Re: [PATCH v5 1/3] sumodule--helper: fix submodule--helper clone usage and check argc count
- From
Jacob Keller <jacob.keller@gmail.com>
- Date
- Feb 29, 2016, 19:34 UTC
- Message-ID
- <CA+P7+xryNo=n0fFn9HNrj78z4XtWwoF=j67+Qy8TbO7reBTO-Q@mail.gmail.com>
- In-Reply-To
- <xmqqr3fvcsoy.fsf@gitster.mtv.corp.google.com>
On Mon, Feb 29, 2016 at 9:49 AM, Junio C Hamano <gitster@pobox.com> wrote:
Show 6 quoted lines
> Jacob Keller <jacob.e.keller@intel.com> writes: >> Fix the usage description to match implementation. Add an argc check to >> enforce no extra arguments. > > The above sounds very sensible. >
Right.
Show 8 quoted lines
>> Fix a bug in the argument passing in >> git-submodule.sh which would pass --reference and --depth as empty >> strings when they were unused, resulting in extra argc after parsing >> options. > > This does make sense but it is an unrelated fix. Perhaps split this > patch into two? >
That actually is required because otherwise adding a check for argc would break the things. I could split them and do this first and then check for argc if you really prefer?
Show 8 quoted lines
>> + if (argc) >> + usage(*git_submodule_helper_usage); >> + > > That asterisk looks very unusual and wanting to be future-proofed > (i.e. who says that only the first entry matters?). Should't this > be calling usage_with_options()? >
I... didn't know usage_with_options was a thing! Hah. I can fix these up.
Thanks, Jake\