> On Dec 1, 2017, at 7:50 AM, Murillo Bernardes <[email protected]> wrote: > > On Fri, Dec 1, 2017 at 11:18 PM, Dirk Hohndel <[email protected] > <mailto:[email protected]>> wrote: > So much fun to wake up to such good news. > Thank you, Murillo! > > I glanced through the diff and (like Miika) am not quite sure why you changed > things to QString (if this is needed, could this be a separate commit?) > > Change to QStrings is not really necessary, just made it quicker for me to > test what I really wanted to test.
OK > And of course the question is "why isn't the call to > connectionListMode.addAddress()" not needed? > > It is needed, still there for newDC. This just removed from the list things > we "know" (maybe not for a fact) are not DC. I haven't understood yet all the > code involved, so I might be missing something. > > Just to make BLE work this change is not really needed, sure. I'm not sure I understand your answer. But then I should stare at the code in context some more. > And likewise "why aren't we calling saveBtDeviceInfo(...) on the other > platforms". > > I only found references to getBtDeviceInfo on a code specific for Mac and iOS > (qt-ble.cpp:300). Not sure how this works on Mac since it is never populated. > > > It's not clear why for each device discovered, instead of just storing the > DeviceInfo we instead use this info to create multiple other structures (like > serviceUuids, btDCs) and later try to reconstruct whatever is needed. I think it is fair to say that this code went through a lot of iterations between the initial ideas and what ended up working across all of the platforms. It may be a good idea to look into cleaning things up and simplifying the data structures, once we have it all working. > I'd love to see a pull request - this makes it easy to get it tested on all > of our platforms. > > I don't think you've submitted code to Subsurface before - it's fairly > straight forward, CONTRIBUTING.md contains the basics and of course feel free > to ask if you need help :-) > > > I'm working on a proper patch and will create a PR, with signed-off patches > and all. Thank you. Much appreciated /D
_______________________________________________ subsurface mailing list [email protected] http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface
