Dear Experts, I create a UIImage for an SFSymbol using [UIImage systemImageNamed], get its CGImage, and look at the sizes of each: UIImageConfiguration* config = [UIImageSymbolConfiguration configurationWithPointSize: 64 weight: UIImageSymbolWeightLight scale: UIImageScaleMedium]; UIImage* img = [UIImage systemImageNamed: @"chevron.compact.down" withConfiguration: config]; CGImageRef c = [img CGImage]; printf("UIImage size %f x %f, CGImage size %f x %f\n", img.size.width, img.size.height, CGImageGetWidth(c), CGImageGetHeight(c)); (Consider that pseudo-code, it's not an exact copy-paste.) Results: UIImage is 70.3333 x 25.6667 and CGImage is 163 x 43. So the aspect ratios (W/H) are 2.74 and 3.79 respectively. That can't be right! I don't expect the UIImage and the CGImage dimensions to be the same, because of the UIImage's scale (which is 3 in this case). But that should be the same for both dimensions. The effect is most pronounced with symbols that have an aspect ratio far from 1, e.g. recordingtape, ellipsis, and this chevron.compact.down. I believe that the CGImage aspect ratios are the correct ones. What is going on here?

Dear Experts, Does monospacedDigitSystemFont work for others? It is supposed to be a font where digits are monospaced, but other characters aren't. This is especially useful for things like counter widgets, where you don't want the total length of the string to fluctuate as it updates. When I try to use it I get the same results as for the regular system font, i.e. digit 1 is narrower than the others. [UIFont monospacedSystemFont...] does work correctly.

Dear Experts, Xcode 14 is giving me a new deprecation warning about sprintf. I understand why this is here (*) but I need to suppress it in some cases. I don't want to disable it everywhere (or lose other deprecation warnings). Specifically, I #include some Boost headers which use it in inline functions. I know that I can use #pragma to disable it, but I don't want to modify the Boost headers. Adding #pragma everywhere that I #include a Boost header is also unappealing. What other options do I have? (*) I'm not enthusiastic about the suggesting in the warning to use snprintf instead; snprintf leaves the destination unterminated on overflow, which can be just as bad as the sprintf behaviour.

Today I've received an email from Apple telling me that the App Store XML API is going away in November. (See e.g. https://help.apple.com/asc/appsspec/en.lproj/static.html). I have been using this API for some years to upload screenshots andIAP details (I have an app with a large number of IAPs). I'd like to thank Apple for giving me three months to implement an alternative. That's much better than the one month warning that they gave me for the last feature that they removed, IAP hosted content. Still, it's frustrating that I need to do a load of work to replace something that has been working fine for years. Having been an iOS developer since 2008 I've got a lot of accumulated "stuff". These days, I seem to spend more time adapting to things that Apple have deprecated than working on new features. And I constantly worry about what is next for the chop.

Is there a way to determine whether the current device has a physical home button or not? I am building part of a UI which is very much like the Files app - it has buttons for "iCloud Documents" and "Documents on this device". For the latter, it will use a suitable symbol from SF Symbols i.e. "iphone" or "ipad" (or something for Macs). Just like the Files app. But I notice that the Files app uses the ".homebutton" symbol variants on older devices that have a physical home button. What is the easiest way to replicate that? I was considering looking at the "iPhoneNN,M" string and comparing it with the model number of the first devices without a physical button, but that is complicated by e.g. the current iPhone SE, which is "iPhone14,6". I don't want to have to maintain a table. Is there some easy way to do this? Thanks.

I believe that if I rename a file, I'm supposed to do that inside a coordinated write on the parent directory - right? So say I have a file /path/to/folder/f1 and I want to rename it (in the same directory) as /path/to/folder/f2. I do a coordinated write (with the forMoving option) on the directory /path/to/folder. That passes me a possibly different path to the directory, right? Say I get /path/to/different. Question: what rename should I actually do? Do I need to construct from and to paths based on the different path that is passed to the accessor? I.e. do I rename from /path/to/folder/f1 to /path/to/folder/f2 or rename from /path/to/different/f1 to /path/to/different/f2 or some other combination? Do NSFileManager's renaming methods do anything special in this regard? Can I use e.g. std::filesystem::rename()? I am interested in both local documents that are visible e.g. in the Files app, and iCloud Drive documents, if that makes any difference. Thanks!

Dear Experts, Is there a way to limit an NSMetadataQuery to one iCloud container, in an app with more than one? It seems to me that setting the scope to NSMetadataQueryUbiquitousDocumentsScope will probably return either the contents of the first (default) container, or all containers Is there something I can put in the predicate? Maybe I have to try to match URLs that start-with the container's own URL?

