On Saturday, June 8, 2002, at 04:49 , John W. Krahn wrote:
[..]
> for ( split/(\d+-\d+(?::\d+)?),?/ ) {
> next unless /\d/;
> @x = /(\d+)/g;
> if ( /-/ ) {
> print "@x", @x == 2 ? " 1\n" : "\n";
> }
> else {
> do {
> @y = ();
> unshift @y, pop @x until $#x == $x[-1] - $x[0];
> print "$x[0] $x[-1] 1\n";
> } while ( @x = @y );
> }
> }
Major SplitFoo Mojo There.... my complements!
for those who may want a little help unwrapping how that plays,
I did a bit of 'insert debug statements' to show how the process
plays out:
cf http://www.wetware.com/drieux/pbl/perlTrick/doWhileLooper.txt
my own old fashion 'pointer arithmetic' style approach is up at
cf http://www.wetware.com/drieux/pbl/perlTrick/walkCommaSep.txt
{ I just happened to notice that this version will be able to
manage the case of 'non positive integers' that are not attempting
to be a part of a 'range'.... }
myFascistHouseMate laughed at me, that I should not whine if
the spec did not call for solving
2,4,6,8
that one need not solve that to
2 8 2
that talkes a bit more work....
cf http://www.wetware.com/drieux/pbl/perlTrick/walkCom2.txt
when they say, there is more than one way - they are clearly
not joking... the complicated part remains how to resolve
which faction to support.... here, well... I still am clueless.
I have opted to avoid the question what to do with
0,1,2,4,6,8
should that be
0 2 1
4 8 2
or what.... pending a cleaner RFP/RFQ....
ciao
drieux
---
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
