"Financial Summary" 정보를 크롤링하는 법을 다뤘었다.
PER(배), PBR(배), 시가총액, 매출액, 유동자산, 부채총계, 자본총계, ROA(%), 당기 순이익, 매출총이익, 자산총액
Measures naver finance 에 matching naver finance tab 적용 기법
PER
PER(배)
기업현황
NCAV, 3P Combo
PBR
PBR(배)
기업현황
Graham, 3P Combo, PBR + GP/A
PSR
시가총액 / 매출액
기업현황
3P Combo
Market cap
시가총액(억)
기업현황
NCAV
Current assets
유동자산
재무분석 (재무상태표)
NCAV
Total liabilities
부채총계(억)
기업현황
NCAV
Total debt/equity
부채총계 / 자본총계
기업현황
Graham
ROA
ROA(%)
기업현황
Graham
Net income
당기 순이익
기업현황
NCAV
GP/A
매출총이익 / 자산총액
재무분석 (포괄손익계산서)
PBR + GP/A
크롤링 결과 (Financial Summary)
PER(배), PBR(배), 매출액, ROA(%), 부채총계, 자본총계, 당기순이익, 자산총계
시가총액 데이터
def get_marketcap(code):
url = 'https://navercomp.wisereport.co.kr/v2/company/c1010001.aspx?cmp_cd={}'.format(code)
raw_data = pd.read_html(url, encoding='utf-8')
data = raw_data[1]
data.columns = ['col', 'value']
marketcap = data[data['col'] == '시가총액']['value'].values[0]
marketcap = marketcap.replace(',', '')
# 단위 변환
float_marketcap = float(marketcap[:-2]) * unit_dict[marketcap[-2:]]
return float_marketcap "개발자도구 - Network" 를 활용해서 재무상태표를 눌렀을 때의 response 를 살펴보자.
재무상태표 URL
def get_balancesheet(code):
bs_encparam = get_encparam_bs(code)
#연간
url = 'https://navercomp.wisereport.co.kr/v2/company/cF3002.aspx?cmp_cd={}&frq=0&rpt=1&finGubun=MAIN&frqTyp=0&cn=&encparam={}'.format(code, bs_encparam)
#분기
#url = 'https://navercomp.wisereport.co.kr/v2/company/cF3002.aspx?cmp_cd={}&frq=1&rpt=1&finGubun=MAIN&frqTyp=1&cn=&encparam={}'.format(code, bs_encparam)
referer_url = 'https://navercomp.wisereport.co.kr/v2/company/c1030001.aspx?cmp_cd={}&cn='.format(code)
response_data = requests.get(url, verify=False, headers={'referer': referer_url})
html = response_data.text
jsonObject = json.loads(html)
jsonObject 구조
jsonObject 의 DATA 구조
포괄손익계산서 URL
재무상태표 URL:
0 &rpt=1 &finGubun=MAIN&frqTyp=0 &cn=&encparam=a0IyMG82R0sxT05OWlZSTlpoZXlhdz09
포괄손익계산서 URL:
0 &rpt=0 &finGubun=MAIN&frqTyp=0 &cn=&encparam=a0IyMG82R0sxT05OWlZSTlpoZXlhdz09
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
@author: better_coco
"""
import requests
import re
import os
import pandas as pd
import json
base_path = os.getcwd()
kospi_df = pd.read_csv(os.path.join(base_path, '검색 결과.csv'))
kospi_list = kospi_df['기호'].values
unit_dict = {'억원': 100000000}
#%% 기업현황
def get_basic_key_values(code):
url = 'https://navercomp.wisereport.co.kr/v2/company/c1010001.aspx?cmp_cd={}'.format(code)
html = requests.get(url).text
re_enc = re.compile("encparam: '(.*)'", re.IGNORECASE)
re_id = re.compile("id: '([a-zA-Z0-9]*)' ?", re.IGNORECASE)
encparam = re_enc.search(html).group(1)
encid = re_id.search(html).group(1)
return encparam, encid
def get_dataframe(code):
encparam, encid = get_basic_key_values(code)
url = 'http://companyinfo.stock.naver.com/v1/company/ajax/cF1001.aspx?cmp_cd={}&fin_type=0&freq_type=A&encparam={}&id={}'.format(code, encparam, encid)
headers = {'Referer': 'HACK'}
html = requests.get(url, headers=headers).text
dfs = pd.read_html(html)
df = dfs[1]
return df
#%% 시총 데이터
def get_marketcap(code):
url = 'https://navercomp.wisereport.co.kr/v2/company/c1010001.aspx?cmp_cd={}'.format(code)
