At 07:54 05.04.01 +1000, Robert Collins wrote:
<snip>
>And hope the user reads it ? :] I suspect a long timeout would be needed.
:-) A timeout that is user-definable then. The 'design' of the info screen
would have to be such that the authentication dialog did not block it of
course.
>> However, this would render all other applications except browsers
>> completely without a clue as to what to do when they are talking to
>>the proxy, but I guess if you add a check for browser id, you could get
>>around that as well.
>Unless the applications you are talking about use the embedded http
>support microsoft supply. That could confuse things. Or an automated
>download of pages with ie.
Ah.. so the (is it ActiveX controls they call it, these programmers? :) MS
IE thingy identifies itself as IE ?
>The technical side is not too hard. When the user first logs in, they are
added to >a in-memory hash table. It would be fairly easy to set a flag
there in the request >indication it was a first login (or even first login
from a given ip) and then send a >internally generated page. Ideally you'd
craft an acl list that is used to test for >when the page is sent (I'd be
happy to write you an acl type for the number of
>requests a given user has had satisfied.).
I am very greatful for any solution, but I fear I do not have the skills
required to follow through. Maybe Murphy here wants to take this further?
Received on Wed Apr 04 2001 - 17:14:40 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:59:09 MST