Facts:
* xtro-sharpie references Mono.Cecil 0.9.6 from a nuget.
* If a local Mono.Cecil.dll can't be found (according to the HintPath in the
csproj, which points to the nuget), then msbuild will look in the system
Mono.
* Mono 5.8 ships Mono.Cecil 0.10.
* Mono.Cecil 0.10 is not source compatible with 0.9.6 (there's a small issue
with interfaces).
* xtro-sharpie's source code works with v0.10.
This all means that xtro-sharpie will build fine as long as the 0.9.6 nuget
has *not* been restored. This can manifest itself confusingly ("msbuild xtro-
sharpie.sln" works fine from the command line, open the solution in VSfM and
it doesn't build anymore, not even from the command line, because VSfM
automatically restored nugets in the background).
Update the source code to work with Mono.Cecil 0.9.6 because there's no 0.10
nuget yet (yet keeping the code for v0.10 clearly marked as such for future
updates to v0.10).
Also bump from 0.9.6.1 to 0.9.6.4 since that's the latest available.
* [MSBuild] Do not set CFBundleDevelopmentRegion if not present.
This is a complicated fix. This is a regression introduced by Apple.
CFLocaleCopyCurrent(), used in the iOS code, will return the value of
the application's CFBundleDevelopmentRegion Info.plist key if all of the
following conditions are true:
* CFBundleDevelopmentRegion is present in the Info.plist
* The CFBundleDevelopmentRegion language is in the list of preferred
languages on the iOS device, but isn't the first one
* There are no localized resources (i.e. no .lproj directory) in the app
for the first preferred locale
This differs from iOS 10 where the presence of the
CFBundleDevelopmentRegion key had no effect.
Note that if the CFBundleDevelopmentRegion key is not present at all,
CFLocaleCopyCurrent() always returns the first preferred locale as it
did in iOS 10.
We are adding the key by default in the plist of the applications which confuses users since they do not see the key in the .plist added by the template. This commit removes it to be more explicit and help users understand the behaviour.
* [generator] Properly set the IsDirectBinding value.
Properly set the IsDirectBinding value to false for models and synthetic types.
This also means we can now stop excluding models when testing if the
IsDirectBinding value is correct.
Also set IsDirectBinding value to true for sealed wrapper types, since those
will always be direct bindings since they can't be subclassed.
https://gist.github.com/rolfbjarne/24028bf944db848fed4083c460d0ec71
* [tests] Add introspection exclusion for XM.
* [introspection] Add back exclusions for Classic, since we can't modify/fix Classic assemblies anymore.
* [generator] Print the correct protocol name with the protocol attribute.
Fixes this test failure:
[FAIL] Foundation.NSUrlDownloadDelegate : ConformsToProtocol(null) failed
because our binding code claimed that our `NSUrlDownloadDelegate` class
implemented the `NSUrlDownloadDelegate` protocol, but since the
`NSUrlDownloadDelegate` protocol doesn't exist (it's `NSURLDownloadDelegate` -
different case), we'd verify against a null protocol (and return true from
`ConformsToProtocol(null)`, which would fail the test).
* [xtro] Only treat interfaces as protocols.
Unfortunately we add [Protocol] to [Model]s as well as on interfaces, but we
must not process those in xtro, since they don't correspond with the actual protocol.
When codesigning iOS Simulator builds, it is important to note
that the codesign command itself cannot include the entitlements,
but the mtouch command *MUST* embed the entitlements using
-Xlinker.
https://bugzilla.xamarin.com/show_bug.cgi?id=59379
MetalPerformanceShadersLibrary is new in macOS 10.13 and only available
for 64bits.
