Re: [PATCH v2 6/7] checkout: split into switch-branch and checkout-files

[Stefan Xenos <sxenos@google.com>]

More thoughts:

git switch-branch should never detach HEAD unless asked to do so
explicitly. That also means that "git switch-branch" shouldn't accept
any of the non-branch tree-ish arguments that would have caused "git
checkout" to do so.
On Wed, Nov 28, 2018 at 3:26 PM Stefan Xenos <sxenos@google.com> wrote:
>
> Although I have no problem with "switch-branch" as a command name,
> some alternative names we might consider for switch-branch might be:
>
> chbranch
> swbranch
> switch
> branch change (as a subcommand for the "branch" command)
>
> I've personally been using "chbranch" as an alias for this
> functionality for some time.
>
>   - Stefan
> On Wed, Nov 28, 2018 at 3:22 PM Stefan Xenos <sxenos@google.com> wrote:
> >
> > > Since the other one is already "checkout-files", maybe this one could just be "checkout-branch".
> >
> > I rather like switch-branch and dislike the word "checkout" since it
> > has been overloaded in git for so long (does it mean moving HEAD or
> > copying files to my working tree?)
> >
> > > nobody will become "sick of" the single "checkout" command that can
> >
> > I have to admit I'm already sick of the checkout command. :-p I can
> > see myself using these two new commands 100% of the time and never
> > missing the old one.
> >
> > Some behaviors I'd expect to see from these commands (I haven't yet
> > checked to see if you've already done this):
> >
> > git checkout-files <tree-ish>
> > should reset all the files in the repository regardless of the current
> > directory - it should produce the same effect as "git reset --hard
> > <tree-ish> && git reset HEAD@{1}". It should also delete
> > locally-created files that aren't present in <tree-ish>, such that the
> > final working tree is exactly identical to what was committed in that
> > tree-ish.
> >
> > git checkout-files foo -- myfile.txt
> > should delete myfile.txt if it is present locally but not present in foo.
> >
> > git checkout-files foo -- .
> > should recursively checkout all files in the current folder and all
> > subfolders, and delete any locally-created files if they're not
> > present in foo.
> >
> > git checkout-files should never move HEAD in any circumstance.
> >
> > Suggestion:
> > If git checkout-files overwrites or deletes any locally-modified files
> > from the workspace or index, those files could be auto-stashed. That
> > would make it easy to restore them in the event of a mistyped command.
> > Auto-stashing could be suppressed with a command-line argument (with
> > alternate behaviors being fail-if-modified or always-overwrite).
> >
> > Idea:
> > If git checkout-files modifies the submodules file, it could also
> > auto-update the submodules. (For example, with something like "git
> > submodule update --init --recursive --progress").
> >
> >   - Stefan
> > On Wed, Nov 28, 2018 at 7:31 AM Duy Nguyen <pclouds@gmail.com> wrote:
> > >
> > > On Wed, Nov 28, 2018 at 7:03 AM Junio C Hamano <gitster@pobox.com> wrote:
> > > >
> > > > Nguyễn Thái Ngọc Duy  <pclouds@gmail.com> writes:
> > > >
> > > > > The good old "git checkout" command is still here and will be until
> > > > > all (or most of users) are sick of it.
> > > >
> > > > Two comments on the goal (the implementation looked reasonable
> > > > assuming the reader agrees with the gaol).
> > > >
> > > > At least to me, the verb "switch" needs two things to switch
> > > > between, i.e. "switch A and B", unless it is "switch to X".
> > > > Either "switch-to-branch" or simply "switch-to", perhaps?
> > > >
> > > > As I already hinted in my response to Stefan (?) about
> > > > checkout-from-tree vs checkout-from-index, a command with multiple
> > > > modes of operation is not confusing to people with the right mental
> > > > model, and I suspect that having two separate commands for "checking
> > > > out a branch" and "checking out paths" that is done by this step
> > > > would help users to form the right mental model.
> > >
> > > Since the other one is already "checkout-files", maybe this one could
> > > just be "checkout-branch".
> > >
> > > > So I tend to think
> > > > these two are "training wheels", and suspect that once they got it,
> > > > nobody will become "sick of" the single "checkout" command that can
> > > > be used to do either.  It's just the matter of being aware what can
> > > > be done (which requires the right mental model) and how to tell Git
> > > > what the user wants it do (two separate commands, operating mode
> > > > option, or just the implied command line syntax---once the user
> > > > knows what s/he is doing, these do not make that much a difference).
> > >
> > > I would hope this becomes better defaults and being used 90% of time.
> > > Even though I know "git checkout" quite well, it still bites me from
> > > time to time. Having the right mental model is one thing. Having to
> > > think a bit every time to write "git checkout" with the right syntax,
> > > and whether you need "--" (that ambiguation problem can still bite you
> > > from time to time), is frankly something I'd rather avoid.
> > > --
> > > Duy