On Tue, 18 Feb 2014 23:03:45 +0900 (JST) [email protected] (YAMAMOTO Takashi) wrote:
>> On Tue, 18 Feb 2014 22:52:38 +0900 (JST) >> [email protected] (YAMAMOTO Takashi) wrote: >> >>>> On Mon, 17 Feb 2014 02:25:55 -0500 >>>> Rob Udechukwu <[email protected]> wrote: >>>> >>>>> Sorry for the delay. Thanks for the help, the only modifications I had to >>>>> make to the code was within the init() and the serialize() function. >>>>> - init() - I am ensuring data"s length is a multiple of 8 as specified in >>>>> 1.3 >>>>> - serialize() - Removed the self.buf for just but >>>>> >>>>> Code listed below. >>>> >>>> Thanks! I converted this to a patch. I'll apply this if it works for >>>> you. >>>> >>>> = >>>> From b7b9ed72b9c51e7495f7a36032dcc1fa89a65110 Mon Sep 17 00:00:00 2001 >>>> From: Rob Udechukwu <[email protected]> >>>> Date: Tue, 18 Feb 2014 22:35:31 +0900 >>>> Subject: [PATCH] of13: support OFPAactionExperimenter with data >>>> >>>> Signed-off-by: Rob Udechukwu <[email protected]> >>>> Signed-off-by: FUJITA Tomonori <[email protected]> >>>> --- >>>> ryu/ofproto/ofproto_v1_3_parser.py | 11 +++++++++-- >>>> 1 file changed, 9 insertions(+), 2 deletions(-) >>>> >>>> diff --git a/ryu/ofproto/ofproto_v1_3_parser.py >>>> b/ryu/ofproto/ofproto_v1_3_parser.py >>>> index 39864d2..06addda 100644 >>>> --- a/ryu/ofproto/ofproto_v1_3_parser.py >>>> +++ b/ryu/ofproto/ofproto_v1_3_parser.py >>>> @@ -3159,19 +3159,26 @@ class OFPActionExperimenter(OFPAction): >>>> experimenter Experimenter ID >>>> ================ >>>> ====================================================== >>>> """ >>>> - def __init__(self, experimenter, type_=None, len_=None): >>>> + def __init__(self, experimenter, data=None, type_=None, len_=None): >>>> super(OFPActionExperimenter, self).__init__() >>>> self.experimenter = experimenter >>>> + self.data = data >>>> + self.len = (utils.round_up(len(data), 8) + >>>> + ofproto.OFP_ACTION_EXPERIMENTER_HEADER_SIZE) >>> >>> how about assert (len(data) % 8) == 0 ? >>> i think it's what the spec says. >> >> Ah, we had better to add a padding to the data if the length is not >> multiples of 8? > > yes, my understanding of the spec is that the actual length of > data should be a multiple of 8. > > automatic padding might confuse users as it breaks serialize/parse > round-trip. i have no strong opinion about this, though. I thought that our code already does such but looks like that it doesn't do padding for data that an user passes via an argument. Simply assert here is fine by me too. I thought that enabling an user to pass any-length data would be nice, and the parser raises an exception if it hits non multiples of 8 (because it's corrupted). ------------------------------------------------------------------------------ Managing the Performance of Cloud-Based Applications Take advantage of what the Cloud has to offer - Avoid Common Pitfalls. Read the Whitepaper. http://pubads.g.doubleclick.net/gampad/clk?id=121054471&iu=/4140/ostg.clktrk _______________________________________________ Ryu-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ryu-devel
