Description

Steps to Reproduce:
1.-5. Download the example project + xliff as created by the following steps: https://www.dropbox.com/s/c1x1yy6qpcby2jy/Bugreport-LocalizationExample.zip?dl=0

• OR -

1. Create a new iOS Single View Application

2. Add a localized string to the code, f.e. in the app delegate:

   NSLog("%@", NSLocalizedString("Example", comment: "Example"))

3. Add a localization in the Project settings, f.e. German

4. Select the project and export xliffs using Editor > Export For Localization

5. Edit the resulting de.xliff file and add a translation for the Example text

1. Select the project and choose Editor > Import Localizations - choose to import de.xliff

Xcode crashes only when the Localizable.strings file doesn't exist already when importing - if one is created manually and localized for Base, it works as expected.

Product Version: Xcode-beta (7A165t)
Created: 2015-07-30 23:08:24.014620
Originated: 2015-07-31T01:08:00
Open Radar Link: http://www.openradar.me/22082589

Description

Summary:
Any request for CloudKit will always get CKError - "Internal Error" (1/4000); "Couldn't get a signing certificate". Some of our devices could still connect with CloudKit, some of them couldn't yesterday. But today, all of our devices can't connect with CloudKit and our users also encounter same issues!

Steps to Reproduce:

1. Launch Loopify App (Apple ID: 1003174079)
2. You will find there is always "Disconnect with Loopify"

All the errors are the same, CKError - "Internal Error" (1/4000); "Couldn't get a signing certificate".

Expected Results:
Without error, get the records from Loopify (based on CloudKit).

Actual Results:
Disconnect with CloudKit!

Same error: CKError - "Internal Error" (1/4000); "Couldn't get a signing certificate".

Version:
iOS 8.3 and iOS 8.4

Configuration:
iPhone 4s / 5s / 6 / 6 Plus, iPod Touch 4th and iPad. Testing on WiFi and 3G.

Product Version: 8.4
Created: 2015-07-31 03:48:13.761630
Originated: 2015-07-31T00:00:00
Open Radar Link: http://www.openradar.me/22086571

Description

Summary:
Core Spotlight indices behave as a write-only variable. For example, it is impossible to inspect their contents to find out what CSSearchableItems are currently indexed in the searchable index.

In case the application works with data that may be changed outside of it (e.g. account state is being synced across devices via a third party service) this may be necessary to find out whether the index contains stale data.

Lack of this API forces developers to reimplement (possibly imperfectly) the index as a local database and make sure the Core Spotlight and their own index remain in sync.

Steps to Reproduce:
Investigate the headers of CSSearchableIndex.

Expected Results:
CSSearchableIndex instance should have a public method such as:

• (void)enumerateSearchableItemsUsingBlock:(void (^ nonnull)(NSArray<CSSearchableItem *> nonnull *searchableItems))block;

Such method would call the block with an array containing the CSSearchableItems currently present in the searchable index.

Actual Results:
No such method exists.

Version:
iOS 9 beta 4

Configuration:
iPhone 6 and iOS Simulator

Product Version: iOS 9.0 (13A4305g)
Created: 2015-07-31 20:39:54.330690
Originated: 2015-07-31T00:00:00
Open Radar Link: http://www.openradar.me/22096010

Description

Summary:
When calling a C function marked as NS_INLINE from Swift, the app crashes when built for Release. It doesn’t crash when built for Debug.

For example
NS_INLINE BOOL myInlineFunction(void) { return UIScreen.mainScreen.bounds.size.height == 568;}

Calling that from Swift leads to
2015-07-30 20:20:42.331 NSInlineWhizBang[82768:5412934] *** NSForwarding: warning: selector (0x1045ccbe2) for message 'mainScreen' does not match selector known to Objective C runtime (0x107d5d8f1)-- abort
2015-07-30 20:20:42.331 NSInlineWhizBang[82768:5412934] +[UIScreen mainScreen]: unrecognized selector sent to class 0x105b24310
2015-07-30 20:20:42.337 NSInlineWhizBang[82768:5412934] *** Terminating app due to uncaught exception 'NSInvalidArgumentException', reason: '+[UIScreen mainScreen]: unrecognized selector sent to class 0x105b24310'

Steps to Reproduce:
Build & run the attached project in Release mode

Expected Results:
app runs

Actual Results:
app crashes

Regression:

Notes:
Can’t reproduce with Xcode 7.

Product Version: Xcode 6.4 (6E35b)
Created: 2015-07-30 18:27:13.142560
Originated: 2015-07-30T20:27:00
Open Radar Link: http://www.openradar.me/22076249

Description

Summary:
I want to order HKQuantitySample by the Quantity when execute HKSampleQuery.
I want something similar as HKPredicateKeyPathQuantity key

The problem:
I want to fetch only few items, set a limit with the highest value, but I can't sort them by value.
To make it work now I need to fetch all the items to the memory and sort them manually.

Code Example:
let type = HKObjectType.quantityTypeForIdentifier(HKQuantityTypeIdentifierHeight)
let sort = [
NSSortDescriptor(key: HKSampleSortIdentifierStartDate, ascending: true),
NSSortDescriptor(key: HKSampleSortIdentifierEndDate, ascending: true),

//I want one more Sort Item here :(
//NSSortDescriptor(key: HKSampleSortIdentifierQuantity, ascending: true)
]

let query = HKSampleQuery(sampleType: type, predicate: nil, limit: 2, sortDescriptors: sort) { query, res, error in
completion(res, error)
}
healthKit.executeQuery(query)

Solution:
Add HKSampleSortIdentifierQuantity

Product Version: Xcode 6.4 (6E35b)
Created: 2015-07-30 08:32:51.811740
Originated: 2015-07-30T10:33:00
Open Radar Link: http://www.openradar.me/22068394

Description

Summary:
If two views are items of a UICollisionBehavior and at the same time of a UIAttachmentBehavior, the UICollisionBehavior does not work between those two views.

Steps to Reproduce:

1. Open the attached Playground
2. Show Assistant Editor

Expected Results:
Both pairs of views should react the same.

Actual Results:
The pair with the UIAttachmentBehavior don't collide with each other.

Version:
Xcode Version 7.0 beta 4 (7A165t)

Notes:

Configuration:
MacBook Air 2015

Attachments:
'AttachmentBehaviorBugPlayground.playground.zip' was successfully uploaded.

