git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: shejialuo <shejialuo@gmail.com>
To: Patrick Steinhardt <ps@pks.im>
Cc: git@vger.kernel.org
Subject: Re: [PATCH 01/16] path: refactor `repo_common_path()` family of functions
Date: Thu, 6 Feb 2025 22:21:24 +0800	[thread overview]
Message-ID: <Z6TFZGNjs3Xfkh9x@ArchLinux> (raw)
In-Reply-To: <20250206-b4-pks-path-drop-the-repository-v1-1-4e77f0313206@pks.im>

On Thu, Feb 06, 2025 at 08:57:57AM +0100, Patrick Steinhardt wrote:
> The functions provided by the "path" subsystem to derive repository
> paths for the commondir, gitdir, worktrees and submodules are quite
> inconsistent. Some functions have a `strbuf_` prefix, others have
> different return values, some don't provide a variant working on top of
> `strbuf`s.
> 
> We're thus about to refactor all of these family of functions so that
> they follow a common pattern:
> 
>   - `repo_*_path()` returns an allocated string.
> 
>   - `repo_*_path_append()` appends the path to the caller-provided
>     buffer while returning a constant pointer to the buffer. This
>     clarifies whether the buffer is being appended to or rewritten,
>     which otherwise wasn't immediately obvious.
> 
>   - `repo_*_path_replace()` replaces contents of the buffer with the
>     computed path, again returning a pointer to the buffer contents.
> 

I want to ask a design question about this. Why do we need to return the
raw pointer to the `struct strbuf` for the last two cases? I somehow
understand why you want to do this. You want to follow a common pattern
for those three functions. But I wonder should we let the caller to
decide whether they want to use the raw pointer?

And in this patch, the return value of the last two cases has never been
used. Until I read the next patch, I have seen the usage of the return
value thus I could understand your motivation.

> diff --git a/path.h b/path.h
> index 5f6c85e5f8..3c75495e1a 100644
> --- a/path.h
> +++ b/path.h
> @@ -25,22 +25,20 @@ char *mkpathdup(const char *fmt, ...)
>  	__attribute__((format (printf, 1, 2)));
>  
>  /*
> - * The `strbuf_git_common_path` family of functions will construct a path into a
> + * The `repo_common_path` family of functions will construct a path into a
>   * repository's common git directory, which is shared by all worktrees.
>   */
> -
> -/*
> - * Constructs a path into the common git directory of repository `repo` and
> - * append it in the provided buffer `sb`.
> - */
> -void strbuf_git_common_path(struct strbuf *sb,
> -			    const struct repository *repo,
> -			    const char *fmt, ...)
> +char *repo_common_path(const struct repository *repo,
> +		       const char *fmt, ...)
> +	__attribute__((format (printf, 2, 3)));
> +const char *repo_common_path_append(const struct repository *repo,
> +				    struct strbuf *sb,
> +				    const char *fmt, ...)
> +	__attribute__((format (printf, 3, 4)));
> +const char *repo_common_path_replace(const struct repository *repo,
> +				     struct strbuf *sb,
> +				     const char *fmt, ...)
>  	__attribute__((format (printf, 3, 4)));
> -void repo_common_pathv(const struct repository *repo,
> -		       struct strbuf *buf,
> -		       const char *fmt,
> -		       va_list args);
>  
>  /*
>   * The `repo_git_path` family of functions will construct a path into a repository's
> @@ -243,6 +241,12 @@ struct strbuf *get_pathname(void);
>  #  include "strbuf.h"
>  #  include "repository.h"
>  
> +/* Internal implementation detail that should not be used. */
> +void repo_common_pathv(const struct repository *repo,
> +		       struct strbuf *buf,
> +		       const char *fmt,
> +		       va_list args);
> +

If we decide to make this as internal implementation, why we don't just
delete this declaration in the header file? Do I miss out something
here?

