On Wed, 10 May 2006 at 10:04pm, Gerald Lai wrote:
> On Wed, 10 May 2006, Ben K. wrote:
>
> >
> > Hi,
> >
> > Is there a way to split a line automatically like awk would?
> >
> > Given "A quick brown fox jumped over ",
> > awk '{print $3}' ... ==> brown
> > or like in perl
> > split(':',$line)...
> >
> > I'd like to do within vim something like
> > :s/{some notation}/\3
> > without having to define the pattern
> > :s/\(\S\+\) \(\S\+\) \(\S\+\) ... /\3 ==> brown
> >
> > using white space, and if needed, by defining my own separator
>
> The regex to search for the <n>th occurence of a <search> is
>
> /^\%(.\{-}\zs<search>\)\{<n>}
>
> Given a line like
>
> :A:quick:brown:fox:jumped:over
>
> you can search for the <n>th word with
>
> /^\%(.\{-}:\zs[^:]*\)\{<n>}
>
> For example, if you want "fox", then do
>
> /^\%(.\{-}:\zs[^:]*\)\{4}
>
> You can easily modify the regex to use whitespace as delimiters.
>
> HTH.
In Vim7, you can use the split function as:
:s/.*/\=split(submatch(0))[2]/
The above will replace the whole currently line with the third word (0
based index).
--
HTH,
Hari
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
