Merge pull request #17 from VisualText/NLP-TUTORIALS-012

NLP-TUTORIALS-012 fixed python certificate problem

Author: David de Hilster

tutorial-13/tutorial-13-a/README.md

@@ -1,3 +1,7 @@
 # Tutorial 13-a
 
-This analyzer parses the URLs from this link: https://state.1keydata.com/ into a URL list. It then has a python script to fetch the webpages and save them in a folder. This folder then can easily be moved into the second analyzer where the pages will be processed.
+This analyzer parses the URLs from this link: https://state.1keydata.com/ into a URL list. It then has a python script to fetch the webpages and save them in a folder. This folder then can easily be moved into the second analyzer where the pages will be processed.
+
+## NOTE
+
+You will have to install BeautifulSoup and certifi before using the python script.

tutorial-13/tutorial-13-a/input/urlfetch.py

@@ -5,12 +5,11 @@
 from bs4 import BeautifulSoup
 from pathlib import Path
 import re
+import certifi
 
 wordsfile = os.path.join(os.path.dirname(__file__), "urls.txt")
 file1 = codecs.open(wordsfile, "r", "utf-8")
 lines = file1.readlines()
-
-urlbase = "https://state.1keydata.com/"
 
 count = 0
 for url in lines:
@@ -31,7 +30,7 @@
     found = False
 
     try:
-        page = urllib.request.urlopen(url)
+        page = urllib.request.urlopen(url, cafile=certifi.where())
     except HTTPError as e:
         print(' Error code: ', e.code)
         file1 = open(os.path.join(os.path.dirname(__file__), "urlorphans.txt"), "a")