raw_data = pd.read_html(url, encoding='utf-8')
data = raw_data[1]
data.columns = ['col', 'value']
marketcap = data[data['col'] == '시가총액']['value'].values[0]
marketcap = marketcap.replace(',', '')
float_marketcap = float(marketcap[:-2]) * unit_dict[marketcap[-2:]]
return float_marketcap
#%%
# 재무상태표 (balance sheet): option = 1
# 포괄손익계산서 (income statement): option = 0
def get_encparam(code):
url = 'https://navercomp.wisereport.co.kr/v2/company/c1010001.aspx?cmp_cd={}'.format(code)
html = requests.get(url).text
regex = re.compile("encparam.*")
mo = regex.search(html)
if mo != None:
encparam = mo.group()[10:].replace("\'", "")
return encparam
def get_balancesheet(code, option):
encparam = get_encparam(code)
url = 'https://navercomp.wisereport.co.kr/v2/company/cF3002.aspx?cmp_cd={}&frq=0&rpt={}&finGubun=MAIN&frqTyp=0&cn=&encparam={}'.format(code, option, encparam)
referer_url = 'https://navercomp.wisereport.co.kr/v2/company/c1030001.aspx?cmp_cd={}&cn='.format(code)
response_data = requests.get(url, verify=False, headers={'referer': referer_url})
html = response_data.text
jsonObject = json.loads(html)
cols = list(map(lambda x: x.replace('<br />', ' '), jsonObject.get('YYMM')))
cols = list(filter(lambda x: '전년대비' not in x, cols))
cols = list(filter(lambda x: '(E)' not in x, cols))
data_list = jsonObject.get('DATA')
tmp_df = []
for ele in data_list:
tmp = []
tmp.append(ele['ACC_NM'])
for idx in range(len(cols)):
tmp.append(ele['DATA{}'.format(idx + 1)])
tmp_df.append(tmp)
cols.insert(0, '항목')
df = pd.DataFrame(tmp_df, columns=cols)
return df
#%% 데이터 취합 함수
def main_code(code):
print(code)
#시가총액
marketcap = get_marketcap(code)
#기본정보
df = get_dataframe(code)
df.columns = df.columns.get_level_values(1)
def get_values(col, current=True):
time_info = df.columns[1:]
current_time = ''
for time in time_info:
if '(E)' not in time:
current_time = time
else:
break
if current == True:
return df[df['주요재무정보'] == col][current_time]
else:
return df[df['주요재무정보'] == col]
per = get_values('PER(배)').values[0]
pbr = get_values('PBR(배)').values[0]
roa = get_values('ROA(%)').values[0]
net_income = get_values('당기순이익').values[0]
total_sales = get_values('매출액').values[0]
total_liabilities = get_values('부채총계').values[0]
total_equity = get_values('자본총계').values[0]
total_assets = get_values('자산총계').values[0]
psr = marketcap / total_sales
debt_ratio = total_liabilities / total_equity
# balance sheet for 유동자산
balance_sheet = get_balancesheet(code, 1)
current_asset = balance_sheet[balance_sheet['항목'] == '유동자산'][balance_sheet.columns[-1]].values[0]
# income statement for 매출총이익
income_statement = get_balancesheet(code, 0)
gross_profit = income_statement[income_statement['항목'] == '매출총이익'][income_statement.columns[-1]].values[0]
gpa = gross_profit / total_assets
return [per, pbr, psr, marketcap, current_asset, total_liabilities, debt_ratio, roa, net_income, gpa]
#%% 전체 함수 call
tmp = []
for code in kospi_list:
try:
tmp.append(main_code(code))
except:
pass
all_df = pd.DataFrame(tmp, columns=['PER', 'PBR', 'PSR', 'MarketCap', 'CurrentAssets', 'TotalLiabilities', 'DebtRatio', 'ROA', 'NetIncome', 'GP/A'])
