Re: [PATCH 1/2] factor: add option for printing in x^y format

[Pádraig Brady]

On 27/04/2022 19:36, Pádraig Brady wrote:
> On 27/04/2022 18:21, Bernhard Voelker wrote:
> > On 4/27/22 16:14, Pádraig Brady wrote:
> > > On 27/04/2022 11:07, Rasmus Villemoes wrote:
> > > > When factoring numbers that have a large 2^n factor, it can be hard to
> > > > eyeball just how many 2's there are. Add an option to print each prime
> > > > power factor in the x^y format (omitting the exponent when it is 1).
> > > >
> > > > * src/factor.c: Add --exponents option for printing in x^y format.
> > > >
> > > > Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
> > >
> > > Thanks for the suggestion.
> > > This is one of those borderline cases.
> > >
> > > Mainly because it's fairly easy to postprocess the output.
> > > Consider for example:
> > >
> > >      $ factor 13 36 | tr ' ' '\n' | uniq -c
> > >            1 13:
> > >            1 13
> > >            1 36:
> > >            2 2
> > >            2 3
> > >
> > > Though for interactive use, the new option is a benefit.
> > >
> > > Given the interactive benefit and a previous request to add this functionality¹
> > > I'm tentatively leaning towards adding this.
> > >
> > > cheers,
> > > Pádraig
> > >
> > > ¹ https://lists.gnu.org/archive/html/coreutils/2017-11/msg00015.html
> >
> > Regarding the postprocess solution:
> > this is a little awk(1) snippet to get the same output as with the patch:
> >
> >     $ factor 13 36 | tr ' ' '\n' | uniq -c | awk '\
> >         BEGIN { n="" }
> >         /:$/  { printf("%s%s", n, $2); n="\n"; next };
> >               { printf(" %s%s", $2, ($1>1?"^"$1:"")); }
> >         END   {printf("%s",n); }
> >         '
> >     13: 13
> >     36: 2^2 3^2
>
> nice!
> > Re. the patch:
> > the implementation is quite clean in 2 places - the LONGINT_OK and the MPZ case 
> > - and
> > doesn't seem to get into the way of any future extensions (null-separated 
> > output, whatever).
> >
> > Therefore, I'm also 70:30 for adding it.
> > Of course, we'd also need a long option and texinfo documentation.
> >
> > Finally, regarding '-e': is there any precedence (or a clashing option) in any 
> > other
> > factor(1) implementation?
>
> We generally avoid new short options where possible.
> This would just need the long option to enable this
> given it's kind of esoteric and not often used interactively.

I see NetBSD has used '-h' for this "human readable" functionality.
I suppose for more compat with that we could have:

  -h, --exponents

We might leave the -h undocumented,
for the reasons stated above.

cheers,
Pádraig