Package: trac-authopenid
Version: 0.4.1-1
Severity: normal
Tags: patch upstream
Control: forwarded -1 https://github.com/dairiki/authopenid-plugin/pull/7

Without the attached patch, when an OpenID provider declines to provide a full
name, trac crashes by accessing the authname variable before it was
initialized.

If no full name is provided, it should fall back to using the OpenID
URL as the user's identity.

The patch attached here should work for 0.4.1-1 -- it's slightly
different from the patch i submitted to upstream's master branch
because upstream has some unreleased changes since 0.4.1.

    --dkg

-- System Information:
Debian Release: 7.0
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable'), (1, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 3.8-trunk-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages trac-authopenid depends on:
ii  python         2.7.3-4
ii  python-openid  2.2.5-3
ii  trac           1.0.1-1

trac-authopenid recommends no packages.

trac-authopenid suggests no packages.

-- debconf-show failed
--- a/authopenid/authopenid.py	2012-06-25 17:06:18.000000000 -0400
+++ b/authopenid/authopenid.py	2013-04-04 16:04:57.542661061 -0400
@@ -584,6 +584,7 @@
                         remote_user = new_user
 
             if allowed:
+                authname = None
                 cookie = hex_entropy()
                 cookie_lifetime = self.trac_auth_cookie_lifetime
 
@@ -604,9 +605,11 @@
                     authname = req.session['name']
                     if self.combined_username:
                         authname = '%s <%s>' % (authname, remote_user)
+                else:
+                    authname = remote_user
 
                 # Possibly lower-case the authname.
-                if self.lowercase_authname:
+                if authname and self.lowercase_authname:
                     authname = authname.lower()
 
                 # Make authname unique in case of collisions

Reply via email to