Re: [PATCH] clone: add `--shallow-submodules` flag

From: Stefan Beller <sbeller@google.com>
Date: Apr 25, 2016, 21:25 UTC
Message-ID: <CAGZ79kbPcefodLd4Jt4tvJJmHZXJGe-AMEgzREWZiUvNeciFSQ@mail.gmail.com>
In-Reply-To: <xmqqd1pdjt7m.fsf@gitster.mtv.corp.google.com>

On Mon, Apr 25, 2016 at 2:04 PM, Junio C Hamano <gitster@pobox.com> wrote:

Show 23 quoted lines

> Stefan Beller <sbeller@google.com> writes:
>
>>   This replaces origin/sb/clone-shallow-passthru.
>> @@ -190,7 +190,11 @@ objects from the source repository into a pack in the cloned repository.
>>
>>  --depth <depth>::
>>       Create a 'shallow' clone with a history truncated to the
>> -     specified number of revisions.
>> +     specified number of revisions. Implies `--single-branch` unless
>> +     `--no-single-branch` is given to fetch the histories near the
>> +     tips of all branches. This implies `--shallow-submodules`. If
>> +     you want to have a shallow superproject clone, but full submodules,
>> +     also pass `--no-shallow-submodules`.
>
> This is not wrong per-se but the early half of the new text seems to
> come from 28a1b569 (docs: clarify that passing --depth to git-clone
> implies --single-branch, 2016-01-06), which was merged to 'master'
> some time ago.
>
> I've resolved the conflicts coming from the duplicates, so no need
> to resend, but the resulting history would become misleading.  I am
> undecided if I should rebasing this on top of 85705cfb (Merge branch
> 'ss/clone-depth-single-doc', 2016-01-20) or later.

I could reroll with another variable name on top of 85705cfb?

Show 18 quoted lines

>
>> diff --git a/t/t5614-clone-submodules.sh b/t/t5614-clone-submodules.sh
>> new file mode 100755
>> index 0000000..825f511
>> --- /dev/null
>> +++ b/t/t5614-clone-submodules.sh
>> @@ -0,0 +1,85 @@
>> +#!/bin/sh
>> +
>> +test_description='Test shallow cloning of repos with submodules'
>> +
>> +. ./test-lib.sh
>> +
>> +p=$(pwd)
>
> A single-letter lower-case global that needs to stay constant for
> the entire file looks like a ticking time bomb screaming to be
> broken by future updates.

I agree. How about `currentdir`, `testdir` or `testtop` instead? That is slightly longer than `D`, `here` or `top`, but is slightly more informative. $TRASH would also work for me.

Talking about time bombs: I see there are many tests in single files (e.g. 7412 has ~80 tests IIRC and adding a test in there I often spend more time figuring out the current state of the testing directory rather than writing the test.

So I'd rather see more testing files with fewer tests rather than a few files with all the tests for one topic. (e.g. these tests could have gone into the clone tests, or the basic submodule tests, but instead I chose a new file to test this.)

> I see $D used for this purpose in many
> scripts, $here and $top in some, and $TRASH in yet some others.
> Perhaps $D may be more appropriate if we wanted to keep this
> ultra-short-and-cryptic while mimicking existing ones.

I was not keen on being ultrashort, rather I was toying around to drop the patch introducing the --no-local option.

Show 78 quoted lines

>
>> +test_expect_success 'setup' '
>> +     git checkout -b master &&
>> +     test_commit commit1 &&
>> +     test_commit commit2 &&
>> +     mkdir sub &&
>> +     (
>> +             cd sub &&
>> +             git init &&
>> +             test_commit subcommit1 &&
>> +             test_commit subcommit2 &&
>> +             test_commit subcommit3
>> +     ) &&
>> +     git submodule add "file://$p/sub" sub &&
>> +     git commit -m "add submodule"
>> +'
>> +
>> +test_expect_success 'nonshallow clone implies nonshallow submodule' '
>> +     test_when_finished "rm -rf super_clone" &&
>> +     git clone --recurse-submodules "file://$p/." super_clone &&
>> +     (
>> +             cd super_clone &&
>> +             git log --oneline >lines &&
>> +             test_line_count = 3 lines
>> +     ) &&
>> +     (
>> +             cd super_clone/sub &&
>> +             git log --oneline >lines &&
>> +             test_line_count = 3 lines
>> +     )
>> +'
>> +
>> +test_expect_success 'shallow clone implies shallow submodule' '
>> +     test_when_finished "rm -rf super_clone" &&
>> +     git clone --recurse-submodules --depth 2 "file://$p/." super_clone &&
>> +     (
>> +             cd super_clone &&
>> +             git log --oneline >lines &&
>> +             test_line_count = 2 lines
>> +     ) &&
>> +     (
>> +             cd super_clone/sub &&
>> +             git log --oneline >lines &&
>> +             test_line_count = 1 lines
>> +     )
>> +'
>> +
>> +test_expect_success 'shallow clone with non shallow submodule' '
>> +     test_when_finished "rm -rf super_clone" &&
>> +     git clone --recurse-submodules --depth 2 --no-shallow-submodules "file://$p/." super_clone &&
>> +     (
>> +             cd super_clone &&
>> +             git log --oneline >lines &&
>> +             test_line_count = 2 lines
>> +     ) &&
>> +     (
>> +             cd super_clone/sub &&
>> +             git log --oneline >lines &&
>> +             test_line_count = 3 lines
>> +     )
>> +'
>> +
>> +test_expect_success 'non shallow clone with shallow submodule' '
>> +     test_when_finished "rm -rf super_clone" &&
>> +     git clone --recurse-submodules --no-local --shallow-submodules "file://$p/." super_clone &&
>> +     (
>> +             cd super_clone &&
>> +             git log --oneline >lines &&
>> +             test_line_count = 3 lines
>> +     ) &&
>> +     (
>> +             cd super_clone/sub &&
>> +             git log --oneline >lines &&
>> +             test_line_count = 1 lines
>> +     )
>> +'
>> +
>> +test_done