On 03/03/2010 01:19 AM, Martin Nagy wrote:
this seems to be a patch for code that isn't even in the repository yet.
BTW, how is one supposed to apply this patch? Saving the whole email
didn't work for me, I had to hand edit it and git-am still complained.

Right, this was superseded by "0003 Fix JSON binary encode and decode errors (rebase)".

The patch didn't apply because the original "Fix JSON binary encode and decode errors" patch didn't apply because of a merge conflict.

Patches have to be applied in order and when one gets skipped either because of review delay or merge conflict it can create problems.

FWIW from this point forward all my patch submissions will be numbered. The numbering will be done via the --start-number argument to git format-patch combined with the -n argument to add the patch number to the subject-prefix. Thus for example my patch number 3 will show up with a subject which looks like this:

[PATCH 3/3] xxx xxx xxx

This is standard git notation. The first number specifies the beginning patch number in the sequence. The second number specifies the last patch number in the sequence. When there is only one patch in the email sequence both numbers are the same. This will always be the case for us because we don't generate multiple patches in a single email (at least not yet, however this is something the kernel team and others do)

So lets say 3 is my next patch number the command to generate the patch email would look like this:

git format-patch -n -start-number 3 -1

FWIW, I have a script which automates this, it maintains the sequence number in a file in the same directory which all my patches are written to. It's also capable of emailing the patch.

John Dennis <jden...@redhat.com>

Looking to carve out IT costs?

Freeipa-devel mailing list

Reply via email to