Product Version: Xcode Version 7.0 beta 4 (7A165t)
Created: 2015-07-30 20:17:09.573760
Originated: 2015-07-30T22:14:00
Open Radar Link: http://www.openradar.me/22078638

Description

Summary:
Attempting to install an enterprise build over the top of an App Store build does not work. Phone console throws an error.

Steps to Reproduce:
Install an app store app and then make a build of the same app using an enterprise profile. Upload the profile to a server and configure a .plist so that the phone can visit and download. When visiting the file to download the console just throws

Jul 31 09:16:44 iPod-Touch-iOS-8-MC itunesstored[107] : ExternalDownloadManifest: Skipping download and install of: org.redcross.gdpcnewzealand

The phone offers no option to install and the requested app is not installed.

Expected Results:
App should install

Actual Results:
User is given no option to install and error is thrown in console

Regression:
This is happening on both and iPhone 6 and iPod Touch running iOS 8.4

Notes:
Installing IPA via Xcode devices window works as expected

Product Version:
Created: 2015-07-31 08:23:33.119170
Originated: 2015-07-31T09:23:00
Open Radar Link: http://www.openradar.me/22088607

Description

Summary:
When an accessibility element containing “purchases” is read with VoiceOver, it sounds like “per-cases”.

Steps to Reproduce:

• go to http://www.komendera.com/radars/accessibility-purchases.html
• select “purchases” while VoiceOver is running

Expected Results:
pronounced correctly

Actual Results:
pronounced like “per-cases”.

Regression:

Notes:
“purchase” is pronounced correctly

Product Version: iOS 8.4
Created: 2015-07-31 07:53:40.467210
Originated: 2015-07-31T09:51:00
Open Radar Link: http://www.openradar.me/22088316

Description

Summary:
CIDetector’s face detection is not accurate. In a picture containing 7 faces — 4 of which are quite vertical and facing front — it detects 0. It needs much improvement.

Steps to Reproduce:
Run sample app and read the console logs. It has nothing because none faces are detected.

Expected Results:
At least it should detect the 4th and 7th faces.

Actual Results:
It detects none.

Sample Code:
http://cl.ly/2N2i2x3E3127

Product Version: 9.0 beta 4
Created: 2015-08-01 01:59:40.453240
Originated: 2015-07-31T20:59:00
Open Radar Link: http://www.openradar.me/22102128

Description

Summary:
Customizing the animation of the push-navigations of a UINavigationController by implementing the UIViewControllerAnimatedTransitioning-delegate no longer works as in iOS 8.4.
The animation still works as expected, but when the iPad gets rotated when the second VC is visible, and then popped back to the previous one, the size of the view is not getting updated.

Steps to Reproduce:

1. Start the attached project on an iOS9-device / the simulator
2. Tap the “Tap me”-button
3. Rotate the device
4. Tap on “< Back”
5. One part of the screen should now be black, as the white-main-view didn’t get resized. There is an example-screenshot inside the project-folder.

You can redo those steps on an iOS 8.3 device / simulator and should be able to see the difference.

Expected Results:
The size of the first VC should be correct after tapping the back-button.

Actual Results:
The view did not get resized.

Regression:
In at least iOS 8.3 and iOS 8.4 it works as expected.

Notes:
Provide additional information, such as references to related problems, workarounds and relevant attachments.

Product Version: iOS 9b4
Created: 2015-07-31 11:27:23.333410
Originated: 2015-07-31T13:27:00
Open Radar Link: http://www.openradar.me/22090161

Description

Summary:
When providing the name of a code signing identity to use to codesign with there is no option or flag to ignore expired certificates. For example, I had an expired certificate with the same name as the renewed certificate and code signing throws an error because it finds two certificates with the same name. If the ignore expired option existed this command would complete as it would just use the new identity.

Steps to Reproduce:
Have two identities in the keychain with the same name but have one expired. Attempt to coding with a command such as

codesign -f -s "iPhone Distribution: 3 Sided Cube Design Ltd (25H7BM6YWK)" --entitlements /Users/matthewcheetham/Desktop/Payload/ARC\ Flood.app/archived-expanded-entitlements.xcent /Users/matthewcheetham/Desktop/Payload/ARC\ Flood.app/ARC\ Flood

See the error;

Distribution: 3 Sided Cube Design Ltd: ambiguous (matches "iPhone Distribution: 3 Sided Cube Design Ltd" and "iPhone Distribution: 3 Sided Cube Design Ltd" in /Users/matthewcheetham/Library/Keychains/login.keychain)

Expected Results:
Codesign should have a flag to ignore expired and succeed with signing.

Actual Results:
Codesigning fails.

Product Version: 10.10.4
Created: 2015-07-31 14:07:55.844290
Originated: 2015-07-31T15:07:00
Open Radar Link: http://www.openradar.me/22091226

Description

Summary:
When reading an article in News app, having tapped through to an article, the user cannot get back to the main screen, and is forced to force-quit the app.

Steps to Reproduce:
Open News app
Select an article from for you
Read the article and scroll down
Tap on screen
Strange partial appearance of keyboard at bottom
Lack of interface

Expected Results:
Normal interface should appear

Actual Results:
Partial appearance of keyboard

Version:
iOS 9 13A4305G

Notes:

Configuration:
IPad Air 1

Attachments:
'image.png' was successfully uploaded.

Product Version: 9 13a4305G
Created: 2015-08-01 07:07:55.848410
Originated: 2015-01-08T00:00:00
Open Radar Link: http://www.openradar.me/22103156

Description

Summary:
Spaces behaves incorrectly in 10.11b4 and b5. For example:

• Spaces behaves as if “When switching to an application, Switch to a Space with open windows for the application” was on, even though it isn’t.
• Two spaces: App A is active on Space 1, App B has an open window on Space 1 and Space 2. Switching from space 1 to space 2 will activate app B, and then incorrectly orders front the window for app B which is on space 1, forcing you back to Space A.

Steps to Reproduce:

1. Disable “When switching to an application, Switch to a Space with open windows for the application”
2. Open Safari. Open a window on Space 1, and another one on Space 2.
3. Activate Finder by clicking on the desktop.
4. Swipe right to go to space 2.

Expected Results:
5. The Safari window on Space 2 becomes key

Actual Results:
5. The Safari window on Space 1 becomes key, and transports me back to Space 1.

Regression:
This was introduced in beta 4. Beta 3 did not behave like this.

