[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: git copies of cvs modules available
On 2009-11-13 07:42 +0000 (Fri), Thomas Adam wrote:
> 2009/11/13 Curt Sampson <cjs%netbsd.org@localhost>:
> > One case is doing doing a single checkout of HEAD, where a git clone
> You could do a --shallow clone here, but that's by-the-by.
That's a good point, thanks for reminding me. Though there are certain
limitations on how a shallow clone can be used, some of which are not
relevant to a comparision of svn and git (i.e., being unable to clone,
since that's something that makes no sense in svn) and some of which are
(being unable to push, which is like having an svn check from which you
> The history is really important.
When you make statements like this, you start to sound like a git fanboy
rather than someone trying to make a rational comparison. (I presume you
really are the latter, and are just letting your tone slip a bit.)
The history is *not* always really important, you have it available with
svn anyway, so long as you have a network connection or an rsync copy of
the repo, and there's no question that svn is more efficient in its use
of disk space in this regard.
> > The other, and even bigger case is partitial checkouts. Subversion will
> > let you check out just src/bin/cat, and our build system will let you
> > build it standing alone. Git does not do this and possibly never will.
> Umm, that's not true.
Yes it is.
> There's sparse-checkout support for Git which is currently being
> worked on.
Well, we'll see. Nothing's been accepted into git yet, and there are no
current plans to have this out for any particular release, as far as I
can tell. We've had patches floating around to do something like this
for at least a year, but that nothing has been acceptable to date seems
to indicate that this is not a trivial problem to solve.
Curt Sampson <cjs%starling-software.com@localhost> +81 90 7737 2974
Functional programming in all senses of the word:
Main Index |
Thread Index |