[Outreachy] [PATCH 0/1]

[Outreachy] [PATCH 0/1]

[Caleb Tillman]

*** BLURB HERE ***

Caleb Tillman (1):
  Microproject - Use debug-friendly test_path_is_* functions.

 t/t0000-basic.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

-- 
2.25.1

Microproject for Outreachy. Replace test -f with test_path_is_file in t/t0000-basic.sh.

[PATCH 1/1] Microproject - Use debug-friendly test_path_is_* functions.

[Caleb Tillman]

t0000-basic.sh - Replace an instance of test -f with test_path_is_file.

Signed-off-by: Caleb Tillman <caleb.tillman@gmail.com>
---
 t/t0000-basic.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/t/t0000-basic.sh b/t/t0000-basic.sh
index 923281af93..eb99892a87 100755
--- a/t/t0000-basic.sh
+++ b/t/t0000-basic.sh
@@ -1191,7 +1191,7 @@ test_expect_success 'writing this tree with --missing-ok' '
 test_expect_success 'git read-tree followed by write-tree should be idempotent' '
 	rm -f .git/index &&
 	git read-tree $tree &&
-	test -f .git/index &&
+	test_path_is_file .git/index &&
 	newtree=$(git write-tree) &&
 	test "$newtree" = "$tree"
 '
-- 
2.25.1

Re: [Outreachy] [PATCH 0/1]

[Christian Couder]

On Sat, Oct 17, 2020 at 8:06 AM Caleb Tillman <caleb.tillman@gmail.com> wrote:
>
> *** BLURB HERE ***

Cover letters like this are about giving context for a series of
patches. As there is only one patch (1/1) in the series, the context
in the patch itself should suffice, so there is no real need for a
cover letter like this if there is only one patch.

When there is a real need for a cover letter though, the title of the
cover letter should explain what the patch series is about and "***
BLURB HERE ***" in the body of the cover letter should be replaced by
text that explains the context and the reason for the patch series.

Thanks!

Re: [PATCH 1/1] Microproject - Use debug-friendly test_path_is_* functions.

[Christian Couder]

On Sat, Oct 17, 2020 at 8:05 AM Caleb Tillman <caleb.tillman@gmail.com> wrote:
>
> t0000-basic.sh - Replace an instance of test -f with test_path_is_file.

This line seems redundant with the subject.

We suggest that all the messages sent by Outreachy applicants or
interns have "[Outreachy]" at the beginning of their subject. This way
we can identify these emails more easily and prioritize them. Also
using "[...]" ensures that "Outreachy" doesn't appear in the commits
when the patches are applied.

For Outreachy patches related to a test script, we suggest a subject like:

[Outreachy][PATCH X/Y] tZZZZ: do something

where X, Y and ZZZZ are numbers and tZZZZ is the identifier of the test script.

So in this case the subject should be something like:

[Outreachy][PATCH 1/1] t0000: replace an instance of test -f with
test_path_is_file

Note that there is no "Microproject" in the subject and that there is
no uppercase letter used after "PATCH".

The body of the message should explain the reason or the goal of the
patch. Here the reason is that test_path_is_file provides an
(hopefully helpful) error message when it fails, so it should make the
reason for a test failure easier to diagnose.

> Signed-off-by: Caleb Tillman <caleb.tillman@gmail.com>
> ---

Here, after the line with "---" and before the stats below, you can
add comments that will not appear in the commit message when the patch
will be applied. For example here you can say that the patch is your
microproject.

If you really want to make it more prominent, another option is to use
[Outreachy-Microproject] in the subject instead of [Outreachy] but it
makes the subject line longer for information that we are anyway not
likely to miss.

>  t/t0000-basic.sh | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/t/t0000-basic.sh b/t/t0000-basic.sh
> index 923281af93..eb99892a87 100755
> --- a/t/t0000-basic.sh
> +++ b/t/t0000-basic.sh
> @@ -1191,7 +1191,7 @@ test_expect_success 'writing this tree with --missing-ok' '
>  test_expect_success 'git read-tree followed by write-tree should be idempotent' '
>         rm -f .git/index &&
>         git read-tree $tree &&
> -       test -f .git/index &&
> +       test_path_is_file .git/index &&
>         newtree=$(git write-tree) &&
>         test "$newtree" = "$tree"
>  '

This looks good to me.

Thanks!