git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH 01/15] refs.c make ref_transaction_create a wrapper to ref_transaction_update

From
Junio C Hamano <gitster@pobox.com>
Date
Jul 23, 2014, 23:07 UTC
Message-ID
<xmqq4my7brof.fsf@gitster.dls.corp.google.com>
In-Reply-To
<1406135035-26441-2-git-send-email-sahlberg@google.com>
Ronnie Sahlberg <sahlberg@google.com> writes:
Show 39 quoted lines
> Signed-off-by: Ronnie Sahlberg <sahlberg@google.com>
> ---
>  refs.c | 18 ++----------------
>  refs.h |  7 ++++---
>  2 files changed, 6 insertions(+), 19 deletions(-)
>
> diff --git a/refs.c b/refs.c
> index 6dcb920..8f2aa3a 100644
> --- a/refs.c
> +++ b/refs.c
> @@ -3490,28 +3490,14 @@ int ref_transaction_create(struct ref_transaction *transaction,
>  			   int flags, const char *msg,
>  			   struct strbuf *err)
>  {
> -	struct ref_update *update;
> -
>  	if (transaction->state != REF_TRANSACTION_OPEN)
>  		die("BUG: create called for transaction that is not open");
>  
>  	if (!new_sha1 || is_null_sha1(new_sha1))
>  		die("BUG: create ref with null new_sha1");
>  
> -	if (check_refname_format(refname, REFNAME_ALLOW_ONELEVEL)) {
> -		strbuf_addf(err, "Bad refname: %s", refname);
> -		return -1;
> -	}
> -
> -	update = add_update(transaction, refname);
> -
> -	hashcpy(update->new_sha1, new_sha1);
> -	hashclr(update->old_sha1);
> -	update->flags = flags;
> -	update->have_old = 1;
> -	if (msg)
> -		update->msg = xstrdup(msg);
> -	return 0;
> +	return ref_transaction_update(transaction, refname, new_sha1,
> +				      null_sha1, flags, 1, msg, err);
>  }

Makes sense, but at the same time makes me wonder why no code is moved to ref_transaction_update() while removing code from here, which would only mean that code in ref_transaction_update() was added redundantly in the first place.

An ideal series would have had only "update" code in _update() when the function is added, and later with a patch like this would lose code from _create() while adding some code to _update(), I would think.

Or if all the code in _update() was necessary from day one, then perhaps this change should have been part of the same patch.

It's not a big deal either way, though.
Thanks.
Show 19 quoted lines
>  int ref_transaction_delete(struct ref_transaction *transaction,
> diff --git a/refs.h b/refs.h
> index b0476c1..1c08cfd 100644
> --- a/refs.h
> +++ b/refs.h
> @@ -276,9 +276,10 @@ struct ref_transaction *ref_transaction_begin(struct strbuf *err);
>  /*
>   * Add a reference update to transaction.  new_sha1 is the value that
>   * the reference should have after the update, or zeros if it should
> - * be deleted.  If have_old is true, then old_sha1 holds the value
> - * that the reference should have had before the update, or zeros if
> - * it must not have existed beforehand.
> + * be deleted.  If have_old is true and old_sha is not the null_sha1
> + * then the previous value of the ref must match or the update will fail.
> + * If have_old is true and old_sha1 is the null_sha1 then the ref must not
> + * already exist and a new ref will be created with new_sha1.
>   * Function returns 0 on success and non-zero on failure. A failure to update
>   * means that the transaction as a whole has failed and will need to be
>   * rolled back.
Previous: Ronnie SahlbergNext: Ronnie Sahlberg
Message 3 of 19 in “ref-transactions for reflogs”
  1. 00/15 ref-transactions for reflogsRonnie Sahlberg, Jul 23, 2014
  2. 01/15 refs.c make ref_transaction_create a wrapper to ref_transaction_updateRonnie Sahlberg, Jul 23, 2014
  3. Junio C HamanoJul 23, 2014
  4. 02/15 refs.c: make ref_transaction_delete a wrapper for ref_transaction_updateRonnie Sahlberg, Jul 23, 2014
  5. 03/15 refs.c: rename the transaction functionsRonnie Sahlberg, Jul 23, 2014
  6. 04/15 refs.c: add a new update_type field to ref_updateRonnie Sahlberg, Jul 23, 2014
  7. 05/15 refs.c: add a function to append a reflog entry to a fdRonnie Sahlberg, Jul 23, 2014
  8. 06/15 lockfile.c: make hold_lock_file_for_append preserve meaningful errnoRonnie Sahlberg, Jul 23, 2014
  9. 07/15 refs.c: add a transaction function to append a reflog entryRonnie Sahlberg, Jul 23, 2014
  10. 08/15 refs.c: add a flag to allow reflog updates to truncate the logRonnie Sahlberg, Jul 23, 2014
  11. 09/15 refs.c: only write reflog update if msg is non-NULLRonnie Sahlberg, Jul 23, 2014
  12. 10/15 refs.c: allow multiple reflog updates during a single transactionRonnie Sahlberg, Jul 23, 2014
  13. 11/15 reflog.c: use a reflog transaction when writing during expireRonnie Sahlberg, Jul 23, 2014
  14. 12/15 refs.c: rename log_ref_setup to create_reflogRonnie Sahlberg, Jul 23, 2014
  15. 13/15 refs.c: make unlock_ref/close_ref/commit_ref staticRonnie Sahlberg, Jul 23, 2014
  16. 14/15 refs.c: remove lock_any_ref_for_updateRonnie Sahlberg, Jul 23, 2014
  17. 15/15 refs.c: allow deleting refs with a broken sha1Ronnie Sahlberg, Jul 23, 2014
  18. Eric SunshineJul 23, 2014
  19. Junio C HamanoJul 23, 2014

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.