mns-scheduler 1.0.8.7__py3-none-any.whl → 1.4.3.2__py3-none-any.whl

mns_scheduler/k_line/clean/daily/daily_k_line_clean_common_service.py

@@ -2,7 +2,7 @@ import sys
 import os
 
 file_path = os.path.abspath(__file__)
-end = file_path.index('mns') + 16
+end = file_path.index('mns') + 17
 project_path = file_path[0:end]
 sys.path.append(project_path)
 import pandas as pd
@@ -85,6 +85,12 @@ def init_day_line_data(k_line_info, stock_qfq_daily):
     k_line_info.loc[:, 'avg_thirty_last'] = 0
     k_line_info.loc[:, 'avg_sixty_last'] = 0
 
+    k_line_info.loc[:, 'slope_five_last'] = 0
+    k_line_info.loc[:, 'slope_ten_last'] = 0
+    k_line_info.loc[:, 'slope_twenty_last'] = 0
+    k_line_info.loc[:, 'slope_thirty_last'] = 0
+    k_line_info.loc[:, 'slope_sixty_last'] = 0
+
     k_line_info.loc[:, 'std_amount_ten'] = 0
     k_line_info.loc[:, 'mean_amount_ten'] = 0
     k_line_info.loc[:, 'std_amount_thirty'] = 0
@@ -105,6 +111,8 @@ def init_day_line_data(k_line_info, stock_qfq_daily):
 
     k_line_info.loc[:, 'diff_close_from_open_last'] = 0
 
+    k_line_info.loc[:, 'max_chg_30'] = 0
+
     return k_line_info
 
 
@@ -116,8 +124,10 @@ def calculate_30_day_max_chg(stock_qfq_daily, k_line_info):
     deal_days = stock_qfq_daily_30.shape[0]
 
     if deal_days < 30:
+        # 排除第一天
         stock_qfq_daily_30 = stock_qfq_daily_30[0: deal_days - 1]
-    if deal_days == 0:
+    if deal_days == 0 or deal_days == 1:
+        k_line_info['max_chg_30'] = 0
         return k_line_info
 
     stock_qfq_daily_30['date_time'] = pd.to_datetime(stock_qfq_daily_30['date'])
@@ -196,24 +206,31 @@ def calculate_exchange_and_k_line_avg_param(stock_qfq_daily):
 
     # 收盘价格与均线差值
     stock_qfq_daily['close_difference_five'] = round(
-        100 * (stock_qfq_daily['close'] - stock_qfq_daily['avg_five']) / stock_qfq_daily['close'],
+        100 * (stock_qfq_daily['close'] - stock_qfq_daily['avg_five']) / stock_qfq_daily['avg_five'],
         2)
 
     stock_qfq_daily['close_difference_ten'] = round(
-        100 * (stock_qfq_daily['close'] - stock_qfq_daily['avg_ten']) / stock_qfq_daily['close'],
+        100 * (stock_qfq_daily['close'] - stock_qfq_daily['avg_ten']) / stock_qfq_daily['avg_ten'],
         2)
 
     stock_qfq_daily['close_difference_twenty'] = round(
-        100 * (stock_qfq_daily['close'] - stock_qfq_daily['avg_twenty']) / stock_qfq_daily['close'],
+        100 * (stock_qfq_daily['close'] - stock_qfq_daily['avg_twenty']) / stock_qfq_daily['avg_twenty'],
         2)
 
     stock_qfq_daily['close_difference_thirty'] = round(
-        100 * (stock_qfq_daily['close'] - stock_qfq_daily['avg_thirty']) / stock_qfq_daily['close'],
+        100 * (stock_qfq_daily['close'] - stock_qfq_daily['avg_thirty']) / stock_qfq_daily['avg_thirty'],
         2)
 
