[rrd-developers] Re: Some minor changes proposal

Tobias Weingartner weingart at cs.ualberta.ca
Fri Oct 1 18:17:44 MEST 1999


On Friday, October 1, Alex van den Bogaerdt wrote:
> 
> It is alright to create aliases for AVERAGE and LAST (AVG and LST),
> and you prefer to include INT over DIV.
> 
> What about both INT and DIV and give the user a choise ?
> 
> And since you mentioned INT, perhaps also include ROUND ?
> It can be "x,/,INT" vs. "x,DIV", and "x,0.5,+,INT" vs. "x,ROUND".

My reason for disagreeing with the "LST" alias, is the fact that it's
not really a standard abreviation.  Simply having a TLA for every OP
is IMHO not the way to go.  I can see the point to AVERAGE => AVG, as
I'm a lazy bastard.  But how are you going to go with ROUND?  What
about RANDOM?  Are they both going to be RND?

Also, I could make as strong an argument to not use "INT", but instead
introduce a new operator called "#", which will give you the integer
part of the number on the top of the stack [push(int(pop(TOS)))].  Then
the above could be replaced with: "x,/,#" and "x,0.5,+,#".  This way
keeping things in the same vein as other +,-,/,%,*,...  Of course, this
would introduce a slightly non-standard operator, one I'd be quite willing
to see not implemented...


Just my $0.03CDN worth...

--Toby.
*----------------------------------------------------------------------------*
| Tobias Weingartner | Email: weingart at cs.ualberta.ca | Nothing here yet     |
| Apt B 7707-110 St. |--------------------------------| Nor here yet...      |
| Edmonton, AB       | Unix Guru, Admin, Systems-Dude | Nor here...          |
| T6G 1G3     ---------------------------------------------------------------| 
| Canada     / %SYSTEM-F-ANARCHISM, The operating system has been overthrown |
*----------------------------------------------------------------------------*

--
* To unsubscribe from the rrd-developers mailing list, send a message with the
  subject: unsubscribe to rrd-developers-request at list.ee.ethz.ch



More information about the rrd-developers mailing list