Notes:
Here is a video recording of this behavior:
https://dl.dropboxusercontent.com/u/6775/Spaces%20bug%2010.11b4-5.m4v

Product Version: 10.11 Beta 5 (15A235d)
Created: 2015-07-31 04:59:37.189510
Originated: 2015-07-30T21:59:00
Open Radar Link: http://www.openradar.me/22086967

Description

Summary:
When dragging a window in OS X El Capitan, clicking on the window moves application frame unexpectedly and sometimes offscreen, where the window can no longer be manipulated or moved.

Steps to Reproduce:

1. Using a MacBook Pro with El Capitan installed, connect a Thunderbolt Display.
2. On the MacBook display, move a window to the center of the screen.
3. Click on the application frame.

Expected Results:
The window does not move until the mouse cursor drags the window to a new location.

Actual Results:
The window moves up and to the left. See attachment.

Product Version: 10.11 Beta (15A235d)
Created: 2015-08-01 22:03:32.463030
Originated: 2015-08-01T17:01:00
Open Radar Link: http://www.openradar.me/22105831

Description

Summary:
Functions have a curried syntax allowing convenient definition of functions returning functions:

func add(x: Int)(_ y: Int) -> Int {
return x + y
}

The same syntax would be useful for defining initializers:

struct Add {
init(_ x: Int)(_ y: Int) {
self.x = x
self.y = y
}
let x: Int
let y: Int
}

The above initializer would be equivalent to a static function:

struct Add {
static func init_(x: Int)(_ y: Int) -> Add {
return Add(x: x, y: y)
}
let x: Int
let y: Int
}

For example it can be partially applied, returning a function which initializes new instances of the type:

let incremented = [0, 1, 2, 3, 4].map(Add(1)) // equivalently: .map(Add.init(1))

This would also allow failable initialization in the usual way:

struct Pair<A, B> {
init?(x: A?)(y: B?) {
guard let x = x else { return nil }
guard let y = y else { return nil }
left = x
right = y
}
let left: A
let right: B
}

This would yield an initializer of type A? -> B? -> Pair<A, B>?, i.e. the semantics are again identical to a curried static function defined over an uncurried (failable) initializer:

struct Pair<A, B> {
static func init_(x: A?)(y: B?) -> Pair {
return Pair(x: x, y: y)
}

init?(x: A?, y: B?) {
    guard let x = x else { return nil }
    guard let y = y else { return nil }
    left = x
    right = y
}
let left: A
let right: B

}

This syntax would enable more idiomatic curried construction of Swift types than is currently possible. For example, my motivating case is the family of loc functions defined here:

https://github.com/antitypical/Manifold/blob/289e8444349c9d3bf7d02cefde7db08966134f69/Manifold/Location.swift#L46-L79

Taking the first function as an example, we could define it in extant Swift 2 syntax as a failable initializer, something like so (🚨 untested 🚨):

init?(_ weave: (A -> Location?) -> A -> Location?, _ up: A -> Location?, _ a: A) {
func into(t1: A) -> Location? {
return Location(it: t1, down: weave(into), up: up, left: const(nil), right: const(nil))
}
self = into(a)
}

However, in doing so we would lose the ability to partially apply it, which is depended upon elsewhere in the codebase.

Steps to Reproduce:
N/A

Expected Results:
N/A

Actual Results:
N/A

Regression:
N/A

Notes:
N/A

Product Version: Xcode-beta (7A165t)
Created: 2015-08-02 20:07:37.152060
Originated: 2015-08-02T16:07:00
Open Radar Link: http://www.openradar.me/22108354

Description

For quick cave man's debugging sessions with NSLog() it would be nice if messages in the debugger output of Xcode would somehow (e.g. by a different color) be marked when they were emitted from the main thread. Thread number "303" used to be a strong indicator for being the main thread, but that stopped working some time ago:
2014-02-06 16:18:26.309 MyApp[1286:303] Message from NSLog called on main thread

Product Version:
Created: 2015-08-02 11:20:17.654480
Originated: 2015-08-02T00:00:00
Open Radar Link: http://www.openradar.me/22107374

Description

Summary:
The battery status when having the MacBook Pro running on battery always shows "20:00 Remaining” and doesn’t update.

Steps to Reproduce:

• unplug Mac
• click the battery icon in the menu bar

Expected Results:
Accurate remaining time on battery shown

Actual Results:
20 hours remaining shown

Regression:
This was more accurate with 10.10

Notes:
Maybe you folks did magic and I really do have 20 hours on my battery? If so, Elon Musk would like to talk to you.

Product Version: Mac OS X 10.11 (15A235d)
Created: 2015-08-01 20:56:24.323720
Originated: 2015-08-01T22:56:00
Open Radar Link: http://www.openradar.me/22105625

Description

Summary:
If you want to use CoreText in watchOS (for instance, to render small-caps text, as I do in the attached sample project), you can’t use it directly, as you can’t import CoreText.

Steps to Reproduce:

1. In a watchOS 2 project, import CoreText using:
   Swift: import CoreText
   Objective-C: @import CoreText;
2. Build

Expected Results:
The app builds and you can use constants defined in CoreText headers.

Actual Results:
The compiler throws an error that it can't find CoreText.

Version:
Xcode Version 7.0 beta 4 (7A165t)

Notes:
According to this URL, as of Beta 4, CoreText is in WatchKit:

https://developer.apple.com/library/prerelease/watchos/releasenotes/General/watchOS2SeedAPIDiffs/frameworks/CoreText.html

Since there’s no <CoreText/CoreText.h> to be found, I tried this:

import <CoreText/SFNTLayoutTypes.h>

since that header was explicitly added in the diff, but it still wouldn’t work.

Product Version: Xcode Version 7.0 beta 4 (7A165t)
Created: 2015-08-02 03:09:49.471790
Originated: 2015-08-01T23:08:00
Open Radar Link: http://www.openradar.me/22106516

Description

Summary:
iPad: Using UIPrintInteractionController's delegate methods to determine when the lifecycle of the controller is ending, an app using the print controller is unable to determine if the popover was dismissed by invoking the "Print" button OR if the popover is dismissed because the user tapped elsewhere in the parent view.

• (void)printInteractionControllerWillDismissPrinterOptions:(UIPrintInteractionController *)printInteractionController;
• (void)printInteractionControllerDidDismissPrinterOptions:(UIPrintInteractionController *)printInteractionController;

both return the printInteractionController, but there are no properties of that object available to say how the dismissal took place.