     stock_qfq_daily['close_difference_sixty'] = round(
-        100 * (stock_qfq_daily['close'] - stock_qfq_daily['avg_sixty']) / stock_qfq_daily['close'],
+        100 * (stock_qfq_daily['close'] - stock_qfq_daily['avg_sixty']) / stock_qfq_daily['avg_sixty'],
         2)
+    if (('slope_' + 'five' not in stock_qfq_daily.columns)
+            or ('slope_' + 'ten' not in stock_qfq_daily.columns) \
+            or ('slope_' + 'twenty' not in stock_qfq_daily.columns) \
+            or ('slope_' + 'thirty' not in stock_qfq_daily.columns) \
+            or ('slope_' + 'sixty' not in stock_qfq_daily.columns)):
+        # 计算均线斜率
+        stock_qfq_daily = calculate_slope(stock_qfq_daily, windows=['five', 'ten', 'twenty', 'thirty', 'sixty'])
 
     stock_qfq_daily = stock_qfq_daily[[
         "symbol",
@@ -251,6 +268,11 @@ def calculate_exchange_and_k_line_avg_param(stock_qfq_daily):
         "avg_twenty",
         'avg_thirty',
         'avg_sixty',
+        "slope_five",
+        "slope_ten",
+        "slope_twenty",
+        'slope_thirty',
+        'slope_sixty',
         "classification",
         "_id",
         "date"
@@ -261,6 +283,34 @@ def calculate_exchange_and_k_line_avg_param(stock_qfq_daily):
     return stock_qfq_daily
 
 
+def fix_avg_slope_name(k_line_info, stock_qfq_daily):
+    stock_qfq_daily_one = stock_qfq_daily.iloc[0:1]
+    k_line_info['slope_five_last'] = stock_qfq_daily_one['slope_five']
+    k_line_info['slope_ten_last'] = stock_qfq_daily_one['slope_ten']
+    k_line_info['slope_twenty_last'] = stock_qfq_daily_one['slope_twenty']
+    k_line_info['slope_thirty_last'] = stock_qfq_daily_one['slope_thirty']
+    k_line_info['slope_sixty_last'] = stock_qfq_daily_one['slope_sixty']
+    return k_line_info
+
+
+# 计算均线
+def calculate_moving_averages(data, windows=['five', 'ten', 'twenty', 'thirty', 'sixty']):
+    for window in windows:
+        data[f'avg_{window}', window] = data['close'].rolling(window=window).mean()
+    return data
+
+
+# 计算均线斜率
+def calculate_slope(data, windows=['five', 'ten', 'twenty', 'thirty', 'sixty']):
+    for window in windows:
+        if 'slope_' + window in data.columns:
+            continue
+        data[f'slope_{window}'] = data[f'avg_{window}'].diff() / data[f'avg_{window}'].shift(1)
+        data[f'slope_{window}'] = data[f'slope_{window}'] * 100
+        data[f'slope_{window}'] = round(data[f'slope_{window}'], 2)
+    return data
+
+
 # 设置当天k线形态 下一个交易日判断当前交易日k线形态
 def set_k_line_patterns(stock_qfq_daily_one):
     open = list(stock_qfq_daily_one['open'])[0]

mns_scheduler/k_line/clean/daily/daily_k_line_service.py

@@ -2,7 +2,7 @@ import sys
 import os
 
 file_path = os.path.abspath(__file__)
-end = file_path.index('mns') + 16
+end = file_path.index('mns') + 17
 project_path = file_path[0:end]
 sys.path.append(project_path)
 from mns_common.db.MongodbUtil import MongodbUtil
@@ -10,6 +10,7 @@ import mns_common.utils.date_handle_util as date_handle_util
 import mns_scheduler.k_line.hot_stocks.recent_hot_stocks_clean_service as recent_hot_stocks_clean_service
 import mns_scheduler.k_line.clean.daily.daily_k_line_clean_common_service as daily_k_line_clean_common_service
 from mns_common.component.classify.symbol_classify_param import stock_type_classify_param
+import mns_scheduler.k_line.clean.week_month.sub_new_week_month_k_line_service as sub_new_week_month_k_line_service
 
