From cfcb876bc19c6fde7343b8d2da3bf008e02b522f Mon Sep 17 00:00:00 2001
From: Gale Andrews <gale@audacityteam.org>
Date: Mon, 4 Apr 2016 18:29:06 +0100
Subject: [PATCH] Buanzo's fix to include a User-Agent HTTP request header

This means that we don't get 403 error when dumping the Manual with the server's security features in place.
---
 scripts/mw2html_audacity/mw2html.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/scripts/mw2html_audacity/mw2html.py b/scripts/mw2html_audacity/mw2html.py
index b5279896d..5f877ec9f 100644
--- a/scripts/mw2html_audacity/mw2html.py
+++ b/scripts/mw2html_audacity/mw2html.py
@@ -60,6 +60,7 @@ footer_text = ''
 counter = 0
 errors = 0
 conn = None
+headers = {"User-Agent": "mw2html.py/Audacity"}
 domain = ''
 
 MONOBOOK_SKIN = 'monobook'    # Constant identifier for Monobook.
@@ -517,7 +518,7 @@ def split_section(url):
 
 def url_open(url):
     # download a file and retrieve its content and mimetype
-    global conn, domain, counter, redir_cache, errors
+    global conn, domain, counter, redir_cache, errors, headers
 
     l_redir = []
     redirect = url
@@ -547,7 +548,7 @@ def url_open(url):
             #increment httplib requests counter
             counter += 1
             try:
-                conn.request("GET", rel_url)
+                conn.request("GET", rel_url,headers=headers)
                 r = conn.getresponse()
                 print 'Status', r.status, r.reason, 'accessing', rel_url
                 if r.status == 404:
@@ -898,7 +899,7 @@ def run(out=sys.stdout):
     """
     Code interface.
     """
-    global conn, domain, counter, redir_cache, config
+    global conn, domain, counter, redir_cache, config, headers
 
     if urlparse.urlparse(config.rooturl)[1].lower().endswith('wikipedia.org'):
         out.write('Please do not use robots with the Wikipedia site.\n')