[Mailman-Users] v2.1.1 and Archive download problem

[Richard Barrett]

On Thursday, August 14, 2003, at 08:40  pm, Anders Norrbring wrote:

> I have the precis same problem, seems like the URL is not converted  
> right,
> in my IE6 it comes up with:
>
> mhtml:http://www.the-server.net/pipermail/fetishclub_info.mbox/ 
> fetishclub_in
> fo.mbox
>
> Which of course is totally wrong..
>

In the case of my Mailman machine, for a public archive, the Apache  
server was returning the HTTP response header Content-Type:  
message/rfc822 for .mbox files.

IE6's behaviour when trying to render a .mbox file appears to be due to  
this header value.

This Content-Type header results from a combination of the Apache  
server's mime.types and magic configuration files. Change those so that  
Apache returned a Content-Type: text/plain header for a .mbox file and  
IE6 renders the .mbox as text OK.

In the case of a private archive, the $prefix/Mailman/Cgi/private.py  
returns a Content-Type: text/plain header which IE6 renders as text  
without problems.

It might be possible to adjust some IE setting to persuade it to  
respond more appropriately to the Content-Type: message/rfc822. But its  
is late and I am tired and I hate non-Open Software so I leave this as  
a task for the reader.

This is not a Mailman problem per se.

> Anders Norrbring
>
>
> -> We are using version 2.1.1 and when someone clicks on the "download  
> the
> -> full
> -> raw archive" on the archives page, the archive is not downloaded  
> under
> -> IE.
> -> It downloads okay using Netscape or Opera.
> ->
> -> I've tried downloading this list's archives and it works fine.   
> Therefore
> -> I'm thinking this is either a configuration problem or an upgrade
> -> problem.
> -> Has anyone experienced this?  Got a fix?
> ->
> -> Ken
>
-----------------------------------------------------------------------
Richard Barrett                               http://www.openinfo.co.uk