 mongodb_util = MongodbUtil('27017')
 # 普通股日线查询数据 60
@@ -17,7 +18,7 @@ NORMAL_DAILY_K_LINE_NUMBER = 60
 
 
 def handle_day_line(k_line_info, str_day, symbol, deal_days):
-    sub_stock_new_max_deal_days = stock_type_classify_param['sub_stock_new_max_deal_days']
+    sub_stock_new_max_deal_days = stock_type_classify_param['sub_new_stock_max_deal_days']
     if deal_days > sub_stock_new_max_deal_days:
         return handle_day_line_normal(k_line_info, str_day, symbol, deal_days)
 
@@ -54,15 +55,25 @@ def handle_day_line_sub_new(k_line_info, str_day, symbol, deal_days):
     stock_qfq_daily = stock_qfq_daily.iloc[0:deal_days - 1]
     # 计算换手平均值 k线 5 10 20 30 60均线
     stock_qfq_daily = daily_k_line_clean_common_service.calculate_exchange_and_k_line_avg_param(stock_qfq_daily)
+    # 计算30天最大涨幅
+    k_line_info = daily_k_line_clean_common_service.calculate_30_day_max_chg(stock_qfq_daily, k_line_info)
     # 设置五日k线和
     k_line_info = daily_k_line_clean_common_service.set_sum_five_chg(k_line_info, deal_days)
     # 计算当前交易日开盘时的涨幅
     k_line_info = daily_k_line_clean_common_service.calculate_open_chg(stock_qfq_daily, k_line_info)
     # 计算 昨日最高点到开盘涨幅差值 and   # 昨日最高点到当日收盘涨幅之间的差值 and  # 昨日收盘到当日开盘涨幅之间的差值
     k_line_info = daily_k_line_clean_common_service.calculate_chg_diff_value(k_line_info)
+    # 计算 月线
+    k_line_info = sub_new_week_month_k_line_service.handle_month_line(k_line_info, stock_qfq_daily, deal_days)
+    # 计算周线
+    k_line_info = sub_new_week_month_k_line_service.handle_week_line(k_line_info, stock_qfq_daily, deal_days)
+
     # 排除最近有三板以上的股票 todo
     # 计算最近热门大涨的股票
     recent_hot_stocks_clean_service.calculate_recent_hot_stocks(stock_qfq_daily, symbol, str_day)
+    # 修改 avg name
+    k_line_info = daily_k_line_clean_common_service.fix_avg_slope_name(k_line_info, stock_qfq_daily)
+
     return k_line_info
 
 
@@ -99,9 +110,11 @@ def handle_day_line_normal(k_line_info, str_day, symbol, deal_days):
     # 排除最近有三板以上的股票 todo
     # 计算最近热门大涨的股票
     recent_hot_stocks_clean_service.calculate_recent_hot_stocks(stock_qfq_daily, symbol, str_day)
-
+    # 修改 avg name
+    k_line_info = daily_k_line_clean_common_service.fix_avg_slope_name(k_line_info, stock_qfq_daily)
     return k_line_info
 
+
 # if __name__ == '__main__':
 #     query1 = {"symbol": '301596', 'date': {"$lte": date_handle_util.no_slash_date('2024-05-31')}}
 #     stock_qfq_daily_301596 = mongodb_util.descend_query(query1, 'stock_qfq_daily', 'date', 15)

mns_scheduler/k_line/clean/k_line_info_clean_impl.py

@@ -16,7 +16,7 @@ mongodb_util = MongodbUtil('27017')
 
