feat: 🌈 output all stock hold fund detail

.gitignore

@@ -6,8 +6,10 @@ code-record/*
 *$py.class
 .vscode/
 *fund_morning_star.csv
+src/temp_update_*
 abnormal
 demo/
+outcome/stocks/
 output/json/
 !output/xlsx/
 .env

src/fund_info/statistic.py

@@ -14,7 +14,7 @@ import sys
 from pprint import pprint
 sys.path.append('../')
 sys.path.append(os.getcwd() + '/src')
-from utils.index import get_quarter_index
+from utils.index import get_quarter_index, fisrt_match_condition_from_list
 from sql_model.fund_query import FundQuery
 
 
@@ -50,12 +50,16 @@ class FundStatistic:
         code_dict = dict()
         for result in results:
             # print(result)
-            for index in range(1, len(result), 2):
+            for index in range(3, len(result), 3):
                 code = result[index]
                 name = result[index + 1]  # 仅以股票名称为key，兼容港股，A股
-                # key = str(code) + '-' + str(name)
-
-                key = str(name)
+                if code == None or name == None:
+                    #print('index', index, 'code', code, 'name', name)
+                    #print('基金名称', result[1],'基金代码', result[0])
+                    continue
+                key = fisrt_match_condition_from_list(list(code_dict), code)
+                if key == None and code and name:
+                    key = str(code) + '-' + str(name)
                 if(key in code_dict and code != None):
                     code_dict[key] = code_dict[key] + 1
                 else:
@@ -66,19 +70,67 @@ class FundStatistic:
             if value > filter_count and key != None:
                 filer_dict[key] = value
                 # print(key + ":" + str(value))
-        list = sorted(filer_dict.items(), key=lambda x: x[1], reverse=True)
-        return list
+        return sorted(filer_dict.items(), key=lambda x: x[1], reverse=True)
+
+    def all_stock_fund_count_and_details(self, *, quarter_index=None, fund_code_pool=None, filter_count=100):
+        """查询某一个季度基金的十大持仓，并对持仓股票进行汇总统计，并根据filter_count进行过滤
+
+        Args:
+            quarter_index (string, optional): [description]. Defaults to None.取self.quarter_index
+            fund_code_pool (string[], optional): [description]. Defaults to None. 传入查询的基金池，为None默认查询全部
+            filter_count (int, optional): [description]. Defaults to 100. 过滤门槛，过滤掉一些持仓低的股票
+
+        Returns:
+            tuple[]: 每只股票的名称，以及对应持仓基金个数的list
+        """
 
+        quarter_index = quarter_index if quarter_index else self.quarter_index
+        results = self.each_query.select_top_10_stock(
+            quarter_index,
+            fund_code_pool
+        )
+        code_dict = dict()
+        for result in results:
+            #print(result)
+            fund_info = {
+                '基金代码': result[0],
+                '基金名称': result[1],
+                '股票总仓位': result[2],
+            }
+            for index in range(3, len(result), 3):
+                code = result[index]
+                name = result[index + 1]  # 仅以股票名称为key，兼容港股，A股
+                portion = result[index + 2]  # 仅以股票名称为key，兼容港股，A股
+                if code == None or name == None:
+                    continue
+                key = fisrt_match_condition_from_list(list(code_dict), code)
+                if key == None and code and name:
+                    key = str(code) + '-' + str(name)
+                #key = str(name)
+                if(key in code_dict and code != None):
+                    code_dict[key]['count'] = code_dict[key]['count'] + 1
+                    code_dict[key]['fund_list'].append({
+                        **fund_info,
+                        '仓位占比': portion,
+                        '仓位排名': int(index / 3)
+                    })
+                else:
+                    code_dict[key] = {
+                        'count': 1,
+                        'fund_list': [{
+                        **fund_info,
+                        '仓位占比': portion,
+                        '仓位排名': int(index / 3)
+                    }]
+                    }
+        #for key, value in code_dict.items(): 
+        #    print('key, value', key, value)
+        return sorted(code_dict.items(), key=lambda x: x[1]['count'], reverse=True)
     # 分组查询特定股票的每个季度基金持有总数
-    def item_stock_fund_count(self, stock_name, fund_code_pool=None):
-        return self.each_query.select_special_stock_fund_count(stock_name, fund_code_pool)
+    def item_stock_fund_count(self, stock_code, fund_code_pool=None):
+        return self.each_query.select_special_stock_fund_count(stock_code, fund_code_pool)
 