Steps to Reproduce:
Use UIPrintInteractionController and handle the delegate methods

• (void)printInteractionControllerWillDismissPrinterOptions:(UIPrintInteractionController *)printInteractionController;
• (void)printInteractionControllerDidDismissPrinterOptions:(UIPrintInteractionController *)printInteractionController;

In either case, the properties of printInteractionController are equivalent no matter how the PIC was dismissed (print button OR tap outside popover).

Expected Results:
Either:

• delegate methods return another parameter "willPrint = YES/NO"
• additional delegate methods to describe the printing lifecycle, like

"...didDismissWithPrintButton"

Actual Results:
A user of the PrintInteractionController doesn't know how the controller was dismissed.

Version:
iOS 7.x, iOS 8.x

Notes:

Configuration:
iPad (any) iPad Simulator (any)

Attachments:

Product Version: 8.x
Created: 2015-07-31 18:55:29.952870
Originated: 2015-07-31T00:00:00
Open Radar Link: http://www.openradar.me/22094417

Description

Summary:
When in the middle of an Apple Music playlist, choosing to add another playlist to Up Next has them playing next, instead of added to the end

Steps to Reproduce:

1. Start a “For You” Apple Music curated playlist (“Playlist A”)
2. Let several songs play
3. Navigate to another curated Apple Music playlist (“Playlist B”)
4. Tap the “…” button for the whole playlist
5. Tap “Add to Up Next”
6. Tap on the mini player
7. Tap on the Up Next button

Expected Results:
The songs get added to the end of the Up Next queue, so all of the songs from B are after the remaining songs from A

Actual Results:
After step 5, the HUD’s image shows the arrow pointing to the bottom of the stack, but then B’s songs are listed as “Up Next”, followed by “Resume: Playlist A” and the rest of A’s songs

Product Version: 8.4
Created: 2015-07-30 21:57:17.084140
Originated: 2015-07-30T17:57:00
Open Radar Link: http://www.openradar.me/22081001

Description

Summary:
the following code crashes the compiler. I am not sure that the generic generator causes the crash, but it didn’t begin crashing until I implemented it.

Steps to Reproduce:
open the provided file in a swift project.

Expected Results:
Successful build or explicit errors

Actual Results:
Command-b hangs and source kit intermittently crashes.

Product Version: Xcode-beta (7A165t)
Created: 2015-08-02 12:11:38.858900
Originated: 2015-08-02T17:41:00
Open Radar Link: http://www.openradar.me/22107434

Description

Summary:
The source included below causes the swift compiler to trap.

Steps to Reproduce:
enum Stack {
typealias Element = T
case Empty
indirect case Cons(Element, Stack)
}

extension Stack {
    var isEmpty: Bool {
        switch self {
        case .Empty:
            return true
        case .Cons:
            return false
        }
    }

    var head:Element? {
        if case let .Cons(theHead, _) = self {
            return theHead
        } else {
            return nil
        }
    }

    var tail:Stack? {
        if case let .Cons(_, theTail) = self {
            return theTail
        } else {
            return nil
        }
    }
}

Expected Results:
Either successful compilation or errors indicating the problem

Actual Results:
Command failed due to signal: Abort trap: 6

Product Version: Xcode-beta (7A165t)
Created: 2015-08-01 10:52:23.258840
Originated: 2015-08-01T16:22:00
Open Radar Link: http://www.openradar.me/22103992

Description

Summary:
I have an empty project with many playgrounds files
I've added a Application an iOS Application target. After that I see the error

Unable to create target for stub executable. unable to find executable for '/var/folders/99/l2t7jb415pzds1sgb03t47jc0000gn/T/com.apple.dt.Xcode.pg/applications/TOC-38535-240.app/TOC'

Steps to Reproduce:

• Create empty project.
• Add a playground with few pages.
• Add am iOS application target.
• Try opening playgrounds.

Result, an error.

Product Version: Xcode-beta (7A165t)
Created: 2015-08-02T13:18:00.558020
Originated: 2015-08-02T15:17:00
Open Radar Link: http://www.openradar.me/22107493

Description

Summary:
We’re using SecKeyEncrypt, SecKeyRawVerify and SecKeyRawVerify/SecKeyRawSign on both iOS and OS X. They are all declared to be available on OS X starting with 10.7 - yet I can’t find them in any of the Mac headers.

OSStatus SecKeyEncrypt(
SecKeyRef key,
SecPadding padding,
const uint8_t *plainText,
size_t plainTextLen,
uint8_t *cipherText,
size_t *cipherTextLen)
__OSX_AVAILABLE_STARTING(__MAC_10_7, __IPHONE_2_0);

OSStatus SecKeyRawVerify(
SecKeyRef key,
SecPadding padding,
const uint8_t *signedData,
size_t signedDataLen,
const uint8_t *sig,
size_t sigLen)
__OSX_AVAILABLE_STARTING(__MAC_10_7, __IPHONE_2_0);

OSStatus SecKeyRawSign(
SecKeyRef key,
SecPadding padding,
const uint8_t *dataToSign,
size_t dataToSignLen,
uint8_t *sig,
size_t *sigLen)
__OSX_AVAILABLE_STARTING(__MAC_10_7, __IPHONE_2_0);

Right now we just manually declare them as we assume this is an oversight int he headers. If this is not the case, then it’s a bug in the iOS documentation. I checked with 8.4 and 9.0b4 - both have the same headers declaring this as available since 10.7. It also works; we use it for (internal) code on both iOS and the Mac.

Product Version: 10.10
Created: 2015-07-30 07:36:57.168270
Originated: 2015-07-30T09:36:00
Open Radar Link: http://www.openradar.me/22067843

Description

Summary:
Many times on long car trips, I like to select a playlist from the For You section. CarPlay does not allow me to scroll this so I have to use my phone while driving which is unsafe.

Steps to Reproduce:

1. Decide to find a For You Playlist
2. Notice that CarPlay does not let you browse ForYou Playlists

Expected Results:

1. Decide to find a For You Playlist
2. Click For You in CarPlay
3. See a scrollable quick list of For You Stations

Actual Results:

1. Decide to find a For You Playlist
2. Notice that CarPlay does not let you browse ForYou Playlists

Regression:
none

Notes:
none

Product Version: iOS 8.4, iOS 9
Created: 2015-08-02 18:37:39.438660
Originated: 2015-08-02T11:37:00
Open Radar Link: http://www.openradar.me/22108127

Description

Summary:

On OS X 10.10.x and later, disabling Gatekeeper does not mean it is permanently off. After a set amount of time (currently 30 days), Gatekeeper will automatically re-enable itself with the "Allow apps downloaded from: Mac App Store and identified developers" setting.