 
 # 日线 周线 月线 成交量  筹码信息
-def calculate_k_line_info(str_day, symbol, diff_days):
+def calculate_k_line_info(str_day, symbol, diff_days, stock_qfq_year_df):
     k_line_info = pd.DataFrame([[
         str_day,
         symbol, diff_days]],
@@ -27,7 +27,8 @@ def calculate_k_line_info(str_day, symbol, diff_days):
     # 交易天数
     deal_days = k_line_common_service_api.get_deal_days(str_day, symbol)
     # 处理周线 月线
-    k_line_info = week_month_k_line_service.handle_month_week_line(k_line_info, str_day, symbol, deal_days)
+    k_line_info = week_month_k_line_service.handle_month_week_line(k_line_info, str_day, symbol,
+                                                                   deal_days, stock_qfq_year_df)
     # 处理日线
     k_line_info = daily_k_line_service.handle_day_line(k_line_info, str_day, symbol, deal_days)
     return k_line_info

mns_scheduler/k_line/clean/k_line_info_clean_task.py

@@ -9,15 +9,16 @@ sys.path.append(project_path)
 import pandas as pd
 import mns_common.utils.date_handle_util as date_handle_util
 from mns_common.db.MongodbUtil import MongodbUtil
-import mns_common.api.em.east_money_stock_api as east_money_stock_api
+import mns_common.component.em.em_stock_info_api as em_stock_info_api
 from loguru import logger
 import threading
 import mns_common.component.trade_date.trade_date_common_service_api as trade_date_common_service_api
 import mns_common.component.common_service_fun_api as common_service_fun_api
-import mns_scheduler.k_line.sync.daily_week_month_line_sync as daily_week_month_line_sync_api
+import mns_scheduler.k_line.month_week_daily.daily_week_month_line_sync as daily_week_month_line_sync_api
 import mns_scheduler.k_line.clean.k_line_info_clean_impl as k_line_info_clean_impl
 import mns_common.utils.data_frame_util as data_frame_util
 import mns_common.component.company.company_common_service_api as company_common_service_api
+import mns_common.constant.db_name_constant as db_name_constant
 
 K_LINE_CLEAN_DB_NAME = 'k_line_clean_fail_name'
 
@@ -67,10 +68,17 @@ def sync_k_line_info(str_day, symbol_list):
 
                     diff_days = list(company_info_df['diff_days'])[0]
 
-                    k_line_result = k_line_info_clean_impl.calculate_k_line_info(str_day, symbol, diff_days)
+                    now_year = int(str_day[0:4])
+                    last_year = now_year - 1
+
+                    query_year_line = {'symbol': symbol, 'year': {"$in": [str(now_year), str(last_year)]}}
+                    stock_qfq_year_df = mongodb_util.find_query_data(db_name_constant.STOCK_QFQ_YEAR, query_year_line)
+
+                    k_line_result = k_line_info_clean_impl.calculate_k_line_info(str_day, symbol, diff_days,
+                                                                                 stock_qfq_year_df)
                     save_k_line_data(symbol, str_day, k_line_result)
                     if result_k_line_list_df is None:
-                        result_k_line_list_df = result
+                        result_k_line_list_df = k_line_result
                     else:
                         result_k_line_list_df = pd.concat([result_k_line_list_df, k_line_result])
 
@@ -87,18 +95,27 @@ def handle_fail_data(str_day, real_time_quotes_now):
     k_line_fail_df = mongodb_util.find_query_data(K_LINE_CLEAN_DB_NAME, query)
     if data_frame_util.is_not_empty(k_line_fail_df):
         fail_data_df = real_time_quotes_now.loc[real_time_quotes_now['symbol'].isin(k_line_fail_df['symbol'])]
-        single_threaded_sync_task(fail_data_df, str_day, 88)
+        now_year = int(str_day[0:4])
+        last_year = now_year - 1
+
+        query_year_line = {'year': {"$in": [str(now_year), str(last_year)]}}
+        stock_qfq_year_df = mongodb_util.find_query_data(db_name_constant.STOCK_QFQ_YEAR, query_year_line)
+        single_threaded_sync_task(fail_data_df, str_day, 88, stock_qfq_year_df)
 
