As an example of usage, this bundle was created by calling darcs send
--reverse and answering y then d. This is much more darcsish than using 'w' until
the wanted patch appears, or using --match.
Wed Nov 11 17:32:50 CET 2009 Florent Becker <florent.becker@ens-lyon.org>
* Add --dont-ignore-times counterpart to --ignore-times
Wed Nov 11 18:27:39 CET 2009 Florent Becker <florent.becker@ens-lyon.org>
* add --dont-delete-log-file option
Wed Nov 11 19:23:01 CET 2009 Florent Becker <florent.becker@ens-lyon.org>
* add a --traditional counterpart to --unified
Wed Nov 11 20:28:04 CET 2009 Florent Becker <florent.becker@ens-lyon.org>
* add --apply-on-disk as counterpart to --store-in-memory
Wed Nov 11 20:54:43 CET 2009 Florent Becker <florent.becker@ens-lyon.org>
* add --no-happy-forwarding
Mon Nov 2 17:55:51 CET 2009 Florent Becker <florent.becker@ens-lyon.org>
* Refactor Darcs.Commands.Pull
Fri Nov 13 15:33:46 CET 2009 Florent Becker <florent.becker@ens-lyon.org>
* add --reverse to all patch-selecting commands
For instance, darcs send --reverse allows a sometimes more natural
goal-driven choice of patches: "goal" patches are offered first,
then their dependencies are added automatically.
This patch is probably wrong, even if I have not yet been able to find a
counter-example. The preselection of patches (patches_to_consider) in the Last
and FirstReversed cases seems fishy (it was not used before). Tips by anyone who
understands SelectChanges welcome.
On Fri, Nov 20, 2009 at 2:13 PM, Florent Becker <bugs@darcs.net> wrote:
>
> Florent Becker <florent.becker@ens-lyon.org> added the comment:
>
> This patch is probably wrong, even if I have not yet been able to find a
> counter-example. The preselection of patches (patches_to_consider) in the
> Last
> and FirstReversed cases seems fishy (it was not used before). Tips by
> anyone who
> understands SelectChanges welcome.
>
I have created Issue1698 for the task of improving the SelectChanges code.