From 3d5c114f14a716c94e930e3293881697d9b97356 Mon Sep 17 00:00:00 2001
From: Bastian Kleineidam <calvin@debian.org>
Date: Sun, 7 Mar 2010 09:36:21 +0100
Subject: [PATCH] Warn on permament redirections even when URL is outside of
 domain filter.

---
 doc/changelog.txt            |  5 ++++-
 linkcheck/checker/httpurl.py | 19 +++++++++++--------
 2 files changed, 15 insertions(+), 9 deletions(-)

diff --git a/doc/changelog.txt b/doc/changelog.txt
index 78f54c34..e1569ca2 100644
--- a/doc/changelog.txt
+++ b/doc/changelog.txt
@@ -7,10 +7,13 @@ Fixes:
 - dns: use /dev/urandom instead of /dev/random to get initial seed
   on Linux machines since the last one can block indefinitely.
   Closes: SF bug #2901667
-- checking: Retry if server closed connection and sent an empty
+- http: Retry if server closed connection and sent an empty
   status line. Fixes the "BadStatusLine" errors.
 - installer: Add commandline executable to Windows installer.
   Closes: SF bug #2903257
+- http: warn about permanent redirections even when redirected URL is
+  outside of the domain filter.
+  Closes: SF bug #2920182
 
 5.1 "Let the right one in" (released 04.08.2009)
 
diff --git a/linkcheck/checker/httpurl.py b/linkcheck/checker/httpurl.py
index c7502436..f09902bd 100644
--- a/linkcheck/checker/httpurl.py
+++ b/linkcheck/checker/httpurl.py
@@ -327,6 +327,7 @@ Use URL `%(newurl)s' instead for checking.""") % {
             self.set_extern(redirected)
             if self.extern[0] and self.extern[0]:
                 if set_result:
+                    self.check301status(response)
                     self.add_info(
                           _("The redirected URL is outside of the domain "
                             "filter, checked only syntax."))
@@ -359,14 +360,8 @@ Use URL `%(newurl)s' instead for checking.""") % {
                 self.aliases.append(redirected)
             # note: urlparts has to be a list
             self.urlparts = urlparts
-            if response.status == 301:
-                if not self.has301status:
-                    if set_result:
-                        self.add_warning(
-                           _("HTTP 301 (moved permanent) encountered: you"
-                             " should update this link."),
-                           tag=WARN_HTTP_MOVED_PERMANENT)
-                    self.has301status = True
+            if set_result:
+                self.check301status(response)
             # check cache again on the changed URL
             if self.aggregate.urlqueue.checked_redirect(redirected, self):
                 return -1, response
@@ -392,6 +387,14 @@ Use URL `%(newurl)s' instead for checking.""") % {
             tries += 1
         return tries, response
 
+    def check301status (self, response):
+        """If response page has been permanently moved add a warning."""
+        if response.status == 301 and not self.has301status:
+            self.add_warning(_("HTTP 301 (moved permanent) encountered: you"
+                               " should update this link."),
+                             tag=WARN_HTTP_MOVED_PERMANENT)
+            self.has301status = True
+
     def get_alias_cache_data (self):
         """
         Return all data values that should be put in the cache,