From: René Scharfe <rene.scharfe@lsrfire.ath.cx>
Date: Fri, 09 Nov 2007 00:49:19 GMT
Subject: Re: [PATCH 3/3] pretty=format: Avoid some expensive calculations when not needed
Message-ID: <4733AE8F.4080403@lsrfire.ath.cx>
In-Reply-To: <7vhcjxaire.fsf@gitster.siamese.dyndns.org>

Junio C Hamano schrieb:
> René Scharfe <rene.scharfe@lsrfire.ath.cx> writes:
> 
>> I haven't seen any comments on strbuf_expand.  Is it too far out?
>> Here it is again, adjusted for current master and with the changes
>> to strbuf.[ch] coming first:
> 
> Numbers talk ;-).

:-)  I just hope someone else has compared the times, too..

> In your previous round, you alluded that the strbuf_expand()
> interface could allow caching of the return value of fn(), but I
> do not think strbuf_expand() in this patch has anything to
> directly support that notion.
> 
> Nor I would expect to --- fn() could keep the really expensive
> information cached, keyed with context value, if it wanted to,
> but in practice for the purpose of format_commit_item() I do not
> offhand see anything cacheable and reusable, unless the user did
> stupid things (e.g. use more than one %h in the format string).

Yes, that's what I arrived at, too, when I actually wrote and used
the function.

> I added a few paragraphs to describe the API in the commit log
> message, and rewrote "# master" to "(master)" etc.

Thanks!  I'll send a slightly enhanced version in a two-patch
series -- with an actual commit message, including your API
desciption -- in just a few seconds.

René