[Groff] Re: .substring bug - indicies don't work as documented(?)

[Bernd Warken]

On Sat, Oct 27, 2001 at 07:23:07PM +0200, Werner LEMBERG wrote:
> 
> Now you'll get the following consistent results:
> 
>   for s=abcdefg .substring s  0   yields "g"
>   for s=abcdefg .substring s  0 1 yields "abcdefg"
>   for s=abcdefg .substring s  0 2 yields "bcdefg"
>   for s=abcdefg .substring s  0 3 yields "cdefg"
>   for s=abcdefg .substring s  1 1 yields "a"
>   for s=abcdefg .substring s  1 2 yields "ab"
>   for s=abcdefg .substring s  1 3 yields "abc"
>   for s=abcdefg .substring s  2 1 yields "ab"
>   for s=abcdefg .substring s  2 2 yields "b"
>   for s=abcdefg .substring s  2 3 yields "bc"
>   for s=abcdefg .substring s -2   yields "efg"
>   for s=abcdefg .substring s -1   yields "fg"
>   for s=abcdefg .substring s  0   yields "g"
>   for s=abcdefg .substring s  1   yields "abcdefg"
>   for s=abcdefg .substring s  2   yields "bcdefg"
>
BTW how about defining a single argument meaning the corresponding
character, e.g. ".substring s 1" the first letter and ".substring -1"
the last letter.

It is illogical to start with 1 from the front and with 0 from the end.
Moreover, 0 means everything up to behind the last character; so the last
character should be -1 anyway.

Also 2 args in reversed order could be made into reversing the output.

It seems to be fairly unused so far, so it's still time to make sound.

Bernd Warken