[838] in Coldmud discussion meeting
tcd & etc
daemon@ATHENA.MIT.EDU (Sun Oct 29 15:12:51 1995
)
Date: Sun, 29 Oct 1995 13:07:42 -0700
From: 869683 Gillespie Brandon James <brandon@avon.declab.usu.edu>
To: coldstuff@pippin.ece.usu.edu
In lieu of optimization I have changed how the .namef method functions.
Before it was rather configurable, taking optional methods/arguments on
the fly. However, this was not very optimal, and since I could not find
a single place in the entire db which actually took advantage of this
(everywhere simply uses one of the 'symbol shortcuts which expands to
something in the normal format, like .namef(['name], "(", ['activity], ")");
Because of this .namef() simply takes one flag, which specifies what
type of formatted name to give, handled by a switch. The flag is required
and therefore caused some problems (which are mostly fixed now). This
is much faster, trimming off ~200+ ticks for each call (not much, unless
you consider something such as a who listing with ~100 people on it).
The functionality of .namef _IS_ still available however, in obj.build_name().
This method simply accepts a variable number of lists or strings, with
the strings being concatenated onto the result, and lists being assumed
to be a method + arguments, which is expected to return a string.
The reason for mailing to this list is because of its impact on ColdCore.
I noticed there were some who attempted to connect to tCD through the night,
it is functional again, as my conversion in the above caused some problems
which I have fixed.
-Brandon Gillespie-