 
 # 多线程同步任务
 def multi_threaded_k_line_sync(str_day):
     # 退市代码
-    de_list_company_symbols = company_common_service_api.get_de_list_company()
-    real_time_quotes_now = east_money_stock_api.get_real_time_quotes_all_stocks()
-    real_time_quotes_now = real_time_quotes_now.loc[~(real_time_quotes_now['symbol'].isin(de_list_company_symbols))]
 
-    # 将list_date列中的所有NaN值设置为99990909
-    real_time_quotes_now['list_date'].fillna(20990909.0, inplace=True)
+    de_list_company_df = mongodb_util.find_all_data(db_name_constant.DE_LIST_STOCK)
+    de_list_company_df = de_list_company_df.loc[de_list_company_df['de_list_date'] < str_day]
+    real_time_quotes_now = em_stock_info_api.get_a_stock_info()
+    real_time_quotes_now = real_time_quotes_now.loc[
+        ~(real_time_quotes_now['symbol'].isin(de_list_company_df['symbol']))]
+
+    # 将list_date列中的所有NaN值设置为19890604
+    real_time_quotes_now['list_date'].fillna(19890604, inplace=True)
+    real_time_quotes_now['list_date'] = real_time_quotes_now['list_date'].replace(99990909, 19890604)
 
     # 将日期数值转换为日期时间格式
     real_time_quotes_now['list_date_01'] = pd.to_datetime(real_time_quotes_now['list_date'], format='%Y%m%d')
@@ -111,8 +128,15 @@ def multi_threaded_k_line_sync(str_day):
 
     #  exclude b symbol
     real_time_quotes_now = common_service_fun_api.exclude_b_symbol(real_time_quotes_now.copy())
-    # exclude amount==0 symbol
-    real_time_quotes_now = common_service_fun_api.exclude_amount_zero_stock(real_time_quotes_now)
+
+    real_time_quotes_now = real_time_quotes_now.loc[real_time_quotes_now['list_date_01'] <= now_date]
+
+    now_year = int(str_day[0:4])
+    last_year = now_year - 1
+
+    query_year_line = {'year': {"$in": [str(now_year), str(last_year)]}}
+    stock_qfq_year_df = mongodb_util.find_query_data(db_name_constant.STOCK_QFQ_YEAR, query_year_line)
+
     total_count = real_time_quotes_now.shape[0]
     global result
     result = pd.DataFrame()  # 重新初始化 result 变量
@@ -125,9 +149,9 @@ def multi_threaded_k_line_sync(str_day):
 
         end_count = (page + 1) * MAX_PAGE_NUMBER
         begin_count = page * MAX_PAGE_NUMBER
-        page_df = real_time_quotes_now.loc[begin_count:end_count]
+        page_df = real_time_quotes_now.iloc[begin_count:end_count]
 
-        thread = threading.Thread(target=single_threaded_sync_task, args=(page_df, str_day, page))
+        thread = threading.Thread(target=single_threaded_sync_task, args=(page_df, str_day, page, stock_qfq_year_df))
         threads.append(thread)
         thread.start()
 
@@ -141,11 +165,12 @@ def multi_threaded_k_line_sync(str_day):
 
 
 # 单线程同步任务
-def single_threaded_sync_task(page_df, str_day, page):
+def single_threaded_sync_task(page_df, str_day, page, stock_qfq_year_df):
     global result
     for stock_one in page_df.itertuples():
         try:
-            k_line_df = k_line_info_clean_impl.calculate_k_line_info(str_day, stock_one.symbol, stock_one.diff_days)
+            k_line_df = k_line_info_clean_impl.calculate_k_line_info(str_day, stock_one.symbol, stock_one.diff_days,
+                                                                     stock_qfq_year_df)
             save_k_line_data(stock_one.symbol, str_day, k_line_df)
             if k_line_df is None:
                 result = k_line_df
@@ -176,19 +201,5 @@ def create_k_line_index():
     mongodb_util.create_index('k_line_info', [("str_day", 1), ("symbol", 1)])
 
