我正在尝试从本网站上的 PDF 中抓取标题。 但是，我得到了标题和链接。 为什么以及如何解决这个问题？

Question

I want to scrape the titles of the PDFs on this website. However, I get the titles and the links. How can I fix this?

publications=[]
text=[]
for i in np.arange(12,19):
    response=requests.get('https://occ.ca/our- 
 publications/page/{}/'.format(i), headers={'User-Agent': 'Mozilla'})

if response.status_code == 200:
    soup = BeautifulSoup(response.text, 'lxml')
    pdfs = soup.findAll('div', {"class": "publicationoverlay"})

    links = [pdf.find('a').attrs['href'] for pdf in pdfs]
    publications.extend(links)
    text.extend(pdfs)

Any help would be much appreciated.

Answer 1

You want the .text though split on \t (to exclude child a text) and strip. I use Session for efficiency.

import requests
from bs4 import BeautifulSoup 
import numpy as np

publications=[]
text=[]

with requests.Session() as s:

    for i in np.arange(12,19):

        response= s.get('https://occ.ca/our-publications/page/{}/'.format(i), headers={'User-Agent': 'Mozilla'})

        if response.status_code == 200:
            soup = BeautifulSoup(response.text, 'lxml')
            pdfs = soup.findAll('div', {"class": "publicationoverlay"})
            text.extend([pdf.text.strip().split('\t')[0] for pdf in pdfs])

You could also use decompose to remove child a tags after getting href and before taking the.text of parent

import requests
from bs4 import BeautifulSoup 
import numpy as np

publications=[]
text=[]
links = []

with requests.Session() as s:

    for i in np.arange(12,19):

        response= s.get('https://occ.ca/our-publications/page/{}/'.format(i), headers={'User-Agent': 'Mozilla'})

        if response.status_code == 200:
            soup = BeautifulSoup(response.text, 'lxml')
            for a in soup.select('.publicationoverlay a'):
                links.extend([a['href']])
                a.decompose()
            pdfs = soup.findAll('div', {"class": "publicationoverlay"})
            text.extend([pdf.text.strip() for pdf in pdfs])

print(list(zip(links, text)))