MOO-cows Mailing List Archive
LPMOO and binary db sillyness
Date: Fri, 1 Dec 1995 18:41:51 PST
From: Marc <email@example.com>
Content-Type: text/plain; charset=US-ASCII
You know, I had a couple laughs at the reasons why binary db's are
bigger, had a few REAL LARGE ones in fact...
Eventually, I assumed somebody would give the -real- reason, but I guess
everybody who's clueful enough to know thought the same thing, so I drew
the lucky straw....
------ PAY ATTENTION:
LPMOO binary databases are so significantly higher then the standard MOO
database for two -PRIMARY- reasons, there are others:
1) The LPMOO has to store original MOO source code -and- the
LPC translated equivalent. Remember folks, LPMOO is not a native
interpreter like MOO, it compiles it to the native LPC language and
then interpretes -that-, it needs to keep the original MOO source
around so you can edit it, list it.. etc..
2) There is overhead in LPMOO itself. Again, LPMOO is not `native',
there is significant overhead in the database to provide the
functionality of the conversion from MOO to DGD itself. All MOO
builtins have to be reimplemented
Ok, for some reason I assumed this was all in the LPMOO docs, so I was
getting huffy cuz people weren't looking, but I wasn't able to find it
myself, still, some of the answers were outright incredibly wrong.
Oh, and (on an ASCII based system) an A is saved as decimal 65 regardless
of if it is saved in a text file or a binary database. In fact, that's
kind of a silly distinction as the OS couldn't care less. `Text' files
are just binary files using only a subset of characters, with \n
(newlines) spread usefully thruout the document.
firstname.lastname@example.org Finger for PGP key and Geek Code
Subject Index |