Ignore punctuation when addressing word n of a string [more info]

[Ken Ray]

> Okay, so "word" automatically assumes white space as 
> delimiter.  I can't skim out the punctuation, so can someone 
> suggest how I do the following. (This example is closer my 
> real life problem).

Mark, here's a way to strip out punctuation - the regEx way:

on mouseUp
  put "now, that is a string!" into tString
  put GetWords(tString) into fld 1
end mouseUp

function GetWords pWhat
  local tWordNoPuncs
  put "" into tWordList
  repeat for each word tWord in pWhat
    get matchText(tWord,"([A-Za-z]*)",tWordNoPuncs)
    put tWordNoPuncs&cr after tWordList
  end repeat
  delete last char of tWordList
  return tWordList
end GetWords

> Within a bunch of text, I want to replace any string of pattern 
> 
> 00-000
> 
> (where 0 is any integer and the dash is a literal dash) with 
> a string of 
> 
> <A HREF="#00-000">00-000</A>
> 
> In other words, I am looking to bracket the pattern string 
> with an HTML index tag that uses the same literal string as a 
> named target.  Short of doing a char-by-char crawl, how can I 
> accomplish this search and replace?  

And for this one, try this:

on mouseUp
  put "Yada 55-432 Yada" into tString
  put HREFIt(tString) into fld 1
end mouseUp

function HREFIt pWhat
  local tText
  get matchText(pWhat,"([0-9][0-9][-][0-9][0-9][0-9])",tText)
  put "<A HREF=" & quote & tText & quote & ">" & \
    tText & "</A>" into tReplaceVal
  replace tText with tReplaceVal in pWhat
  return pWhat
end HREFIt


Have fun!

Ken Ray
Sons of Thunder Software
Email: kray at sonsothunder.com
Web Site: http://www.sonsothunder.com/