Hello Maurice.

12 Feb 05 09:24, you wrote to me:


 RT>> A MSGID is a valued item here, especially when it comes to dupe
 RT>> checking, so
 RT>> if you are able to generate them fast enough, then please do so.

 MK> Generating them fast isn't a big issue but making them meaningful is a
 MK> tad tricky given the accepted "standard", if I can be so bold as to
 MK> call it a "standard".

Generating them fast enough is, it doesn't help that you generate a msgid, and 
two or more messages are tossed to that msgid.

Next is generating a msgid for each msg tossed, ie. not creating 100 msgid;s 
for 10 new messages, a waste of valuable resources.

Thereafter one needs to make it "meaningfull" :-))

How I see it ...

 MK> If based on time and/or date (is there a difference?) then it slows
 MK> things down and given the current limitation makes it even trickier
 MK> and slower.  That is why I thought a base ID at the start and
 MK> incrementation within a loop for multiple messages might be a better
 MK> objective as it wouldn't add any extra calls and is faster then
 MK> calling localtime for each one, which will produce dupes that aren't
 MK> dupes if limited to seconds on a fast machine.  This way it won't
 MK> matter even if the machine is slow enough to produce unique IDs using
 MK> seconds.  This method does speed things up and should produce unique
 MK> IDs for whatever length of years as bits allocated to the "year"
 MK> employed.

That leaves the question of how does one generate/create/format a 
"meaningfull" base ID, and still leave enough "space to add
"uniqueness"

Russell

--- GoldED+/LNX 1.1.4.7