CODEWITHSUNDEEP
pythondatetimebeautifulsoup
python.toddler
python.toddler

Reputation: 19

Scraping based on date with BeautifulSoup

I am very new to python programming. Emphasis on VERY. I am trying to set up my first web scraping project (for news article curation).

I have already managed to scrape the news site and to create a loop that organizes the results how I want them. My issue is that I plan on scraping the web page once a day, but only for the publications that were published that same day. I don't want all of them because that would mean I would get a lot of repetition.

I know that it has something to do with converting the date via the datetime module (with an if statement), but for the life of me I couldn't find a way to make it work.

In the html, this is an example of how the date is displayed:

<time datetime="2019-02-24T10:30:46+00:00">Feb 24, 2019 at 10:30</time>

This is what I have so far:

from urllib.request import urlopen as uReq
from bs4 import BeautifulSoup as soup
from datetime import datetime

my_url = "https://www.coindesk.com/category/business-news/legal"

# Opening up the website, grabbing the page
uFeedOne = uReq(my_url, timeout=5)
page_one = uFeedOne.read()
uFeedOne.close()

# html parser
page_soup1 = soup(page_one, "html.parser")

# grabs each publication block
containers = page_soup1.findAll("a", {"class": "stream-article"} )

for container in containers:
  link = container.attrs['href']
  publication_date = "published on " + container.time.text
  title = container.h3.text
  description = "(CoinDesk)-- " +  container.p.text

  print("link: " + link)
  print("publication_date: " + publication_date)
  print("title: " + title)
  print("description: " + description)

Upvotes: 2

Views: 6360

Answers (2)

Manish Dash
Manish Dash

Reputation: 2073

Your time tag has a datetime attribute that is giving a much better datetime representation than the text. Use that.

You can use the dateutil package to parse the string. Following is a sample code:

from urllib.request import urlopen as uReq
from bs4 import BeautifulSoup as soup
from datetime import datetime, timedelta
from dateutil import parser
import pytz

my_url = "https://www.coindesk.com/category/business-news/legal"

# Opening up the website, grabbing the page
uFeedOne = uReq(my_url, timeout=5)
page_one = uFeedOne.read()
uFeedOne.close()

# html parser
page_soup1 = soup(page_one, "html.parser")

# grabs each publication block
containers = page_soup1.findAll("a", {"class": "stream-article"} )

for container in containers:
  ## get todays date.
  ## I have taken an offset as the site has older articles than today.
  today =  datetime.now() - timedelta(days=5)
  link = container.attrs['href']

  ## The actual datetime string is in the datetime attribute of the time tag.
  date_time = container.time['datetime']

  ## we will use the dateutil package to parse the ISO-formatted date.
  date = parser.parse(date_time)

  ## This date is UTC localised but the datetime.now() gives a "naive" date
  ## So we have to localize before comparison
  utc=pytz.UTC
  today = utc.localize(today)

  ## simple comparison
  if date >= today:
      print("article date", date)
      print("yesterday", today," \n")
      publication_date = "published on " + container.time.text
      title = container.h3.text.encode('utf-8')
      description = "(CoinDesk)-- " +  container.p.text

      print("link: " + link)
      print("publication_date: " + publication_date)
      print("title: ", title)
      print("description: " + description)

Upvotes: 2

Manik Tharaka
Manik Tharaka

Reputation: 308

The date is represented in the ISO 8601 format. Extract the datetime attribute as a string from the time tag. If you are using python 3.7 you can use the datetime.datetime.fromisoformat method to convert this to a datetime object and then do the comparison. If you are using an older version of python I think the easiest approach would be see this question and the first answer provided.

Upvotes: 0

Related Questions