Re: [exim-dev] [PATCH] allow to use server responses in clie…

Top Page
Delete this message
Reply to this message
Author: Philip Hazel
Date:  
To: Johannes Berg
CC: exim-dev
Subject: Re: [exim-dev] [PATCH] allow to use server responses in client plaintext authenticator
On Mon, 20 Feb 2006, Johannes Berg wrote:

> The following patch allows the use of $0, $1, ... in the client_send
> variable and introduces a new option called client_ignore_invalid_base64
> that controls whether base64 decoding errors on server responses are
> ignored (current behaviour since it isn't even decoded) or not.


I have put this patch on my list of things to consider, and will look at
it when it gets to the top. With any luck, this will be later this week.
One thing you should know is that there has been a new invention of
variables called $auth1 - $auth3 which can be used instead of $1 - $3 in
authentication server configurations. This was done because the
numerical variables can get reset in certain complicated expansions.

If I like this patch, I think the data should end up in $auth1, $auth2,
etc rather than $0, $1, etc.

You can see the latest code in the current snapshot, which is nowadays
updated automatically daily at around 03h15 UTC:

ftp://ftp.csx.cam.ac.uk/pub/software/email/exim/Testing/exim-snapshot.tar.bz2


-- 
Philip Hazel            University of Cambridge Computing Service
Get the Exim 4 book:    http://www.uit.co.uk/exim-book