https://wikidocs.net/21879

파이썬에는 웹과의 소통을 편하게 해주는 requests라는 라이브러리

웹 scraping

• TAG 셀렉터
• ID 셀렉터
• Class 셀렉터

id 있는경우

import requests
url_hynix = "https://finance.naver.com/item/main.nhn?code=000660"
response = requests.get(url_hynix)
html = response.text

# conda install bs4 html5lib
from bs4 import BeautifulSoup
soup = BeautifulSoup(html, "html5lib")

tags = soup.select("#_per")

# soup.select("#_per")
# tags                 # [10.58]
# tags[0]              # 10.58
# tags[0].text         # '10.58'

finance.naver.com/item/main.nhn?code="
def get_per(code):
    url=url_finance+code
    html= requests.get(url).text
    soup = BeautifulSoup(html, "html5lib")
    tags = soup.select("#_per")
    tag = tags[0]
    return float(tag.text)

print(get_per("000660"))

id 없는 경우

import requests
from bs4 import BeautifulSoup

url_hynix = "https://finance.naver.com/item/main.nhn?code=000660"
response = requests.get(url_hynix)
html = response.text
soup = BeautifulSoup(html, "html5lib")

tags = soup.select("#tab_con1 > div:nth-child(3) > table > tbody > tr.strong > td > em")

tag = tags[0]
print(tag.text)

for tag in tags:
    print(tag.text)

웹 API

RestfulAPI

https://api.korbit.co.kr/v1/ticker/detailed?currency_pair=btc_krw

https://api.korbit.co.kr/v1/ticker/detailed?currency_pair=xrp_krw

Json –> Dictionary

import requests
url = "https://api.korbit.co.kr/v1/ticker/detailed?currency_pair=btc_krw"
response = requests.get(url)

# result = response.text      # string
result = response.json()    # dictionary

import datetime as dt
daytime = dt.datetime.fromtimestamp(result['timestamp']/1000)
daytime.year
daytime.second
daytime.microsecond
daytime.weekday()  # 0:월요일 1:화요일...
daytime.strftime(f"%B %m월 %d일 %A")  # 'March 03월 14일 Monday'

x= daytime.strptime("2017-01-02 14:44", "%Y-%m-%d %H:%M")
x.hour
x.minute

브라우저에서만 볼수 있는 경우

Ajax (Asynchronous JavaScript and XML)로 데이터를 요청하는 것이기 때문

import requests
from bs4 import BeautifulSoup as bs
import pandas as pd

url='https://finance.naver.com/item/sise_day.nhn?code=066570'
# 브라우저 인 것 처럼 요청을 보냅니다. (브라우저가 아니면 응답을 하지 않기 때문에 )
headers = {'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/88.0.4324.96 Safari/537.36'}
res = requests.get(url, headers=headers)

html = bs(res.text, 'lxml')
html_table = html.select("table")
tables = pd.read_html(str(html_table))

tables[0].dropna()
tables[1]