def GetTopVolume():
cssSelector = '#divStockList'
url = f'https://goodinfo.tw/tw/StockList.asp?RPT_TIME=&MARKET_CAT=熱門排行&INDUSTRY_CAT=日成交張數創近期新高日數@@成交張數@@日成交張數創近期新高日數'
print(url)
try:
df = Utils.GetDataFrameByCssSelector(url, cssSelector)
#return df
except:
time.sleep(random.randint(20, 30))
df = Utils.GetDataFrameByCssSelector(url, cssSelector)
print(df)
#df.columns = df.columns.get_level_values(1)
df.columns = df.columns.get_level_values(0)
df = df.drop_duplicates(keep=False, inplace=False)
#gain = pd.to_numeric(df['漲跌 價'], errors='coerce') > 0
#market = df['市 場'] == '市'
length = df['代號'].astype(str).map(len) == 4
#df = df[gain & length]
df = df[length]
df.to_csv(f'{Utils.GetRootPath()}\Data\Daily\日成交張數創近期新高日數.csv',
encoding='utf_8_sig')
return df['代號'].values
def GetFinData(stockId):
url = f'https://goodinfo.tw/StockInfo/StockFinDetail.asp?RPT_CAT=XX_M_QUAR_ACC&STOCK_ID={stockId}'
css_selector = '#txtFinBody'
try:
df = Utils.GetDataFrameByCssSelector(url, css_selector)
except:
time.sleep(random.randint(20, 30))
df = Utils.GetDataFrameByCssSelector(url, css_selector)
#print(df)
return df
def GetDividend(stockId):
url = f'https://goodinfo.tw/tw/StockDividendPolicy.asp?STOCK_ID={stockId}'
cssSelector = '#divDetail'
try:
df = Utils.GetDataFrameByCssSelector(url, cssSelector)
df.columns = df.columns.get_level_values(3)
except:
time.sleep(random.randint(20, 30))
df = Utils.GetDataFrameByCssSelector(url, cssSelector)
df.columns = df.columns.get_level_values(3)
# column replace space
df.columns = df.columns.str.replace(' ', '')
# filter not ∟
df = df[df['股利發放年度'] != '∟']
#print(df)
# 年度大於2022, 移除第一列
firstRow = df.iloc[0, :]
if int(firstRow['股利發放年度']) > datetime.now().year:
df = df.iloc[1: , :]
rowsCount = 5
# 年度(取前5筆, index重新排序)
year = pd.to_numeric(df.iloc[:, 0], errors='coerce').dropna(how='any',axis=0).head(rowsCount).astype(int).reset_index(drop=True)
#print(year)
# 現金(取前5筆, index重新排序)
cash = pd.to_numeric(df.iloc[:, 3], errors='coerce').dropna(how='any',axis=0).head(rowsCount).reset_index(drop=True)
#print(cash)
# 股票(取前5筆, index重新排序)
stock = pd.to_numeric(df.iloc[:, 6], errors='coerce').dropna(how='any',axis=0).head(rowsCount).reset_index(drop=True)
#print(stock)
data = []
for index in range(0, rowsCount):
data.append(str(cash[index]).rjust(6) + ' / ' + str(stock[index]).rjust(6))
print(data)
df = pd.DataFrame([data], columns=year)
return df
def GetAllDividend():
cssSelector = '#divStockList'
for rankIndex in range(0, 6):
url = f'https://goodinfo.tw/tw/StockList.asp?SHEET=股利政策&MARKET_CAT=熱門排行&INDUSTRY_CAT=合計股利&RANK={str(rankIndex)}'
print(url)
# 休息10~20秒
time.sleep(random.randint(10, 20))
try:
df = Utils.GetDataFrameByCssSelector(url, cssSelector)
#return df
except:
time.sleep(random.randint(20, 30))
df = Utils.GetDataFrameByCssSelector(url, cssSelector)
print(df)
#df.columns = df.columns.get_level_values(1)
df.columns = df.columns.get_level_values(0)
df = df.drop_duplicates(keep=False, inplace=False) #移除重複標題
#gain = pd.to_numeric(df['漲跌 價'], errors='coerce') > 0
#market = df['市 場'] == '市'
print(df)
length = df['代號'].astype(str).map(len) == 4
#df = df[gain & length]
df = df[length]
filePath = f'{Utils.GetRootPath()}\Data\Yearly\合計股利.csv'
if rankIndex == 0:
df.to_csv(filePath, encoding='utf_8_sig')
else:
df.to_csv(filePath, mode='a', header=False, encoding='utf_8_sig')
