Note for pre 1.1 users: SWISH has changed considerably since version 1 .0! The format is different and configuration variables have changed. 1.0 SWISH users should reindex their files using version 1.2.1. Please read the list of changes since version 1.0.
[Index] [Next Chapter]
What is SWISH?SWISH stands for Simple Web Indexing System for Humans. With it, you can index directories of files and search the generated indexes.
For an example of swish can do, try various keyword searches on my home page. Everything you see there were indexed by swish. When you do a search, it's the swish program that's doing the actual searching (but hidden behind a web-based interface called w4ais).
SWISH vs Other ProgramsSWISH was created to fill the need of the growing number of Web administrators on the Internet - many current indexing systems are not well documented, are hard to use and install, and are too complex for their own good. Here are some pros and cons regarding SWISH:
- It's simple.
We've tried to make SWISH as simple as possible while keeping some of the things that people look for in an indexer. The drawback is that you can't do many things that full-featured indexers and searching programs can do, such as stemming (searching for different versions of a word) or the use of synonyms.
- It's made for Web sites.
In indexing HTML files, SWISH can ignore data in tags and give higher relevance to information in header and title tags. Titles are extracted from HTML files and appear in the search results. SWISH can automatically search your whole Web site for you in one pass, if it's under one directory. You can also search for words that exist in HTML titles, META tags, comments, and emphasized tags, and 8-bit HTML characters can be indexed, converted, and searched.
- It's fairly nice on disk space and is pretty fast.
Index files consist of only one file, so they can be transported around and easily maintained. The SWISH source is not large and generated indexes average out to around half the size of comparable WAIS indexes, or 1 to 5% of the size of your original HTML data. Searching is as fast as or better than using a non-commercial WAIS-based solution.
- You can fix the source.
I encourage people to send in patches and suggestions on how to make SWISH better. Although it's not in the public domain, I am always more than happy to integrate contributed code into the distribution. Please note the license concerning its use. If you do have licensing questions, you should theoretically contact Jay Weber at email@example.com. Good luck!
Platforms SupportedEverything was written in pretty vanilla C, so it should work just about anywhere (at least anywhere with UNIX-style file I/O and directory names). It has been compiled with gcc or cc (and often both) on the following platforms by the current maintainer:
Others, including the original author, have reported successful compilation and usage on these platforms as well:
- AIX 4.2
- FreeBSD 2.2.2
- HP/UX 9 and 10
- Linux 1.2.8 and 2.0.34
- Solaris 2.4 and 2.5
If you have successfully compiled SWISH for platforms other than these, please let me firstname.lastname@example.org know.
- BSDI 1.1
- IRIX 5.2 and 5.3
- OSF/1 2.0
- SCO x.x
- SunOS 4.1.3
Great! How do I get started?If you don't already have it, you will need a current copy of the source from
http://www.rru.com/~meo/useful/www.html#swish. Unpack the source and compile and install swish, per the directions in the SWISH Installation Guide. Then, configure SWISH per the instructions in the SWISH Configuration Guide. Setting up indexes and searching them is covered in the SWISH Users Guide.
You will probably also want to install the w4ais package to provide a nice, web-based front end, unless you plan to use SWISH in a non-web environment.
[Index] [Next Chapter]
Last update: 18/Aug/1998