Thanks,
Jialuo

  parent reply	other threads:[~2025-02-06 14:19 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-06  7:57 [PATCH 00/16] path: remove dependency on `the_repository` Patrick Steinhardt
2025-02-06  7:57 ` [PATCH 01/16] path: refactor `repo_common_path()` family of functions Patrick Steinhardt
2025-02-06 11:17   ` Karthik Nayak
2025-02-07  6:16     ` Patrick Steinhardt
2025-02-06 14:21   ` shejialuo [this message]
2025-02-07  6:16     ` Patrick Steinhardt
2025-02-06  7:57 ` [PATCH 02/16] path: refactor `repo_git_path()` " Patrick Steinhardt
2025-02-06 11:53   ` Karthik Nayak
2025-02-07  6:15     ` Patrick Steinhardt
2025-02-06  7:57 ` [PATCH 03/16] path: refactor `repo_worktree_path()` " Patrick Steinhardt
2025-02-06  7:58 ` [PATCH 04/16] submodule: refactor `submodule_to_gitdir()` to accept a repo Patrick Steinhardt
2025-02-06  7:58 ` [PATCH 05/16] path: refactor `repo_submodule_path()` family of functions Patrick Steinhardt
2025-02-06 12:05   ` Karthik Nayak
2025-02-07  6:16     ` Patrick Steinhardt
2025-02-07  7:04       ` Karthik Nayak
2025-02-06 15:03   ` shejialuo
2025-02-06  7:58 ` [PATCH 06/16] path: drop unused `strbuf_git_path()` function Patrick Steinhardt
2025-02-06  7:58 ` [PATCH 07/16] path: drop `git_pathdup()` in favor of `repo_git_path()` Patrick Steinhardt
2025-02-06  7:58 ` [PATCH 08/16] path: drop `git_path_buf()` in favor of `repo_git_path_replace()` Patrick Steinhardt
2025-02-06  7:58 ` [PATCH 09/16] worktree: return allocated string from `get_worktree_git_dir()` Patrick Steinhardt
2025-02-07  7:15   ` Karthik Nayak
2025-02-07 10:49     ` Patrick Steinhardt
2025-02-06  7:58 ` [PATCH 10/16] path: drop `git_common_path()` in favor of `repo_common_path()` Patrick Steinhardt
2025-02-06 15:54   ` shejialuo
2025-02-07  6:16     ` Patrick Steinhardt
2025-02-06  7:58 ` [PATCH 11/16] rerere: let `rerere_path()` write paths into a caller-provided buffer Patrick Steinhardt
2025-02-06  7:58 ` [PATCH 12/16] path: drop `git_path()` in favor of `repo_git_path()` Patrick Steinhardt
2025-02-06 16:01   ` shejialuo
2025-02-07  6:16     ` Patrick Steinhardt
2025-02-06  7:58 ` [PATCH 13/16] repo-settings: introduce function to clear struct Patrick Steinhardt
2025-02-06  7:58 ` [PATCH 14/16] environment: move access to "core.hooksPath" into repo settings Patrick Steinhardt
2025-02-06  7:58 ` [PATCH 15/16] environment: move access to "core.sharedRepository" " Patrick Steinhardt
2025-02-06  7:58 ` [PATCH 16/16] path: adjust last remaining users of `the_repository` Patrick Steinhardt
2025-02-06 16:14 ` [PATCH 00/16] path: remove dependency on `the_repository` shejialuo
2025-02-07  6:16   ` Patrick Steinhardt
2025-02-07  8:17 ` Karthik Nayak
2025-02-07 11:03 ` [PATCH v2 " Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 01/16] path: refactor `repo_common_path()` family of functions Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 02/16] path: refactor `repo_git_path()` " Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 03/16] path: refactor `repo_worktree_path()` " Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 04/16] submodule: refactor `submodule_to_gitdir()` to accept a repo Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 05/16] path: refactor `repo_submodule_path()` family of functions Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 06/16] path: drop unused `strbuf_git_path()` function Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 07/16] path: drop `git_pathdup()` in favor of `repo_git_path()` Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 08/16] path: drop `git_path_buf()` in favor of `repo_git_path_replace()` Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 09/16] worktree: return allocated string from `get_worktree_git_dir()` Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 10/16] path: drop `git_common_path()` in favor of `repo_common_path()` Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 11/16] rerere: let `rerere_path()` write paths into a caller-provided buffer Patrick Steinhardt
2025-02-22  7:20     ` Jeff King
2025-02-24 16:14       ` Junio C Hamano
2025-02-24 22:19         ` Jeff King
2025-02-24 22:50           ` Junio C Hamano
2025-02-24 23:10             ` Jeff King
2025-02-24 23:14               ` Junio C Hamano
2025-02-25  6:24                 ` Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 12/16] path: drop `git_path()` in favor of `repo_git_path()` Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 13/16] repo-settings: introduce function to clear struct Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 14/16] environment: move access to "core.hooksPath" into repo settings Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 15/16] environment: move access to "core.sharedRepository" " Patrick Steinhardt
2025-02-07 11:03   ` [PATCH v2 16/16] path: adjust last remaining users of `the_repository` Patrick Steinhardt
2025-02-07 11:44   ` [PATCH v2 00/16] path: remove dependency on `the_repository` Karthik Nayak
2025-02-08 15:31   ` shejialuo
2025-02-10 18:32     ` Junio C Hamano
2025-02-11 10:03       ` shejialuo

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=Z6TFZGNjs3Xfkh9x@ArchLinux \
    --to=shejialuo@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=ps@pks.im \
    /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).