[v2,0/7] remote rename/remove: improve handling of configuration values
mbox series

Message ID cover.1579857394.git.bert.wesarg@googlemail.com
Headers show
Series
  • remote rename/remove: improve handling of configuration values
Related show

Message

Bert Wesarg Jan. 24, 2020, 9:25 a.m. UTC
While fixing that 'git remote rename X Y' does not rename the values for
'branch.*.pushRemote', it opened the possibility to more improvements in
this area:

 - 'remote rename' did not accept single-letter abbreviations for
   'branch.*.rebase' like 'pull --rebase' does

 - minor clean-ups the config callback

 - patch 5 will be replaced by/rebased on Matthew's work in 'config: allow user to
   know scope of config options', once 'config_scope_name' is available

 - gently handling the rename of 'remote.pushDefault'

Changes since v1:
 * avoid mixed declarations and statements
 * 'git remote remove' learned similar treatment

Bert Wesarg (7):
  pull --rebase/remote rename: document and honor single-letter
    abbreviations rebase types
  remote: clean-up by returning early to avoid one indentation
  remote: clean-up config callback
  remote rename/remove: handle branch.<name>.pushRemote config values
  [RFC] config: make `scope_name` global as `config_scope_name`
  config: provide access to the current line number
  remote rename/remove: gently handle remote.pushDefault config

 Documentation/config/branch.txt |   7 +-
 Documentation/config/pull.txt   |   7 +-
 Makefile                        |   1 +
 builtin/pull.c                  |  29 +----
 builtin/remote.c                | 188 ++++++++++++++++++++++----------
 config.c                        |  24 ++++
 config.h                        |   2 +
 rebase.c                        |  35 ++++++
 rebase.h                        |  15 +++
 t/helper/test-config.c          |  18 +--
 t/t1308-config-set.sh           |  14 ++-
 t/t5505-remote.sh               |  88 ++++++++++++++-
 12 files changed, 322 insertions(+), 106 deletions(-)
 create mode 100644 rebase.c
 create mode 100644 rebase.h

Comments

Junio C Hamano Jan. 24, 2020, 9 p.m. UTC | #1
All steps looked quite sensibly done.

>  - patch 5 will be replaced by/rebased on Matthew's work in 'config: allow user to
>    know scope of config options', once 'config_scope_name' is available

I expect that Matthew's topic would become solid enough after one
more reroll to name the function back to config_scope_name(); after
that, let's drop the step and instead fork this topic off of Matthew's
topic to queue the remaining patches on top.

Thanks.
Matt Rogers Jan. 25, 2020, 12:39 a.m. UTC | #2
Yeah, I just resubmitted so you should be good to go

On Fri, Jan 24, 2020 at 4:10 PM Junio C Hamano <gitster@pobox.com> wrote:
>
> All steps looked quite sensibly done.
>
> >  - patch 5 will be replaced by/rebased on Matthew's work in 'config: allow user to
> >    know scope of config options', once 'config_scope_name' is available
>
> I expect that Matthew's topic would become solid enough after one
> more reroll to name the function back to config_scope_name(); after
> that, let's drop the step and instead fork this topic off of Matthew's
> topic to queue the remaining patches on top.
>
> Thanks.
Bert Wesarg Jan. 27, 2020, 6:50 a.m. UTC | #3
Junio,

On Sat, Jan 25, 2020 at 1:39 AM Matt Rogers <mattr94@gmail.com> wrote:
>
> Yeah, I just resubmitted so you should be good to go

I can resend the rebased topic at any time. And as I don't see that
you picked up by latest re-roll (cover in
cover.1579857394.git.bert.wesarg@googlemail.com) in into
bw/remote-rename-update-config yet, I think it makes sense to do this
now.

Best,
Bert

>
> On Fri, Jan 24, 2020 at 4:10 PM Junio C Hamano <gitster@pobox.com> wrote:
> >
> > All steps looked quite sensibly done.
> >
> > >  - patch 5 will be replaced by/rebased on Matthew's work in 'config: allow user to
> > >    know scope of config options', once 'config_scope_name' is available
> >
> > I expect that Matthew's topic would become solid enough after one
> > more reroll to name the function back to config_scope_name(); after
> > that, let's drop the step and instead fork this topic off of Matthew's
> > topic to queue the remaining patches on top.
> >
> > Thanks.
>
>
>
> --
> Matthew Rogers