Re: [libvirt] [PATCH 2/6] Add DBus helper methods for creating reply messages

On 03/20/2014 11:28 AM, Daniel P. Berrange wrote: > On Thu, Mar 20, 2014 at 06:16:08PM +0100, Michal Privoznik wrote: >> On 20.03.2014 13:28, Daniel P. Berrange wrote: >>> The test suites often have to create DBus method reply messages >>> with payloads. Create two helpers for simplifying the process >>> of creating replies with payloads. >>> >>> Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com&gt; >>> --- >>> src/libvirt_private.syms | 2 ++ >>> src/util/virdbus.c | 60 ++++++++++++++++++++++++++++++++++++++++++++++++ >>> src/util/virdbus.h | 5 ++++ >>> 3 files changed, 67 insertions(+) > > >>> + ret = 0; >>> + cleanup: >> >> Indentation's off. > > Do we actually have an indentation rule for labels ? Most code starts it in the first column, with no leading space. In some respects, emacs doesn't handle it well (it assumes anything in the first column is a function name, so it tries to treat the label as a function name when generating changelog templates); on the other hand, when I hit TAB on a label, emacs reindents it to the first column (that is, our .dir-locals.el requests c-file-style "K&R", and apparently that style includes putting labels at one indentation layer less than the rest of the code; so if you are labelling something that indents four spaces, the label gets indented 0 spaces). I've been going by the general rule of thumb that if emacs reindents something, then my style wasn't consistent with the bulk of the code; but I agree that HACKING doesn't actually mention this, and not everyone uses emacs.