Hi Harald,
OK for trunk? And for backports to 8/7?
I played around with your patch and found a few problems with
ICEs, but these were all pre-existing as far as I could determine;
I have submitted PR89266 for what I discovered.
I a bit concerned about the case of like a(i:i-2) returning negative
lengths. Could you maybe set the length to zero if it is
calculated to be negative?
I'd say that the patch is OK for trunk with that change, even though
we are technically in regression-only mode. It is fairly localized,
and should not have ill effects. Regarding backport to the other
open branches - well, it does not fix a regression, so I'd be
inclined not to backport it.
Regards
Thomas
