Files
OSGKeyboard/OSGKeyboardHostSupport/Services/CustomLanguageModelManager.swift
T
rockyandClaude Opus 5 0023741455 feat(app): add crash visibility, memory relief, and clipboard reply work
Squashes the working tree that had accumulated on cloud-agent. Two
strands of work, kept in one commit at the maintainer's request.

Crash and stability (this session):

- Subscribe to MetricKit so crash, hang, CPU, and disk-write diagnostics
  are recorded on device. Keyboard-extension memory kills are the most
  likely way a user loses the keyboard mid-sentence and iOS does not
  classify them as crashes, so they never reach App Store Connect.
- Upload those reports to POST /v1/analytics/diagnostics from internal
  builds only (local Debug and TestFlight), reusing the analytics
  installation identifier and skipping entirely when analytics are off.
  App Store builds keep everything on device.
- Shed the Chinese engine, English lexicon, and in-flight pipeline work
  at the 40 MiB and 48 MiB budget thresholds instead of waiting for a
  system memory warning that jetsam often preempts.
- Install audio taps with the input node's live format through one
  shared adaptive resampler, so an audio-route change mid-recording no
  longer raises an uncatchable format-mismatch exception. Covers the
  dictation preview and the Mac recorder, which still used the pattern
  the Flow capture path had already moved away from.
- Harden the librime bridge: never pass a nil string across a nonnull
  boundary when a user dictionary holds malformed UTF-8, null-check the
  engine API vector, and distrust a candidate count without its array.
- Guard the expanded candidate grid against a list that shrinks while
  the collection view is still laying out against the previous count.
- Resolve localized text without the App Group so the screen shown when
  that container is missing renders its message instead of crashing on
  launch, which is the one situation it exists to handle.
- Gate CI on the App Group and Keychain wiring that the keyboard
  extension hard-depends on, with a self-test proving the check fails.

Clipboard, reply, and semantics (pre-existing working tree):

- Unify the reply center, refresh keyboard AI features, and add native
  event and reminder exporters.
- Expand clipboard semantic routing with authored corpora and a
  product-blind holdout, plus the scripts that build and validate them.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-09-18 17:31:29 -07:00

