[Python] Python으로 웹 스크래퍼 만들기 (~2.5)

Jeongyun Heo·2020년 12월 22일
0
post-thumbnail

노마드 코더 Python으로 웹 스크래퍼 만들기
https://nomadcoders.co/python-for-beginners

✍️  해야할 것

  1. 사이트 접근

  2. 페이지 몇 개인지 알아내기

  3. 페이지 하나씩 들어가기


indeed와 stackoverflow에서 scraping 해 오기

마지막에 모든 결과를 엑셀 시트에 보여준다

파이썬에서는 라이브러리 또는 패키지를 import 할 수 있다

📕  requests 라이브러리

Python HTTP 라이브러리. 웹 페이지의 HTML을 가져오는 모듈
r = requests.get(url) 해당 url 사이트에 get 요청을 보내는 것
요청이 성공하면 200을 응답한다.

package로 가서 requests 찾아서 설치하기

설치가 완료되면

✍️  import requests

requests를 import 해준다

request를 만들고 그 결과를 변수에 넣는다.

indeed_result 라는 변수를 만들고

.get 괄호 안에 indeed 주소 복사해서 붙여넣기

import requests

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

실행해본다

오류가 안 났다

이제 출력을 해본다

import requests

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

print(indeed_result)  ⬅️ 추가

200이라고 응답이 왔는데 okay라는 뜻이다.

Indeed의 html을 살펴본다.

requests로 이것 저것 가지고 올 수 있는데

✍️   indeed_result.text

우리가 가져올 것은 text이다.

indeed_result.text 라고 입력한다.

출력해본다.

import requests

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

print(indeed_result.text) ⬅️ indeed_result.text로 변경

뭐가 엄청 많이 출력된다.

html 전부를 가지고 온 것이다.

✍️  import BeautifulSoup

이제 가지고 온 html에서 원하는 정보를 추출해야 한다.

페이지 숫자를 가져와야 한다.

수동으로 하려면 시간이 오래 걸리므로

구글에 beautifulsoup 검색

html에서 정보를 추출할 때 유용한 package이다.

repl.it packages에 가서 beautifulsoup4 설치

beautifulsoup4 Screen-scraping library 설치

📕  Beautiful Soup Documentation
https://www.crummy.com/software/BeautifulSoup/bs4/doc/

import requests
from bs4 import BeautifulSoup ⬅️ 추가

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

print(indeed_result.text)

실행해본다.

오류 안 남

✍️  BeautifulSoup(html_doc, 'html.parser')

이제 변수를 만들어준다.

BeautifulSoup(html_doc, 'html.parser')

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser") ⬅️

print(indeed_result.text)

실행해보기

오류 안 남

✍️  indeed 페이지 가져오기

아래와 같이 beautifulsoup으로 데이터 구조를 탐색할 수 있다.

📕 find와 find_all의 차이점

soup.find_all('a')
매개변수로 tag 입력
해당되는 모든 tag들 리스트로 반환

soup.find('a')
매개변수로 tag 입력
해당되는 첫 번째 tag만 가져온다.
그 첫 번째 tag 안에 들어 있는 tag들도 다 가지고 옴.
리스트로 돌려주지 않는다.

class명이 pagination인 div를 찾는다.

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

print(pagination)

✍️  pagination.find_all('a')

이제 pagination에서 a를 찾는다

출력해준다

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

pages = pagination.find_all('a')

print(pages)

✍️  page.find("span")

indeed html을 보면

a 안에 span이 들어있다

각 a 안에 들어있는 span을 찾는다

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

pages = pagination.find_all('a')

for page in pages:
  print(page.find("span"))

✍️  spans.append(page.find("span"))

마지막에 출력된 np(next page)를 없애줘야 한다.

빈 리스트를 만든다.

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

pages = pagination.find_all('a')
spans = []  ⬅️ 빈 리스트를 만들어준다.
for page in pages:
  print(page.find("span"))

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

pages = pagination.find_all('a')
spans = []
for page in pages:
    spans.append(page.find("span")) ⬅️ .append 사용
print(spans)

리스트에 span이 모두 들어가 있다.

✍️  마지막에 있는 next 빼기 👉 spans[:-1]

마지막 거 빼주기

a[-1] ⬅️ 리스트 a의 마지막 요솟값

리스트의 슬라이싱
a[시작 번호:끝 번호] ⬅️ 끝 번호에 해당하는 것은 포함하지 않는다

시작 번호 생략시 처음부터 시작

a[:-1] ⬅️ 시작 번호가 생략되어 있으므로 처음부터 시작. 끝 번호는 -1. -1은 마지막 요솟값. 끝 번호에 해당하는 것은 포함하지 않으므로 마지막 요솟값은 포함하지 않는다.

