bug#33749: 26.1; input-decode-map to empty vector should preserve echo area

[Stefan Monnier <monnier@IRO.UMontreal.CA>]

>> I think The Right Thing to do is likely to move this code to
>> read_key_sequence, more specifically, move it to the point where we
>> *know* we really do have an event.
>> IOW, I think the patch below might be a better option (where we test
>> `indec.start > 0` to make sure some *decoded* event was read).
> Where in that patch is the variable that I asked for, which would
> trigger this special behavior?

Nowhere ;-)

Mostly because I don't know where we could set this variable (other
than globally).

If you want me to add a global var to control whether we use the old or
the new behavior, I'm fine with that.

Note also that the current patch is likely not quite right yet anyway:
I mostly posted it for discussion.  E.g. I think where I put it
currently, it fails to be run for single-key commands.


        Stefan