[RFC][PATCH 0/4] Speed improvement in fetch/clone

[John Arbash Meinel]

Goffredo Baroncelli wrote:
> On Thursday 15 December 2005 05:18, you (Robert Collins) wrote:
> 
> 
>>It would be nice to have code that is changed + the tests for it in the
>>same patches :)..
> 
> The mail was sent as RFC; my target was first focusing on the function 
> file_involved( ), then on their application.
> So I tried to sent the code in the cleaner mode. And on the basis of my experience
> the best is to send the change in little patches ( one for mail ).
> 
> However I noted that on this mailing list the code is not very often attached to 
> the mail, instead is sent the url of the repo ( which is useful ). The thing which
> made me strange, is that so it isn't impossible to see quickly the code...
> And cloning the bzr repo ( or its modification ) via http ( which is the most
> frequent transport ) require too much time...
> 
> Please, tell me how you ( and the other in this ML ) prefer to receive the code.
> 
> Thanks for you suggestions
> Goffredo
> 

In general, I believe we prefer to at least receive the branch that you
have done your work on. That way we can have a local copy for doing
merge, etc.

Second, attaching all patches is prefered to several messages, and is
preferred to having the patches inline.

At least I think that is the general consensus.

For small changes, a patch is very nice for code review. For larger
changes, it is difficult to get the feeling for context, etc. So it is
easier to have a branch that you can run merge/annotate/diff etc.

John
=:->

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 256 bytes
Desc: OpenPGP digital signature
Url : https://lists.ubuntu.com/archives/bazaar/attachments/20051215/32b09e59/attachment.pgp