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

pages = pagination.find_all('a')
spans = []
for page in pages:
    spans.append(page.find("span"))
print(spans[:-1]) ⬅️ 마지막 요소 포함하지 않는다.

마지막 값 제외됨

spans = spans[:-1] 로 바꿔준다.

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

pages = pagination.find_all('a')
spans = []
for page in pages:
    spans.append(page.find("span"))
spans = spans[:-1]  ⬅️ 미리 만들어둔 빈 리스트에 넣음

indeed 페이지 추출 완료

pages ➡️ links
page ➡️ link
spans ➡️ pages
수정하기

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

links = pagination.find_all('a')
pages = []
for link in links:
    pages.append(link.find("span"))
pages = pages[:-1]

✍️  페이지 숫자만 가져오기 👉 link.find("span").string

텍스트만 가져온다.

span을 찾은 다음 그 안에 있는 string만 가져온다.

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

links = pagination.find_all('a')
pages = []
for link in links:
    pages.append(link.find("span").string)
pages = pages[:-1]
print(pages)

👉
['2', '3', '4', '5', '6', '7', '8', '9', '10', '11', '12', '13', '14', '15', '16', '17', '18', '19', '20']

텍스트만 추출되었다.

span 말고 link에서 .string 실행해도 동일한 결과를 얻는다.

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

links = pagination.find_all('a')
pages = []
for link in links:
    pages.append(link.string)  ⬅️ link에서 string 가져오기
pages = pages[:-1]
print(pages)

👉
['2', '3', '4', '5', '6', '7', '8', '9', '10', '11', '12', '13', '14', '15', '16', '17', '18', '19', '20']
결과는 동일하다.

예를 들어 anchor가 있고 이 요소 안에 다른 요소가 있고 그 요소에 srting이 오직 하나 있다면 그냥 anchor에서 string method를 실행하면 된다.

✍️  페이지 숫자 str에서 정수로 바꿔주기 👉 int(link.string)

페이지 숫자가 str로 되어 있으므로 int로 바꿔준다.

int는 문자열로 되어 있는 숫자를 정수형으로 바꿔주는데
np(next page)는 숫자가 아니기 때문에 오류 남

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

links = pagination.find_all('a')
pages = []
for link in links[:-1]:  ⬅️ 처음부터 [:-1] 써줌
    pages.append(int(link.string))
# pages = pages[:-1]  ⬅️ 지움
print(pages)

👉
[2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]

✍️  마지막 페이지 숫자 가져오기 👉 max_page = pages[-1]

마지막 페이지를 찾는다.

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

links = pagination.find_all('a')
pages = []
for link in links[:-1]:
    pages.append(int(link.string))
print(pages[-1])  ⬅️ [-1] 입력. 뒤에서 첫 번째 요소.

👉 20

indeed의 마지막 페이지를 변수에 넣는다.

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

links = pagination.find_all('a')
pages = []
for link in links[:-1]:
    pages.append(int(link.string))

max_page = pages[-1]  ⬅️ 변수를 만들어서 넣어준다.

✍️  range(max_page)

최대 페이지 수 만큼 request를 보내야 함

range 함수를 사용한다.

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

links = pagination.find_all('a')
pages = []
for link in links[:-1]:
    pages.append(int(link.string))

max_page = pages[-1]  ⬅️ 20
print(range(max_page))  ⬅️ range(20)

👉
range(0, 20) ⬅️ 0 이상 20 미만 (0 ~ 19)

range(시작 숫자, 끝 숫자) ⬅️ 끝 숫자는 포함되지 않는다.

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

links = pagination.find_all('a')
pages = []
for link in links[:-1]:
    pages.append(int(link.string))

max_page = pages[-1]  ⬅️ 20

for n in range(max_page):  ⬅️ range(20) 0부터 20 미만의 숫자
    print(n)
  
👉
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

✍️  f"start={n*50}"

페이지 숫자에 50을 곱해준다

0부터 19까지 페이지 넘버가 하나씩 차례로 대입되면서
페이지 넘버 * 50 문장을 반복적으로 수행한다.

n 변수에 50을 곱해준다.

import requests
from bs4 import BeautifulSoup

indeed_result = requests.get("https://www.indeed.com/jobs?q=python&limit=50")

indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

pagination = indeed_soup.find("div", {"class": "pagination"})

links = pagination.find_all('a')
pages = []
for link in links[:-1]:
    pages.append(int(link.string))

max_page = pages[-1]

for n in range(max_page):
    print(f"start={n*50}")  ⬅️ n(0~19)*50
    
👉
start=0
start=50
start=100
start=150
start=200
start=250
start=300
start=350
start=400
start=450
start=500
start=550
start=600
start=650
start=700
start=750
start=800
start=850
start=900
start=950

1 페이지 ⬅️ start=0
https://www.indeed.com/jobs?q=python&limit=50&start=0

20 페이지 ⬅️ start=950
https://www.indeed.com/jobs?q=python&limit=50&start=950

✍️  def extract_indeed_pages():

여태까지 작성한 걸 새로운 파일을 만들어서 따로 넣어준다.

