9.0.2

Repackages upstream Google ML Kit 9.0.0, same as 9.0.1. This release
completes the API coverage.

Added

MLKitDigitalInkRecognition and MLKitEntityExtraction — the only two
user-facing ML Kit iOS APIs this package did not wrap. Coverage is now
complete: 19 library products.

Product	What it does
MLKitDigitalInkRecognition	Recognise handwriting and drawn strokes
MLKitEntityExtraction	Find addresses, dates, phone numbers and the like in text

Both fetch their models from Google on first use rather than shipping them, so
they need network access the first time you use them. Digital ink recognition
also needs MLKitDigitalInkRecognition_resource.bundle added to your app
target, like the other modules.

If you are still on 9.0.0 or 9.0.0-1

Everything 9.0.1
fixed applies here. In short: text recognition, pose detection and selfie
segmentation ship their models at all for the first time; a single product can
be adopted without undefined symbols; the simulator slice covers arm64 so
Apple Silicon works without EXCLUDED_ARCHS; and App Store Connect accepts the
upload.

.package(url: "https://github.com/d-date/google-mlkit-swiftpm", from: "9.0.2")

Two steps stay manual, both documented in the README: -ObjC -all_load in
Other Linker Flags, and the resource bundles for your modules
(./scripts/download_bundles.sh 9.0.2).

Verified

• Five runtime tests pass on the arm64 Simulator, including the two new modules
• The Example app links all 19 products, launches on the Simulator and archives
  for device with all 19 model bundles present and nothing linked dynamically
• All 33 XCFrameworks carry ios-arm64_x86_64-simulator
• An App Store Connect upload passed Apple's validation with no errors and
  reached VALID

🤖 Generated with Claude Code