How to recognize which word from a list?

Xeratul · 06-10-2013, 01:38 AM

Hi,

I have a list of words which are my database, and I would
like that linux (aspell, ...or other non x11) can find
out which matching word it might be.

List={ "scanner" , "printer" , "hardware" , "processor" }

Into my input, if I give : "scahnner" or "trasher", it will give me the solution (closer "wording") :
scanner

Which program could do similar thing? I thought aspell, but nope

Thank you in advance if you have any possible idea.
Regards

Ygrex · 06-10-2013, 04:30 AM

probably this algorithm can be helpful: http://en.wikipedia.org/wiki/Levenshtein_distance

chrism01 · 06-10-2013, 06:56 AM

You could try http://perldoc.perl.org/perlfaq4.htm...of-a-string%3F

David the H. · 06-11-2013, 12:52 PM

TRE is an approximate matching regex engine. It comes with tre-agrep, for direct use.

http://laurikari.net/tre/

There's not a whole lot of documentation for it though. I haven't yet figured out more than a few basic configurations so far myself.

There are several other approximate matching solutions out there, I believe, including an older agrep.

Edit: fstrcmp might also come in useful, although it appears to be more of a string-to-string comparison tool. Again, it provides a simple shell interface in addition to the library itself.

http://fstrcmp.sourceforge.net/