Text 28, 40 rader
Skriven 2005-02-08 09:57:38 av Maurice Kinal (1:261/38.9)
     Kommentar till text 24 av tony summerfelt (1:249/100)
Ärende: txt2pkt.pl
==================
Hey tony!

Feb 08 11:38 05, tony summerfelt wrote to Maurice Kinal:

 ts> i take it the scripts i worked on aren't readily available anywhere?

Which scripts are we talking about?

 ts> i've been working on converting them to ruby...

That I know nothing about.  Don't they, whoever they are, use that for embedded
type stuff?  I've heard of it but haven't ever looked at it.  Perl is good. 
:-)

 ts> i'm also in the process of finishing up an all encompassing log 
 ts> module for ruby. it's part of the port of tkblog to ruby...

I was thinking the XML-RPC server would handle that aspect.  Given that it is
doing the conversions and compliance work it seems logical the logs would be
there.  Also it could generate a html report of what transpired to the user if
that is something they wanted.  That way each user could be assured that their
particular session actually had some real effect and would have something to
show for their work.

What is bothering me at the moment is the MSGID generation.  I have figured out
how to avoid different users on a multiuser system creating the same MSGID in
the same area (a very real possibility) but am still pondering a different
routine then one based on day,hour,min,sec seeing that the txt2pkt.pl takes
only 0.053 seconds on average to do one message.  As is that means that any one
user has the potential of creating roughly 20 different messages per session
with the same MSGID, which would cause Fido compliant tossers upstream to barf
them out as dupes which they aren't.  I don't want to  slow this down any so
I'll have to consider something else, other then randomness that is.  Any
insight there you'd care to share?

Life is good,
Maurice

--- Msged/LNX 6.1.2
 * Origin: The Pointy Stick Society X1V - Louis Who? (1:261/38.9)