함수 이름을 extract_indeed_pages 라고 정의해준다.

import requests
from bs4 import BeautifulSoup


def extract_indeed_pages():  ⬅️ 함수이름 새로 만들어줌
    indeed_result = requests.get(
        "https://www.indeed.com/jobs?q=python&limit=50")

    indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

    pagination = indeed_soup.find("div", {"class": "pagination"})

    links = pagination.find_all('a')
    pages = []
    for link in links[:-1]:
        pages.append(int(link.string))

    max_page = pages[-1]

✍️  URL 변수로 만들기

.get 안에 있는 링크를 꺼내서 변수로 만들어 준다.

URL = "https://www.indeed.com/jobs?q=python&limit=50"

나중에 계속 불러와서 쓸 수 있도록 변수로 만들어 준다.

.get 안에는 링크주소 대신 변수이름을 적어준다.

import requests
from bs4 import BeautifulSoup

URL = "https://www.indeed.com/jobs?q=python&limit=50"

def extract_indeed_pages():
    indeed_result = requests.get(URL)  ⬅️ 변수로 적어줌

    indeed_soup = BeautifulSoup(indeed_result.text, "html.parser")

    pagination = indeed_soup.find("div", {"class": "pagination"})

    links = pagination.find_all('a')
    pages = []
    for link in links[:-1]:
        pages.append(int(link.string))

    max_page = pages[-1]

새로 만든 파일 이름이 indeed.py이므로 변수명에 indeed를 굳이 안 넣어도 되므로 빼준다.

import requests
from bs4 import BeautifulSoup

URL = "https://www.indeed.com/jobs?q=python&limit=50"


def extract_indeed_pages():
    result = requests.get(URL)

    soup = BeautifulSoup(result.text, "html.parser")

    pagination = soup.find("div", {"class": "pagination"})

    links = pagination.find_all('a')
    pages = []
    for link in links[:-1]:
        pages.append(int(link.string))

    max_page = pages[-1]

✍️  return max_page

마지막에 return max_page 추가

import requests
from bs4 import BeautifulSoup

URL = "https://www.indeed.com/jobs?q=python&limit=50"


def extract_indeed_pages():
    result = requests.get(URL)

    soup = BeautifulSoup(result.text, "html.parser")

    pagination = soup.find("div", {"class": "pagination"})

    links = pagination.find_all('a')
    pages = []
    for link in links[:-1]:
        pages.append(int(link.string))

    max_page = pages[-1]
    return max_page

✍️  main.py에 import 해주기

main.py로 돌아와서 새로 만든 함수를 import 해 준다.

indeed.py에 있는 extract_indeed_pages 함수를 import 해 준다.

from indeed import extract_indeed_pages

indeed 마지막 페이지 변수를 만들고
extract_indeed_pages 함수를 호출한 결괏값을 넣어준다.

max_indeed_pages = extract_indeed_pages()

from indeed import extract_indeed_pages

max_indeed_pages = extract_indeed_pages()

print(max_indeed_pages)

👉 20

✍️  def extract_indeed_jobs(last_page):

indeed.py에 또 다른 함수를 만든다.

indeed page를 입력 받아서 페이지 수 만큼 request를 만드는 함수

함수이름은 extract_indeed_jobs로 한다.

def extract_indeed_jobs(last_page):

마지막 페이지를 매개변수로 받는다.

import requests
from bs4 import BeautifulSoup

URL = "https://www.indeed.com/jobs?q=python&limit=50"


def extract_indeed_pages():
    result = requests.get(URL)

    soup = BeautifulSoup(result.text, "html.parser")

    pagination = soup.find("div", {"class": "pagination"})

    links = pagination.find_all('a')
    pages = []
    for link in links[:-1]:
        pages.append(int(link.string))

    max_page = pages[-1]
    return max_page


def extract_indeed_jobs(last_page):  ⬅️ 매개변수로 마지막 페이지 받음
    for page in range(last_page):
        print(f"&start={page*50}")

✍️  LIMIT 변수 생성

URL변수에서 맨 끝에 있는 50을 따로 변수로 만들어 준다.

