return from WebPage to indicate whether a link was actually crawlable and only actually crawl it if it was

crawler.py

@@ -6,6 +6,7 @@ Need a docstring.
 import argparse
 import jinja2
 import os
+import asyncio
 from datetime import datetime
 from utils.helpers import (UrlPool, WebPage, RobotsTxt, sanitise_url)
 
@@ -57,16 +58,18 @@ def process_pool(base_url=None, uncrawled_urls=None, crawled_urls=None, robots=N
         # create a WebPage object for the URL
         current_page = WebPage(url=new_url, base_url=base_url, robots=robots)
         try:
-            current_page.run()
-            _urls = current_page.list_urls()
-            crawled_urls.add_to_pool(new_url)
+            succeeded = current_page.run()
         except Exception as e:
             print(e)
 
-        for url in _urls:
-            sanitised_url = sanitise_url(url=url)
-            if sanitised_url not in crawled_urls.pool:
-                uncrawled_urls.add_to_pool(url)
+        if succeeded:
+            _urls = current_page.list_urls()
+            crawled_urls.add_to_pool(new_url)
+
+            for url in _urls:
+                sanitised_url = sanitise_url(url=url)
+                if sanitised_url not in crawled_urls.pool:
+                    uncrawled_urls.add_to_pool(url)
 
         print('{0} URLs crawled, {1} remaining'.format(len(crawled_urls.pool),
                 len(uncrawled_urls.pool)))

test_helpers.py

@@ -1,37 +1,35 @@
 #!/usr/bin/env python
 
 import unittest
-from utils.helpers import (clean_base_url)
+from utils.helpers import (sanitise_url)
 
 class TestUrls(unittest.TestCase):
 
-    base_url = "github.com"
-
     base_url_list = (('eu.httpbin.org', 'http://eu.httpbin.org'),
                      ('www.simonweald.com', 'http://www.simonweald.com'),
                      ('http://www.github.com/', 'http://www.github.com'),
                      ('https://www.github.com', 'https://www.github.com'))
 
-    valid_urls = ["https://www.github.com", "http://www.github.com",
-                  "github.com", "/some/url/", "index.html"]
+    urls_to_clean = (('https://www.github.com/', 'https://www.github.com/'),
+                     ('https://github.com/?foo=bar', 'https://github.com/'),
+                     ('https://github.com/#anchor', 'https://github.com/'))
 
 
-    def test_clean_base_url(self):
+    def test_sanitise_base_url(self):
         '''
         Tests whether a URL's protocol can be discovered if not provided.
         '''
         for url, target in self.base_url_list:
-            result = clean_base_url(url)
+            result = sanitise_url(url, base_url=True)
             self.assertEqual(result, target)
 
-    # def test_url_validation(self):
-    #     '''
-    #     Passes when given a valid URL. A valid URL is qualified
-    #     by being local to the domain to be crawled.
-    #     '''
-    #     for url in self.valid_urls:
-    #         result = url_validation(self.base_url, url)
-    #         self.assertTrue(result)
+    def test_sanitise_url(self):
+        '''
+        Tests whether a URL's protocol can be discovered if not provided.
+        '''
+        for url, target in self.urls_to_clean:
+            result = sanitise_url(url)
+            self.assertEqual(result, target)
 
 
 if __name__ == '__main__':

utils/helpers.py

@@ -4,6 +4,7 @@ Utilities to provide various misc functions.
 '''
 
 from bs4 import BeautifulSoup
+import aiohttp
 import urllib.request
 import urllib.robotparser
 import urllib.error
@@ -104,10 +105,9 @@ class WebPage(object):
         pool if they start with the base URL.
         '''
         for url in self.discovered_hrefs:
-            if url.startswith(self.url):
-                if self.robots.check(url):
-                    sanitised_url = sanitise_url(url=url)
-                    self.urls_to_crawl.add(sanitised_url)
+            if url.startswith(self.base_url) and self.robots.check(url):
+                sanitised_url = sanitise_url(url=url)
+                self.urls_to_crawl.add(sanitised_url)
 
 
     def list_urls(self):
@@ -132,6 +132,9 @@ class WebPage(object):
         if self.source:
             self.find_links()
             self.parse_urls()
+            return True
+        else:
+            return False
 
 
 class RobotsTxt(object):
@@ -140,6 +143,9 @@ class RobotsTxt(object):
     '''
 
     def __init__(self, base_url=None):
+        '''
+        Manually retrieve robots.txt to allow us to set the user-agent.
+        '''
         self.base_url = base_url
         self.headers = {'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:61.0) Gecko/20100101 Firefox/61.0'}