Is it possible to stop Xcode (8) from automatically generating a Swift interface when showing an Objective-C header? For example by setting a user defaults value.
In theory it’s a really nice feature of Xcode. Unfortunately it doesn’t work reliably. For me it even seem to fail most of the time. As I haven’t been able to discern why it happens, I’d like to opt out of it, until the bugs Xcode/SourceKit have been resolved.
The default behaviour should be dependent on your project settings. If you're in a Swift project it will want to show you Swift interfaces. Also if there is an Objective-C interface that Xcode generate Swift from then using Cmd+Alt+Enter you should be able to look at the Objective-C header "Counterpart" of the Swift code there for comparison.
Related
I'm trying to use the SwiftyChrono swift library from an Objective-C project. Since it's using a struct instead of a class (amongst many other Swift only features), I'm unable to make it work.
Given I am not in the position of moving all our Objective-C code to Swift, what are my options? Would writing a second Swift framework work, one that sits in between our app and SwiftyChrono and serves as a wrapper? Would that even work? I'm guessing if this new Swift framework was dumb enough to only perform a single function in 'Swift Land', it would play well with Objective-C?
I haven't dealt with inter-op and it really feels like a losing battle. Any help would be appreciated.
What is the use of bridging header?
Is it just for using Objective-C and Swift code in the same project?
Should we avoid using bridging header?
Say, if there are two third party library which are very similar; one of them is in Objective-C and other is in Swift. Should we use the Swift library or use Objective-C library. Are there any downside of using bridging headers?
Apple has written a great book that covers this in depth. It can be found here:
https://developer.apple.com/library/ios/documentation/Swift/Conceptual/BuildingCocoaApps/MixandMatch.html
I will quote it to answer your questions:
"What is the use of bridging header?
Is it just for using Objective-C and Swift code in the same project?"
To import a set of Objective-C files in the same app target as your Swift code, you rely on an Objective-C bridging header to expose those files to Swift. Xcode offers to create this header file when you add a Swift file to an existing Objective-C app, or an Objective-C file to an existing Swift app.
The answer to this question is yes. It is just there to make Swift and Objective-C work together in the same project.
"Should we avoid using bridging header? Say, if there are two third party library which are very similar; one of them is in Objective-C and other is in Swift. Should we use the Swift library or use Objective-C library. Are there any downside of using bridging headers?"
There are always tradeoffs. The first answer to this is no you should not avoid using a bridging header; however, as far as third party libraries you have to look at many factors. Which one has more functionality? Is it being maintained and/or added to frequently?
Using an Objective-C library will also add things to be aware of and work around. From the book:
Troubleshooting Tips and Reminders
Treat your Swift and Objective-C files as the same collection of code, and watch out for naming collisions.
If you’re working with frameworks, make sure the Defines Module (DEFINES_MODULE) build setting under Packaging is set to “Yes".
If you’re working with the Objective-C bridging header, make sure the Objective-C Bridging Header (SWIFT_OBJC_BRIDGING_HEADER) build setting under Swift Compiler - Code Generation is set to a path to the bridging header file relative to your project (for example, “MyApp/MyApp-Bridging-Header.h").
Xcode uses your product module name (PRODUCT_MODULE_NAME)—not your target name (TARGET_NAME)—when naming the Objective-C bridging header and the generated header for your Swift code. For information on product module naming, see Naming Your Product Module.
To be accessible and usable in Objective-C, a Swift class must be a descendant of an Objective-C class or it must be marked #objc.
When you bring Swift code into Objective-C, remember that Objective-C won’t be able to translate certain features that are specific to Swift. For a list, see Using Swift from Objective-C.
If you use your own Objective-C types in your Swift code, make sure to import the Objective-C headers for those types before importing the Swift generated header into the Objective-C .m file you want to use your Swift code from.
Swift declarations marked with the private modifier do not appear in the generated header. Private declarations are not exposed to Objective-C unless they are explicitly marked with #IBAction, #IBOutlet, or #objc as well.
For app targets, declarations marked with the internal modifier appear in the generated header if the app target has an Objective-C bridging header.
For framework targets, only declarations with the public modifier appear in the generated header. You can still use Swift methods and properties that are marked with the internal modifier from within the Objective-C part of your framework, as long they are declared within a class that inherits from an Objective-C class. For more information on access-level modifiers, see Access Control in The Swift Programming Language (Swift 2.2).
