def get_all_news(stock_index, save_dir, end_year):
#stk = '000917'
make_dir(save_dir)
k = 1
old_date = ''
for i in range(1, 100):
content1 = html_content(stock_index, i).find_url_content()
if len(content1) > 0:
#k=0
for url in content1:
#k += 1
with eventlet.Timeout(5.5, False):
date, text_out = news_in_html_url(url)
if date[0:4] == str(end_year):
sys.exit(0)
if date == old_date:
k += 1
if date != old_date:
k = 1
saveFileTxt = saveFile(stock_index=stock_index,
save_dir=save_dir,
k=k,
date=date).news_report_save_dir()
content_to_txt(saveFileTxt, text_out)
old_date = date
time.sleep(30)
else:
break ## if no content break
def download_VG_many (cart_id, map_name, storing_name, number):
# Spesify where to store the files
directory = "../Downloaded_cartoons/" + map_name
functions.make_dir(directory) # Function that check if the directory exists. If not - create it
for i in range(number):
# Make a new url
date_ = date.today() - timedelta(int(i+1))
date_ = str(date_)
td = date_.split('-')
url_string_array = ["http://www.heltnormalt.no/img/",cart_id,td[0],"/",td[1],"/",td[2],".jpg"]
picture_url = ''.join(url_string_array)
# Make a new name for the file with timestamp
name_string_array = [directory,storing_name,"_",date_,"_VG",".jpg"]
tag = ''.join(name_string_array)
# Downlad the picture
if (functions.download_file(picture_url, tag) == 0):
break
if ((i+1)%10 == 0):
print ("finished with picture # " , i+1)
print ("*** Finished " + storing_name + ". Downloaded " + str(i) + " pictures")
return
def do_download(stock_index, year1, season1):
#print(stock_index)
dir_name1 = os.path.join(dir_day_history_wy, stock_index)
make_dir(dir_name1)
try:
df1 = download_data(stock_index, year1, season1)
if df1.shape[0] > 1:
file_name = save_file_name(dir_name1, stock_index, str(year1),
str(season1))
df1.to_csv(file_name, index=0)
except:
print(stock_index + " not download")
pass
def get_dazong_stock_list(dazong_date = "2020-02-20"):
data_dir = data_dict.get("dazongjiaoyi")
#dazong_date = "2020-02-20"
file_name = "dazongjiaoyi_%s.csv"%(dazong_date)
df1 = pd.read_csv(os.path.join(data_dir,file_name))
df2 = df1.groupby(["SECUCODE","SNAME"])['TVAL'].sum().reset_index()
#df2.sort_values("TVAL")
df2['SECUCODE'] = [str(x).zfill(6) for x in df2['SECUCODE'].tolist() ]
df2['trade_date'] = dazong_date
save_dir = data_dict.get("tmp")
save_dir = os.path.join(save_dir,"dazong_data")
make_dir(save_dir)
save_file = "dazongjiaoyi_stock_list_%s.csv"%(dazong_date)
save_file = os.path.join(save_dir,save_file)
df3 = df2[df2['TVAL']>1000]
df2[['SECUCODE','trade_date','TVAL']].to_csv(save_file,index=0)
def do_loop_download(stock_index):
for stock_index in stk_index_list:
print(stock_index)
dir_name1 = os.path.join(dir_day_history_wy,stock_index)
make_dir(dir_name1)
for year1 in year:
for season1 in season:
try:
df1 = download_data(stock_index,year1,season1)
if df1.shape[0]>1:
file_name = save_file_name(dir_name1,stock_index,str(year1),str(season1))
#print(file_name)
df1.to_csv(file_name,index=0)
except:
pass
time.sleep(5)
df3 = df2.groupby(["SNAME", "SECUCODE", "BUYERNAME", "Zyl",
"PRICE"])['TVAL'].sum().reset_index()
df_sname = df2.groupby(["SNAME", "SECUCODE"])['TVAL'].sum().reset_index()
df4 = sort_data(df3,
["SNAME", "SECUCODE", "BUYERNAME", "Zyl", "PRICE", "TVAL"])
df5 = sort_data(df_sname, ["SNAME", "SECUCODE", "TVAL"])
df4['stock_date'] = now_date
df5['stock_date'] = now_date
#df4 = df3.sort_values("TVAL",ascending=False)[["SNAME","SECUCODE","BUYERNAME","Zyl","PRICE","TVAL"]]
#df4['SECUCODE'] = [str(x).zfill(6) for x in df4['SECUCODE'].tolist()]
print(df4.head(30))
print(df5.head(30))
return df4, df5
if __name__ == "__main__":
now_date, now_date_time = get_the_datetime()
now_date = now_date.replace("_", "-")
now_date = "2020-08-03"
data_dir = data_dict.get("dazongjiaoyi")
file_in = "dazongjiaoyi_%s.csv" % (now_date)
df1 = pd.read_csv(os.path.join(data_dir, file_in))
report_dir = data_dict.get("daily_report")
save_dir = os.path.join(report_dir, now_date)
make_dir(save_dir)
df4, df5 = zyl_postive(df1, now_date)
out_file = "dazongjiaoyi_report_%s.csv" % (now_date)
file_name = os.path.join(save_dir, out_file)
df4.round(2).to_csv(file_name, index=0)
def save_to_daily_report(self):
report_dir = data_dict.get("daily_report")
save_dir = os.path.join(report_dir, self.now_date)
make_dir(save_dir)
return save_dir
df1['stock_index'] = stock_index
return df1
def save_file_name(dir_name,stock_index,year,season):
file_name1 = '_'.join([stock_index,year,season])
file_name = os.path.join(dir_name,file_name1)+'.csv'
return file_name
dir_day_history_wy = data_dict.get("day_history_wangyi")
stk_index_list = stk_index_list_gen()
year = [str(x) for x in range(1998,2020)]
season = [str(x) for x in [1,2,3,4]]
for stock_index in stk_index_list:
print(stock_index)
dir_name1 = os.path.join(dir_day_history_wy,stock_index)
make_dir(dir_name1)
for year1 in year:
for season1 in season:
try:
df1 = download_data(stock_index,year1,season1)
if df1.shape[0]>1:
file_name = save_file_name(dir_name,stock_index,year,season)
#print(file_name)
df1.to_csv(file_name,index=0)
except:
pass
time.sleep(3)
'''
stock_index = '601398'
year = 1880
season = 2