     def select_fund_pool(self, *, morning_star_rating_5="", morning_star_rating_3="", **args):
-        # print("morning_star_rating_5", morning_star_rating_5)
-        # print("morning_star_rating_3", morning_star_rating_3)
-        # print("args", args)
-        # morning_star_rating_5 = 5
-
         return self.each_query.select_certain_condition_funds(
             morning_star_rating_5=morning_star_rating_5,
             morning_star_rating_3=morning_star_rating_3,

src/fund_statistic.py

@@ -9,27 +9,58 @@ Author: luxuemin2108@gmail.com
 Copyright (c) 2020 Camel Lu
 '''
 import time
+import re
 from pprint import pprint
 import pandas as pd
 from fund_info.statistic import FundStatistic
+from utils.index import get_last_quarter_str
+from openpyxl import load_workbook
+import os
 
+def get_fund_code_pool():
+    # fund_code_pool = ['000001', '160133', '360014', '420002',
+    #                   '420102', '000409', '000418', '000746',
+    #                   '000751', '000884', '000991', '001043',
+    #                   '001054', '001104', '001410', '001473',
+    #                   '519714', '000003', '000011', '000029']
+    morning_star_rating_5_condition = {
+        'value': 4,
+        'operator': '>='
+    }
+    morning_star_rating_3_condition = {
+        'value': 5,
+        'operator': '='
+    }
+    last_year_time = time.localtime(time.time() - 365 * 24 * 3600)
+    last_year_date = time.strftime('%Y-%m-%d', last_year_time)
+    condition_dict = {
+        'morning_star_rating_5': morning_star_rating_5_condition,
+        'morning_star_rating_3': morning_star_rating_3_condition,
+        # 'manager_start_date': '2020-05-25'
+    }
+    fund_code_pool = each_statistic.select_fund_pool(
+    #    **condition_dict,
+    #)
+    return fund_code_pool
 
 def stocks_compare(stock_list, fund_code_pool=None):
     each_statistic = FundStatistic()
     filter_list = []
     for stock in stock_list:
-        stock_name = stock[0]
+        stock_code = stock[0].split('-', 1)[0]
+        stock_name = stock[0].split('-', 1)[1]
         stock_sum = stock[1]
 
         stock_quarter_count_tuple = each_statistic.item_stock_fund_count(
-            stock_name,
+            stock_code,
             fund_code_pool
         )
         try:
             last_count_tuple = stock_quarter_count_tuple[len(
                 stock_quarter_count_tuple) - 2]
         except:
-            print('该股票查询异常', stock_name)
+            print("stock_quarter_count_tuple", stock_quarter_count_tuple)
+            print('该股票查询异常', stock_code, stock_name)
             continue
 
         last_stock_sum = last_count_tuple[0]  # 选出上一个季度的
@@ -55,48 +86,61 @@ def stocks_compare(stock_list, fund_code_pool=None):
         # print(item_tuple)
     return filter_list
 
-
-if __name__ == '__main__':
-    each_statistic = FundStatistic()
-
-    # fund_code_pool = ['000001', '160133', '360014', '420002',
-    #                   '420102', '000409', '000418', '000746',
-    #                   '000751', '000884', '000991', '001043',
-    #                   '001054', '001104', '001410', '001473',
-    #                   '519714', '000003', '000011', '000029']
-    morning_star_rating_5_condition = {
-        'value': 4,
-        'operator': '>='
-    }
-    morning_star_rating_3_condition = {
-        'value': 5,
-        'operator': '='
-    }
-    last_year_time = time.localtime(time.time() - 365 * 24 * 3600)
-    last_year_date = time.strftime('%Y-%m-%d', last_year_time)
-    condition_dict = {
-        'morning_star_rating_5': morning_star_rating_5_condition,
-        'morning_star_rating_3': morning_star_rating_3_condition,
-        # 'manager_start_date': '2020-05-25'
-    }
-    fund_code_pool = each_statistic.select_fund_pool(
-        **condition_dict,
-    )
-    # pprint(fund_code_pool)
+def top100_stock(each_statistic, threshold=80):
+    quarter_index = get_last_quarter_str()
+    last_quarter_index = get_last_quarter_str(2)
     stock_top_list = each_statistic.all_stock_fund_count(
-        quarter_index="2021-Q1",
-        fund_code_pool=fund_code_pool,
-        filter_count=5)
+        quarter_index=quarter_index,
+        filter_count=threshold)[:100]
     # print('2020-Q4 top 100 股票')
     pprint(stock_top_list)
-    print(len(stock_top_list))
+    #print(len(stock_top_list))
+
+    filter_list = stocks_compare(stock_top_list)
+
+    #pprint(filter_list)
+    pprint(len(filter_list))
+    df_filter_list = pd.DataFrame(filter_list, columns=[
+        '名称', quarter_index + '持有数量（只）', last_quarter_index + '持有数量（只）', '环比', '环比百分比', '升Or降'])
+    df_filter_list.to_excel(
+        './outcome/strategy/top100.xlsx', sheet_name=quarter_index + '基金重仓股T100')
 