# 去除重複標頭
#sum_df[sum_df.ne(sum_df.columns).any(1)].to_csv(f'{Utils.GetRootPath()}\Data\Monthly\董監持股比例.csv',encoding='utf_8_sig')
print('執行完成')
def WriteData():
cssSelector = '#divStockList'
sum_df = pd.DataFrame()
for rankIndex in range(0, 5):
url = f'https://goodinfo.tw/tw/StockList.asp?SHEET=董監持股&MARKET_CAT=熱門排行&INDUSTRY_CAT=全體董監持股比例&RANK={str(rankIndex)}'
print(url)
try:
time.sleep(random.randint(5, 10))
df = Utils.GetDataFrameByCssSelector(url, cssSelector)
print(df)
sum_df = pd.concat([sum_df, df], axis=0)
#df.columns = df.columns.get_level_values(1)
except:
time.sleep(random.randint(20, 30))
df = Utils.GetDataFrameByCssSelector(url, cssSelector)
print(df)
#df.columns = df.columns.get_level_values(1)
# 去除重複標頭
sum_df[sum_df.ne(sum_df.columns).any(1)].to_csv(
f'{Utils.GetRootPath()}\Data\Monthly\董監持股比例.csv', encoding='utf_8_sig')
def GetPE(stockId):
url = f'https://goodinfo.tw/StockInfo/ShowK_ChartFlow.asp?RPT_CAT=PER&STOCK_ID={stockId}&CHT_CAT=WEEK'
css_selector = '#divK_ChartFlowDetail'
try:
df = Utils.GetDataFrameByCssSelector(url, css_selector)
# 取前兩列後面倒數6欄資料
firtRowDf = df.iloc[0,-6:]
#print(firtRowDf)
except:
time.sleep(random.randint(20, 30))
df = Utils.GetDataFrameByCssSelector(url, css_selector)
# 取前兩列後面倒數6欄資料
firtRowDf = df.iloc[0,-6:]
#print(firtRowDf)
#dataframe轉成dictionary 參考 https://stackoverflow.com/questions/45452935/pandas-how-to-get-series-to-dict
dictionaries = [dict(key=re.findall(r'[0-9]+[.]?[0-9]*', str(k))[0], value=v) for k, v in firtRowDf.items()]
#print(data)
# 轉換成dataframe
data = []
headers = ['本益比-級距1倍數', '本益比-級距1價格',
'本益比-級距2倍數', '本益比-級距2價格',
'本益比-級距3倍數', '本益比-級距3價格',
'本益比-級距4倍數', '本益比-級距4價格',
'本益比-級距5倍數', '本益比-級距5價格',
'本益比-級距6倍數', '本益比-級距6價格']
for entry in dictionaries:
#print(entry)
data.append(entry['key'])
data.append(entry['value'])
##print(headers)
#print(data)
df = pd.DataFrame([data], columns=headers)
return df
def GetTransaction(stockId):
url = f'https://goodinfo.tw/tw/ShowK_Chart.asp?STOCK_ID={stockId}&CHT_CAT2=DATE'
cssSelector = '#divPriceDetail'
try:
df = Utils.GetDataFrameByCssSelector(url, cssSelector)
df.columns = df.columns.get_level_values(1)
except:
time.sleep(random.randint(20, 30))
df = Utils.GetDataFrameByCssSelector(url, cssSelector)
df.columns = df.columns.get_level_values(1)
# 印出全部的rows
#pd.set_option('display.max_rows', df.shape[0]+1)
#print(df)
headers = ['收盤', '張數', '外資 持股 (%)', '券資 比 (%)']
smaPeroids = [1, 5, 20, 60]
dict = {}
for header in headers:
try:
#print(header)
entry = ''
for period in smaPeroids:
#print(df[header])
data = pd.to_numeric(df[header], errors='coerce').dropna(
how='any', axis=0).head(period)
#print(data)
sma = round(data.mean(), 2)
#print(sma)
entry += ('' if entry == '' else ' / ') + str(sma).rjust(8)
#print(header.replace(' ', ''))
#print(entry)
if header == '收盤':
data = [x.strip() for x in entry.split('/')]
prefixIcon = ''
if float(data[0]) > float(data[1]) and float(data[0]) > float(
data[2]):
prefixIcon = '👍'
elif float(data[0]) < float(data[3]):
prefixIcon = '👎'
entry = prefixIcon + entry
# 成交量 > 5ma 3倍
if header == '張數':
data = [x.strip() for x in entry.split('/')]
if (float(data[0]) / float(data[1]) > 3.0):
entry = '🏆' + entry
dict.update({
header.replace(' ', '') + '(' + 'ma / '.join(
map(str, smaPeroids)) + 'ma)':
str(entry)
})
except:
dict.update({
header.replace(' ', '') + '(' + 'ma / '.join(
map(str, smaPeroids)) + 'ma)':
''
})
#print(dict)
result = pd.DataFrame([dict])
return result