[PENNMUSH] Converting a Win32 Database over to Unix

[K Moon]

On Mon, 24 Jan 2005, Chris Tankersley wrote:

> It looks like the problem I had was I chose the wrong compression type
> when I compiled under Fedora. Did the no compression option, copied
> the DB over, ran dos2unix on the DBs and it loaded just fine.  Now
> that the DB is running, is there any way to turn compression on?
>
> Chris

Yay.  Moving on... if you're talking about in memory compression, that's 
all handled on DB startup/shutdown, and is never written out to disk 
(except maybe via the chunk allocator?).  The main issue you need to be 
aware of in this case is whether or not you have 8-bit characters in your 
DB, and whether or not your compression type is 8-bit clean.  AFAIK, 
that's the only kind of compression which is a compile time option, and I 
don't even know if you have as many choices in 1.8.

Now, if you're talking about file compression (which would make more 
sense), you can just shut down your server, change the settings in the 
mush.cnf, and then (re)compress your databases with the appropriate 
commands before restarting.  For example, if you're using gzip, you can do 
something like gzip indb to get indb.gz.  If you have indb.Z, you'd do 
uncompress indb.Z first to get indb, then gzip indb to get indb.gz.  Etc.