-    # filter_list = stocks_compare(stock_top_list, fund_code_pool)
+def all_stock(each_statistic, threshold=0):
+    quarter_index = "2021-Q1"
+    stock_list = each_statistic.all_stock_fund_count_and_details(
+        quarter_index=quarter_index,
+        filter_count=threshold)
+    for i in range(0, len(stock_list)):
+        stock = stock_list[i]
+        stock_name_code = stock[0]
+        stock_code = stock_name_code.split('-', 1)[0]
+        path = 'other'
+        if bool(re.search("^\d{5}$", stock_code)):
+            path = 'hk'
+        elif bool(re.search("^\d{6}$", stock_code)):
+            path = 'a'
+        hold_fund_count = stock[1]['count']
+        hold_fund_list = stock[1]['fund_list']
+        df_list = pd.DataFrame(hold_fund_list)
+        stock_name_code = stock_name_code.replace('/', '-')
+        path = './outcome/stocks/' + path + '/' + stock_name_code + '.xlsx'
+        path = path.replace('\/', '-')
+        if os.path.exists(path):
+            writer = pd.ExcelWriter(path, engine='openpyxl')
+            book = load_workbook(path)
+            writer.book = book
+            df_list.to_excel(
+                writer, sheet_name=quarter_index)
+            writer.save()
+            writer.close()
+        else:
+            df_list.to_excel(
+                path, sheet_name=quarter_index)
+
+if __name__ == '__main__':
+    each_statistic = FundStatistic()
+    # 获取重仓股
+    top100_stock(each_statistic)
 
-    # pprint(filter_list)
-    # pprint(len(filter_list))
-    # df_filter_list = pd.DataFrame(filter_list, columns=[
-    #     '名称', '2021-Q1持有数量（只）', '2020-Q4持有数量（只）', '环比', '环比百分比', '升Or降'])
-    # print(df_filter_list)
-    # df_filter_list.to_excel(
-    #     './output/xlsx/top100-2021_q1_vs_2020_q4.xlsx', sheet_name='top100')
+    # 所有股票的基金持仓细节
+    #all_stock(each_statistic)

src/fund_strategy.py

@@ -9,12 +9,15 @@ Copyright (c) 2021 Camel Lu
 '''
 from sql_model.fund_query import FundQuery
 import pandas as pd
-from openpyxl import load_workbook
+from openpyxl import load_workbook,Workbook
+from openpyxl.utils import get_column_letter
+from utils.index import get_last_quarter_str
 from pprint import pprint
 
