find-broken-http-links

brokenlinks.py

#!/usr/bin/env python3
# Broken Links on www Site
import sys
from bs4 import BeautifulSoup
from urllib.parse import urlparse
from urllib.parse import urljoin

searched_links = []
broken_links = []

def getLinksFromHTML(html):
    def getLink(el):
        return el["href"]
    return list(map(getLink, BeautifulSoup(html, features="html.parser").select("a[href]")))

def find_broken_links(domainToSearch, URL, parentURL):
    if (not (URL in searched_links)) and (not URL.startswith("mailto:")) and (not ("javascript:" in URL)) and (not URL.endswith(".png")) and (not URL.endswith(".jpg")) and (not URL.endswith(".jpeg")):
        try:
            requestObj = requests.get(URL);
            searched_links.append(URL)
            if(requestObj.status_code == 404):
                broken_links.append("BROKEN: link " + URL + " from " + parentURL)
                print(broken_links[-1])
            else:
                print("NOT BROKEN: link " + URL + " from " + parentURL)
                if urlparse(URL).netloc == domainToSearch:
                    for link in getLinksFromHTML(requestObj.text):
                        find_broken_links(domainToSearch, urljoin(URL, link), URL)
        except Exception as e:
            print("ERROR: " + str(e));
            searched_links.append(domainToSearch)

find_broken_links(urlparse(sys.argv[1]).netloc, sys.argv[1], "")

print("\n--- DONE! ---\n")
print("The following links were broken:")

for link in broken_links:
    print ("\t" + link)

shell.nix

{ pkgs ? import <nixpkgs> {} }:
let
  my-python-packages = ps: with ps; [
    requests
    beautifulsoup4
    uritools
    # other python packages
  ];
  my-python = pkgs.python3.withPackages my-python-packages;
in pkgs.mkShell {

  buildInputs = [
    my-python
    pkgs.neofetch
    pkgs.figlet
  ];
  shellHook =
  ''
    neofetch
    figlet "Hello python3"
  '';
}

usage:
$ nix-shell
$ brokenLinks.py https://docs.cardano.org/introduction > ./report.txt 2>&1