git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Ævar Arnfjörð Bjarmason" <avarab@gmail.com>
To: Jonathan Tan <jonathantanmy@google.com>
Cc: git@vger.kernel.org, gitster@pobox.com, jabolopes@google.com
Subject: Re: [PATCH v2] sparse-checkout: create leading directory
Date: Fri, 21 Jan 2022 19:48:46 +0100	[thread overview]
Message-ID: <220121.8635lgkj17.gmgdl@evledraar.gmail.com> (raw)
In-Reply-To: <20220121174441.3991963-1-jonathantanmy@google.com>


On Fri, Jan 21 2022, Jonathan Tan wrote:

> When creating the sparse-checkout file, Git does not create the leading
> directory, "$GIT_DIR/info", if it does not exist. This causes problems
> if the repository does not have that directory. Therefore, ensure that
> the leading directory is created.
>
> This is the only "open" in builtin/sparse-checkout.c that does not have
> a leading directory check. (The other one in write_patterns_and_update()
> does.)
>
> Note that the test needs to explicitly specify a template when running
> "git init" because the default template used in the tests has the
> "info/" directory included.
>
> Helped-by: Jose Lopes <jabolopes@google.com>
> Signed-off-by: Jonathan Tan <jonathantanmy@google.com>
> ---
> Changes from v1:
>  - made test simpler

This partial application of the fix-up I suggested in
https://lore.kernel.org/git/220120.86mtjqks1b.gmgdl@evledraar.gmail.com/
leaves the now-unused "blank-template"

>  - added attribution to Jose Lopes for finding and making the first
>    draft of this patch (after confirming with them)
>
> Ævar mentioned "git sparse-checkout add" but I think that that is a
> different problem - in the "git sparse-checkout init" case, we could get
> into this case with a template that does not have .git/info, but in the
> "git sparse-checkout add" case, the user would have had to explicitly
> remove the info directory. So I'll limit this patch to the "init" case,
> for now.
> ---
>  builtin/sparse-checkout.c          | 3 +++
>  t/t1091-sparse-checkout-builtin.sh | 6 ++++++
>  2 files changed, 9 insertions(+)
>
> diff --git a/builtin/sparse-checkout.c b/builtin/sparse-checkout.c
> index 679c107036..2b0e1db2d2 100644
> --- a/builtin/sparse-checkout.c
> +++ b/builtin/sparse-checkout.c
> @@ -471,6 +471,9 @@ static int sparse_checkout_init(int argc, const char **argv)
>  		FILE *fp;
>  
>  		/* assume we are in a fresh repo, but update the sparse-checkout file */
> +		if (safe_create_leading_directories(sparse_filename))
> +			die(_("unable to create leading directories of %s"),
> +			    sparse_filename);
>  		fp = xfopen(sparse_filename, "w");
>  		if (!fp)
>  			die(_("failed to open '%s'"), sparse_filename);
> diff --git a/t/t1091-sparse-checkout-builtin.sh b/t/t1091-sparse-checkout-builtin.sh
> index 42776984fe..3189d3da96 100755
> --- a/t/t1091-sparse-checkout-builtin.sh
> +++ b/t/t1091-sparse-checkout-builtin.sh
> @@ -79,6 +79,12 @@ test_expect_success 'git sparse-checkout init' '
>  	check_files repo a
>  '
>  
> +test_expect_success 'git sparse-checkout init in empty repo' '
> +	test_when_finished rm -rf empty-repo blank-template &&
> +	git init --template= empty-repo &&
> +	git -C empty-repo sparse-checkout init
> +'

I agree that it's a slightly different problem, but I was just
advocating for us testing what happened in these cases.

The below fix-up does that. I think we should use warning_errno() there
instead of some specutalite "file may not exist", but with/without this
patch these tests show that only the "init" case was broken.

As a more general issue I don't understand why "add" and "init" need to
be conceptually different operations. If what defines a sparse checkout
is just that it has that file and the 2 default patterns, which unless
I'm missing something is the case. Why isn't "add" merely an "init"
that'll optimistically add whatever pattern you asked for, in addition
to doing an "init" if you didn't already?

Then "add" and "init" will share the same error recovery behavior, and
you won't needlessly have to run "init/add x" just to start using
sparse-checkout with a pattern of "x".

But I've never *actually* used this command, so maybe I'm just missing
something obvious...

diff --git a/t/t1091-sparse-checkout-builtin.sh b/t/t1091-sparse-checkout-builtin.sh
index 3189d3da965..6b56d9d177f 100755
--- a/t/t1091-sparse-checkout-builtin.sh
+++ b/t/t1091-sparse-checkout-builtin.sh
@@ -80,11 +80,37 @@ test_expect_success 'git sparse-checkout init' '
 '
 
 test_expect_success 'git sparse-checkout init in empty repo' '
-	test_when_finished rm -rf empty-repo blank-template &&
+	test_when_finished rm -rf empty-repo &&
 	git init --template= empty-repo &&
 	git -C empty-repo sparse-checkout init
 '
 
+test_expect_success 'git sparse-checkout add -- info/sparse-checkout missing' '
+	test_when_finished "rm -rf empty" &&
+	git init --template= empty &&
+	git -C empty sparse-checkout init &&
+	rm -rf empty/.git/info &&
+
+	cat >expect <<-\EOF &&
+	fatal: unable to load existing sparse-checkout patterns
+	EOF
+	test_expect_code 128 git -C empty sparse-checkout add bar 2>actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git sparse-checkout list -- info/sparse-checkout missing' '
+	test_when_finished "rm -rf empty" &&
+	git init --template= empty &&
+	git -C empty sparse-checkout init &&
+	rm -rf empty/.git/info &&
+
+	cat >expect <<-\EOF &&
+	warning: this worktree is not sparse (sparse-checkout file may not exist)
+	EOF
+	git -C empty sparse-checkout list 2>actual &&
+	test_cmp expect actual
+'
+
 test_expect_success 'git sparse-checkout list after init' '
 	git -C repo sparse-checkout list >actual &&
 	cat >expect <<-\EOF &&

  reply	other threads:[~2022-01-21 19:04 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-20 18:55 [PATCH] sparse-checkout: create leading directory Jonathan Tan
2022-01-20 20:26 ` Junio C Hamano
2022-01-20 21:30 ` Ævar Arnfjörð Bjarmason
2022-01-22  7:58   ` SZEDER Gábor
2022-01-21 17:44 ` [PATCH v2] " Jonathan Tan
2022-01-21 18:48   ` Ævar Arnfjörð Bjarmason [this message]
2022-01-22  5:21     ` Elijah Newren
2022-01-22 12:06       ` Ævar Arnfjörð Bjarmason
2022-01-22 17:29         ` Elijah Newren
2022-01-24 11:17           ` Ævar Arnfjörð Bjarmason
2022-01-24 18:25             ` Jonathan Tan
2022-01-25  5:37             ` Elijah Newren
2022-01-22  2:33   ` Elijah Newren
2022-01-24 18:22     ` Jonathan Tan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=220121.8635lgkj17.gmgdl@evledraar.gmail.com \
    --to=avarab@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=jabolopes@google.com \
    --cc=jonathantanmy@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).