Re: Gawk 3.04

[Aharon Robbins]

Hi.  Thanks for the suggestion.  Unfortunately, given the state of the
gawk internals, this is very difficult to do. It also makes the semantics
of tolower/toupper etc much less orthonogal.

This task is relatively easy to do with an awk function, split(), substr(),
and some local variables:

function upcasewords(s,         a, n, i, result)
{
        n = split(s, a, " ")    # split s into array a
        for (i = 1; i <= n; i++)
                a[i] = toupper(substr(a[i], 1, 1)) substr(a[1], 2)

        result = a[1]
        for (i = 2; i <= n; i++)
                result = result " " a[i]

        return result
}

Usage would be:

        s = upcasewords("xxx yyy zzz")

Hope that helps,

Arnold

> From: address@hidden
> Date: Thu, 1 Feb 2001 02:00:12 GMT
> Subject: Gawk 3.04
> To: address@hidden
> To: address@hidden
> To: "Gonzalez, Lissette" <address@hidden>
>
> it's been a occuring theme for me to run into this problem which - if solved
> - would add alot of power/functionality to gawk:
>
>     s="xxx yyy zzz"
>     gsub(/^./, toupper("&"), s)
>     gsub(/ ./, toupper("&"), s)
>
> as an over-simplistic example of intending to capitalize the 1st char of all
> words in a sentence. but gawk doesn't seem to allow the matched r.e. to be
> passed to any string returning functions for processing.  this would be a
> very powerful enhancement without breaking any existing programs - these
> types of operations are very unwieldy to do in other ways.  it also
> has a feel of being "sensible", since "&" should be holding the
> match throughout the middle parameter of g/sub functions. ?
>
> thank you for all your work. :)
>
> Chris.