On 11/19/2015 06:10 AM, Daniel P. Berrange wrote: > On Thu, Nov 19, 2015 at 01:03:39PM +0000, Dr. David Alan Gilbert wrote: >> * Markus Armbruster (arm...@redhat.com) wrote: >>> "Please keep this list in alphabetical order" has been more honoured >>> in the breach than in the observance. Clean up.
Whose alphabet? ... >>> >>> While there, drop a redundant struct declaration. >>> >>> Signed-off-by: Markus Armbruster <arm...@redhat.com> >> Reviewed-by: Eric Blake <ebl...@redhat.com> >> Hmph, it's only a year since I sorted this file. >> At least one of these changes is from a misordering within one patch! > > Which just goes to show that any such rules about sorting are > not worth the hassle unless you also provide some make rule to > validate them as part of 'make test'. It shouldn't be too > hard to hook it up - a make rule that runs this: > > grep typedef include/qemu/typedefs.h > orig.txt > sort orig.txt > sorted.txt ...This would have to be 'LC_ALL=C sort orig.txt ...' to ensure that difference in locale doesn't change the order (whether due to case-insensitivity, or due to ignoring whitespace). Picking on one of the changes in the commit: $ printf 'typedef struct PropertyInfo PropertyInfo;\ntypedef struct Property Property;\n' | sort # LANG=en_US.UTF-8 typedef struct PropertyInfo PropertyInfo; typedef struct Property Property; $ printf 'typedef struct PropertyInfo PropertyInfo;\ntypedef struct Property Property;\n' | LC_ALL=C sort typedef struct Property Property; typedef struct PropertyInfo PropertyInfo; > diff orig.txt sorted.txt > /dev/null > > as part of 'make test' would suffice Also, Markus probably noticed the problem because I have a pending patch that will be adding QObject to the list, and my addition will conflict since it falls next to the fixup for Range vs. QEMUTimerListGroup. Whichever one goes in first gets to watch the other rebase :) -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
