Re: [swift-corelibs-dev] [swift-dev] Swift Encoders on Linux

2017-05-23 Thread Tony Parker via swift-corelibs-dev
Hi David, I think the struct Measurement should probably be very similar, but the class NSMeasurement is probably pretty different. Regardless, if there is something you think needs to be ported over (I would believe that there are a few changes made there that we haven’t copied yet), please

Re: [swift-corelibs-dev] [swift-dev] Swift Encoders on Linux

2017-05-23 Thread David Hart via swift-corelibs-dev
Hi Tony, Looking forward to all this on Linux! On the same topic, I looked at the Measurement Swift type and did a diff between the shims in the swift repo and the version in corelibs-foundation and saw quite a few differences. Is this something that can be improved or it is a consequence of

Re: [swift-corelibs-dev] Debugging failures seen only in CI

2017-05-23 Thread Tony Parker via swift-corelibs-dev
I'm ok with that if it helps us get to the bottom of this once and for all. I wonder if it would be possible to enable the logging conditionally, so it only appears in CI? Maybe via a runtime call ("turn logging on now"). - Tony > On May 23, 2017, at 2:25 AM, Pushkar N Kulkarni

Re: [swift-corelibs-dev] Swift Encoders on Linux

2017-05-23 Thread Tony Parker via swift-corelibs-dev
Hi Ankit, Our intention is to share as much of the implementation as possible. There are some details that need to be worked out around bridging of types like NSNumber (since bridging doesn't exist on Linux we may need to avoid using 'as' in a few places), but in general it could be as easy as

[swift-corelibs-dev] Swift Encoders on Linux

2017-05-23 Thread Ankit Aggarwal via swift-corelibs-dev
Hi, It looks like SE-0167 is implemented and can be used in the latest macOS snapshot (yay!). Will corelibs-foundation need a completely separate implementation? Is the corelibs-foundation implementation going to be community driven? PS: I think these are awesome APIs and the Package Manager