From 5e13a78f668d5a296800da714eed62daf471c4e4 Mon Sep 17 00:00:00 2001
From: Bastian Kleineidam <calvin@debian.org>
Date: Fri, 9 Mar 2012 11:54:18 +0100
Subject: [PATCH] Fix non-ascii HTTP header debugging.

---
 doc/changelog.txt            | 3 +++
 linkcheck/checker/httpurl.py | 8 ++++++--
 linkcheck/i18n.py            | 2 +-
 linkcheck/strformat.py       | 6 +++---
 4 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/doc/changelog.txt b/doc/changelog.txt
index 9d01495a..15560461 100644
--- a/doc/changelog.txt
+++ b/doc/changelog.txt
@@ -6,6 +6,9 @@ Fixes:
   Closes: SF bug #3495407
 - checking: Fix non-ascii HTTP header handling.
   Closes: SF bug #3495621
+- checking: Fix non-ascii HTTP header debugging.
+  Closes: SF bug #3488675
+
 
 7.5 "Kukushka" (released 13.02.2012)
 
diff --git a/linkcheck/checker/httpurl.py b/linkcheck/checker/httpurl.py
index 56749582..bc0ad799 100644
--- a/linkcheck/checker/httpurl.py
+++ b/linkcheck/checker/httpurl.py
@@ -37,6 +37,9 @@ from .const import WARN_HTTP_ROBOTS_DENIED, \
     WARN_HTTP_DECOMPRESS_ERROR, WARN_HTTP_UNSUPPORTED_ENCODING, \
     WARN_HTTP_AUTH_UNKNOWN
 
+# assumed HTTP header encoding
+HEADER_ENCODING = "iso-8859-1"
+
 # helper alias
 unicode_safe = strformat.unicode_safe
 
@@ -197,7 +200,8 @@ class HttpUrl (internpaturl.InternPatternUrl, proxysupport.ProxySupport):
                 response.reason = unicode_safe(response.reason)
             log.debug(LOG_CHECK,
                 "Response: %s %s", response.status, response.reason)
-            log.debug(LOG_CHECK, "Headers: %s", self.headers)
+            uheaders = unicode_safe(self.headers, encoding=HEADER_ENCODING)
+            log.debug(LOG_CHECK, "Headers: %s", uheaders)
             # proxy enforcement (overrides standard proxy)
             if response.status == 305 and self.headers:
                 oldproxy = (self.proxy, self.proxyauth)
@@ -453,7 +457,7 @@ class HttpUrl (internpaturl.InternPatternUrl, proxysupport.ProxySupport):
         value = self.headers.get(name)
         if value is None:
             return default
-        return value.decode("iso-8859-1", "replace")
+        return unicode_safe(value, encoding=HEADER_ENCODING)
 
     def get_alias_cache_data (self):
         """
diff --git a/linkcheck/i18n.py b/linkcheck/i18n.py
index 92297caf..536afd62 100644
--- a/linkcheck/i18n.py
+++ b/linkcheck/i18n.py
@@ -1,5 +1,5 @@
 # -*- coding: iso-8859-1 -*-
-# Copyright (C) 2000-2011 Bastian Kleineidam
+# Copyright (C) 2000-2012 Bastian Kleineidam
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
diff --git a/linkcheck/strformat.py b/linkcheck/strformat.py
index bff4ce2a..42e87909 100644
--- a/linkcheck/strformat.py
+++ b/linkcheck/strformat.py
@@ -1,5 +1,5 @@
 # -*- coding: iso-8859-1 -*-
-# Copyright (C) 2000-2011 Bastian Kleineidam
+# Copyright (C) 2000-2012 Bastian Kleineidam
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
@@ -34,7 +34,7 @@ import pydoc
 from . import i18n
 
 
-def unicode_safe (s, encoding=i18n.default_encoding):
+def unicode_safe (s, encoding=i18n.default_encoding, errors='replace'):
     """Get unicode string without raising encoding errors. Unknown
     characters of the given encoding will be ignored.
 
@@ -48,7 +48,7 @@ def unicode_safe (s, encoding=i18n.default_encoding):
     if isinstance(s, unicode):
         # s is already unicode, nothing to do
         return s
-    return unicode(str(s), encoding, "ignore")
+    return unicode(str(s), encoding, errors)
 
 
 def ascii_safe (s):