[patch] 0.15.2 segfault with encoded header text

[David Relson]

On Mon, 08 Sep 2003 07:18:16 +0200
Stefan Bellon <sbellon at sbellon.de> wrote:

> David Relson wrote:
> 
> > Here's a patch to fix the segfault you encountered.  Keep testing
> > and see what else you can find !!!
> 
> The message in question gets processed without problems with that
> patch applied. I'm currently testing my 50 MB spam archive (which
> takes a little time :-). I let you know if other problems occur.

Great!  I've enhanced the code a bit so that it checks what's between
the encoded tokens.  If that's all whitespace, it's discarded. 
Otherwise it's kept (including spacing).  Thus "=?charset?q?Stefan?=  
=?charset?q?Bellon?=" becomes "StefanBellon", while
"=?charset?q?Stefan?=X.  =?charset?q?Bellon?=" becomes "StefanX. Bellon"
(which is parsed as "StefanX." and "Bellon").  The enhanced code will be
in 0.15.3 and is available from the CVS repository (if you need it).