After doing some research, it looks like Gatekeeper’s automatic re-enablement function can be disabled by running the following command with root privileges:

defaults write /Library/Preferences/com.apple.security GKAutoRearm -bool false

This would allow Gatekeeper to be set to "Allow apps downloaded from: Anywhere" and have it stay that way.

However, it does not appear that I can manage this with a profile based on the CFPreferencesCopyValue variable being used.

https://github.com/aosm/security_systemkeychain/blob/master/syspolicyd/syspolicyd.cpp#L298

Instead, it looks like CFPreferencesCopyAppValue would need to be used. (see attached screenshot).

Steps to Reproduce:

1. Install profile
2. Check value for com.apple.security GKAutoRearm

Expected Results:

com.apple.security GKAutoRearm is set to False

Actual Results:

com.apple.security GKAutoRearm is set to True

Regression:

Running the defaults command listed above sets the following value:

GKAutoRearm is set to False

Notes:

I have more information on this issue available here:

https://derflounder.wordpress.com/2015/07/31/gatekeeper-automatically-re-enables-after-30-days-on-yosemite-and-later/

I have a sample management profile available here:

https://github.com/rtrouton/profiles/tree/master/DisableGatekeeperAutomaticReenablement

Product Version: Mac OS X 10.10.4 (14E46)
Created: 2015-07-31 18:43:14.732970
Originated: 2015-07-31T14:43:00
Open Radar Link: http://www.openradar.me/22094327

Description

Summary:
The documentation for [NSTabViewController tabStyle] lists four possible enum values, the last one being NSTabViewControllerTabStyleUnspecified which the description ‘A style that indicates the the tab view controller does not provide the tab selection UI. Your app provides the control (such as an NSSegmentedControl or NSPopUpButton) for navigating between tabs. You can bind an existing control to the tab view controller object so that interactions with the control automatically change tabs.’

When I change this to the Unspecified style in code IB the end result still displays the NSSegmentedControl in a similar way to .TabsOnTop but not exactly the same.

If I change the tabStyle programmatically in viewDidLoad then this bug is not exhibited

Steps to Reproduce:
1.) Add a new NSTabViewController to your layout in a storyboard.
2.) Change the Tab View Controller ‘Style’ property to ‘Unspecified’.
3.) Run the app.

Expected Results:
At this point you should not see the NSSegmented Control.

Actual Results:
You do see the NSSegmentedControl.

Notes:
The provided attachment is a very simple project demonstrating this problem. If you uncomment the line in ‘TestTabViewController’ you will see it work as I would expect from IB.

Product Version: 10.10.10
Created: 2015-07-31 17:30:16.437680
Originated: 2015-07-31T00:00:00
Open Radar Link: http://www.openradar.me/22093439

Description

This morning, I was late for a meeting and forgot to unlock my watch after putting it on my wrist.

After the meeting, I had breakfast and started working.

At lunch, I looked at my wrist and realized that the watch was locked. I also realized that I hadn’t gotten any notifications for almost five hours; including reminders to stand at 50 minutes past the hour.

That one moment when waking up led to me sitting on my ass for far too much time. Luckily, I didn’t have any additional appointments in the morning or I would have missed those, too.

The Apple Watch knows that it’s locked. It also knows that I’m wearing it. Why can’t it remind me that it’s locked and not presenting notifications?

Product Version: 1.0
Created: 2015-07-31 20:26:55.093710
Originated: 2015-07-31T13:26:00
Open Radar Link: http://www.openradar.me/22095844

Description

Attempting to use libtool to build an iOS library generates the following error:

 building for OSX, but linking in object file built for iOS, for architecture i386

The ld “linker” command has both -macosx_version_min and -ios_version_min flags, attempting to pass the latter of these generates a different error:

$ libtool -dynamic -ios_version_min 8.0 …

error: /Applications/Xcode7-beta4.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/libtool: can't open file: 8.0 (No such file or directory)

Product Version: Xcode7-beta4 (7A165t)
Created: 2015-08-01 16:02:45.783670
Originated: 2015-08-01T10:02:00
Open Radar Link: http://www.openradar.me/22104678

Description

Summary:
Attempts to create new messages in Mail via its AppleScript scripting bridge fail.

Steps to Reproduce:

1. Create a new Automator workflow
2. Add a single action from the Mail group: New Mail Message
3. Run the workflow

Expected Results:
Mail.app should become active and create a new mail message window.

Actual Results:
Mail.app does not respond at all, and Automator reports an error with the workflow: “-[SBProxyByClass setSender:]: object has not been added to a container yet; selector not recognized”

Notes:

Mail.app also fails to respond if an actual AppleScript is used, such as this one:

tell application "Mail"
set t to make new outgoing message
set visible of t to true
activate
end tell

Product Version: Mail 9.0 (3073)
Created: 2015-07-31 15:37:46.173570
Originated: 2015-07-31T11:37:00
Open Radar Link: http://www.openradar.me/22091989

Description

Summary:
The Xcode 7 Beta 4 Release notes state the following known issue:

“Code Coverage
• Xcode does not show code coverage information for source files in static libraries.
Workaround: Add the source files directly to application or framework targets. (15605406)”

I have code and tests within frameworks, and their code coverage is not properly reported in the Report Navigator tab. When viewing the code coverage results in the actual source files, I can verify nearly the entire source file is covered, but in the Report Navigator it shows up as zero percent.

The workaround listed in the release notes does not appear to be accurate, and I currently do not get accurate coverage results for my framework targets.

Product Version: Xcode 7 Beta 4
Created: 2015-07-31 16:08:58.086160
Originated: 2015-07-31T11:08:00
Open Radar Link: http://www.openradar.me/22092373

Description

Summary:
Xcode 7 Beta 4: Interface Builder crashes if the root view has translatesAutoresizingMaskIntoConstraints="NO" property. It does not crash when there is no translatesAutoresizingMaskIntoConstraints property in the root level view tag.

Steps to Reproduce:
In Xcode 7 Beta 4, open View1.xib (which has translatesAutoresizingMaskIntoConstraints="NO” in its root level view tag) in the attached project.

