Re: [RFC/PATCH v2 3/3] status: introduce status.displayCommentChar to disable display of #

[Junio C Hamano <gitster@pobox.com>]

Matthieu Moy <Matthieu.Moy@imag.fr> writes:

> diff --git a/Documentation/config.txt b/Documentation/config.txt
> index ec57a15..dacf4b9 100644
> --- a/Documentation/config.txt
> +++ b/Documentation/config.txt
> @@ -2118,6 +2118,11 @@ status.branch::
>  	Set to true to enable --branch by default in linkgit:git-status[1].
>  	The option --no-branch takes precedence over this variable.
>  
> +status.displayCommentChar::
> +	If set to false, linkgit:git-status[1] will not prefix each
> +	line with the comment char (`core.commentchar`, i.e. `#` by
> +	default). Defaults to true.

We prefix core.commentchar followed by a single space for non-empty
lines; is that worth mentioning, I wonder?

Also I envision that we would extend core.commentchar to be more
than a single character.  Is "displayCommentChar" still a good name
for this setting?

"status.omitCommentPrefix" that defaults to false might be a better
setting, I suspect.  I further suspect that in the longer term, we
may want to consider flipping its default to true (for Git 2.0, it
is too late).  I do not think we need these comment prefix in the
human readable "status" output---after all, there is no line that is
not commented (other than "nothing added to commit" at the end) in
the current output, so the comment prefix only wastes the screen
real estate.

What are our plans to help existing scripts people have written over
time, especially before "status -s" was invented, that will be
broken by use of this?

> @@ -663,17 +666,18 @@ static void wt_status_print_submodule_summary(struct wt_status *s, int uncommitt
>  	char summary_limit[64];
>  	char index[PATH_MAX];
>  	const char *env[] = { NULL, NULL };
> -	const char *argv[8];
> +	const char *argv[9];
>  
>  	env[0] =	index;
>  	argv[0] =	"submodule";
>  	argv[1] =	"summary";
>  	argv[2] =	uncommitted ? "--files" : "--cached";
>  	argv[3] =	"--for-status";
> -	argv[4] =	"--summary-limit";
> -	argv[5] =	summary_limit;
> -	argv[6] =	uncommitted ? NULL : (s->amend ? "HEAD^" : "HEAD");
> -	argv[7] =	NULL;
> +	argv[4] =	s->display_comment_char ? "--display-comment-char" : "--no-display-comment-char";
> +	argv[5] =	"--summary-limit";
> +	argv[6] =	summary_limit;
> +	argv[7] =	uncommitted ? NULL : (s->amend ? "HEAD^" : "HEAD");
> +	argv[8] =	NULL;

Not strictly your fault, but we should lose these funny indent after
"=" and use argv_array API here.