[518] in Coldmud discussion meeting
Re: To-Do list..
daemon@ATHENA.MIT.EDU (Fri Nov 4 18:12:19 1994
)
From: brandon@riverheights.declab.usu.edu
To: coldstuff@MIT.EDU
In-Reply-To: Your message of "Fri, 04 Nov 94 13:01:50 PST."
<199411042101.AA00504@crl4.crl.com>
Date: Fri, 04 Nov 94 16:05:38 -0700
< When DB size exceeds cache, objects/parameters/changes disappear spuriously
Just for reference, I've been running 0.12.0 in Ultrix for at least 2 (?)
months, with no problem. It is probably an OS specific thing (Linux? If so,
somebody try to track down the problem in dbm and mail the slackware guys?)
< todbref(int) => toobj(int)
The more I think about it, I think we should change all the 'to*()' function
syntaxes to be more standard, rather than using abbreviations. For instance:
tostr() => to_string()
toint() => to_int()
toliteral() => to_literal()
toobj() => to_objnum()
(etc)
< ebin/ -- directory where execute() programs are located
Actually (cosmetic here), I'd prefer executables/ or exe_bin/, or hmm,
execute_bin/ (hsm), perhaps execute/ (sigh).
< [pipes] Anyway, what do people think?
I like the idea of doing it through the same method as a connection, less
duplicity in-db (since it is doing the same thing).
< buffer_to_strings() => ascii_buffer_to_strings()
< strings_to_buffer() => strings_to_ascii_buffer()
I would prefer:
buffer_to_strings(buf) => buffer_to_strings(buf, ['type])
strings_to_buffer(buf) => strings_to_buffer(buf, ['type])
Where 'type defines what type of conversion to do (default 'ascii).
< "locked" compilation flag for nethods
"private" would be good too, in my opinion (would remove a lot of perm
checking which simply checks to make sure it is called by this object).
Also, I would (still) like to see the assignment operator change from '='
to ':='. I believe it would seriously help lessen confusion in the language,
in relation to people learning the language.
-Brandon