Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!elroy.jpl.nasa.gov!ames!uhccux!munnari.oz.au!yoyo.aarnet.edu.au!frodo.cc.flinders.edu.au!gamgee.cc.flinders.edu.au!psjmt
From: psjmt@cc.flinders.edu.au (James Tizard)
Newsgroups: comp.databases
Subject: Should I use dbm
Message-ID: <1991Mar26.052521.14653@frodo.cc.flinders.edu.au>
Date: 26 Mar 91 05:25:21 GMT
Sender: @frodo.cc.flinders.edu.au
Followup-To: psjmt@research.cc.flinders.edu.au (James Tizard)
Distribution: aus,usa
Organization: Cognitive Psychology Lab, Flinders University, South Australia
Lines: 30
Originator: psjmt@gamgee.cc.flinders.edu.au

Hello to all database experts from one who isn't...
I am writing an EEG simulation program that generates LOTS of binary
time series data (probably > 20 Mb per session). I wish to store
these time series in a database that is accessible to both DOS & UNIX.
I am considering using the standard UNIX dbm routines, since

	a) They are free (i.e GNU gdbm)
	b) The same code compiles on UNIX & MS-DOS (it does too !)
	c) They seem to fill my needs for retrieval of binary records on
	   a single key.
	d) The perl language can treat dbm files as associative arrays

What I am hoping is that some people who know about these things may wish
to express an opinion on my choice. The sorts of things I would be interested
in would be

. Are there any hidden gotchas in dbm (or ndbm)
. Are there any tools other than perl that can handle dbm format databases
. Is it easy to convert to other formats if necessary
. Are there any other free or PD databases that I should consider
. Any other comments

Thanks in advance

-----------------------------------------------------------------------
 _--_|\        James Tizard                             
/      \       Cognitive Psychology Laboratory
\_.-*._/       Psychology Department                    
      v        Flinders University of South Australia   
               psjmt@cc.flinders.edu.au