Re^4: Efficient way to replace a set of values with another set of values

[roboticus]

in reply to Re^3: Efficient way to replace a set of values with another set of values
in thread Efficient way to replace a set of values with another set of values

space_monk:

OK, your solution is a bit better .... at least until you get to "27. This is a really long list!".

;^D

...roboticus

When your only tool is a hammer, all problems look like your thumb.

[ColonelPanic]

Ha ha. I spent an inordinate amount of time trying to make a one-liner solution for arbitrarily large numbers (where it would go to "aa" after "z"). However, it turned out that today is Friday and I can't think that hard.

[tobyink]

You were looking for:

s/^(\d+)/$n=$1;$l="a";$l++while--$n;$l/meg

Doesn't cope with 0 especially well.

[AnomalousMonk]

... until you get to "27. This is a really long list!".

Maybe something like (and 'zz' could easily be 'zzz'):

>perl -wMstrict -e
"my @lines = (
   '1. This is just a sample.',
   '2. This is to check',
   '3. How a set of values',
   '4. can be replaced by another',
   '5. set of values and that too',
   '6. in the most efficient way.',
   '99. And also handle large numbers.',
   '999. Oops, too big!',
   '99 bottles of beer on the wall does not translate!',
   );
 ;;
 my %xlate;
 { my @alphas = (undef, 'a' .. 'zz');
   @xlate{ 1 .. $#alphas } = @alphas[ 1 .. $#alphas ];
   }
 ;;
 s{ \A (\d+) (?= [.]) }
  { exists $xlate{$1}
    ? $xlate{$1} : (warn(qq{bad '$1' in '$_'}), '??')
    }xmse
  for @lines;
 ;;
 print qq{$_\n} for @lines;
"
bad '999' in '999. Oops, too big!' at -e line 1.
a. This is just a sample.
b. This is to check
c. How a set of values
d. can be replaced by another
e. set of values and that too
f. in the most efficient way.
cu. And also handle large numbers.
??. Oops, too big!
99 bottles of beer on the wall does not translate!