Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!wuarchive!zaphod.mps.ohio-state.edu!lavaca.uh.edu!uhnix1!starsoft!david From: david@starsoft.UUCP (Dave Lowrey) Newsgroups: comp.sys.amiga Subject: Re: Fish Contents On-Line Index Message-ID: Date: 25 Aug 90 23:52:54 GMT References: <1870@lpami.wimsey.bc.ca> <1990Aug18.054937.11246@zorch.SF-Bay.ORG> <81414@aerospace.AERO.ORG> <1990Aug21.123905.9773@murdoch.acc.Virginia.EDU> <2680.26d1d649@csc.anu.oz> Lines: 59 >In article <2680.26d1d649@csc.anu.oz> myb100@csc.anu.oz writes: >tmb@davinci.acc.Virginia.EDU (Thomas M. Breeden) writes: >>>>[discussion on Fish Discs on CD-ROM, + some index] >> >> Is there any current interest in an on-line "every word" index to the >> Fish Contents files (ie the program abstracts)? (Or does this facility >> already exist?) >> > >There is already an excellent program called Aquarium on Fish#301 which >does this: It has a database file containg *all* the Contents listing that >Fred puts out, and also a keyword listing for every entry on each disc. > I am working on, and have a very basic version of, just such a program. It takes it's input records, and "keywords" them, placing the records and keyword info in a 'database'. You then have a seperate "searcher" program that takes keywords and finds all of the records that contain the keywords. Since the keyword data is already built, it finds the records extremely FAST. Aquarium has to search the text each time you specify a new keyword. Keywords can be "anded and ored", and can contain wildcards (disk*). It is set up to work with any type of data, but I am using a fish-disk contents file for testing. The original text file is aprox. 1/2 meg in size. The database takes up a little more than a 1&1/2 megs. Projected release date of version one is the end of September. I am looking for "beta" testers who would have use for such a program. I need people who have databases that this could work with. You would have to build the input to my database-builder, but that is a very simple format. I see this product of use for searching customer, problem/fix, documentation, source, and other types of databases. The data needs to be static, but you can re-build the index database any time you wish. The fish database takes about 10 minutes to run on my 2500. It also takes up several megs of memory to run the build program, but the serach program doesn't take much (should run in 512K). If you think you would like to give it a try (for other than Fish Disk data), please drop me a line. Those who would like to use it with Fish Disk data should wait for the "official" release. -- ---------------------------------------------------------------------------- These words be mine. The company doesn't care, because I am the company! :-) Dave Lowrey | david@starsoft or {uhnix1,moray}!starsoft!david Starbound Software Group | Houston, TX | "Dare to be stupid!" -- Weird Al Yankovich