[Kde-scm-interest] atomicity, again

[Boyd Stephen Smith Jr.]

In <200906151757.53601.thiago at kde.org>, Thiago Macieira wrote:
>Jeff Mitchell wrote:
>>Thomas Capricelli wrote:
>>> http://www.selenic.com/blog/mercurial/sharedandsubrepos.html
>>Git has a similar feature (submodules) although I believe there are some
>>drawbacks with it; Thiago probably knows more.
>Git submodules work fine for what they're meant to be: tracking the state
>of a sub-module for a given state of its parent module. However, it's not
>meant to serve as an atomic commit.
>
>Reading the link above, it seems that Mercurial's solution is exactly like
>Git's. It does not solve the problem of atomic commits across multiple
>repositories.

Odd, I read the link, and it seemed as though commit command(s) in Mercurial 
would recur into subrepositories and perform a commit there and then use 
that new commit id for updating the "parent" repository.

At a structural level, there's little to no difference using this wouldn't 
make any atomicity guarantees[1] that git doesn't.

It does seem to make it harder to forget to make your commit atomic, but 
it's been little while since I looked into git's submodule handling as I 
don't use it myself and couldn't continue to follow the git mailing list.
-- 
Boyd Stephen Smith Jr.           	 ,= ,-_-. =.
bss at iguanasuicide.net            	((_/)o o(\_))
ICQ: 514984 YM/AIM: DaTwinkDaddy 	 `-'(. .)`-'
http://iguanasuicide.net/        	     \_/

[1] Whatever those are in the case.  I'm not really sure what "atomic 
commits across multiple repositories" means in this context.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
Url : http://mail.kde.org/pipermail/kde-scm-interest/attachments/20090615/bd5e4769/attachment.sig