def get_zz500s():
"""
获取中证500成份股
Return
--------
DataFrame
code :股票代码
name :股票名称
"""
from webdata.stock.fundamental import get_stock_basics
try:
# df = pd.read_excel(ct.HS300_CLASSIFY_URL_FTP%(ct.P_TYPE['ftp'], ct.DOMAINS['idxip'],
# ct.PAGES['zz500b']), parse_cols=[0,1])
# df.columns = ct.FOR_CLASSIFY_B_COLS
# df['code'] = df['code'].map(lambda x :str(x).zfill(6))
wt = pd.read_excel(
ct.HS300_CLASSIFY_URL_FTP %
(ct.P_TYPE['ftp'], ct.DOMAINS['idxip'], ct.PAGES['zz500wt']),
parse_cols=[0, 3, 6])
wt.columns = ct.FOR_CLASSIFY_W_COLS
wt['code'] = wt['code'].map(lambda x: str(x).zfill(6))
df = get_stock_basics()[['name']]
df = df.reset_index()
return pd.merge(df, wt)
except Exception as er:
print(str(er))
def get_gem_classified():
"""
获取创业板股票
Return
--------
DataFrame
code :股票代码
name :股票名称
"""
df = fd.get_stock_basics()
df.reset_index(level=0, inplace=True)
df = df[ct.FOR_CLASSIFY_B_COLS]
df = df.ix[df.code.str[0] == '3']
df = df.sort('code').reset_index(drop=True)
return df
def get_area_classified():
"""
获取地域分类数据
Return
--------
DataFrame
code :股票代码
name :股票名称
area :地域名称
"""
df = fd.get_stock_basics()
df = df[['name', 'area']]
df.reset_index(level=0, inplace=True)
df = df.sort('area').reset_index(drop=True)
return df
def get_st_classified():
"""
获取风险警示板股票
Return
--------
DataFrame
code :股票代码
name :股票名称
"""
df = fd.get_stock_basics()
df.reset_index(level=0, inplace=True)
df = df[ct.FOR_CLASSIFY_B_COLS]
df = df.ix[df.name.str.contains('ST')]
df = df.sort('code').reset_index(drop=True)
return df
def get_sme_classified():
"""
获取中小板股票
Return
--------
DataFrame
code :股票代码
name :股票名称
"""
df = fd.get_stock_basics()
df.reset_index(level=0, inplace=True)
df = df[ct.FOR_CLASSIFY_B_COLS]
df = df.ix[df.code.str[0:3] == '002']
df = df.sort_values(by='code').reset_index(drop=True)
return df
def get_open_h_hdf5(code, h5):
"""
获取个股全部历史交易数据
"""
if code[0] == '0' or code[0] == '3' or code[0] == '2':
label = 'M/sz' + code
elif code[0] == '6' or code[0] == '9':
label = 'M/ss' + code
try:
df = h5[label]
tem = str(df.index[-1])[0:10]
if tem < today:
#if datetime.datetime.today().isoweekday() in [1,2,3,4,5]:
t = time.strptime(tem, '%Y-%m-%d')
y, m, d = t[0:3]
tt = datetime.datetime(y, m, d)
bd = tt + datetime.timedelta(days=1)
bday = bd.strftime('%Y-%m-%d')
df1 = wt.get_h_data(code, start=bday, end=today)
df = df.append(df1)
#df.index=pd.to_datetime(df.index)
#df=df.sort_index(ascending=True)
#print(df)
h5[label] = df
except:
tem = wf.get_stock_basics()
date = tem.loc[code]['timeToMarket']
t = time.strptime(str(date), '%Y%m%d')
startt = time.strftime('%Y-%m-%d', t)
df = wt.get_h_data(code, start=startt, end=today)
if df is not None:
#df.index=pd.to_datetime(df.index)
#df=df.sort_index(ascending=True)
h5[label] = df
finally:
pass
return df
def get_hs300s():
"""
获取沪深300当前成份股及所占权重
Return
--------
DataFrame
code :股票代码
name :股票名称
date :日期
weight:权重
"""
from webdata.stock.fundamental import get_stock_basics
try:
wt = pd.read_excel(
ct.HS300_CLASSIFY_URL_FTP %
(ct.P_TYPE['ftp'], ct.DOMAINS['idxip'], ct.PAGES['hs300w']),
parse_cols=[0, 3, 6])
wt.columns = ct.FOR_CLASSIFY_W_COLS
wt['code'] = wt['code'].map(lambda x: str(x).zfill(6))
df = get_stock_basics()[['name']]
df = df.reset_index()
return pd.merge(df, wt)
except Exception as er:
print(str(er))
def get_h_csv(code, index=False, autype='qfq'):
"""
获取历史复权数据，分为前复权和后复权数据，接口提供股票上市以来所有历史数据，
默认为前复权。如果不设定开始和结束日期，则返回近一年的复权数据，从性能上考虑，
推荐设定开始日期和结束日期，而且最好不要超过三年以上，获取全部历史数据，
请分年段分步获取，取到数据后，请及时在本地存储。
index:False(提取非指数的数据;True(提取指数的数据)
autype：主要是提取复权类型，None 表示不复权；qfq 前复权；hfq 后复权
"""
if index:
if code[0] == '0':
tick = 'SS_' + code
else:
tick = 'SZ_' + code
fn = './stockdata/data/history/' + tick + '.csv'
if not os.path.exists(fn):
df = wt.get_h_data(code, index=True, start='1995-01-01', end=today)
df = df.sort_index(ascending=True)
df.to_csv(fn)
else:
df = pd.read_csv(fn, index_col='date')
tem = str(df.index[-1])[0:10]
#dftem=dftem.set_index('date')
if tem < bftoday:
print('\nUpdating data from %s for %s:' % (tem, code))
t = time.strptime(tem, "%Y-%m-%d")
y, m, d = t[0:3]
tt = datetime.datetime(y, m, d)
bd = tt + datetime.timedelta(days=1)
if bd.weekday() == 5:
bd = bd + datetime.timedelta(days=2)
bday = bd.strftime('%Y-%m-%d')
all_data = wt.get_h_data(code,
index=True,
start=bday,
end=today)
if all_data is not None:
all_data = all_data.sort_index(ascending=True)
all_data.to_csv(fn, header=None, mode='a')
df = df.append(all_data)
df.index = pd.to_datetime(df.index)
else:
h5path = './stockdata/data/history/' + code + '.csv'
if not os.path.exists(h5path):
tem = wf.get_stock_basics()
date = tem.loc[code]['timeToMarket']
t = time.strptime(str(date), '%Y%m%d')
startt = time.strftime('%Y-%m-%d', t)
df = wt.get_h_data(code, autype=autype, start=startt, end=today)
df.to_csv(h5path)
else:
df = pd.read_csv(h5path, index_col='date')
tem = str(df.index[-1])[0:10]
if tem < today:
#if datetime.datetime.today().isoweekday() in [1,2,3,4,5]:
t = time.strptime(tem, '%Y-%m-%d')
y, m, d = t[0:3]
tt = datetime.datetime(y, m, d)
bd = tt + datetime.timedelta(days=1)
bday = bd.strftime('%Y-%m-%d')
df1 = wt.get_h_data(code, autype=autype, start=bday, end=today)
if df1 is not None:
df1 = df1.sort_index(ascending=True)
df = df.append(df1)
df1.to_csv(h5path, mode='a', header=None)
df.index = pd.to_datetime(df.index)
return df
