On 12/20/18 3:32 PM, Gregg Reynolds wrote: > Have been looking at nw monitoring code. > > It's astoundingly complicated. But in the end, I see one thing happening > for all transports: update the list of local endpoints. For TCP there's > more, because of cloud, but it is not yet implementated. > > On my analysis we can eliminate almost all of this code. E.g. there's a lot > of code devoted to maintaining an interface list. Which is never actually > used. Example: > CAIPGetInterfaceInformation > > is a waste of time and can be eliminated. Again, the interface list it > maintains serves no purpose afaik, so can be eliminated.
this code seems to be more complicated than it needs to almost anywhere. asking a lot, but are you able to tell if: this was an attempt to start implementing interfaces, which then turned out to be the wrong way and was thus not pursued; or if there should be interface lists but iotivity doesn't implement them properly? From my previous life I seem to recall there was a mismatch between where the OCF specs were going with interfaces (at least "default interface"), and what iotivity did; I don't know if that ever got resolved. interfaces as envisioned by OCF are a stretch in applying the concept you see in some object-oriented languages to the OCF API model, and didn't ever get explained well. but that's just my opinion as an outsider-gadfly type these days :) -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#10084): https://lists.iotivity.org/g/iotivity-dev/message/10084 Mute This Topic: https://lists.iotivity.org/mt/28815690/21656 Group Owner: iotivity-dev+ow...@lists.iotivity.org Unsubscribe: https://lists.iotivity.org/g/iotivity-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