Bonus - Steps to generate such xib file using Xcode 6.4:

1. In a storyboard scene, create a fully constrained view
2. Select the view and copy it to clipboard
3. Create a new xib file, delete all existing views, and paste the view.
4. Examine the xib file using a file editor and verify that translatesAutoresizingMaskIntoConstraints=“NO” is in the root level view tag.
5. Save and open this xib file in Xcode 7 Beta 4. Xcode will crash.

If you try the above steps in Xcode 7 Beta 4, it will crash after step 3.

Expected Results:
Xcode/ibtool should not crash

Actual Results:
But totally does

Regression:
Not happening in Xcode 6.4 or Xcode 7 beta 3

Notes:
Workaround: manually remove translatesAutoresizingMaskIntoConstraints=“NO” in the root level view tag from the xib file.
Project can also be downloaded from here: https://www.dropbox.com/s/cpl15wwpcoo8jxu/IBCrasher.zip?dl=0

Product Version: Xcode 7 Beta 4
Created: 2015-07-30T09:42:35.695500
Originated: 2015-07-30T02:42:00
Open Radar Link: http://www.openradar.me/22069117

Description

Summary:
On iOS 8, it was possible to layout a UITableView in code so that it would fill the screen, including under a UINavigatonBar. Unless you set the automaticallyAdjustsScrollViewInsets property of the UIViewController otherwise, the insets on the tableview were changed so that content would not appear below the navigation bar until the tableview was scrolled up.

In iOS 9, that functionality still appears to work when using storyboards, but it no longer works when adding the table view as a subview and applying constraints to it in code.

Steps to Reproduce:

1. Run the enclosed project on an iOS 8 device. Hit the plus sign to add content to the table view. Note that the content appears immediately below the navigation bar.
2. Run the same enclosed project on an iOS 9 beta 4 device. Hit the plus sign to add content to the table view. Note that the content is hidden by the navigation bar.

Expected Results:
I would expect the behavior to be the same on both iOS 8 and iOS 9

Actual Results:
On iOS 9, the content in the tableview are not offset by the height of the navigation bar.

Version:
iOS 9.0 (13A4305g)

Notes:
I left a second storyboard with the TableView laid out there so it is easy to see that it works on iOS 9.

Configuration:
iPod touch, 5th generation running iOS 9; verified correct behavior on an iPhone 5s running iOS 8.4

Attachments:
'TableViewTest.zip', 'iOS 8.4 - visible content.png' and 'iOS 9.0 beta 4 - content hidden.png' were successfully uploaded.

Product Version: iOS 9 beta 4
Created: 2015-08-02 03:32:35.060200
Originated: 2015-08-01T00:00:00
Open Radar Link: http://www.openradar.me/22106545

Description

Summary:
When querying for a secure field in a UI test, the test always fails because it finds multiple versions with the same label, even when there is only one field.

Product Version: Xcode-beta (7A165t)
Created: 2015-07-31 22:19:21.485660
Originated: 2015-07-31T15:19:00
Open Radar Link: http://www.openradar.me/22098492

Description

Summary:
Provide a descriptive summary of the issue.

Steps to Reproduce:

1. Viewing “My Music”, “Artists” view/group/sort
2. Select “Compilations” from left source pane.
3. Scroll through the compilation albums.

Observe:

Expected Results:
Album cells containing track listings are scrolled into view with consistent vertical spacing from each other.

Actual Results:
While first few albums are laid out fairly close to each other, eventually the albums become separated vertically by whitespace. It appears as though the cell sizing might be getting calculated from all the tracks laid out in one column, but then displayed in more than one. (Four columns in my fullscreen case, but observed with two columns in smaller window.)

I have not observed this in any of the other selections (Recently Added, All Artists, individual artists), only Compilations.

Product Version: iTunes 12.2.1 (12.2.1.16)
Created: 2015-07-30 21:39:50.421020
Originated: 2015-07-30T14:39:00
Open Radar Link: http://www.openradar.me/22080544

Description

Summary:
Table views instantiated from XIB files and Storyboards compiled using Interface Builder in Xcode 7 beta 4 (7A165t) have white (default) background at runtime, regardless of background color set for them in Interface Builder. This issue happens when running any iOS version on device/simulator.

Steps to Reproduce:

1. Create an empty app with a view controller that manages a view with a table view, using a XIB file or Storyboard to construct the interface.
2. Set table view color to red in Interface Builder, without changing anything in code.
3. Run the app.

Expected Results:
Table view has a red background (apart from the area covered by cells).

Actual Results:
Table view has a white (default) background color.

Version:
Xcode 7 beta 4 (7A165t) running OS X 10.10.4 (14E46)

Notes:
This is a regression from previous Xcode versions. I have no information regarding whether it appeared in Xcode 7 beta 4 or in earlier seeds. A workaround for this issue is to set the background color manually in code, but that's kinda missing the point of Interface Builder.
The attached project demonstrates the issue.

Configuration:
Any device or simulator.

Attachments:
'TableBackgroundColorDemo.zip' was successfully uploaded.

Product Version: Xcode 7 beta 4 (7A165t) running OS X 10.10.4 (14E46)
Created: 2015-08-02 09:57:59.807740
Originated: 2015-08-02T00:00:00
Open Radar Link: http://www.openradar.me/22107245

Description

Summary:
In rdar://problem/22108390, I described allowing self to be assigned a value of Optional type in failable value type initializers, automatically returning nil if self was assigned nil.

I personally think that would have better semantics than the semantics of return in a failable initializer—it’s inconsistent and confusing that you can return only nil but can assign only non-nil—but return is already established, and assignment to self isn’t a thing for classes anyway, so one might argue that this conflates things unnecessarily.

