OpenEuphoria

RE: Fast "locate" function

• Posted by Matthew Lewis <matthewwalkerlewis at YAHOO.COM> Dec 10, 2001
• 352 views

> -----Original Message-----
> From: euman at bellsouth.net [mailto:euman at bellsouth.net]

> This is the best Ive seen in a while....

Thanks. :)

 
> function find_all_2(object test, sequence data)
>         integer ix, jx, len
>         sequence result
>         result = {}
>         ix = 1
>         len = length(data)
>         jx = find( test, data[ix..len] )
>         while jx do
>                 result &= ix+jx-1
>                 ix += jx
>                 jx = find( test, data[ix..len] )
>         end while
>         return result
> end function

However, when I wrote this, I knew I was looking for one-element items, not
a sequence (as you seem to be interested in).  You could easily convert to
find multiple occurences of your data: 

function find_all_3(object test, sequence data)
        integer ix, jx, len, t_len -- add t_len for the length of your test
object
        sequence result
        result = {}
        -- add this to get the length of the test object
        if atom(test) then
            t_len = 1
        else
            t_len = length( test )
        end if

        ix = 1
        len = length(data)
        jx = match( test, data[ix..len] ) -- change find() to match()
        while jx do
                result &= ix+jx-1
                ix += jx + t_len  -- add the length of the test object to
                                  -- continue searching past the end of 
                                  -- the last match
                jx = match( test, data[ix..len] ) -- change find() to
match()
        end while
        return result
end function

The function returns a sequence of indices pointing to the first elements of
the starts of your test objects within the sequence.  I've found myself
using this 'double' indexing a lot when using sequences.

Matt Lewis