Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.mira.net.au!vic.news.telstra.net!act.news.telstra.net!psgrain!news.uoregon.edu!arclight.uoregon.edu!usenet.eel.ufl.edu!hookup!news.mathworks.com!news.kei.com!nntp.coast.net!news.sprintlink.net!news.interpath.net!usenet From: kpneal@interpath.com (Kevin P. Neal) Newsgroups: comp.os.linux.development.apps,comp.os.bsd.freebsd.misc,gnu.misc.discuss,comp.windows.x.apps,comp.unix.bsd.misc Subject: Re: Portable database library for Linux/FreeBSD Date: 27 Apr 1996 03:57:20 GMT Organization: Sweaty-Hot Apartment RetroComputing Lines: 68 Message-ID: <4ls5v0$764@news.interpath.net> References: <4lk4cj$jko@sidhe.memra.com> NNTP-Posting-Host: raleigh-5.interpath.net Mime-Version: 1.0 Content-Type: Text/Plain; charset=US-ASCII X-Newsreader: WinVN 0.99.7 Xref: euryale.cc.adfa.oz.au comp.os.linux.development.apps:14976 gnu.misc.discuss:24697 comp.windows.x.apps:11274 comp.unix.bsd.misc:861 In article <4lk4cj$jko@sidhe.memra.com>, michael@memra.com says... >I would like to see some coordinated development of apps for Linux/FreeBSD >that are designed first, then written, so that they have some staying >power and are attractive to a wider base of users than stuff like xfig and >ez. One of my goals is to have all those apps use a set of standard file >formats so that anyone who has a better idea can simply adopt the file >format (say for a word processor document) and write their own app to use >it. This would allow for much more diversity of apps than the MS Windows >environment because the data files would be portable from one system to >another. > >For instance, I have some ideas for a WP document format that >would make it both upwards and downwards compatible so that the first word >processor to use it will still be able to read and edit documents created >by the latest gee-whiz word processor. > Ok, I suggest looking at the Electronic Arts Interchange File Format, known as "EA IFF 85" or just IFF. It organizes data into chunks. You have a FORM, a 32-bit size, and then the actual form type (another four letters). Documents are created with specificially designed and documented form types. These forms contain in them chunks that are also designed and documented well. There is a standard Amiga library that parses IFF files. The library doesn't understand the contents of the chunks, but you can give the library a file and ask it for the first chunk of, say, type WORD (a real FORM type), and it will give it to you. A very nice design, very extendable. The spec says that any app that doesn't understand a chunk type should ignore it and not rewrite it. This means that FreeWord for Unix V1 can write files. Then the next version, FreeWord for Unix V2 can add more features, and write the files back out. The new can read the old. The newer features will be written as new chunks that the old will ignore, so the old could read the new files. There you go. This file format is general enough to be used for just about anything. I wouldn't recommend it for a database, but any other day to day item could (and probably already is used in this) be fine for this. >If anyone is interested in doing something like this in a collaborative >project like GNU, X Consortium, FreeBSD team, Linux developpers, then send >me some email and/or follow up to this post. Tell you what, you work on apps, and I can have a generic Free Unix IFF parser library written and put up on the web by the beginning to mid part of June, 1996. I'll even have docs for it. Email me here at Interpath, or check my web page for updates on my progress. Lemmie know what you think of my idea. -- XCOMM Kevin P. Neal, Sophomore, Comp. Sci. \ kpneal@interpath.com XCOMM Frue, Secret Agent of Smerp (shh!) \ kpneal@eos.ncsu.edu XCOMM Visit the House of RetroComputing at / Perm. Email: XCOMM http://www4.ncsu.edu/~kpneal/www/ / kevinneal@bix.com