83 lines
2.6 KiB
Python
83 lines
2.6 KiB
Python
from bs4 import BeautifulSoup
|
|
import requests
|
|
import datefinder # another date finding library
|
|
import re
|
|
from datetime import datetime
|
|
from datetime import timedelta
|
|
import time
|
|
import pandas as pd
|
|
|
|
# this function is brute force looping through the whole directory and checking dates
|
|
# it may sound horrible, but for certain distros, i believe it's indeed the best solution
|
|
|
|
# lists
|
|
urls=[]
|
|
|
|
home_site = "http://ftp.netbsd.org/pub"
|
|
|
|
# function created
|
|
def scrape(site):
|
|
|
|
# getting the request from url
|
|
r = requests.get(site)
|
|
|
|
# converting the text
|
|
s = BeautifulSoup(r.text,"html.parser")
|
|
|
|
for i in s.find_all("a"): # for a href directories
|
|
href = i.attrs['href']
|
|
|
|
if href.endswith("/") and href != "../" and href != "/":
|
|
"""if home_site+href in urls: # avoids the link to parent directory
|
|
continue"""
|
|
if href == "//ftp.netbsd.org/": # netbsd specific code
|
|
continue
|
|
site_next = site+href
|
|
|
|
if site_next not in urls:
|
|
urls.append(site_next)
|
|
print(site_next)
|
|
# calling it self
|
|
scrape(site_next)
|
|
|
|
def get_latest_date(web_dir):
|
|
page = requests.get(site).text
|
|
|
|
str_dates = re.findall(r'(\d{2}-\w{3}-\d{4} \d{2}:\d{2})|(\d{4}-\d{2}-\d{2} \d{2}:\d{2})', page)
|
|
dates = [list(datefinder.find_dates(date))[0] for date in str_dates]
|
|
|
|
# for date in dates:
|
|
# print(date)
|
|
|
|
return(max(dates))
|
|
|
|
# main function
|
|
if __name__ =="__main__":
|
|
|
|
# website to be scrape
|
|
site="http://ftp.netbsd.org/pub/NetBSD/"
|
|
# works on: https://www.x.org/releases/
|
|
# https://mirror.csclub.uwaterloo.ca/linuxmint/ #works wonders for linuxmint
|
|
# unfortunately, linuxmint does not have a public repo, the worldwide mirror LayerOnline on https://linuxmint.com/mirrors.php seems like the best choice
|
|
|
|
# calling function
|
|
scrape(site)
|
|
|
|
latest_date = get_latest_date(urls[0])
|
|
# get_latest_date(urls[0])
|
|
for dir in urls:
|
|
latest_date2 = get_latest_date(dir)
|
|
if (latest_date2 >= latest_date):
|
|
latest_date = latest_date2
|
|
|
|
print(latest_date)
|
|
|
|
"""page = requests.get("http://rsync-mxlinux.org/mirmon/index.html").text
|
|
indexOfFile = page.find("mirror.csclub.uwaterloo.ca")
|
|
|
|
m = re.search(r'(\d+ hours)|(\d+(\.)?\d+ days)', page[indexOfFile:]) # solution from: https://stackoverflow.com/questions/21074100/how-to-convert-standard-timedelta-string-to-timedelta-object/21074460
|
|
|
|
duration = pd.to_timedelta(m.group(0))
|
|
|
|
print (duration <= pd.to_timedelta(86400, unit='s'))"""
|
|
|