Que : What is the use of bridging header?
Its correct to say, Bridging header allows user to use Objective-C classes/files in their swift code in same project.
A Swift bridging header allows you to communicate with your old Objective-C classes from your Swift classes. You will need one if you plan to keep portions of your codebase in Objective-C. It should be noted that even if you decide to convert all of your code to Swift, some classes or libraries you may use such as SVProgressHUD haven’t been rewritten in Swift and you will need to use a bridging header to use them.
Que : Should we avoid using bridging header?
Considering your question there are 2 possible cases.
case 1 : Lets say your project is developed in Objective-C and now you are developing new features using swift in it, in this case you have to have BridgingHeader as you need access of your Objective-C classes in swift code.
case 2 : If your project is developed in swift then there is no need to have Bridging header, as well if its in only Objective-C and you are not planning to move it in swift then also you don't need it.
Read more about Using swift with cocoa and Objective-C in apple documentation.
Following apple document image indicates usage of Bridging header
No, there are no downsides to using Obj-c code in your Swift project. Bridging header only exposes your Obj-c files to Swift. The two languages can coexist in the same project with no problems, as you can expose your Swift code to the Obj-c just as easily too - xCode will generate a header for all of your public Swift declarations. Although everything is possible, if you start a new project you should stick to one language so the project is easier to understand. For example if you decide on Swift you should only use Obj-c for libraries that are not available in Swift.
The bridging header allows the use of Swift and Objective-C in the same project. There are no downsides to having a bridging header in your project as the two languages can work well together within the same app.
Removing a bridging header from a project after it has been added may cause errors, as it is referenced in other places in the project when it is created.
If you only intend to use one of the two languages, a bridging header is unnecessary. On the other hand, if you are using both Swift and Objective-C, a bridging header is required and will not cause any issues.
Here is a link to find more information on the subject:
https://developer.apple.com/library/ios/documentation/Swift/Conceptual/BuildingCocoaApps/MixandMatch.html
I hope that answered your question. Good luck with your project!
What is the use of bridging header?
You have already got that answer. You're right.
Should we avoid using bridging header?
No. Its good when a third party library developed in Obj-C and may not available in Swift yet. You should use bridging header to have a best library for your app.
It depends which on you choose. In case of networking? If your project is in Obj-C based you can use AFNetworking or with the case of Swift you can use AlamoFire, you can still use AFNetworking in Swift but its not suggestable.
Bridging headers are a great way to get Objective - C code into your Swift project. If you have two libraries, one that is in Swift and one that is in Objective - C, choose the one that will offer more functionality in your app. If they offer the same functionality, I'd just go with the Swift library -> My reasoning: if the Objective-C library isn't widely used and there aren't many tutorials on how to convert the Objective - C code into Swift, it can be very time consuming to figure it out on your own. If you use a Swift library, the code is already formatted in the correct language, saving you time and potentially money (depending on if this is a hobby for you or not). As far as any downsides to using a bridging header, their really isn't! With so many libraries written in Objective-C, you almost need a bridging header in your app. Take, for example, Reachability (Here is a video on implementation in Swift). This is a library that Apple created to handle network interruptions in your app. This is a great tool for developers and requires a bridging header. Here is a great YouTube video on how to use a bridging header, but if you add a header file into your Swift file, Xcode typically asks to crete one for you. Hope this helps!
After I updated to 4.3.3 Xcode started to jump to interfaces instead implementations. Same with open quickly it doesn't show me implementation files, only headers.
Is there same way to fix it?
I don't know of a way to choose whether Xcode jumps directly to either the interface or the implementation.
However, if you are looking at the interface, you can jump to the implementation by command-clicking the name in the interface. If you are looking at the implementation, you can jump to the interface by command-clicking the name in the implementation.
This works for both class names and selectors. Thus you are never more than two command-clicks from both the interface and the implementation (if Xcode knows where to find them).
