implement parsing of robots.txt

crawler.py

@@ -4,16 +4,16 @@ Need a docstring.
 '''
 
 import argparse
-from utils.helpers import (UrlPool, WebPage, sanitise_url)
+from utils.helpers import (UrlPool, WebPage, RobotsTxt, sanitise_url)
 from pprint import pprint
 
 
-def init_crawler(base_url=None):
+def init_crawler(base_url=None, robots=None):
     '''
     needs a docstring
     '''
     uncrawled_urls, crawled_urls = UrlPool(), UrlPool()
-    initial_page = WebPage(url=base_url, base_url=base_url)
+    initial_page = WebPage(url=base_url, base_url=base_url, robots=robots)
 
     try:
         initial_page.run()
@@ -45,7 +45,7 @@ def init_crawler(base_url=None):
     return(uncrawled_urls, crawled_urls)
 
 
-def process_pool(base_url=None, uncrawled_urls=None, crawled_urls=None):
+def process_pool(base_url=None, uncrawled_urls=None, crawled_urls=None, robots=None):
     '''
     Needs a docstring
     '''
@@ -53,7 +53,7 @@ def process_pool(base_url=None, uncrawled_urls=None, crawled_urls=None):
         # pop url from pool
         new_url = uncrawled_urls.remove_from_pool()
         # create a WebPage object for the URL
-        current_page = WebPage(url=new_url, base_url=base_url)
+        current_page = WebPage(url=new_url, base_url=base_url, robots=robots)
         try:
             current_page.run()
             _urls = current_page.list_urls()
@@ -72,9 +72,10 @@ def run(args=None):
     needs a docstring.
     '''
     base_url = sanitise_url(args.url, base_url=True)
+    robots = RobotsTxt(base_url=base_url)
 
-    uncrawled_urls, crawled_urls = init_crawler(base_url)
+    uncrawled_urls, crawled_urls = init_crawler(base_url, robots)
-    process_pool(base_url, uncrawled_urls, crawled_urls)
+    process_pool(base_url, uncrawled_urls, crawled_urls, robots)
 
     pprint(crawled_urls.pool)
     print('{0} URLs crawled'.format(len(crawled_urls.pool)))

notes.md

@@ -9,4 +9,4 @@
   * better url checking to get bare domain
   * ~~remove base url from initial urls with and without trailing slash~~
   * investigate using [tldextract](https://github.com/john-kurkowski/tldextract) to match urls
-  * implement parsing of [robots.txt](http://docs.w3cub.com/python~3.6/library/urllib.robotparser/)
+  * ~~implement parsing of [robots.txt](http://docs.w3cub.com/python~3.6/library/urllib.robotparser/)~~

utils/helpers.py

@@ -3,8 +3,9 @@
 Utilities to provide various misc functions.
 '''
 
-import urllib.request
 from bs4 import BeautifulSoup
+import urllib.request
+import urllib.robotparser
 from urllib.parse import (urljoin, urlsplit)
 
 
@@ -47,9 +48,10 @@ class WebPage(object):
 
     headers = {'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:61.0) Gecko/20100101 Firefox/61.0'}
 
-    def __init__(self, url=None, base_url=None):
+    def __init__(self, url=None, base_url=None, robots=None):
         self.url = url
         self.base_url = base_url
+        self.robots = robots
 
 
     def get_source(self):
@@ -90,6 +92,7 @@ class WebPage(object):
 
         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)
 
@@ -119,6 +122,31 @@ class WebPage(object):
             print(e)
 
 
+class RobotsTxt(object):
+    '''
+    needs a docstring
+    '''
+
+    def __init__(self, base_url=None):
+        self.base_url = base_url
+
+        robots = urllib.robotparser.RobotFileParser()
+        robots.set_url(urljoin(self.base_url, 'robots.txt'))
+        try:
+            robots.read()
+        except Exception as e:
+            print(e)
+
+        self.robots = robots
+
+
+    def check(self, url):
+        '''
+        needs a docstring
+        '''
+        return self.robots.can_fetch("*", url)
+
+
 def sanitise_url(url, base_url=False):
     '''
     If `base_url` is True, we attempt to standardise `url` to ensure it can be