On 04.11.26 22:38, Jelte Fennema-Nio wrote:
On Wed, 4 Nov 2026 at 12:08, Peter Eisentraut <[email protected]> wrote:
If we want to make this behavior change, then this might be better done
as a separate patch before the refactoring.
Agreed. I had somehow missed that there were different sigjump targets
and not all of them set the cancel_requested flag. Attached is a new
patchset with this behaviour change done first explicitely. Because I
agree it's a bug that Ctrl+C in a promt doesn't stop the script.
I have committed the 0001 patch and I will continue looking at the rest.
