RE: fix for bug 10994 breaks ido customizations in major way

["Drew Adams" <drew.adams@oracle.com>]

> > (let ((t1 (propertize "aaa" 'aaa 12))
> >       (t2 (propertize "aaa" 'aaa 11)))
> >   (ido-completing-read "?: " (list t1 t2 "sfd")))
> >
> > works as expected. And the above patch breaks that.
> 
> That would be a horrible UI.  Luckily AFAICT, it hasn't happened.
> That's why I say there is no actual valid use-case for repeating the
> same string in completions.

I don't care much about Ido, so I don't have an opinion about the question at
hand.

However, I will say that it is NOT at all the case that there is nothing out
there, in the wild, that takes advantage of completion candidates that have the
same text (same string chars) but that have other differences, whether the
latter be text properties, symbol plists, associated files/buffers, associated
buffer positions or urls or other locations - or whatever.

Icicles is an example.  There are _loads_ of commands in Icicles that take
advantage of this.  And there are many different ways that the differences,
beyond the candidate text, can be communicated to users.  And there are
different ways that users can choose among such otherwise identical candidates.

The possibilities are too numerous to mention here.  Suffice it to say that,
yes, there are Emacs applications that take advantage of candidate strings (and
symbols) that have text (and symbol) properties.

And yes, there are also applications that take advantage of the fact that a
completion candidate can be a structure (typically an alist element that is a
cons), of which the displayed string is only one component.  Matching can
involve any number of such components, as can the value returned as the user's
completion choice.

FWIW, wrt removing duplicates (and duplication can be defined various ways in
different contexts, as just mentioned), Icicles lets you hit a key (`C-$')
during completion to toggle whether to remove duplicates (or to otherwise
transform the set of matching candidates, depending on the command and user
settings).

HTH.