[RFC PATCH 0/7] merge: learn --autostash

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Alban reported[1] that he expected merge to have an --autostash option,
just like rebase. Since there's not really any reason why merge can't
have it, let's implement it in this patchset.

The actual logic isn't too bad. That change can be found in the last
patch. The remainder is refactoring so that the change can be made
easily. One thing I would like some attention on is if apply_autostash()
is called in the right place. It's called somewhere above
close_object_store() and also in the --abort case, just like in rebase,
but I'm not quite sure if that's right.

Since this is an RFC, we're missing tests and documentation but I've
tested it and it seems to work reasonably well.

[1]: https://github.com/gitgitgadget/git/issues/394

Denton Liu (7):
  Makefile: alphabetically sort += lists
  autostash: extract read_one() from rebase
  autostash: extract apply_autostash() from rebase
  autostash: extract reset_head() from rebase
  autostash: extract perform_autostash() from rebase
  autostash.c: undefine USE_THE_INDEX_COMPATIBILITY_MACROS
  merge: teach --autostash option

 Makefile         |  75 +++++++--------
 autostash.c      | 239 ++++++++++++++++++++++++++++++++++++++++++++++
 autostash.h      |  24 +++++
 builtin/merge.c  |  13 +++
 builtin/pull.c   |   9 +-
 builtin/rebase.c | 240 +----------------------------------------------
 t/t5520-pull.sh  |   8 --
 7 files changed, 323 insertions(+), 285 deletions(-)
 create mode 100644 autostash.c
 create mode 100644 autostash.h

-- 
2.23.0.897.g0a19638b1e




[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]

  Powered by Linux