 
-def clean_history_data():
-    query = {"$and": [{"trade_date": {"$gte": '2023-11-06'}}, {"trade_date": {"$lte": '2023-11-16'}}]}
-    trade_date_list_df = mongodb_util.find_query_data('trade_date_list', query)
-    trade_date_list_df = trade_date_list_df.sort_values(by=['trade_date'], ascending=False)
-    for trade_data_one in trade_date_list_df.itertuples():
-        try:
-            sync_k_line_info_task(trade_data_one.trade_date)
-            logger.info("清洗数据到:{}", trade_data_one.trade_date)
-        except BaseException as e:
-            logger.error("发生异常:{},{}", trade_data_one.trade_date, e)
-
-
 if __name__ == '__main__':
-    sync_k_line_info("2024-06-03", None)
-    # clean_history_data()
-    # sync_k_line_info("2023-12-22")
+    sync_k_line_info_task('2025-11-13')

mns_scheduler/k_line/clean/week_month/__init__.py

@@ -2,6 +2,6 @@ import sys
 import os
 
 file_path = os.path.abspath(__file__)
-end = file_path.index('mns') + 16
+end = file_path.index('mns') + 17
 project_path = file_path[0:end]
 sys.path.append(project_path)

mns_scheduler/k_line/clean/week_month/normal_week_month_k_line_service.py

@@ -2,22 +2,27 @@ import sys
 import os
 
 file_path = os.path.abspath(__file__)
-end = file_path.index('mns') + 16
+end = file_path.index('mns') + 17
 project_path = file_path[0:end]
 sys.path.append(project_path)
 from mns_common.db.MongodbUtil import MongodbUtil
 import mns_common.utils.date_handle_util as date_handle_util
 from mns_common.component.classify.symbol_classify_param import stock_type_classify_param
+import mns_common.utils.data_frame_util as data_frame_util
+
+# import functools
 
 mongodb_util = MongodbUtil('27017')
 
 
 # 处理月线 周线 todo 暂时简单计算周线之和
-def handle_month_week_line(k_line_info, str_day, symbol, deal_days):
-    sub_stock_new_max_deal_days = stock_type_classify_param['sub_stock_new_max_deal_days']
+def handle_month_week_line(k_line_info, str_day, symbol, deal_days,
+                           stock_qfq_year_df):
+    sub_stock_new_max_deal_days = stock_type_classify_param['sub_new_stock_max_deal_days']
     if deal_days > sub_stock_new_max_deal_days:
         k_line_info = handle_month_line(k_line_info, str_day, symbol)
         k_line_info = handle_week_line(k_line_info, str_day, symbol)
+        k_line_info = set_year_k_line(k_line_info, symbol, stock_qfq_year_df, str_day)
     else:
         k_line_info['week01'] = 0
         k_line_info['week02'] = 0
@@ -33,37 +38,130 @@ def handle_month_week_line(k_line_info, str_day, symbol, deal_days):
         k_line_info['month02'] = 0
         k_line_info['month01_date'] = '19890729'
         k_line_info['month02_date'] = '19890729'
+        # 上市交易日不到100天的默认设置年线数据为0
+        k_line_info['now_year_chg'] = 0
+        k_line_info['now_year_open_from_high_chg'] = 0
+        k_line_info['now_year_low_from_high_chg'] = 0
+
+        k_line_info['last_year_chg'] = 0
+        k_line_info['last_year_open_from_high_chg'] = 0
+        k_line_info['last_year_low_from_high_chg'] = 0
+
+    return k_line_info
+
+
+# 年线数据设置
+def set_year_k_line(k_line_info, symbol, stock_qfq_year_df, str_day):
+    now_year = int(str_day[0:4])
+    last_year = str(now_year - 1)
+    stock_qfq_now_year_df = stock_qfq_year_df.loc[(stock_qfq_year_df['symbol'] == symbol)
+                                                  & (stock_qfq_year_df['year'] == str(now_year))]
+    if data_frame_util.is_empty(stock_qfq_now_year_df):
+        k_line_info['now_year_chg'] = 0
+        k_line_info['now_year_open_from_high_chg'] = 0
+        k_line_info['now_year_low_from_high_chg'] = 0
+    else:
+        k_line_info['now_year_chg'] = list(stock_qfq_now_year_df['chg'])[0]
+        k_line_info['now_year_open_from_high_chg'] = list(stock_qfq_now_year_df['open_to_high_pct'])[0]
+        k_line_info['now_year_low_from_high_chg'] = list(stock_qfq_now_year_df['low_to_high_pct'])[0]
+
+    stock_qfq_last_year_df = stock_qfq_year_df.loc[(stock_qfq_year_df['symbol'] == symbol)
+                                                   & (stock_qfq_year_df['year'] == last_year)]
+
+    if data_frame_util.is_empty(stock_qfq_last_year_df):
+        k_line_info['last_year_chg'] = 0
+        k_line_info['last_year_open_from_high_chg'] = 0
+        k_line_info['last_year_low_from_high_chg'] = 0
+    else:
+        k_line_info['last_year_chg'] = list(stock_qfq_last_year_df['chg'])[0]
+        k_line_info['last_year_open_from_high_chg'] = list(stock_qfq_last_year_df['open_to_high_pct'])[0]
+        k_line_info['last_year_low_from_high_chg'] = list(stock_qfq_last_year_df['low_to_high_pct'])[0]
+
     return k_line_info
 
