bug#44365: 27.1; intern-soft given a symbol never returns nil

[Robert Weiner <rsw@gnu.org>]

[-- Attachment #1: Type: text/plain, Size: 1269 bytes --]

Hi Lars:

I would have only sent intern-soft a string if it were not for the doc
string which says it takes a string or a symbol.

"Return the canonical symbol named NAME, or nil if none exists.
NAME may be a string or a symbol.  If it is a symbol, that exact
symbol is searched for."

So, the doc string must need changing if your view is correct.

Bob

On Sun, Nov 1, 2020 at 8:19 AM Lars Ingebrigtsen <larsi@gnus.org> wrote:

> Robert Weiner <rsw@gnu.org> writes:
>
> > If I first call intern-soft with the symbol-name of an unbound symbol,
> > then it returns nil properly.  But if I instead give intern-soft the
> > symbol, it always returns the symbol rather than nil.  Maybe it
> > is because the internal call to symbol-name interns the symbol before it
> can test
> > whether it is bound.  Isn't this a bug?
>
> So your test case is
>
> (intern-soft 'does-not-exist)
>
> ?
>
> That's not a valid use case for the function -- the reader will intern
> `does-not-exist', and so the `intern-soft' call will just return the
> interned symbol.  The function is meant to be used like:
>
> (intern-soft "does-not-exist")
>
> So I'm closing this bug reprt.
>
> --
> (domestic pets only, the antidote for overdose, milk.)
>    bloggy blog: http://lars.ingebrigtsen.no
>

[-- Attachment #2: Type: text/html, Size: 2678 bytes --]