Sorry about that, I'm beginning in sending patch for the kernel. I tried 
to use the git send-email command, but I saw after that the command just 
send email to the people I added, but not the mailing list. To be sure 
to (re)send the last patch, I attached it to be sure to send it 
"correctly", but not according to the recommended practice.

I will resend the whole patch list to the mailing list.

The patch you told was not intended to be self-consistant but a final 
cleanup after the two patches I tried to send before. I splitted them 
like that in order to keep them human readable as git tried to intrelace 
the additions and removals, even if they were not related together.

Sorry again

Benjamin

On 03/08/2010 09:27 PM, Michael Poole wrote:

Benjamin Tissoires writes:

quoted

quoted

Well ... I don't see any patch titled "HID: deletion of unused code in
hid-magicmouse" neither in my inbox nor in my linux-input@ folder. Could
you perhaps please resend?

Thanks,

Of course, I attached it. Do I need to send it in mbox format or is it ok ?

Cheers,
Benjamin

The recommended practice for sending patches is (from
Documentation/SubmittingPatches) "7) No MIME, no links, no compression,
no attachments.  Just plain text."

I'm a bit mystified, though -- if magicmouse_setup_input() is from "the
old implementation", where is the new implementation?  Jiri's
hid/for-next and hid/master trees definitely use the input device that
it sets up.  The "magicmouse_input_mapping" function that shows up at
the end of your patch doesn't exist in those trees.  The reason I didn't
use input_mapping() initially was because it only allows setting up one
input field, and the touch records generate multiple types of events.

Michael Poole