Re: [NEW] zone files parser

From: Jesse Becker (
Date: 10/29/02

On Tue, 29 Oct 2002, Torgny Bjers wrote:
> From: "Andrew Ermishin" <shicoy@MAIL.RU>
> > How about writing a new parser for zone-files? Something like this:
> > NUMBER_FIELD number
> > STRING_FIELD "string"
> > "or continue string"


> Well, the only real thing you actually accomplish by this is to increase the
> size of your world files.  If you want truly readable zone files you should
> go with XML instead, although that will eat memory and diskspace too, but it
> will be far easier to write offline rooms with, among other things.

XML would just serve to make the files even larger.  XML is a great idea
(in theory?), but it tends to merely add a *lot* of bloat to your data
files.  FWIW, the 'gnucash' project is a great example of this (it's a
Quicken-like program, very good--I recommend it).  They use XML files for
storage, which is good since you can hand edit them if you need to (and I
have, once, and it was my own fault).  The downside is that the files are
*huge*, even though they compress well (which of course defeats the
purpose of plain text in the first place...)

You also have a very valid point about memory usage, since CM would have
to have an XML parser either built into it, meaning there's more code to
maintain, or link against libxml, which adds a library dependency.  I'd
think that disk space for most muds isn't much of an issue these days
(unless you have to run under a truly draconian quota).  How much space to
really large and busy muds take up (excluding object files)?  I expect
it's much less than 30 megs (which is more than twice the space I'm

That said, a more readable format for mobs/objs/rooms would be a welcome
addition in my book--just like ascii pfiles is, IMO, a great improvement
over the original binary file format.


   | FAQ: |
   | Archives: |
   | Newbie List:   |

This archive was generated by hypermail 2b30 : 06/25/03 PDT