Re: [PATCH]: remove extra-verbosity from git-clone (http)

From: Junio C Hamano <junkio@cox.net>
Date: Mar 19, 2007, 23:45 UTC
Message-ID: <7vslc0er1e.fsf@assigned-by-dhcp.cox.net>
In-Reply-To: <20070319230436.GB12555@sequoia.sous-sol.org>

Chris Wright <chrisw@sous-sol.org> writes:

Show 26 quoted lines

> * Fernando Herrera (fherrera@onirica.com) wrote:
>> the attached patch removes the extra verbosity when cloning a http
>> repository. I noticed this after some git upgrade in fedora. Please,
>> consider applying it, getting tons of "got XXX", "walk XXX" lines by
>> default is a little bit annoying.
>> 
>> Salu2
>> 
>> PS: Thanks for this so great piece os software!
>> 
>> --- git-clone.sh.orig	2007-02-15 19:03:51.000000000 +0200
>> +++ git-clone.sh	2007-02-15 19:04:06.000000000 +0200
>> @@ -59,7 +59,7 @@
>>  		else
>>  			tname=$name
>>  		fi
>> -		git-http-fetch -v -a -w "$tname" "$name" "$1/" || exit 1
>> +		git-http-fetch -a -w "$tname" "$name" "$1/" || exit 1
>>  	done <"$clone_tmp/refs"
>>  	rm -fr "$clone_tmp"
>>  	http_fetch "$1/HEAD" "$GIT_DIR/REMOTE_HEAD" ||
>
> Was there any problem with this patch from Fernando?  I don't care much
> either way (I simply never use http fetch myself ;-).  Just doing some
> old bugzilla cleanup and noticed this still was neither picked up nor
> replied to.  So, in case it just fell through the cracks...

I saw it. The -v option used not to be there, but we added it with an explicit purpose of not boring people fetching over http connection (which used to be slow especially before packfiles were introduced). I do not care much either way, but I suspect there actually are people who want to see that "assuring look" of something happening. I dunno.

I am certainly *not* opposed to have an explicit option or a per-user configuration to make it not pass -v to http-fetch. I am just wondering if we want to change the default. Especially people new to git may wonder what is happening without _any_ feedback if we turned it off by default.