+    # 处理月线
+
 
-# 处理月线
 def handle_month_line(k_line_info, str_day, symbol):
-    month_begin_day = str_day[0:7] + '-01'
+    now_year = int(str_day[0:4])
+    last_year = str(now_year - 1)
+
+    now_month_begin_day = str_day[0:7] + '-01'
+
+    last_year_begin_day = last_year + '-01-01'
+    # now_year_begin_day = str(now_year) + '-01-01'
+
     query = {"symbol": symbol,
-             'date': {"$lt": date_handle_util.no_slash_date(month_begin_day)}}
-    stock_hfq_monthly = mongodb_util.descend_query(query, 'stock_qfq_monthly', 'date', 2)
-    month_num = stock_hfq_monthly.shape[0]
-    k_line_info['month_num'] = month_num
-    if month_num > 0:
-        k_line_info['sum_month'] = round(sum(stock_hfq_monthly['chg']), 2)
-    else:
+             'date': {"$gte": date_handle_util.no_slash_date(last_year_begin_day)}}
+    stock_hfq_monthly_all = mongodb_util.find_query_data('stock_qfq_monthly', query)
+    if data_frame_util.is_empty(stock_hfq_monthly_all):
         k_line_info['sum_month'] = 0
 
-    if month_num == 0:
-        k_line_info['month01'] = 0
-        k_line_info['month02'] = 0
-        k_line_info['month01_date'] = '19890729'
-        k_line_info['month02_date'] = '19890729'
-    elif month_num == 1:
-        k_line_info['month01'] = stock_hfq_monthly.iloc[0].chg
-        k_line_info['month02'] = 0
-        k_line_info['month01_date'] = stock_hfq_monthly.iloc[0].date
-        k_line_info['month02_date'] = '19890729'
-    elif month_num == 2:
-        k_line_info['month01'] = stock_hfq_monthly.iloc[0].chg
-        k_line_info['month02'] = stock_hfq_monthly.iloc[1].chg
-        k_line_info['month01_date'] = stock_hfq_monthly.iloc[0].date
-        k_line_info['month02_date'] = stock_hfq_monthly.iloc[1].date
+    else:
+        stock_hfq_monthly_all = stock_hfq_monthly_all.sort_values(by=['date'], ascending=False)
+        stock_hfq_monthly_all = stock_hfq_monthly_all.loc[
+            stock_hfq_monthly_all['date'] <= date_handle_util.no_slash_date(now_month_begin_day)]
+
+        # stock_hfq_monthly_last_year = stock_hfq_monthly_all.loc[
+        #     stock_hfq_monthly_all['date'] < date_handle_util.no_slash_date(now_year_begin_day)]
+        #
+        # stock_hfq_monthly_now_year = stock_hfq_monthly_all.loc[
+        #     stock_hfq_monthly_all['date'] > date_handle_util.no_slash_date(now_year_begin_day)]
+
+        # 最近两个月k线
+        before_two_month_stock_hfq_monthly = stock_hfq_monthly_all.iloc[0:2]
+        month_num = before_two_month_stock_hfq_monthly.shape[0]
+        k_line_info['month_num'] = month_num
+
+        if month_num == 0:
+            k_line_info['sum_month'] = 0
+            k_line_info['month01'] = 0
+            k_line_info['month02'] = 0
+            k_line_info['month01_date'] = '19890729'
