Hi Junio

On 30/05/2024 18:09, Junio C Hamano wrote:

"Phillip Wood via GitGitGadget" [off-list ref] writes:

quoted

  1:  1bcf92c6105 ! 1:  91c6f2f1b45 rebase -i: pass struct replay_opts to parse_insn_line()
      @@ builtin/rebase.c: static int edit_todo_file(unsigned flags)
       @@ builtin/rebase.c: static int do_interactive_rebase(struct rebase_options *opts, unsigned flags)
        		error(_("could not generate todo list"));
        	else {
      - 		discard_index(&the_index);
      + 		discard_index(the_repository->index);
       -		if (todo_list_parse_insn_buffer(the_repository, todo_list.buf.buf,
       -						&todo_list))
       +		if (todo_list_parse_insn_buffer(the_repository, &replay,

OK.  It would probably have been unnecessary to rebase only for this
update.

I agree the conflict is easy to solve but GitGitGadget gets upset and 
refuses to run the ci if it cannot merge the branch into master

quoted

      + ## Documentation/config/advice.txt ##
      +@@ Documentation/config/advice.txt: advice.*::
      + 		`pushNonFFCurrent`, `pushNonFFMatching`, `pushAlreadyExists`,
      + 		`pushFetchFirst`, `pushNeedsForce`, and `pushRefNeedsUpdate`
      + 		simultaneously.
      ++	rebaseTodoError::
      ++		Shown when there is an error after editing the rebase todo list.

This thing is new.  It is unclear to me if this description is clear
enough to readers that we are checking the edited todo list for
errors.

I'm happy to improve it if you have any suggestions - I had hoped "after 
editing" would be clear enough.

 It is clear enough from the actual code change, and the
readers come to this list after advise_if_enabled() triggers and
reports that the rebaseTodoError knob allows them to squelch it, so
it probably is OK.

Thanks, will replace.  Let's see if we see comments from others and
then mark it for 'next' soonish.

Thanks

Phillip