User Profile

Hello. I recently released a new version (2.6) of my CloudKit syncing CoreData based app. The new version uses v18 of my data model. V18 is exactly like v17, except for a new optional string field in one of the entities. I diffed the two model versions to be extra sure. During my test I did not encounter any crash, and I also tested automatic migration, installing 2.55 and building 2.6 on top of it. No troubles. However my crash logger reports a number of crashes among my users. It's not a huge numbers (10 users over about 900 or more) but I am surprised it crashes at all. The crash happens when I call NSPersistentCloudKitContainer.loadPersistentStores, I get this error in the completion block: The model used to open the store is incompatible with the one used to create the store" I always assumed these slight changes (new field added to an existing model) did not require a manual migration. Any suggestions?

I have a complex project; it's got 10 targets and 5 languages so far (plus base localization). I have followed Apple documentation to extract the XLIFF files to pass to translators (Product / Export Localizations). The export seems to work, but if I inspect the generated xloc files I find no xliff: the "Localized Contents" folder is simply empty. This applies to all languages. No xliffs at all. So far I kinda managed by sending localizable.strings sources to translators, but they have been (rightly) complaining because the .strings format doesn't integrate well with their tools. I believe there is something wrong in my project, something that needs repairing, but I don't know where to start from. By the way, this is a commercial app, already generating revenue, so I am quite cautious when manipulating project settings and all. Do you have any suggestions? Cheers, Davide

Hello, I have a strange bug. An extra tab bar button appears in the lower left corner of the tab bar; it has a label but no image. It duplicates the second tab bar button. There is an identical issue in stack overflow, but none of the suggestions there apply to my case; but you can look at the image there to see what happens. https://stackoverflow.com/questions/45757172/uitabbar-item-title-repeated-in-the-bottom-left-corner-visual-bug To give you more details: this app has a storyboard with a UITabBarController subclass at its root. This subclass does nothing, except managing some global notification. No subclassing of uitabbar. The tab bar controller has 4 items All items are UINavigation controllers, whose root are different controllers items 1 and 3 are pure UIKit, so we have a navigation controller and then some UIViewController subclasses items 2 and 4 have a UINavigationController with a UIHostingController inside, and of course then it's SwiftUI To reproduce the bug I need to: select items 2 or 4, which contain List elements select a list item (NavigationLink) = a detail controller opens navigate back from the detail controller I have checked with debug view heirarchy and logs, I have verified that when I do these operations an extra UIBarButtonItem gets added. Quite interestingly, if I do this on item 2 and then on item 4 I get two extra UIBarBUttons; but if I repeat the operation over and over again I don't get anything more. So ax extra button gets added once for each operation on items 2 and 4. It looks like some weird interaction between UINavigationController, UITabBarController and SwiftUI views. Or maybe an iOs bug, who knows. I have implemented a hackish way to traverse the hierarchy and remove the extra views, but I would like to remove the bug at its root. Do you have any suggestion?

Hello, I am dabbing my feet in core ML and I am not so sure about the best way to tackle a language processing issue. I need to tokenize based on multiple words items. IE: "I like Ferrari Testarossa" In order to match "Ferrari Testarossa" I have tried two segmentation strategies: ["I","like","Ferrari Testarossa"] ["person", "like", "car"] and ["I","like","Ferrari", "Testarossa"] ["person", "like", "car", "car"] After generating the model, I realized that NSLinguisticTagger can only divide into words OR sentences (or bigger), nothing in between. So with either strategies, If I type "I love Lamborghini Contact" I get ["I","love","Lamborghini","Contact"] ["person", "like","car","car"] The first segmentation strategy (put tokens with 2 words in the same item) would be more suited to my UX, but I can deal with the second, it is quite trivial to join tags. I was just wondering, which is the best way to feed the model generator, Machine Learning-wise? Is there a difference in feeding the model "Ferrari Testarossa" and "Ferrari", "Testarossa" ?