나중에 20개씩 보기, 30개씩 보기로 바꿔줄 수도 있기 때문.

LIMIT = 50
URL = f"https://www.indeed.com/jobs?q=python&limit={LIMIT}"

import requests
from bs4 import BeautifulSoup

LIMIT = 50  ⬅️ 나중에 바꿀 수도 있어서 따로 빼주었다.
URL = f"https://www.indeed.com/jobs?q=python&limit={LIMIT}"


def extract_indeed_pages():
    result = requests.get(URL)

    soup = BeautifulSoup(result.text, "html.parser")

    pagination = soup.find("div", {"class": "pagination"})

    links = pagination.find_all('a')
    pages = []
    for link in links[:-1]:
        pages.append(int(link.string))

    max_page = pages[-1]
    return max_page


def extract_indeed_jobs(last_page):
    for page in range(last_page):
        print(f"&start={page*LIMIT}")  ⬅️ 50 대신 LIMIT변수 사용

✍️  main.py에 import extract_indeed_jobs

main.py로 가서 extract_indeed_jobs 함수를 추가로 import 해준다.

from indeed import extract_indeed_pages, extract_indeed_jobs

콤마 사용해서 추가하면 됨

변수 이름 수정
max_indeed_pages ➡️ last_indeed_page

from indeed import extract_indeed_pages, extract_indeed_jobs

last_indeed_page = extract_indeed_pages()

extract_indeed_jobs(last_indeed_page)

👉
&start=0
&start=50
&start=100
&start=150
&start=200
&start=250
&start=300
&start=350
&start=400
&start=450
&start=500
&start=550
&start=600
&start=650
&start=700
&start=750
&start=800
&start=850
&start=900
&start=950

✍️  requests.get(f"{URL}&start={page*LIMIT}")

이제 URL 변수랑 위에 나온 결괏값이랑 합쳐준다.

result 라는 변수 만들고
result = requests.get(f"{URL}&start={page*LIMIT}")

.status_code 출력
print(result.status_code)

import requests
from bs4 import BeautifulSoup

LIMIT = 50
URL = f"https://www.indeed.com/jobs?q=python&limit={LIMIT}"


def extract_indeed_pages():
    result = requests.get(URL)
    soup = BeautifulSoup(result.text, "html.parser")
    pagination = soup.find("div", {"class": "pagination"})
    links = pagination.find_all('a')
    pages = []
    for link in links[:-1]:
        pages.append(int(link.string))
    max_page = pages[-1]
    return max_page


def extract_indeed_jobs(last_page):
    for page in range(last_page):
        result = requests.get(f"{URL}&start={page*LIMIT}")
        print(result.status_code)
        
👉
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
200
# 200이 20개 출력됨

✍️  jobs = [ ]

indeed.py에 추출한 일자리 정보를 담을 빈 리스트를 만든다.

# indeed.py
import requests
from bs4 import BeautifulSoup

LIMIT = 50
URL = f"https://www.indeed.com/jobs?q=python&limit={LIMIT}"


def extract_indeed_pages():
    result = requests.get(URL)
    soup = BeautifulSoup(result.text, "html.parser")
    pagination = soup.find("div", {"class": "pagination"})
    links = pagination.find_all('a')
    pages = []
    for link in links[:-1]:
        pages.append(int(link.string))
    max_page = pages[-1]
    return max_page


def extract_indeed_jobs(last_page):
    jobs = []  ⬅️ # 추출한 일자리를 담을 빈 리스트 생성
    for page in range(last_page):
        result = requests.get(f"{URL}&start={page*LIMIT}")
        print(result.status_code)
    return jobs  ⬅️ # 최종적으로 일자리 담은 리스트 반환

1개의 댓글

comment-user-thumbnail
2021년 1월 11일

질문드려도 될까요? 제가 main.py에서 코드를 복사한다음에 indeed.py 파일 만들어서 그대로 복사 붙여 넣기해서 def extract_indeed_pages(): 함수를 만들었습니다 그리고 다시 main.py에 있던 원래 코드는 지우고 indeed.py를 실행하면 아무런 반응이 없습니다.. 그래서 main.py에 있던 원래 코드를 남겨두면 실행이 됩니다.. 왜 그럴까요 ㅠㅠㅠ 여기서 이틀째 진도를 못나가고 있어요
확인해보니까 indeed.py에서 run을 해도 main.py가 실행이 되네요 왜그럴까요

답글 달기