Dear Experts, NSCoordinatedFile's coordinatedRead and coordinatedWrite methods supply a possibly-modified path to the accessor block. (I think it is only modified if there are concurrent accesses, so none of this is easy to test.) If I enumerate the contents of a directory inside a coordinatedRead on the directory, the paths that I get for the directory entries will include this possibly-modified parent path. (Unless maybe NSFileManager's enumeration methods have some magic to avoid that.) Is it OK for me to keep these "special" pathnames beyond the end of the coordinatedRead? If not, perhaps I need to get only the leafname from the directory enumeration and append that to the original directory path. Thoughts anyone?

Dear All, Does anyone have any suggestions for an iPad text editor app that is well-behaved with respect to NSFileCoordinator / NSFilePresenter / UIDocument etc.? Specifically, I'd like something that I can run in iPad split screen mode alongside my test app. I'd like to be able to make changes to files using the text editor and see my test app pick them up, and vice-versa. I am currently trying an app called "Quick Text" which mostly works but it doesn't seem to pick up changes that I make in my test app unless I close and re-open the file at its end. I'm also soon going to reach its limit of "10 edits per week"! I have a couple of others to try, but maybe someone here has a suggestion?

Dear Experts, Should I use NSFileCoordinator coordinateWritingItemAtURL: around a call to NSFileManager trashItemAtURL: ? What options should I pass - NSFileCoordinatorWritingForDeleting, or ...ForMoving, or something else? I've attempted various combinations. If I don't use the NSFileCoordinator, then the change is not reflected in the Files app (open alongside my test app in iPad split screen mode). All attempts using NSFileCoordinator result in a deadlock inside trashItemAtURL - it is stuck waiting for a semaphore. I do have a FilePresenter for this file, but I am passing it to the NSFileCoordinator's init so that should not lead to deadlock. I believe I am doing everything on the main thread. The files are local, in the app's Documents directory. Any ideas anyone?

Dear Experts, We are required to remove our NSFilePresenters from the NSFileCoordinator when the app goes to the background and re-add them when it returns to the foreground. Does this mean that when we foreground, we need to manually check if our files have been modified, deleted, or moved in the meantime? Or is there some mechanism in NSFilePresenter/NSFileCoordinator that will report any such changes to us? Detecting that a file has been modified or deleted is not difficult, but I don't think it's possible to detect that it has moved. Changes to directories are also difficult. Any advice? What does UIDocument do in this case?

Dear Experts, If a user with an older device tries to download an app, and the current version of the app requires an iOS version which their device does not support, the App Store will offer them an older most-recent-compatible version instead. I've just had an enquiry from a user of my app for whom this has not worked. Specifically he has an iPad 2. He foolishly deleted my app from the device, and now the App Store will not let him re-install it. Does anyone know why this may be? Is it simply that this App Store feature does not work for devices as old as the iPad 2? Thanks.

I’ve just tried to reply to a post and I get an “unauthorized” error page back. I’m logged in. I’m posting from an iPad, which is unusual, I normally use my Mac. This post is in part to see if I can post at all, or whether it was just that one reply that is. not working.

Dear Experts, I have been looking at the documentation for system colours at https://developer.apple.com/design/human-interface-guidelines/ios/visual-design/color/ https://developer.apple.com/documentation/uikit/uicolor/standard_colors/ https://developer.apple.com/documentation/uikit/uicolor/ui_element_colors It's all reasonably clear... until I actually look at some UIKit screens and try to work out which colour is being used where. For example, what colour is the background of a navigation bar? It does not seem to be one of the standard colours. Attached is a screenshot from "UIKit Lab", which is a free UIKit catalogue/demo app. Scrolling through its list of standard colours, none of them seems to match the background of the navigation bar. It lies somewhere between systemBackground and secondarySystemBackground. Is this perhaps because the base colour has been modified by a material effect, or something? Thanks, Phil.

Dear Experts, Where is the best place to persist an App Attest key id? The docs ( https://developer.apple.com/documentation/devicecheck/establishing_your_app_s_integrity ) say “Record the identifier in persistent storage — for example, by writing it to a file”. That is what I have done, but I have encountered a problem. If a user gets a new device and restores a backup of an old device onto it, the new device will try to use the key id from the old device - which is of course wrong. One solution is to detect the error when the invalid key is used and to generate a new one. Is that the best approach? I am wondering if there is some part of the filesystem that does not survive the backup/restore process, but is otherwise persistent? It should be more persistent than a cache file. (Also looking at the docs again I now see that I am supposed to store distinct keys for each “user”. What is meant by “user” in this case?) Thanks.