threads / patch / 60371

patch, 2 partsdoc/git-repack: fix syntax for `-g` shorthand option

Subject: [PATCH 1/2] doc/git-repack: fix syntax for `-g` shorthand option

## tl;dr

5 messages between Oct 16, 2023 and Oct 30, 2023. Diffs are folded; open one to read it.

replies: 4people: 2as markdown or json

Patrick Steinhardt· Oct 16, 2023, 07:19 UTC · lore

[PATCH 0/2] doc/git-repack: small fixes for geometric repacks

Hi,

this small series addresses two issues in our documentation for the git-repack(1) command that relate to geometric repacks.

Patrick
Patrick Steinhardt (2):
  doc/git-repack: fix syntax for `-g` shorthand option
  doc/git-repack: don't mention nonexistent "--unpacked" option
 Documentation/git-repack.txt | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)
-- 
2.42.0
Patrick Steinhardt· Oct 16, 2023, 07:19 UTC · re: Patrick Steinhardt · lore

The `-g` switch is a shorthand for `--geometric=` and allows the user to specify the geometric. The documentation is wrong though and indicates that the syntax for the shorthand is `-g=<factor>`. In fact though, the option must be specified without the equals sign via `-g<factor>`.

Fix the syntax accordingly.
Signed-off-by: Patrick Steinhardt <ps@pks.im>
---
 Documentation/git-repack.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Show changes to Documentation/git-repack.txt +1 −1
diff --git a/Documentation/git-repack.txt b/Documentation/git-repack.txt
index 8545a32667..dfd2a59c50 100644
--- a/Documentation/git-repack.txt
+++ b/Documentation/git-repack.txt
@@ -209,7 +209,7 @@ depth is 4095.
 	Pass the `--delta-islands` option to `git-pack-objects`, see
 	linkgit:git-pack-objects[1].
 
--g=<factor>::
+-g<factor>::
 --geometric=<factor>::
 	Arrange resulting pack structure so that each successive pack
 	contains at least `<factor>` times the number of objects as the
-- 
2.42.0
Patrick Steinhardt· Oct 16, 2023, 07:19 UTC · re: Patrick Steinhardt · lore

[PATCH 2/2] doc/git-repack: don't mention nonexistent "--unpacked" option

The documentation for geometric repacking mentions a "--unpacked" option that supposedly changes how loose objects are rolled up. This option has never existed, and the implied behaviour, namely to include all unpacked objects into the resulting packfile, is in fact the default behaviour.

Correct the documentation to not mention this option.
Signed-off-by: Patrick Steinhardt <ps@pks.im>
---
 Documentation/git-repack.txt | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)
Show changes to Documentation/git-repack.txt +2 −5
diff --git a/Documentation/git-repack.txt b/Documentation/git-repack.txt
index dfd2a59c50..d61078b697 100644
--- a/Documentation/git-repack.txt
+++ b/Documentation/git-repack.txt
@@ -226,11 +226,8 @@ uniquely by the set of packs being "rolled-up"; in other words, the
 packs determined to need to be combined in order to restore a geometric
 progression.
 +
-When `--unpacked` is specified, loose objects are implicitly included in
-this "roll-up", without respect to their reachability. This is subject
-to change in the future. This option (implying a drastically different
-repack mode) is not guaranteed to work with all other combinations of
-option to `git repack`.
+Loose objects are implicitly included in this "roll-up", without respect to
+their reachability. This is subject to change in the future.
 +
 When writing a multi-pack bitmap, `git repack` selects the largest resulting
 pack as the preferred pack for object selection by the MIDX (see
-- 
2.42.0
Taylor Blau· Oct 30, 2023, 21:51 UTC · re: Patrick Steinhardt · lore

Re: [PATCH 2/2] doc/git-repack: don't mention nonexistent "--unpacked" option

On Mon, Oct 16, 2023 at 09:19:56AM +0200, Patrick Steinhardt wrote:
Show 28 quoted lines
> The documentation for geometric repacking mentions a "--unpacked" option
> that supposedly changes how loose objects are rolled up. This option has
> never existed, and the implied behaviour, namely to include all unpacked
> objects into the resulting packfile, is in fact the default behaviour.
>
> Correct the documentation to not mention this option.
>
> Signed-off-by: Patrick Steinhardt <ps@pks.im>
> ---
>  Documentation/git-repack.txt | 7 ++-----
>  1 file changed, 2 insertions(+), 5 deletions(-)
>
> diff --git a/Documentation/git-repack.txt b/Documentation/git-repack.txt
> index dfd2a59c50..d61078b697 100644
> --- a/Documentation/git-repack.txt
> +++ b/Documentation/git-repack.txt
> @@ -226,11 +226,8 @@ uniquely by the set of packs being "rolled-up"; in other words, the
>  packs determined to need to be combined in order to restore a geometric
>  progression.
>  +
> -When `--unpacked` is specified, loose objects are implicitly included in
> -this "roll-up", without respect to their reachability. This is subject
> -to change in the future. This option (implying a drastically different
> -repack mode) is not guaranteed to work with all other combinations of
> -option to `git repack`.
> +Loose objects are implicitly included in this "roll-up", without respect to
> +their reachability. This is subject to change in the future.
>  +

Oops. This refers to the "--unpacked" option that pack-objects takes, not repack. I agree that mentioning "--unpacked" is too low-level a detail for this user-facing documentation, so even something like:

    When `repack` passes `--unpacked` down to `pack-objects` (which is
    the default) ...
would be too much detail for this man page.
I am very happy with the patch here as an alternative.

Thanks, Taylor

Taylor Blau· Oct 30, 2023, 21:52 UTC · re: Patrick Steinhardt · lore

Re: [PATCH 0/2] doc/git-repack: small fixes for geometric repacks

On Mon, Oct 16, 2023 at 09:19:46AM +0200, Patrick Steinhardt wrote:
Show 6 quoted lines
> Patrick Steinhardt (2):
>   doc/git-repack: fix syntax for `-g` shorthand option
>   doc/git-repack: don't mention nonexistent "--unpacked" option
>
>  Documentation/git-repack.txt | 9 +++------
>  1 file changed, 3 insertions(+), 6 deletions(-)

Sorry that this one fell off of my queue. These are both looking good to me, thanks for finding and fixing them!

Thanks, Taylor

← back to recent threads