-if __name__ == '__main__':
-    each_query = FundQuery()
-    quarter_index = '2020-Q4'
+# 输出高分基金
+def output_high_score_funds(each_query,quarter_index=None):
+    if quarter_index == None:
+       quarter_index = get_last_quarter_str()
     high_score_funds = each_query.select_high_score_funds(
         quarter_index=quarter_index)
     columns_bk = ['代码', '名称', '季度', '总资产', '现任基金经理管理起始时间', '投资风格', '三月最大回撤', '六月最大回撤', '夏普比率', '阿尔法系数', '贝塔系数',
@@ -34,7 +37,24 @@ if __name__ == '__main__':
     writer = pd.ExcelWriter(path, engine='openpyxl')
     book = load_workbook(path)
     writer.book = book
-
     df_high_score_funds.to_excel(writer, sheet_name=quarter_index)
     writer.save()
     writer.close()
+
+if __name__ == '__main__':
+    #each_query = FundQuery()
+    #quarter_index = '2020-Q4'
+    #output_high_score_funds()
+    dest_filename = 'empty_book.xlsx'
+    #wb = Workbook(dest_filename)
+    wb = load_workbook(filename = 'empty_book.xlsx')
+    ws = wb.active
+    print("ws", ws)
+    #ws.merge_cells('A2:D2')
+    ws.merge_cells(start_row=2, start_column=1, end_row=4, end_column=4)
+    ws.merge_cells('J17:J20')
+    ws.column_dimensions.group('A','D', hidden=True)
+    ws.row_dimensions.group(1,10, hidden=True)
+    wb.save(dest_filename)
+    #ws.unmerge_cells('A2:D2')
+

src/sql_model/fund_query.py

@@ -224,12 +224,13 @@ class FundQuery:
         c_class_result = self.cursor.fetchone()
         return c_class_result
 
+    #获取基金十大持仓以及代码，名称
     def select_top_10_stock(self, quarter_index=None, fund_code_pool=None):
         stock_sql_join = ''
         for index in range(10):
             stock_sql_join = stock_sql_join + \
-                "t.top_stock_%s_code, t.top_stock_%s_name" % (
-                    str(index), str(index)) + ","
+                "t.top_stock_%s_code, t.top_stock_%s_name, t.top_stock_%s_portion" % (
+                    str(index), str(index), str(index)) + ","
         stock_sql_join = stock_sql_join[0:-1]
         fund_code_list_sql = ''
         # 判断是否传入fund_code_pool
@@ -238,7 +239,7 @@ class FundQuery:
                 return ()
             list_str = ', '.join(fund_code_pool)
             fund_code_list_sql = "AND t.fund_code IN (" + list_str + ")"
-        sql_query_quarter = "SELECT t.fund_code," + stock_sql_join + \
+        sql_query_quarter = "SELECT t.fund_code, t.fund_name, t.stock_position_total, " + stock_sql_join + \
             " FROM fund_morning_stock_info as t WHERE t.quarter_index = %s AND t.stock_position_total > 20 " + \
             fund_code_list_sql + \
             ";"  # 大于20%股票持仓基金
@@ -249,13 +250,13 @@ class FundQuery:
         return results
 
     # 分组查询特定股票的每个季度基金持有总数
-    def select_special_stock_fund_count(self, stock_name, fund_code_pool=None):
+    def select_special_stock_fund_count(self, stock_code, fund_code_pool=None):
         stock_sql_join = '('
         for index in range(10):
-            escape_name = stock_name.replace("'", "\\'")
+            escape_code = stock_code.replace("'", "\\'")
             stock_sql_join = stock_sql_join + \
-                "t.top_stock_{0}_name = '{1}' or ".format(
-                    str(index), escape_name)
+                "t.top_stock_{0}_code = '{1}' or ".format(
+                    str(index), escape_code)
         stock_sql_join = stock_sql_join[0:-3] + ')'
         fund_code_list_sql = ''
         # 判断是否传入fund_code_pool

src/utils/index.py

@@ -62,8 +62,8 @@ def get_quarter_index(input_date):
     return index
 
 
-def get_last_quarter_str():
-    last_quarter_time = time.localtime(time.time() - 3 * 30 * 24 * 3600)
+def get_last_quarter_str(last_index=1):
+    last_quarter_time = time.localtime(time.time() - last_index * 3 * 30 * 24 * 3600)
     year = time.strftime("%Y", last_quarter_time)
     date = time.strftime("%m-%d", last_quarter_time)
     index = get_quarter_index(date)
@@ -75,3 +75,10 @@ def get_quarter_date(quarter_index_str):
     boundary_date_list = ['03-31', '06-30', '09-30', '12-31']
     quarter_index = quarter_index_str.split('-')[1][1:]
     return year + '-' + boundary_date_list[int(quarter_index) - 1]
+
+def fisrt_match_condition_from_list(list, code):
+  for item in list:
+    stock_code = item.split('-', 1)[0]
+    is_exist = code == stock_code
+    if is_exist:
+      return item