git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jeff King <peff@peff.net>
To: Dave Olszewski <cxreg@pobox.com>
Cc: git@vger.kernel.org, gitster@pobox.com
Subject: Re: [PATCH] push: allow --follow-tags to be set by config push.followTags
Date: Mon, 16 Feb 2015 00:20:49 -0500	[thread overview]
Message-ID: <20150216052049.GA5031@peff.net> (raw)
In-Reply-To: <1424055690-32631-1-git-send-email-cxreg@pobox.com>

On Sun, Feb 15, 2015 at 07:01:30PM -0800, Dave Olszewski wrote:

> +push.followTags::
> +	If set to true enable '--follow-tags' option by default.  You
> +	may override this configuration at time of push by specifying
> +	'--no-follow-tags'.

Thanks, this is something I've considered implementing myself, as I have
one repo that is frequently migrating tags from one remote to another,
and I often forget to specify the option.

> diff --git a/builtin/push.c b/builtin/push.c
> index fc771a9..47f0119 100644
> --- a/builtin/push.c
> +++ b/builtin/push.c
> @@ -525,6 +525,11 @@ int cmd_push(int argc, const char **argv, const char *prefix)
>  
>  	packet_trace_identity("push");
>  	git_config(git_push_config, NULL);
> +
> +	/* set TRANSPORT_PUSH_FOLLOW_TAGS in flags so that --no-follow-tags may unset it */
> +	if (push_follow_tags)
> +		flags |= TRANSPORT_PUSH_FOLLOW_TAGS;

You can see above that we use git_push_config to load our config...

> --- a/config.c
> +++ b/config.c
> @@ -977,6 +977,11 @@ static int git_default_push_config(const char *var, const char *value)
>  		return 0;
>  	}
>  
> +	if (!strcmp(var, "push.followtags")) {
> +		push_follow_tags = git_config_bool(var, value);
> +		return 0;
> +	}

But here you are adding to git_default_push_config, which is in another
file.

I'm trying to figure out why git_default_push_config exists at all. The
major difference from git_push_config is that the "default" variant will
get loaded for _all_ commands, not just "push". So if it affected
variables that were used by other commands, it would be needed. But all
it sets is push_default, which seems to be specific to builtin/push.c.

So I suspect it can be removed entirely, and folded into
git_config_push. But that's outside the scope of your patch.

What _is_ in the scope of your patch is that I think the new option you
are adding could go into git_push_config; it is definitely only about
the push command itself. And then you could declare it as:

  static int push_follow_tags;

without having to worry about making it an extern that is available
everywhere.

> diff --git a/transport.c b/transport.c
> index 0694a7c..ff5f63d 100644
> --- a/transport.c
> +++ b/transport.c
> @@ -1148,7 +1148,7 @@ int transport_push(struct transport *transport,
>  			match_flags |= MATCH_REFS_MIRROR;
>  		if (flags & TRANSPORT_PUSH_PRUNE)
>  			match_flags |= MATCH_REFS_PRUNE;
> -		if (flags & TRANSPORT_PUSH_FOLLOW_TAGS)
> +		if ((flags & TRANSPORT_PUSH_FOLLOW_TAGS))
>  			match_flags |= MATCH_REFS_FOLLOW_TAGS;

This looks like just noise in the diff (I guess leftover from some
debugging you were doing). Is that correct?

-Peff

  reply	other threads:[~2015-02-16  5:20 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-02-16  3:01 [PATCH] push: allow --follow-tags to be set by config push.followTags Dave Olszewski
2015-02-16  5:20 ` Jeff King [this message]
2015-02-16  5:45   ` [PATCH 0/2] clean up push config callbacks Jeff King
2015-02-16  5:46     ` [PATCH 1/2] git_push_config: drop cargo-culted wt_status pointer Jeff King
2015-02-16  5:47     ` [PATCH 2/2] builtin/push.c: make push_default a static variable Jeff King
2015-02-16  5:57       ` Junio C Hamano
2015-02-17 10:46       ` Jeff King
2015-02-17 17:45         ` Junio C Hamano
2015-02-17 18:23           ` Jeff King
2015-02-17 22:16             ` Junio C Hamano
2015-02-18 18:50               ` Jeff King
2015-02-18 19:08                 ` Junio C Hamano
2015-02-18 19:25                   ` Jeff King
2015-02-18 19:50                     ` Junio C Hamano
2015-02-18 20:03                       ` Jeff King
2015-02-16  5:54     ` [PATCH 3/2] push: allow --follow-tags to be set by config push.followTags Jeff King
2015-02-16  6:02       ` Junio C Hamano
2015-02-16  6:10         ` [PATCH 0/3] cleaner bit-setting in cmd_push Jeff King
2015-02-16  6:12           ` [PATCH 1/3] cmd_push: set "atomic" bit directly Jeff King
2015-02-16  6:13           ` [PATCH 2/3] cmd_push: pass "flags" pointer to config callback Jeff King
2015-02-16  7:05             ` Junio C Hamano
2015-02-16  7:16               ` Jeff King
2015-02-16  6:16           ` [PATCH 3/3] push: allow --follow-tags to be set by config push.followTags Jeff King
2015-03-14  6:06             ` Junio C Hamano
2015-03-14 17:34               ` Jeff King
2015-03-14 17:50               ` Dave Olszewski
2015-03-14 22:08                 ` Junio C Hamano
2015-02-16  6:11         ` [PATCH 3/2] " Junio C Hamano
2015-02-16  6:17           ` Jeff King
  -- strict thread matches above, loose matches on Subject: below --
2015-02-15 23:39 [PATCH] push: allow --follow-tags' " Dave Olszewski
2015-02-16  0:01 ` Junio C Hamano

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=20150216052049.GA5031@peff.net \
    --to=peff@peff.net \
    --cc=cxreg@pobox.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    /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).