from https://wrench.internalx.com/Wrench/WebServices/Download.aspx?workfile_id=19778531
Tests run: 248, Passed: 238, Errors: 0, Failures: 1, Inconclusive: 0
Not run: 9, Invalid: 0, Ignored: 9, Skipped: 0
Elapsed time: 00:00:11.3800000
Errors and Failures:
1) ExpectedLibrariesAreLoaded (Xamarin.Mac.Tests.EveryFrameworkSmokeTests.ExpectedLibrariesAreLoaded)
MetalPerformanceShadersLibrary (/System/Library/Frameworks/MetalPerformanceShaders.framework/MetalPerformanceShaders) failed to load but this was not expected
at Xamarin.Mac.Tests.EveryFrameworkSmokeTests.ExpectedLibrariesAreLoaded () [0x000c5] in /Users/builder/data/lanes/5665/74d2dcad/source/xamarin-macios/tests/apitest/src/EveryFrameworkSmokeTest.cs:99
at (wrapper managed-to-native) System.Reflection.MonoMethod:InternalInvoke (System.Reflection.MonoMethod,object,object[],System.Exception&)
at System.Reflection.MonoMethod.Invoke (System.Object obj, System.Reflection.BindingFlags invokeAttr, System.Reflection.Binder binder, System.Object[] parameters, System.Globalization.CultureInfo culture) [0x00032] in /Library/Frameworks/Xamarin.Mac.framework/Versions/4.1.1.45/src/mono/mcs/class/corlib/System.Reflection/MonoMethod.cs:305
Since tvOS 11 there's not a single font, on that platform, that
has the ligatures that the test verified.
Test remains enabled for other platforms.
reference:
https://bugzilla.xamarin.com/show_bug.cgi?id=58929
* [tests] Fix introspection tests for macOS
It seems that apple forgot to ship SSLSetALPNProtocols and SSLCopyALPNProtocols in macOS
there are already radars filled about this https://bugs.swift.org/browse/SR-6131
So this test will fail once Apple fixes this issue. when this happens we need to do two things, reenable
the API and reenable the [Get|Set]AlpnProtocols tests, the one insides 'StreamDefaults' for the mac.
* Implement feedback
Wrench runs `wrench-mac-xammac_tests`, but since there were no such target,
make would execute the `wrench-%` target, which is disabled when iOS is
disabled.
Thus this strange behavior would be seen on wrench for xammac tests when iOS
is disabled:
/Applications/Xcode92-beta2.app/Contents/Developer/usr/bin/make -C /Users/builder/data/lanes/5665/d2b1b757/source/xamarin-macios/tests wrench-mac-xammac_tests
git clean -xfdq
iOS tests have been disabled [wrench-mac-xammac_tests]
By creating the `wrench-mac-xammac_tests` target, we'll end up doing the right
thing instead.
* [MetalPerformanceShaders] Activate bindings for Xamarin.Mac and add n… (#2816)
* [MetalPerformaceShaders] Several MPSCnnKernel properties should be readonly (#2938)
The subclasses versions of the properties need Override, cannot be removed since it would break visibility for iOS 10
* Remove some [Model] attributes that sholdn't be needed
* Fix introspection test crashes
* More introspection fixes
* NN does not need to be PascalCased
Remove unneeded Models and BaseTypes
* PR Whitespace fixes and renamings
* Paste fail
* More fixes from PR comments
* [MPS] Adds new intro test, fixes ctors and xtro output
* Removes duplicated availability attributes.
* Removes obsoleted API from macOS since mps is new to it.
* Fixes xtro output.
* Adds missing API.
* Fixes parameterless ctors, some of them do not really work, found
by our new intro test and disabled the one that seem to not make
sense due to the presence of DesignatedInitializers.
* Fixes a selector typo.
* Adds new `ShouldNotExposeDefaultCtorTest` to intro.
ShouldNotExposeDefaultCtorTest
==============================
This test checks for types with a parameterless ctor that are subclasses
of `NSObject` and then cheks if the BaseType of said objects also expose
a parameterless ctor (all in .NET land), if this is not the case it reports
them and so they can manually audited. Also this test has the ability to
print alloc/init ObjC code by setting `genObjCTestCode` to `true` so you can
take this code into an Xcode project and easily tests the ctors.
It seems that xtro (sharpie) does not have a complete picture of when a ctor
must be exposed hence the hability to generate this code and manually test.
Right now this test is just enabled for MPS since it is the scope of this PR.
In the future it should be enabled for all other frameworks and the output be
manually audited.
* [MPS] Fixes premature collection possible in bindings (bug 59547) and implements feedback.
https://bugzilla.xamarin.com/show_bug.cgi?id=59547
* Fixes premature collection possible in bindings im MPSKernel.cs
* Fixes MPSImageHistogramTest from using deprecated API.
* Removes renamed selectors and typos from ApiSelectorTest and ApiTypoTest.