Therefore I also propose allowing if let and guard let to assign to self in failable value type initializers, which would allow me to replace this:

init?(_ weave: (A -> Location?) -> A -> Location?, _ up: A -> Location?, _ a: A) {
func into(t1: A) -> Location? {
return Location(it: t1, down: weave(into), up: up, left: const(nil), right: const(nil))
}
guard let location = into(a) else { return nil }
self = location
}

with the briefer, more idiomatic:

init?(_ weave: (A -> Location?) -> A -> Location?, _ up: A -> Location?, _ a: A) {
func into(t1: A) -> Location? {
return Location(it: t1, down: weave(into), up: up, left: const(nil), right: const(nil))
}
guard let self = into(a) else { return nil }
}

Steps to Reproduce:
N/A

Expected Results:
N/A

Actual Results:
N/A

Regression:
N/A

Notes:
If you try this today, you get unhelpful error messages:

/Users/rob/Developer/Projects/Manifold/Manifold/Location.swift:64:9: error: pattern matching in a condition requires the 'case' keyword
guard let self = into(a) else { return nil }
^
case
/Users/rob/Developer/Projects/Manifold/Manifold/Location.swift:64:13: error: binary operator '~=' cannot be applied to operands of type 'Location' and 'Location?'
guard let self = into(a) else { return nil }
^~~~

Taking its advice and using guard case let self = into(a), you get:

/Users/rob/Developer/Projects/Manifold/Manifold/Location.swift:64:14: warning: 'let' pattern has no effect; sub-pattern didn't bind any variables
guard case let self = into(a) else { return nil }
^~~~~~~~

/Users/rob/Developer/Projects/Manifold/Manifold/Location.swift:64:18: error: binary operator '~=' cannot be applied to operands of type 'Location' and 'Location?'
guard case let self = into(a) else { return nil }
^~~~

So you have to bind a pointless temporary either way.

Product Version: Xcode-beta (7A165t)
Created: 2015-08-02 20:26:59.435050
Originated: 2015-08-02T16:26:00
Open Radar Link: http://www.openradar.me/22108400

Description

Take a look at this page in Safari:

http://www.digitaltrends.com/computing/you-can-now-download-the-windows-10-iso-disk-image/

It autoplays a video with sound, but neither the tab nor the header show the “producing audio” icon.

Product Version: Safari 10601.1.39.0.2
Created: 2015-07-31 00:32:50.953320
Originated: 2015-07-30T17:32:00
Open Radar Link: http://www.openradar.me/22084554

Description

Summary:
While trying to see if my mail is still syncing, I noticed that the “Mail Activity” pane on the sidebar of mail displayed no activity, while the “Activity” window shows syncing still happening.

Steps to Reproduce:

1. Start syncing your mail.
2. Open the “Mail Activity” sidebar pane.
3. Open Mail’s “Activity” window.

Expected Results:
Both areas display the same data.

Actual Results:
Not all data is displayed on the “Mail Activity” pane.

Regression:
Unknown

Notes:
See screenshot.

Product Version: Mail 8.2 (2102)
Created: 2015-07-30 14:18:22.550760
Originated: 2015-07-30T10:18:00
Open Radar Link: http://www.openradar.me/22071807

Description

Summary:
Swift segfaults. That’s bad.

Steps to Reproduce:
class A { var o: String? { return "foo" } }
class B : A { override var o: String { return "bar" } }
B().o

Expected Results:
Shouldn’t crash.

Actual Results:
But totally does.

Product Version: Xcode-7-beta.4 (7A165t)
Created: 2015-07-31 00:05:29.039590
Originated: 2015-07-30T17:05:00
Open Radar Link: http://www.openradar.me/22084083

Description

Summary:
Password entries in Terminal now has a keychain icon to indicate that the password field will not display any password. However, since not displaying the password is common knowledge, the icon can be used instead to select a keychain autofill. This would actually be a very useful feature if implemented (for SSH, FTP, etc) but will need to be heavily regulated in case unwanted access to terminal was obtained.

Steps to Reproduce:
N/A

Expected Results:
N/A

Actual Results:
N/A

Version:
10.11 Beta (15A235d)

Notes:

Configuration:
MacBook Pro (Retina, 15-inch, Early 2013)

Attachments:

Product Version: 10.11 (15A235d)
Created: 2015-08-01 03:01:22.118750
Originated: 2015-08-01T10:59:00
Open Radar Link: http://www.openradar.me/22102418

Description

when I try to upload my project which is built with iOS 9 deployment target and uses watchOS 2 in the WatchKit extension, I get a bunch of errors from iTunes Connect - see attached file.

Steps to Reproduce:
archive project
upload to iTunes Connect

Expected Results:
TestFlight should allow distributing apps with watch OS 2 - please :)

Actual Results:
get the attached error messages, such as

• ITMS-90086 - missins 64 bit support
• unsupported architectures
• invalid code signing
• invalid info.plist value
• binary is invalid, not built with Apple's linker

Version:
iOS 9 beta 4, watchOS 2.0 beta 4

Product Version: iOS 9 beta 4, watchOS 2.0 beta 4
Created: 2015-08-01 09:54:50.721460
Originated: 2015-08-01T00:00:00
Open Radar Link: http://www.openradar.me/22103812

Description

Summary:
In rdar://problem/22108354 I wrote a (pseudoSwift) example of rewriting a static function into a failable initializer:

init?(_ weave: (A -> Location?) -> A -> Location?, _ up: A -> Location?, _ a: A) {
func into(t1: A) -> Location? {
return Location(it: t1, down: weave(into), up: up, left: const(nil), right: const(nil))
}
self = into(a)
}

Unfortunately this does not compile, because the assignment of an Optional type to self is illegal. Instead, one has to write:

init?(_ weave: (A -> Location?) -> A -> Location?, _ up: A -> Location?, _ a: A) {
func into(t1: A) -> Location? {
return Location(it: t1, down: weave(into), up: up, left: const(nil), right: const(nil))
}
if let location = into(a) {
self = location
} else {
return nil
}
// note: one can use an equivalent guard clause:
// guard let location = into(a) else { return nil }
// self = location
}

While this has the virtue of actually compiling today, it’s a lot of extra syntax to do essentially the same thing: return nil if the assigned value was nil, or else assign the (non-nil) value.

Steps to Reproduce:
N/A

Expected Results:
N/A

Actual Results:
N/A

Regression:
N/A

Notes:
This would presumably only apply to value types since if I recall correctly assignment to self is only legal in value type initializers.

Product Version: Xcode-beta (7A165t)
Created: 2015-08-02 20:20:02.590900
Originated: 2015-08-02T16:19:00
Open Radar Link: http://www.openradar.me/22108390

Description

Summary:
Upon switching accounts in the iPhone "Music" app, the "Cancel" button does nothing and the app must be force-quit

Steps to Reproduce:
0. Have multiple valid iTunes accounts (I have accounts on multiple international stores, and used USA and Canada for this test)

1. Open Music app
2. Tap "person" icon at upper left
3. Tap "Sign Out"
4. Enter credentials for other iTunes account
5. Acknowledge alert that "You will be switched to the [Country] store"

Expected Results:
Can click "cancel" and return to normal Music app activities, with your purchases from that store streamable

Actual Results:
Cancel button does nothing. Must force-quit the app and relaunch it to start listening to your purchases from this store.

