
+            table_name=f'淘宝店铺数据',
+            start_date=start_date,
+            end_date=end_date,
+            projection=projection,
+        )
+        df['日期'] = pd.to_datetime(df['日期'], format='%Y-%m-%d', errors='ignore')  # 转换日期列
+        df = df[df['日期'] == pd.to_datetime('2024-12-12')]
+
+        df_set['商品id'] = df_set['商品id'].astype('int64')
+        df['商品id'] = df['商品id'].astype('int64')
+        df_set.sort_values('商品id', ascending=False, ignore_index=True, inplace=True)
+
+        def check_year(item_id):
+            for item in df_set.to_dict(orient='records'):
+                if item_id > item['商品id']:
+                    return item['上市年份']
+
+        df['上市年份'] = df['商品id'].apply(lambda x: check_year(x))
+        p = df.pop('上市年份')
+        df.insert(loc=5, column='上市年份', value=p)
+        now = datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+        print(f'{now} 正在更新: mysql ({host}:{port}) {db_name}/{table_name}')
+        set_typ = {
+            '日期': 'date',
+            '店铺id': 'bigint',
+            '店铺名称': 'varchar(255)',
+            '商家id': 'bigint',
+            '商品id': 'bigint',
+            '商品标题': 'varchar(255)',
+            '商品链接': 'varchar(255)',
+            '商品图片': 'varchar(255)',
+            '销量': 'varchar(50)',
+            '页面价': 'int',
+            'data_sku': 'varchar(1000)',
+            '更新时间': 'timestamp',
+            '上市年份': 'varchar(50)',
+        }
+        m_engine.df_to_mysql(
+            df=df,
+            db_name=db_name,
+            table_name=table_name,
+            # icm_update=['日期', '一级来源', '二级来源', '三级来源', '访客数'],  # 增量更新, 在聚合数据中使用，其他不要用
+            move_insert=True,  # 先删除，再插入
+            df_sql=False,  # 值为 True 时使用 df.to_sql 函数上传整个表, 不会排重
+            drop_duplicates=False,  # 值为 True 时检查重复数据再插入，反之直接上传，会比较慢
+            count=None,
+            filename=None,  # 用来追踪处理进度
+            reset_id=True,  # 是否重置自增列
+            set_typ=set_typ,
+        )
+        company_engine.df_to_mysql(
+            df=df,
+            db_name=db_name,
+            table_name=table_name,
+            # icm_update=['日期', '一级来源', '二级来源', '三级来源', '访客数'],  # 增量更新, 在聚合数据中使用，其他不要用
+            move_insert=True,  # 先删除，再插入
+            df_sql=False,  # 值为 True 时使用 df.to_sql 函数上传整个表, 不会排重
+            drop_duplicates=False,  # 值为 True 时检查重复数据再插入，反之直接上传，会比较慢
+            count=None,
+            filename=None,  # 用来追踪处理进度
+            reset_id=True,  # 是否重置自增列
+            set_typ=set_typ,
+        )
+
+
     def spph(self, db_name='聚合数据', table_name='天猫_商品排行'):
         """  """
         start_date, end_date = self.months_data(num=self.months)
@@ -3840,7 +3917,7 @@ if __name__ == '__main__':
     sdq = MysqlDatasQuery()  # 实例化数据处理类
     sdq.months = 1  # 设置数据周期， 1 表示近 2 个月
     sdq.update_service = True  # 调试时加，true: 将数据写入 mysql 服务器
-    sdq.tg_rqbb(db_name='聚合数据', table_name='天猫_人群报表')
+    sdq.item_up()
 
     # string = '30-34岁,35-39岁,40-49岁'
     # d = "~".join(re.findall(r'(\d+)\D.*\D(\d+)岁', string)[0])

mdbq/spider/aikucun.py

@@ -483,7 +483,7 @@ class AikuCunNew:
 
 if __name__ == '__main__':
     get_cookie_aikucun()  # 登录并获取 cookies
-    akucun(date_num=3, headless=True)  # 下载数据
+    akucun(date_num=10, headless=True)  # 下载数据
 
     # a = AikuCunNew(shop_name='aikucun')
     # a.akc()

{mdbq-3.4.0.dist-info → mdbq-3.4.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mdbq
-Version: 3.4.0
+Version: 3.4.2
 Home-page: https://pypi.org/project/mdbq
 Author: xigua, 
 Author-email: 2587125111@qq.com

{mdbq-3.4.0.dist-info → mdbq-3.4.2.dist-info}/RECORD

@@ -2,9 +2,9 @@ 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=2NzHGjGoUy2WG-MxmbilCj6KBAmVah3jqFuEd2zv9XU,32379
+mdbq/aggregation/datashow.py,sha256=k4gUYldnmi_iZJrM7wNtjeenXJl82hUoYcPu6iIL3PU,35864
 mdbq/aggregation/optimize_data.py,sha256=RXIv7cACCgYyehAxMjUYi_S7rVyjIwXKWMaM3nduGtA,3068
-mdbq/aggregation/query_data.py,sha256=FiNZhL5_El2B5ADfCPGUZXsE2iZd3UmGml9Te9qJIpU,175364
+mdbq/aggregation/query_data.py,sha256=9NALeHTP9tblOEPyntLBRtdroLG_qN9qWi34Hg4rXFM,178891
 mdbq/bdup/__init__.py,sha256=AkhsGk81SkG1c8FqDH5tRq-8MZmFobVbN60DTyukYTY,28
 mdbq/bdup/bdup.py,sha256=LAV0TgnQpc-LB-YuJthxb0U42_VkPidzQzAagan46lU,4234
 mdbq/config/__init__.py,sha256=jso1oHcy6cJEfa7udS_9uO5X6kZLoPBF8l3wCYmr5dM,18
@@ -33,8 +33,8 @@ mdbq/pbix/pbix_refresh.py,sha256=JUjKW3bNEyoMVfVfo77UhguvS5AWkixvVhDbw4_MHco,239
 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=eAIITxnbbxsR_EoohJ78CRw2dEdfSHOltfpxBrh0cvc,22207
-mdbq-3.4.0.dist-info/METADATA,sha256=5yMS8KxK8EkUCkrPLlPexLMSb2JYk3pVTL5b2S39uME,243
-mdbq-3.4.0.dist-info/WHEEL,sha256=eOLhNAGa2EW3wWl_TU484h7q1UNgy0JXjjoqKoxAAQc,92
-mdbq-3.4.0.dist-info/top_level.txt,sha256=2FQ-uLnCSB-OwFiWntzmwosW3X2Xqsg0ewh1axsaylA,5
-mdbq-3.4.0.dist-info/RECORD,,
+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,,