I am building an iOS 5 app with ARC using clang on Xcode 4.2/Lion. Good practice for me is to try to get rid of as many warnings as possible but I'm lost on this one. The app has quite a few classes, but a limited number of their header files (8 or 9) give linker warnings like this:
warning: no rule to process file '$(PROJECT_DIR)/TKMyClass.h' of type sourcecode.objj.h for architecture arm7`
I don't see any significant commonalities across these headers: a couple are 3rd party OSS code; the rest are all my own. One is a plain-C header file (with no corresponding .c) containing nothing but constants, #defines and enums; a couple are UIView and UIViewController subclasses created with Xcode's templates; the rest are ordinary Obj-C classes, some of which inherit from NSObject and some of which don't. All (of my classes) were created from scratch within the project at various times. Both older and newer classes give no warnings.
My project uses a mix of C++, Objective-C++ and Objective C classes. The warning-generating classes here are mostly Objective C (their implementations are in .m files) but the two 3rd party classes are implemented in Objective C++ (.mm).
The project otherwise builds fine and runs in the simulator and on iDevices.
To my shame I'm not particularly familiar with the project settings pages in Xcode. I expect the solution is contained in there somewhere but I'm not sure where to start without breaking things.
That means that you have accidentally added header files to be compiled. You need to go into your projects Build Phases and remove all header files from the Compile Sources section.
Check your Architecture (Project and Target): Standard (armv7) - $(ARCHS_STANDARD_32_BIT) and go to Build Pharse (Compile Source) and cheek there should not be any .h file added there.
Coming from an Eclipse / Java background, one of my favorite features is the ability to quickly stub out all the methods required by an interface. In Eclipse, I can choose 'Override / implement' from the source menu to generate stub methods for any method of the Interface.
I'd like to do the same thing in Objective-C. For instance, if I declare a class that implements the 'NSCoding' protocol, I'd like to have Xcode automatically generate the methods required to implement this Protocol. It's frustrating to have to look-up and then copy/paste the signatures of the required methods every Protocol that I'm trying to implement.
I've been trying for awhile to find out if this is possible, but haven't found anything promising yet. Is this possible in XCode?
I believe that Accessorizer will do what you want.
Accessorizer will write the encode and decode methods for ivars passed to it (NSCoding protocol and for NSDocument archiving). It will also generate string constants either static or #define with a custom prefix; copyWithZone:; and other things if you need - all from a simple shortcut via Services or from the toolbar. Accessorizer keyed archiving
Not the direсt answer, just hint:
Out of the box XCode can't.
But AppCode can.
It can do this things automatically (with your permission, of course).
If some methods of protocol marked as #required - AppCode will highlight the implementation and suggest to implement this methods.
#optional methods also available to implement automatically (shortcut: control + I).
Your can create scripts for the scripting menu item in AppleScript, Perl, Python, Ruby, or any other scripting language that go in the scripting menu.
Your could place the insertion point in the .m file and have the script look up the corresponding .h file. Locate the protocols supported and so forth...
MacTech ran an article in 2007 Xcode Menu Scripts
Xcode 3.2 will autocomplete known method implementations. In other words, if the method is declared somewhere (for example, in a protocol), when you start to type it in a .m file, Xcode 3.2 will autocomplete the method signature for you. This isn't quite what you asked for, but it is awfully handy.
I'm also looking for a way to generate method stubs for the protocols in my header file. I checked out Accessorizer and it looks to be a handy tool but unless I missed something I didn't find a way to get it to generate method stubs for a protocol.
Eric, If you found another solution please post what you found. It's amazing to me that XCode doesn't already have this built into the IDE.
Since the accepted answer's given link does not work anymore (and is redirected to an ad), here's another good explanation on how to use accessorizer to create protocol method stubs.
Based on AllanCraig's "Create #property, #synthesize & dealloc from Variable Declaration" ruby script, I made one to generate implementation stubs from interface ones: http://pastebin.com/4T2LTBh6
How to use?
Setup the script on your XCode (Shell Script) and assign a shortcut for it (e.g. CMD+5).
Select lines from your interface file in which you want to generate the implementation, and press the hotkey.
Your .m will contain your selected methods.
I know this is an old question but if you'd like to always see the latest definitions just right click on the class in question and Jump to Definition. Here lyes all the current non-deprecated functions so you aren't relying on a 3rd party to stay up to date.
In My case Below style helps me much, In a sense solved my problem.
Suppose you have following method declaration:
+(DBManager*)getSharedInstance;
From Implementation file you start typing +ge and xcode will automatically choose method
+(DBManager*)getSharedInstance;