Version:
iOS 8.4

Notes:

Configuration:
iPhone 5s

Attachments:

Product Version: 8.4
Created: 2015-08-02 18:26:03.233580
Originated: 2015-08-02T00:00:00
Open Radar Link: http://www.openradar.me/22108095

Description

Summary:
watchOS 2 NSURLSession requests are 10x-50x slower than iOS.

I have an app that makes local network requests. With both the iPhone app and WatchKit 1 app extension these requests were very quick, completing the NSURLSession dataTaskWithRequest:completionHandler: block in under a half a second. On watchOS 2, it's between 10x-50x slower when fetching JSON.

Steps to Reproduce:

1. Open attached sample project "FooProject"
2. Build and run iOS target
3. Press the button to make network request
4. Note the times displayed on screen and in the console
5. Build and run watchOS target
6. Press the button to make network request
7. Note the times displayed on screen and in the console

Expected Results:
NSURLSession network calls on watchOS can be slower than on iOS, but should not 10x-50x slower when fetching JSON.

Actual Results:
watchOS NSURLSession is extremely slow compared to iOS, in both the simulator and actual device.

Actual iOS device, 3 separate runs:

Request time: 0.052136 seconds
Request time: 0.321204 seconds
Request time: 0.044031 seconds

Actual watchOS watch (exact same code, exact same request):

Request time: 3.554821 seconds
Request time: 2.775169 seconds
Request time: 2.781989 seconds

Version:
WatchOS 2.0 (13S5305d)

Notes:

Configuration:

Attachments:
'FooProject.zip' was successfully uploaded.

Product Version: 2.0 beta 4 (13S5305d)
Created: 2015-07-31 00:17:31.302430
Originated: 2015-07-30T20:14:00
Open Radar Link: http://www.openradar.me/22084261

Description

Summary:
When I’m going to open an Application, I’m going to use Spotlight, as it is really convenient and super-efficient. But sometimes, Spotlight tricks me to open an Application I certainly didn’t want to open.

Example:
I type “qui” and Spotlight already shows “QuickRadar.app”, which is the app i want to start. But as my hands type faster than my head realizes it, I type further to “quick” + enter, but right by tapping the “k” Spotlight changes the suggestion to “QuickTime Player”, which is quite confusing and unexpected, as “quick” still matches on “QuickRadar.app”.

So please, don’t change the first proposal as long as the entered characters still are perfectly matching the current proposal.

Steps to Reproduce:
Not sure if this can be easily reproduced on other machines (does Spotlight some kind of learn from my actions?)

Expected Results:
Changing my input from “qui” to “quick” should not change the proposal from “QuickRadar.app” to “QuickTime Player”

Actual Results:
I’m confused and get slightly angry when I open the QuickTime Player instead of being able to quickly file radars.

Regression:
Describe circumstances where the problem occurs or does not occur, such as software versions and/or hardware configurations.

Notes:
Provide additional information, such as references to related problems, workarounds and relevant attachments.

Product Version: OS X 10.10.4
Created: 2015-07-31 11:43:37.955660
Originated: 2015-07-31T13:43:00
Open Radar Link: http://www.openradar.me/22090264

Description

Summary:
The Extension Builder does not appear to see the Safari Developer certificate that I have installed in Keychain Access. This same certificate works on my Yosemite machine.

Steps to Reproduce:

1. Copy cert to freshly upgraded El Capitan beta machine
2. Install cert into Keychain Access (by double-clicking)
3. Open Safari
4. Open the Extension Builder

Expected Results:
I expect to have a valid certificate listed and be able to develop my extension.

Actual Results:
Extension Builder displays the message "No Safari Developer Certificate"

Version:
10.11 (15A178w)

Notes:
I've tried restarting Safari, rebooting the computer, toggling the "Show Develop menu in the menu bar" option, nothing has working yet.

Product Version: 10.11 (15A178w)
Created: 2015-07-31 23:49:33.784180
Originated: 2015-07-31T00:00:00
Open Radar Link: http://www.openradar.me/22100305

Description

Summary:
The CarPlay interface does not have an option for you to love a song on Apple Music.

Steps to Reproduce:

1. Play a song

Expected Results:

1. Play a song
2. Like a song.
3. Click the love icon.

Actual Results:

1. Play a song
2. Like a song.
3. Pick up the phone while driving.
4. Locate a love icon while driving which is unsafe.
5. Click love icon.

Regression:
none

Notes:
none.

Product Version: iOS 8.4, iOS 9
Created: 2015-08-02 18:34:38.847390
Originated: 2015-08-02T11:34:00
Open Radar Link: http://www.openradar.me/22108118

Description

Although NSArray and NSOrderedSet have method indexOfObject:inSortedRange:options:usingComparator:, there is no such method in Swift Array.

By now there are two options to bypass this problem:

1. Implement your own binary search method. You can find more on this here http://stackoverflow.com/a/26679191/1722840.
2. Bridge Swift Array to NSArray and use Foundation method. Though such bridging may be slow (described here: http://swiftdoc.org/swift-2/type/Array/)

Considering all the available options it would be only natural to have such method in Swift.

Product Version: Swift 1.2
Created: 2015-07-31 22:13:23.782710
Originated: 2015-08-01T00:00:00
Open Radar Link: http://www.openradar.me/22098225

Description

Summary:
Video was shot via a Canon 7D Mark II. Some videos import and play fine, and others will import and not play - but sometimes "jerk" one frame, almost as if it's attempting to play but cannot. Please see attached video.

Steps to Reproduce:
1.Shoot pictures and video on DSLR
2. Import pictures and video
3. Notice that some videos will not play.

Expected Results:
Video plays

Actual Results:
Neigh.

Version:
El Capitan beta 5, Photos 1.1 (320.8.0)

Notes:

Configuration:
El Capitan beta 5, Photos 1.1 (320.8.0)

Attachments:
'bug_report.mov' was successfully uploaded.

Product Version: (320.8.0)
Created: 2015-07-31 14:44:47.844410
Originated: 2015-07-31T00:00:00
Open Radar Link: http://www.openradar.me/22091468