+            k_line_info['month02_date'] = '19890729'
+        elif month_num == 1:
+            k_line_info['month01'] = before_two_month_stock_hfq_monthly.iloc[0].chg
+            k_line_info['month02'] = 0
+            k_line_info['month01_date'] = before_two_month_stock_hfq_monthly.iloc[0].date
+            k_line_info['month02_date'] = '19890729'
+            k_line_info['sum_month'] = before_two_month_stock_hfq_monthly.iloc[0].chg
+        elif month_num == 2:
+            k_line_info['month01'] = before_two_month_stock_hfq_monthly.iloc[0].chg
+            k_line_info['month02'] = before_two_month_stock_hfq_monthly.iloc[1].chg
+            k_line_info['month01_date'] = before_two_month_stock_hfq_monthly.iloc[0].date
+            k_line_info['month02_date'] = before_two_month_stock_hfq_monthly.iloc[1].date
+            close_price = before_two_month_stock_hfq_monthly.iloc[0].close
+            open_price = before_two_month_stock_hfq_monthly.iloc[1].last_price
+            sum_chg = round((close_price - open_price) * 100 / open_price, 2)
+            k_line_info['sum_month'] = sum_chg
+        #
+        # last_year_month_number = stock_hfq_monthly_last_year.shape[0]
+        # if last_year_month_number == 0:
+        #     k_line_info['last_year_chg'] = 0
+        # elif last_year_month_number == 1:
+        #     k_line_info['last_year_chg'] = stock_hfq_monthly_last_year.iloc[0].chg
+        # else:
+        #     # chg_list = list(stock_hfq_monthly_last_year['chg'])
+        #     # # 将列表中的每个元素加上1
+        #     # updated_list = [round((x / 100) + 1, 2) for x in chg_list]
+        #     #
+        #     # # 使用 functools.reduce 将列表中的所有元素相乘
+        #     # last_year_chg = functools.reduce(lambda x, y: x * y, updated_list)
+        #     close_price = stock_hfq_monthly_last_year.iloc[0].close
+        #     open_price = stock_hfq_monthly_last_year.iloc[last_year_month_number - 1].last_price
+        #     last_year_chg = round((close_price - open_price) * 100 / open_price, 2)
+        #     k_line_info['last_year_chg'] = last_year_chg
+        #
+        # now_year_month_number = stock_hfq_monthly_now_year.shape[0]
+        # if now_year_month_number == 0:
+        #     k_line_info['now_year_chg'] = 0
+        # elif now_year_month_number == 1:
+        #     k_line_info['now_year_chg'] = stock_hfq_monthly_now_year.iloc[0].chg
+        # else:
+        #     close_price = stock_hfq_monthly_now_year.iloc[0].close
+        #     open_price = stock_hfq_monthly_now_year.iloc[now_year_month_number - 1].last_price
+        #     last_year_chg = round((close_price - open_price) * 100 / open_price, 2)
+        #     k_line_info['now_year_chg'] = last_year_chg
 
     return k_line_info