Skip to main content.
home | support | download

Back to List Archive

RE: Combining SWISH++ and Swish-E

From: Mark Gaulin <gaulin(at)not-real.globalspec.com>
Date: Mon Nov 02 1998 - 16:48:38 GMT
Personally I would like one code base too. Right now swish-e is
the "official" version and swish++ is (correct me if I'm wrong) one
person's effort. In either case it is difficult for any programmers
other than the people directly in charge of the code to make changes.
It would be nice if a distributed, "open source" system was put in place
so that individual contributors could get in, make changes, and
get out. (I have no direct experience with such a system... they might
suck to work in too.) 

And, for the record, I would love to use C++ over C for this, or any
project. I chose to work with swish-e over swish++ because the
front end for swish++ was (at the time) more specific to what the
author wanted. Hey, maybe porting swish-e to C++ would solve those
nagging ansi/not-so-ansi C porting issues we had when 1.3 was
released.

Lastly, an important question is: Why have two competing systems at all?

	Mark

At 08:28 AM 11/2/98 -0800, Earl Robinson wrote:
>Paul et al,
>
>Regarding new versions of Swish[-e|++].. Is anyone planning to move 
>features of swish++ into the swish-e codebase. I'd like to have the 
>featureset of both programs, but since they use different index-file 
>formats, You have to choose one or the other. My problem is that I'm not 
>a coder, so I can't/won't write the web front-ed for swish++, and Swish-e 
>can't handle the size of my document tree without lots of RAM. It's speed 
>starts decreasing exponentially when it starts using swap space.
>
>In other words, I'm saying what alot of people have said. I want to use 
>Swish++ with Autoswish. I'd be willing to help test it, but like I said, 
>I'm not a coder, so my ability to help in that area is very limited.
>
>earl robinson
>
>
>Paul J. Lucas
>
>>	The changes are attached from the 'Changes' file.  This version
>>	includes the aforementioned feature whereby you can dump the
>>	words "around" the query words, e.g.:
>>
>>		$ search -w7 electrical
>>		elapse
>>		elect
>>		electric
>>		electrical
>>		electricity
>>		electrocute
>>		elevator
>>
>>	Above, '7' specifies the window size (at most).  You can
>>	optionally specify the number of characters that must match,
>>	e.g.:
>>
>>		$ search -w7,4 electrical
>>		elect
>>		electric
>>		electrical
>>		electricity
>>		electrocute
>>
>>	Above, "elapse" and "elevator" were not output because the
>>	first 4 characters of the words do not match "elec": the first
>>	4 characters of "electrical".
>>
>>	Creating a web interface using this new feature is left as an
>>	exercise for the reader.  (My off-the-top-of-my-head suggestion
>>	would be to create one <SELECT SIZE=n> ... </SELECT> per query
>>	word and allow the user to refine their query from the words in
>>	the index.  I would be interested in seeing your solutions.)
>>
>>	- Paul
>>
>>
>>NEW FEATURES
>>------------
>>
>>* In "search," a "window" of words can be dumped around the query words.
>>
>>* In "search," the -d option to dump the index for a word now dumps all the
>>  query words instead of a single word.  Additionally, a stop-word used to
>>  print "stop-word"; now it prints "# ignored: " followed by the word.
>>
>>* In "search," the -d option to dump the index for a word now prints the
>>  comment:
>>
>>	# not found: word
>>
>>  if 'word' is not found in the index.
>>
>>
> 
Received on Mon Nov 2 08:58:05 1998