Contentful API breaks Swift Concurrency by calling completion handlers more than once

[tristan-warner-smith]

• contentful.swift version number: 5.5.2
• Xcode version number: 13.3
• Target operating system(s) and version number(s)
  • [X] iOS: 14.5, 15.4
  • [ ] tvOS:
  • [ ] watchOS:
  • [ ] macOS:
• Package manager:
  • [ ] Carthage
  • [ ] Cocoapods

Using the Contentful client SDK results in app crashes when completion-based calls fail to decode.

I just encountered this because I've wrapped fetchArray(of in a Task and this triggers Fatal error: SWIFT TASK CONTINUATION MISUSE: fetchArray(of:) tried to resume its continuation more than once, throwing Unknown error occured during decoding.!

The issue I saw is in Client.swift handleJSON<DecodableType: Decodable> where if there's a decode failure the completion is called twice with potentially different errors. This seems to be related to a desire to churnLinks regardless of the failure - but if it failed to decode anyway, we're done there's no value in resolving other failures, we can just go fix our issue and move on, right?

If a failure happens on the original decode the failure completion is called but not returned so it continues to a second decoded check that fails and again calls completion(.failure(error).

The two completion calls are in Client.swift:476 and 500 in Contentful 5.5.2.

This is just how I came across it but any decoding error should reproduce:

• Create a parent child model - 1 Article has many Pages for example
• Create an article with no pages
• Have your decode be something like: self.pages = try fields.decode([Link].self, forKey: .pages) where it assumes non-optionals

• Wrap Contentful API to be async-awaitable

  extension Contentful.Client {
  // MARK: - Making Contentful fetches async
  
  func fetchArray<ResourceType: EntryDecodable & FieldKeysQueryable>(of resourceType: ResourceType.Type) async throws -> [ResourceType] {
  
      try await withCheckedThrowingContinuation { continuation in
          fetchArray(of: resourceType) { result in
              switch result {
              case .success(let results):
                  continuation.resume(returning: results.items.map { $0 as ResourceType })
              case .failure(let error):
                  continuation.resume(throwing: error)
              }
          }
      }
  }
  
  func fetchArray<EntryType>(
      of entryType: EntryType.Type,
      matching query: QueryOn<EntryType>
  ) async throws -> [EntryType] {
      try await withCheckedThrowingContinuation { continuation in
          fetchArray(of: entryType, matching: query) { result in
              switch result {
              case .success(let results):
                  continuation.resume(returning: results.items.map { $0 as EntryType })
              case .failure(let error):
                  continuation.resume(throwing: error)
              }
          }
      }
  }
  }

• Make a call that will return the parent-child types: let articles = try await client.fetchArray(of: Article.self)
• See fatal as above SWIFT TASK CONTINUATION MISUSE: fetchArray(of:) tried to resume its continuation more than once, throwing Unknown error occured during decoding.!

[nilsen340]

Will this be taken care of in the foreseeable future?

[tristan-warner-smith]

@nilsen340 there seems to be no active triage of issues in this repo, my report is more than a year old so I'd assume not.

That said, I created a PR to fix it over here and it was approved and closed, I didn't check to ensure it made it into master though, but I assume it was so YMMV.

[nilsen340]

@tristan-warner-smith thanks Tristan, it looks like your PR was approved but not merged, it was closed. But it seems they merged a fix later on their own: https://github.com/contentful/contentful.swift/pull/370/files

Not sure if this covers all the cases.

[tristan-warner-smith]

That's enough for me, thanks for checking @nilsen340.