bug#25193: [PATCH] Improve the doc of re-search-forward and re-search-backward.

[Tino Calancha]

On Sun, 2 Apr 2017, npostavs@users.sourceforge.net wrote:

> Tino Calancha <tino.calancha@gmail.com> writes:
>
> > Thanks for the report.
> > You could refer for details to the manual, maybe providing a link to the
> > proper node; but you don't want to refer to the doc string of another
> > function 'B' to document the arguments of the current function 'A'.
> > IMO the doc string of 'A' must introduce all its arguments.
> >
> > Otherwise, i am worry you could go an step further, f.i.
> > `search-forward'/ `search-backward' share the same optional arguments, so:
>
> I think it's okay to point to other functions, as long as we keep the
> chain to length 1.
Maybe OK (see below for the rest of my answer, like in my example :)
> > 1) doc string `re-search-backward': See the doc string of `re-search-forward'
> >    for details.
> > 2)  doc string `re-search-forward': See the doc string of
> >    `search-backward' for details.
> > 3)  doc string `search-backward': See the doc string of
> >    `search-forward' for details.
> > 4)  doc string `search-forward': Wow, you are are very persistent
> >     user! Please see the manual for details, i am a very busy doc string.
>
> :D
>
> But if all of `re-search-backword', re-search-forward',
> `search-backward', would say "see doc string of `search-forward'" I
> think it would be okay.
OK for me if it's just 1 jump, but i slightly prefer self-contained
docstrings.  Don't know what Eli or John might think about it.