bug#61730: 30.0.50; Compiler warnings for delq and delete

[Michael Heerdegen <michael_heerdegen@web.de>]

Eli Zaretskii <eliz@gnu.org> writes:

> > Thus such a warning definitely falls on the beneficial side. Let's do it.
>
> I'd hate to see Emacs's byte compiler becoming the Big Brother who
> knows better.  Richard described a couple of situations where the
> "problematic" usage is completely legitimate and safe

He mentioned only one: "when you know that the elements to be deleted
cannot include the first element", and that's the only one I know.  Are
there more?

Even that case is a burden to the reader and the maintainer, because one
has to think about and verify that this condition is fulfilled.  And IMO
this case is more dangerous than the `mapcar' case which just produces
slightly more inefficient code.

But yes, we would warn about some legitimate and safe calls.  The
advantages outweigh the disadvantages in my opinion.

> please try not to emit the warning in those cases.

I think this is impossible.  The compiler can't know or prove whether
the element to delete is different from the first one.


Michael.