furrykefx
07-01-2004, 09:05 AM
I've been working on a relative search program...I'd say the current incarnation would be version 0.2 if I had to put a number on it. You may be wondering why we need yet another relative searcher, but I haven't found one that has all the features I like...such as the ability to search some large number of large files. And it should support byte skipping. Nothing else I know of allows all this!
Currently my script can search a 15.4 MB file in six seconds (not as fast as Jair's program, but not bad for a Python script). With a search string that uses byte skipping, it takes longer: 44 seconds. (I'm working on optimizing that...) It's not a speed demon but at least it's flexible.
By the way, it's currently all command line. I'll try to add GUI support, too (it'll still support the command line), but if it doesn't work out well I'll axe it. I don't expect much problem with it, though...
I have three questions:
- What should I name it? (It's currently called KSearch, which is kinda boring.)
- Who wants to test it?
- What would you guys like to see in a relative search program?
EDIT: I just added the ability to search for hex strings, so it's in version 0.3 now. Support for custom relative tables will probably come soon (a la Romanji Search; this tool will hopefully supersede it, too); the program has been written with them in mind from the beginning.
By the way, the program's a little difficult to set up on Windows, so in addition to the Python script there will be a precompiled EXE available. Unfortunately it'll be a little large (1.6+ megabytes uncompressed, but maybe 800k compressed...definitely more if I add GUI support).
I'll probably post the script up here soon, but I want to refine it just a little first. And I gotta write the README file.<P ID="edit"><FONT class="small">Edited by furrykefx on 07/01/04 07:36 AM.</FONT></P>
Currently my script can search a 15.4 MB file in six seconds (not as fast as Jair's program, but not bad for a Python script). With a search string that uses byte skipping, it takes longer: 44 seconds. (I'm working on optimizing that...) It's not a speed demon but at least it's flexible.
By the way, it's currently all command line. I'll try to add GUI support, too (it'll still support the command line), but if it doesn't work out well I'll axe it. I don't expect much problem with it, though...
I have three questions:
- What should I name it? (It's currently called KSearch, which is kinda boring.)
- Who wants to test it?
- What would you guys like to see in a relative search program?
EDIT: I just added the ability to search for hex strings, so it's in version 0.3 now. Support for custom relative tables will probably come soon (a la Romanji Search; this tool will hopefully supersede it, too); the program has been written with them in mind from the beginning.
By the way, the program's a little difficult to set up on Windows, so in addition to the Python script there will be a precompiled EXE available. Unfortunately it'll be a little large (1.6+ megabytes uncompressed, but maybe 800k compressed...definitely more if I add GUI support).
I'll probably post the script up here soon, but I want to refine it just a little first. And I gotta write the README file.<P ID="edit"><FONT class="small">Edited by furrykefx on 07/01/04 07:36 AM.</FONT></P>