SpellCheck

[DunbarX at aol.com]

Eric.

Right on all counts.

In HC I absolutely had to keep the dictionary in an external file and load as 
needed. TextEdit limits (30K) would not allow a field to hold the data.

I had a "spellCheck" handler run through the text and present all 
questionable words and their possible corrections in a list in an external window. Pretty 
much like spell checkers generally do. I am just starting to learn 
Transcript. I did not intend this to be pretty, in fact I was just trying to see if I 
can work this Rev thing at all.

I had tried to cover what I thought might be the most common spelling or 
typing mistakes. More work would be needed to cover transposed letters, or much 
more deviant misspellings. All three routines I used are pretty much brute force 
text crunching. I am trying to think of a way to find close or nearly colose 
text matches in a more elegant way; some sort of hash table where the numeric 
value of a word can be matched within a certain range of the numeric values of 
other words. The number of the beast. How do the grownups do it?

Craig Newman


> Not tested your script but two helpful thoughts, I hope:
> 
> 1. Having your dictionary in a field does not sounds so good: you 
> should keep it in an external file, load it when needed in a global 
> for instance and delete this global when done. This is for RAM usage.
> 
> 2. You use answer dialogs: this does not sounds so good too from an 
> ergonomic point of view. Probably could you check last word (when a 
> space or a punctuation is typed) and underline it and/or change its 
> color for instance. Then a pop up menu would display what you call 
> 'possibles' in your script and replace what the user typed if he 
> chooses it
> 


**************
Make your life easier with all your friends, email, and 
favorite sites in one place.  Try it now. (http://www.aol.com/?optin=new-dp&
icid=aolcom40vanity&ncid=emlcntaolcom00000010)