From: Stefan Sperling <stsp@stsp.name>
Subject: Re: Always report stat() error with path
To: Christian Weisgerber <naddy@mips.inka.de>
Cc: gameoftrees@openbsd.org
Date: Fri, 14 Jul 2023 17:12:28 +0200

On Fri, Jul 14, 2023 at 05:04:39PM +0200, Christian Weisgerber wrote:
> Is there a reason to report a stat() error without path here?
> This code has been unchanged since "initial 'got import' implementation".

I don't see a reason not to print this path. ok by me.

> diff /home/naddy/got
> commit - 5fb50fce7978aa218c87ec7ac6c3f4e7bf20e5b3
> path + /home/naddy/got
> blob - 01fe4299c82e150fcd8020915536f48fc73bacbc
> file + got/got.c
> --- got/got.c
> +++ got/got.c
> @@ -463,7 +463,7 @@ edit_logmsg(char **logmsg, const char *editor, const c
>  	}
>  
>  	if (stat(logmsg_path, &st2) == -1)
> -		return got_error_from_errno("stat");
> +		return got_error_from_errno2("stat", logmsg_path);
>  
>  	if (require_modification && st.st_size == st2.st_size &&
>  	    timespeccmp(&st.st_mtim, &st2.st_mtim, ==))
> -- 
> Christian "naddy" Weisgerber                          naddy@mips.inka.de
> 
>