#!/usr/bin/env python
# coding: utf-8

# # Part 0: Import package

# In[1]:


import requests as r
import json
import pandas as pd
from datetime import datetime, date
import matplotlib.pyplot as plt
get_ipython().run_line_magic('matplotlib', 'inline')


# # Part 1: 下載一個月台股個股成交資料

# In[2]:


url = "https://www.twse.com.tw/exchangeReport/STOCK_DAY?response=json&date=20200801&stockNo=2330"


# In[3]:


r.get(url)


# In[4]:


res = r.get(url)


# In[5]:


res.json()


# In[6]:


stock_json = res.json()


# In[7]:


stock_json['data']


# In[8]:


pd.DataFrame.from_dict(stock_json['data'])


# In[9]:


stock_df = pd.DataFrame.from_dict(stock_json['data'])
stock_df.head()


# In[10]:


stock_json['fields']


# In[11]:


stock_df.columns = stock_json['fields']


# In[12]:


stock_df.head()


# # Part 2: 下載歷年台股個股成交資料

# In[13]:


url = "https://www.twse.com.tw/exchangeReport/STOCK_DAY?response=json&date=20200801&stockNo=2330"


# In[14]:


month_list = pd.date_range('2020-06-01','2020-08-01', freq='MS').strftime("%Y%m%d").tolist()    


# In[15]:


for month in month_list:
    print(month) 


# In[16]:


df = pd.DataFrame()
df


# In[17]:


for month in month_list:
    url = "https://www.twse.com.tw/exchangeReport/STOCK_DAY?response=json&date="+ month + "&stockNo=2330"
    res = r.get(url)
    stock_json = res.json()
    stock_df = pd.DataFrame.from_dict(stock_json['data'])
    df = df.append(stock_df, ignore_index = True)


# In[18]:


df.columns = ['日期', '成交股數', '成交金額', '開盤價', '最高價', '最低價', '收盤價', '漲跌價差', '成交筆數']


# In[19]:


df.head()


# In[20]:


def get_stock_data(start_year, start_month, end_year, end_month, stock_code):
    start_date = str(date(start_year, start_month, 1))
    end_date = str(date(end_year, end_month, 1))
    month_list = pd.date_range(start_date, end_date, freq='MS').strftime("%Y%m%d").tolist()
    
    df = pd.DataFrame()
    for month in month_list:
        url = "https://www.twse.com.tw/exchangeReport/STOCK_DAY?response=json&date="+ month + "&stockNo=" + str(stock_code)
        res = r.get(url)
        stock_json = res.json()
        stock_df = pd.DataFrame.from_dict(stock_json['data'])
        df = df.append(stock_df, ignore_index = True)
    
    df.columns = ['日期', '成交股數', '成交金額', '開盤價', '最高價', '最低價', '收盤價', '漲跌價差', '成交筆數']
    return df


# In[21]:


get_stock_data(start_year = 2019, start_month = 6, end_year = 2019, end_month = 8, stock_code = 2330)


# # Part 3: 將下載的個股資料從字串(string)轉型成日期(datetime)及浮點數(float)

# In[22]:


stock = get_stock_data(start_year = 2019, start_month = 6, end_year = 2019, end_month = 8, stock_code = 2330)
stock.head()


# ### 把日期從字串(string)換成時間(datetime)，並將民國年換成西元年

# In[23]:


for row in range(stock.shape[0]):
    date2 = stock.iloc[row,0].split('/')
    stock.iloc[row, 0] = datetime(int(date2[0]) + 1911, int(date2[1]), int(date2[2]))
        
stock.head(10)


# ### 把"成交股數" "成交金額" "開盤價", "最高價", "最低價", "收盤價" "漲跌價差" "成交筆數"帶有逗號的字串(string)換成浮點數(float)

# In[26]:


for col in [1, 2, 3, 4, 5, 6, 8]:
    for row in range(stock.shape[0]):
        stock.iloc[row, col] = float(stock.iloc[row,col].replace(',', ''))
        
stock.head(10)


# # Part 4: 將下載的個股資料另存成csv檔

# In[27]:


stock.to_csv("2330.csv")


# # Part 5: 畫出下載個股的走勢圖

# In[28]:


stock2 = stock[:10]


# In[29]:


stock2 


# In[30]:


fig = plt.figure(figsize = (10, 5))
plt.title('2330 Stock Price')
plt.plot(stock2['日期'], stock2['收盤價'])
plt.plot(stock2['日期'], stock2['開盤價'])
plt.legend(['Close', 'Open'])


# In[ ]: