All of lore.kernel.org
 help / color / mirror / Atom feed
From: Junio C Hamano <gitster@pobox.com>
To: Ronnie Sahlberg <sahlberg@google.com>
Cc: git@vger.kernel.org, mhagger@alum.mit.edu
Subject: Re: [PATCH v2 02/13] refs.c: use a single exit path from transaction commit and handle onerr
Date: Tue, 22 Apr 2014 12:11:01 -0700	[thread overview]
Message-ID: <xmqqbnvt5ge2.fsf@gitster.dls.corp.google.com> (raw)
In-Reply-To: <1398120811-20284-3-git-send-email-sahlberg@google.com> (Ronnie Sahlberg's message of "Mon, 21 Apr 2014 15:53:20 -0700")

Ronnie Sahlberg <sahlberg@google.com> writes:

> diff --git a/refs.c b/refs.c
> index 138ab70..9daf89e 100644
> --- a/refs.c
> +++ b/refs.c
> @@ -3414,12 +3414,12 @@ int ref_transaction_commit(struct ref_transaction *transaction,
>  			   const char *msg, enum action_on_err onerr)
> ...
> +	if (ret) {
> +		const char *str = "Cannot commit transaction.";
> +		switch (onerr) {
> +		case UPDATE_REFS_MSG_ON_ERR: error(str); break;
> +		case UPDATE_REFS_DIE_ON_ERR: die(str); break;
> +		case UPDATE_REFS_QUIET_ON_ERR: break;
> +		}
> +	}
>  	return ret;
>  }

I think this is a response to Michael's earlier review "do different
callers want to give different error messages more suitable for the
situation?".  I suspect that the ideal endgame may end up all
callers passing QUIET_ON_ERR and doing the error message themselves,
e.g. branch.c::craete_branch() may end something like this:

    ...
    if (!dont_change_ref)
        if (ref_transaction_commit(..., UPDATE_REFS_QUIET_ON_ERR))
                die_errno(_("Failed to write branch '%s'"),
                          skip_prefix(ref.buf, "refs/heads/));

And in the meantime until the callers are converted, we may end up
showing the fallback "Cannot commit transaction." but that would be
OK during the development and polishing of this series.

  reply	other threads:[~2014-04-22 19:11 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-04-21 22:53 [PATCH v2 00/13] Use ref transactions from most callers Ronnie Sahlberg
2014-04-21 22:53 ` [PATCH v2 01/13] refs.c: constify the sha arguments for ref_transaction_create|delete|update Ronnie Sahlberg
2014-04-21 22:53 ` [PATCH v2 02/13] refs.c: use a single exit path from transaction commit and handle onerr Ronnie Sahlberg
2014-04-22 19:11   ` Junio C Hamano [this message]
2014-04-22 22:46     ` Ronnie Sahlberg
2014-04-22 19:14   ` Junio C Hamano
2014-04-21 22:53 ` [PATCH v2 03/13] refs.c: change ref_transaction_update() to do error checking and return status Ronnie Sahlberg
2014-04-21 22:53 ` [PATCH v2 04/13] refs.c: change ref_transaction_create " Ronnie Sahlberg
2014-04-21 22:53 ` [PATCH v2 05/13] refs.c: ref_transaction_delete to check for error " Ronnie Sahlberg
2014-04-21 22:53 ` [PATCH v2 06/13] tag.c: use ref transactions when doing updates Ronnie Sahlberg
2014-04-21 22:53 ` [PATCH v2 07/13] replace.c: use the ref transaction functions for updates Ronnie Sahlberg
2014-04-21 22:53 ` [PATCH v2 08/13] commit.c: use ref transactions " Ronnie Sahlberg
2014-04-21 22:53 ` [PATCH v2 09/13] sequencer.c: use ref transactions for all ref updates Ronnie Sahlberg
2014-04-21 22:53 ` [PATCH v2 10/13] fast-import.c: change update_branch to use ref transactions Ronnie Sahlberg
2014-04-21 22:53 ` [PATCH v2 11/13] branch.c: use ref transaction for all ref updates Ronnie Sahlberg
2014-04-21 22:53 ` [PATCH v2 12/13] walker.c: use ref transaction for " Ronnie Sahlberg
2014-04-21 22:53 ` [PATCH v2 13/13] refs.c: change update_ref to use a transaction Ronnie Sahlberg
2014-04-22 19:34 ` [PATCH v2 00/13] Use ref transactions from most callers Junio C Hamano
2014-04-22 22:43   ` Ronnie Sahlberg

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=xmqqbnvt5ge2.fsf@gitster.dls.corp.google.com \
    --to=gitster@pobox.com \
    --cc=git@vger.kernel.org \
    --cc=mhagger@alum.mit.edu \
    --cc=sahlberg@google.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.