My app only offers consumable in-app purchase products. I can successfully purchase a product once but when I try to purchase it again, a popup "This In-App purchase has already been bought. It will be restored for free." appears and the first transaction is restored: the payment queue delegate is notified that a new transaction (with the same identifier as the successful transaction) is updated to the purchased state. I never get a purchased transaction with a new identifier and hence I consider that the purchase failed. Also, each time the app is put in foreground again or at startup, the payment queue delegate is notified that a transaction has been updated (as if it had never been finished). Even though the app properly finishes the transaction each time. Side notes: I guarantee that all purchased or failed transactions are finished (confirmed by the removedTransactions SKPaymentQueueDelegate method being called each time). I even tried to call finishTransaction from the main thread with no success. Right before the subsequent purchase attempts, the transactions queue is empty. I'm also pretty sure I was able to purchase several times the same product last week, with no change in code (same app version from Testflight). I observe this behaviour when building the app with Xcode or when distributed via Testflight. It has not been published yet so I cannot check how it behaves in the Production environment. Do you have any clue on what's going on? Could it be a side effect of using the Sandbox environment? Thanks for your help, Aurélien.

We want to allow the enduser to purchase movies from our VOD app using in-app purchases (IAP). Since our app exposes thousands of movies, we plan to create several generic consumable IAP products, one by price tier (according to the expected selling prices). Each productId can be used to buy different movies. Once a purchase is successful (transaction in purchased state), our app needs to send the transaction id and receipt along with the movie id to our back-office in order to perform checks and eventually mark this specific movie as purchased for this user (our app being multi-platform, the purchased movies have to be available on all platforms, whatever the platform used for the purchase). But since the purchase is solely based on the SKProduct (hence its id), and since there's no way to attach some kind of userInfo to the SKPayment added to the default SKPayementQueue, there's no direct way to match an SKPaymentTransaction to the movie whose purchase triggered its creation, especially when dealing with "Ask to Buy" and highly asynchronous transactions updates. So, as far as I know, I have to maintain some sort of permanent context with the movie id, the product id and the last update date in order to try to get the movie id matching a transaction. But there are lots of things to take care of and a high risk maintaining this parallel context synced with the SKPaymentQueue pending transactions. I am pretty sure other apps have the same IAP strategy so I wonder if I'm missing something obvious. Are there some recommendations from Apple on the way to handle it? Or has anyone been through this road and being able to provide advices? Thanks for your help, Aurélien.

For Q&A and logging purposes, I need to know whether the response to an URLSessionDataTask comes from the URLCache or not. I've read here and there that it was just a matter of comparing the cachedResponse before and after receiving the dataTask response. But it does not work as each call to URLCache.shared.cachedResponse(for: urlRequest) returns a new instance of CachedURLResponse (each one having another new instance of HTTPURLResponse). Am I missing something? Is there another way to tell if the response comes from the cache?

Our iOS apps use to embed and sign third party frameworks built for various architectures (i386 x86\_64 armv7 arm64). When building an archive, a script phase strips unwanted architectures (simulator ones) from the build. Everything was working fine until XCode 12.3 (updated this morning). Now, what seems to be a "Target integrity" check always reports the following error for these embedded frameworks (example with GoogleCast framework): error: Building for iOS, but the linked and embedded framework 'GoogleCast.framework' was built for iOS + iOS Simulator. Also, as these frameworks are synced from a server at build time using a script phase, if I delete the frameworks before compiling, there are no errors and the app is built properly. But if the frameworks are already synced, the compilation systematically fails right away. Finally, other frameworks with the same architectures but that are not embedded don't trigger this error. I've tried to changes some build settings regarding the architectures but with no success. Could you please tell me how to embed these frameworks, keeping in mind they are third-party frameworks? Cheers, Aurélien