* [MPS] Reenable Copy API and DesignatedInitializer xtro feedback
* Implement more feedback
* More feedback
* [msbuild] Added support for building CoreML models
* [msbuild] Make sure to Dispose() processes after they're done
* Updated to use newer MLModel.Create() and MLFeatureValue.Create() methods
* Added SqueezeNet.txt to document SqueezeNet.mlmodel
Make the collection of logs (the `Logs` class) more capable by making it
disposable (which will dispose all contained logs) and give it a Directory
property to state where the logs should be stored. This makes it possible to
simplify a few repeated path calculations. It also allows us to easily dispose
the entire collection of logs when done with a test, as opposed to dispoing
the logs one by one.
Make LogFile more capable:
* Add support for writing byte arrays.
* Add support for logging after disposal: this will still write to the file,
and not keep any files open after finished writing. This fixes a problem
where writing to a log after it was disposed would crash xharness (which is
not all that uncommon, given the async nature of how xharness runs tests).
This makes it possible to get rid of LogStream, and use LogFile instead.
* [msbuild] Pack all iOS MSBuild Task assemblies into a single assembly
* Fixed the build
* Renamed ProcessArgumentBuilder to CommandLineArgumentBuilder
This is needed to prevent symbol conflicts with Xamarin.MacDev's
ProcessArgumentBuilder (which is functionally different from
Xamarin.MacDev.Tasks.Core's class of the same name).
* Fixed ILRepack logic for filtering dll's to repack
* Fixed building of Xamarin.iOS.Tasks.Tests now that X.iOS.Tasks.dll contains all symbols
* Updated Makefile now that only 1 iOS Task assembly needs to be distributed
* ILRepack Xamarin.Mac.Tasks as well
* Fixed up *.targets to specify The One Assembly To Rule Them All
* [xharness] Build MSBuild tests with MSBuild.
* Touch the ilrepack stamp file *after* invoking ILRepack, not before.
* Same for Xamarin.Mac.Tasks
The mmp regression tests are already executed in parallel (using `make -j8` in
the Makefile), and they're quite CPU intensitive, which means that if xharness
runs other tests in addition to the these tests, then those other tests may
fail randomly.
However there's a small window between the time we get a pointer
and the call to the native selector where the memory is not fixed.
During this time the GC can move the memory resulting in hard to
diagnose crashes.
Note: `initWithValues:count:` copies the provided memory so what
happens afterward is not an issue.
commit xamarin/maccore@14c3eca3aa
Author: Rolf Bjarne Kvinge <rolf@xamarin.com>
Date: Fri Nov 24 09:53:13 2017 +0100
Bump mac-api-docs.
commit xamarin/mac-api-docs@057bb83848
Author: Timothy Risi <timothy.risi@xamarin.com>
Date: Thu Nov 23 23:52:21 2017 -0900
Docs should also be generated for Xamarin.Mac (#2)
* Docs should also be generated for Xamarin.Mac
* Remove the push target
commit xamarin/maccore@bb3ebb634f
Author: Rolf Bjarne Kvinge <rolf@xamarin.com>
Date: Thu Nov 23 19:06:51 2017 +0100
[tests/external/device-builds] Use BUILD_WORK_HOST instead of BUILD_HOST so that we calculate the right wrench url.
Dispose logs (so that the corresponding files are closed) when we're done with
them, and also don't open a file log by default (usually we just want a
filename to pass to somebody else), but instead open the file if needed.
This should decrease the number of open file descriptors in xharness, which
sometimes become a problem when running many tests.
* [tests] Add support for passing arguments to XM unit tests from the command line.
* [xharness] Get xml results for mac unit tests and parse it to show failures inline in the html report.
The parsing done by `System.Version` does not accept a major-only string,
e.g. providing "11" would throw an exception.
Since people generally refer version as iOS 11 (and not iOS 11.0) this
is, at best, a nuisance. Xcode toolchain accept "11" as a valid string.
The first part of message was updated to show both the option name and
the (user provided) value.
The 2nd part remain the text of the .net exception message, i.e. what
`Version.Parse` tells you when it validates the string. Seeing the input
value should make it more obvious for other, incorrect version strings.
reference:
https://bugzilla.xamarin.com/show_bug.cgi?id=60280
Vincent Dondain
Chris Hamons
Rolf Bjarne Kvinge
Manuel de la Pena
Jeffrey Stedfast
Sebastien Pouliot
Alex Soto
Timothy Risi