from jqdata import *
import datetime
import pandas as pd
import numpy as np
import re
from collections import Counter

today_date = str(datetime.datetime.now()).split(' ')[0]

# Get the trading days calendar list
td_list = list(get_all_trade_days())

# Get current stock existing as of date specified
get_stock_list = get_all_securities(date=today_date)
stock_index = get_stock_list.index

# Iterating counter that will be used in DataFrame
k=0

# check if today is trading day
# inititialize day counter
d0 = 0

for i in range(len(td_list)):
    if str(td_list[i]) == today_date:
        d0 = i
        print('今天为交易日，日期为：', td_list[d0], '\n')
if d0 == 0:
    for i in range(20):
        date = str((datetime.datetime.now() + datetime.timedelta(days=-i)).strftime("%Y-%m-%d %H:%M:%S")).split(' ')[0]
        for j in range(len(td_list)):
            if str(td_list[j]) == date:
                d0 = j
                print('今日为非交易日,输出上一交易日数据，上一交易日为', td_list[d0], '\n')
        if d0 != 0:
            break
        # Check if today's trading data is updated by database
start_date = td_list[d0]
end_date = td_list[d0]
check_data = get_money_flow('000001.XSHE', start_date, end_date, ["date", "sec_code", "change_pct"])
if len(check_data) == 0:
    start_date = td_list[d0 - 4]
    end_date = td_list[d0 - 1]
    print('今日数据尚无，输出上交易数据,上一交易日为：', end_date, '\n')
else:
    start_date = td_list[d0 - 3]
    end_date = td_list[d0]
# create a dataframe here first to prepare for further analysis

df = pd.DataFrame(np.zeros([9,len(stock_index)]), index=['当日涨停','连板','今日涨幅','上交易日涨幅','行业','概念','市值（亿）','市盈','市净'])

# Iterate all the stocks
for i in range(len(stock_index)):
    # Get rise and fall of all stock
    get_stock_rise_fall = get_money_flow(stock_index[i], start_date, end_date, ["date", "sec_code", "change_pct"])
    # Get the stock with rise more than 9.9 on the end date, also make sure not empty
    if len(get_stock_rise_fall.change_pct) == 4 and get_stock_rise_fall.change_pct[3] >= 9.9 and \
            get_stock_rise_fall.change_pct[3] < 10.1:
        # check for consecutive rise limit
        count = 0
        for j in range(len(get_stock_rise_fall.change_pct)):
            if get_stock_rise_fall.change_pct[j] >= 9.9 and get_stock_rise_fall.change_pct[j] < 10.1:
                count += 1
                # Get industry of the stock
        industry = get_industry(stock_index[i], date=end_date)
        concept = get_concept(stock_index[i], date=str(end_date))
        industry_list = []
        concept_list = []
        for item1 in industry[stock_index[i]]:
            industry_list.append(industry[stock_index[i]][item1]['industry_name'])
        for item2 in concept[stock_index[i]]['jq_concept']:
            concept_list.append(item2['concept_name'])
        # query for company info
        q = query(
            valuation
        ).filter(
            valuation.code == stock_index[i]
        )
        fundamentals = get_fundamentals(q, end_date)

        pe = fundamentals['pe_ratio'][0]
        pb = fundamentals['pb_ratio'][0]
        value = fundamentals['market_cap'][0]
        print('当日涨停: ', stock_index[i], '\n',
              '连板:', count, '今日涨幅: ', get_stock_rise_fall.change_pct[3], '上交易日涨幅:', get_stock_rise_fall.change_pct[2],
              '\n',
              '行业:', industry_list, '\n',
              '概念:', concept_list, '\n',
              '市值（亿）：', value, '市盈：', pe, '市净：', pb, '\n'
              )
# write data into DataFrame        
        df.iloc[0,k] = stock_index[i]
        df.iloc[1,k] = count
        df.iloc[2,k] = get_stock_rise_fall.change_pct[3]
        df.iloc[3,k] = get_stock_rise_fall.change_pct[2]
        df.iloc[4,k] = str(industry_list)
        df.iloc[5,k] = str(concept_list)
        df.iloc[6,k] = value
        df.iloc[7,k] = pe
        df.iloc[8, k] = pb
        k = k+1
print('\n', '当日涨停板输出完成', '\n')

print('\n', '开始涨停板特征分析', '\n')

# draw concept & industry data from DataFrame

industry_str =str(df.iloc[4,:])
concept_str =str(df.iloc[5,:])

# string processing, remove punctuation and letters

industry_str = industry_str.split(' ')
ind_list=[]
for i in range(len(industry_str)):
    industry_str[i] = re.sub("[A-Za-z0-9\!\%\[\]\,\。 \n-:,，、]", "", industry_str[i])
    if industry_str[i] != '':
        ind_list.append(industry_str[i])

# first tag counting
c = Counter(ind_list)
industry1=c.most_common(1)[0][0]
for item in ind_list:
    if item == industry1:
       ind_list.remove(industry1)

# second tag counting   
c = Counter(ind_list)
industry2=c.most_common(1)[0][0]
print('第一行业特征为（最多tag匹配）：',industry1,'\n','第二行业特征为（第二最多tag匹配）：',industry2, '\n')

# similiar processing, tag counting, etc.
concept_str=concept_str.split(' ')
con_list=[]
for i in range(len(concept_str)):
    concept_str[i] = re.sub("[A-Za-z0-9\!\%\[\]\,\。 \n-:,，、]", "", concept_str[i])
    if concept_str[i] != '':
        con_list.append(concept_str[i])
c = Counter(con_list)
concept1=c.most_common(1)[0][0]

for item in con_list:
    if item == concept1:
       con_list.remove(concept1)
c = Counter(con_list)
concept2=c.most_common(1)[0][0]
print('第一概念特征为（最多tag匹配）：',concept1,'\n','第二概念特征为（第二最多tag匹配）：',concept2,'\n')