Re: [swift-corelibs-dev] libdispatch build issue

Cool, I’ll have a look later this week or when it gets more ready ;) btw, we got a pull request that seem good to me (https://github.com/apple/swift-corelibs-libdispatch/pull/13 ) but I don’t have the linux setup to check it’s good, w

Re: [swift-corelibs-dev] libdispatch build issue

-Pierre > On Dec 14, 2015, at 9:18 AM, Pierre Habouzit via swift-corelibs-dev > wrote: > > Cool, I’ll have a look later this week or when it gets more ready ;) > > btw, we got a pull request that seem good to me > (https://github.com/apple/swift-corelibs-libdispatch/pull/13 >

[swift-corelibs-dev] Possible Proposal: Foundation corelibs API change necessary for NSPredicate

I have a pending pull request that needs a little more work around NSPredicates, but in my testing on darwin foundation, I’ve discovered what appears to be an obj-c nullability annotation bug. When constructing a block predicate, the type of the block is this: (AnyObject, [String: AnyObject]?)

Re: [swift-corelibs-dev] Possible Proposal: Foundation corelibs API change necessary for NSPredicate

This might actually just be a bug in our annotations of what is nullable and what is not. I would have to double check but it seems pretty reasonable that it should have been nullable to begin with. > On Dec 14, 2015, at 10:06 AM, Kevin Lundberg via swift-corelibs-dev > wrote: > > I have a pe

Re: [swift-corelibs-dev] Possible Proposal: Foundation corelibs API change necessary for NSPredicate

Great, should I file a radar on bugreport.apple.com for this? And if it’s simple enough to fix that can I apply the same fix to the SwiftFoundation API as a PR without any extra process? -- Kevin Lundberg ke...@klundberg.com > On Dec 14, 2015, at 1:09 PM, Philippe Hausler wrote: > > This migh

Re: [swift-corelibs-dev] Possible Proposal: Foundation corelibs API change necessary for NSPredicate

As a general rule of thumb we ❤️radars. In this case I have already filed a bug with the component owner since it clearly looks like this is perhaps a incorrectly annotated API. So a PR for this is quite reasonable to change to ? since it would be a bit difficult to implement else wise. > On De

Re: [swift-corelibs-dev] [swift-corelibs-libdispatch] Port transform and use libbsd (#13)

> […] > On a previous pull request, Daniel had requested we use #ifdef __APPLE__ > instead of #if HAVE_MACH. Not sure if that also applies to this pull request > too or if HAVE_MACH is ok to use in these files. Hi, I think this explanation is worth sharing and not leaving on an hidden review. T