455 lines
16 KiB
Swift

// CustomLanguageModelManager.swift
// OSGKeyboard · HostSupport
//
// Prepares the bundled SFCustomLanguageModelData asset for the host app's
// iOS SpeechAnalyzer pipeline and caches the compiled LM + Vocab in the
// App Group container. The keyboard extension does not run ASR or load
// these assets.
import Foundation
import os
import Speech
#if canImport(OSGKeyboardShared)
import OSGKeyboardShared
#endif
public final class CustomLanguageModelManager: @unchecked Sendable {
public static let shared = CustomLanguageModelManager()
public enum PrepareState: Equatable, Sendable {
case idle
case preparing
case ready
case failed(String)
}
struct BundledManifest: Decodable {
let version: String
let binBytes: Int
let identifier: String
private enum CodingKeys: String, CodingKey {
case version
case binBytes = "bin_bytes"
case identifier
}
}
private enum Storage {
static let subdirectory = "CustomLanguageModel/v1"
static let fingerprintKey = "customLM.preparedFingerprint"
static let preparedAtKey = "customLM.preparedAt"
static let lastFailureAtKey = "customLM.lastFailureAt"
static let attemptCountKey = "customLM.attemptCount"
static let maxRetryAttempts = 3
/// Backoff after failure attempts 1, 2, and 3 (seconds).
static let backoffIntervals: [TimeInterval] = [30, 120, 600]
}
private let lock = OSAllocatedUnfairLock()
private var cachedConfiguration: SFSpeechLanguageModel.Configuration?
private var state: PrepareState = .idle
private var prepareTask: Task<Void, Never>?
private init() {}
// MARK: - Public API
/// Returns a prepared configuration for Chinese locales when available.
public func configurationForTranscription(locale: Locale) -> SFSpeechLanguageModel.Configuration? {
guard Self.isChineseLocale(locale) else { return nil }
return lock.withLock { () -> SFSpeechLanguageModel.Configuration? in
if let cachedConfiguration {
return cachedConfiguration
}
if let loaded = Self.loadCachedConfigurationFromDisk() {
cachedConfiguration = loaded
state = .ready
return loaded
}
return nil
}
}
public func currentState() -> PrepareState {
lock.withLock { state }
}
/// Fire-and-forget preparation for the host app. Safe to call repeatedly.
/// Retries after exponential backoff when a prior attempt failed.
public func prepareInBackgroundIfNeeded() {
#if os(iOS)
guard AppGroup.isAvailable else { return }
#endif
let shouldStart = lock.withLock { () -> Bool in
if case .preparing = state { return false }
if cachedConfiguration != nil { return false }
if let loaded = Self.loadCachedConfigurationFromDisk() {
cachedConfiguration = loaded
state = .ready
Self.clearRetryState()
return false
}
if prepareTask != nil { return false }
if case .failed = state {
guard Self.canRetryAfterFailure() else { return false }
} else if !Self.canRetryAfterFailure() {
return false
}
state = .preparing
return true
}
guard shouldStart else { return }
prepareTask = Task.detached(priority: .utility) { [weak self] in
guard let self else { return }
defer {
self.lock.withLock { self.prepareTask = nil }
}
do {
_ = try await self.prepareIfNeeded()
} catch {
Self.recordFailure()
self.lock.withLock {
self.state = .failed(error.localizedDescription)
}
Self.log(
"prepare failed (attempt \(Self.storedAttemptCount())): \(error.localizedDescription)"
)
}
}
}
/// Prepares the bundled training asset into the App Group container.
@discardableResult
public func prepareIfNeeded() async throws -> SFSpeechLanguageModel.Configuration? {
if let existing = configurationForTranscription(locale: Locale(identifier: "zh-Hans")) {
lock.withLock { state = .ready }
Self.clearRetryState()
return existing
}
guard Self.canRetryAfterFailure() else {
throw PrepareError.retryBudgetExhausted
}
guard let manifest = Self.bundledManifest() else {
throw PrepareError.missingManifest
}
guard let assetURL = Self.bundledTrainingAssetURL() else {
throw PrepareError.missingTrainingAsset
}
guard let preparedDir = Self.preparedDirectoryURL() else {
throw PrepareError.missingAppGroupContainer
}
let fingerprint = Self.fingerprint(for: manifest)
if Self.storedFingerprint() == fingerprint,
let cached = Self.loadCachedConfigurationFromDisk() {
lock.withLock {
cachedConfiguration = cached
state = .ready
}
Self.clearRetryState()
return cached
}
lock.withLock { state = .preparing }
let languageModelURL = preparedDir.appendingPathComponent("LM")
let vocabularyURL = preparedDir.appendingPathComponent("Vocab")
try Self.removeItemIfExists(at: languageModelURL)
try Self.removeItemIfExists(at: vocabularyURL)
let configuration = SFSpeechLanguageModel.Configuration(
languageModel: languageModelURL,
vocabulary: vocabularyURL
)
Self.log(
"preparing custom LM (\(manifest.binBytes) byte asset)… \(OSGDiag.memoryTag())"
)
OSGDiag.log(
"clm.prepare.begin bytes=\(manifest.binBytes) \(OSGDiag.memoryTag())",
category: "asr"
)
try await Self.prepareLanguageModel(assetURL: assetURL, configuration: configuration)
guard FileManager.default.fileExists(atPath: languageModelURL.path),
FileManager.default.fileExists(atPath: vocabularyURL.path) else {
throw PrepareError.missingPreparedArtifacts
}
Self.persistenceDefaults.set(fingerprint, forKey: Storage.fingerprintKey)
Self.persistenceDefaults.set(Date().timeIntervalSince1970, forKey: Storage.preparedAtKey)
Self.clearRetryState()
lock.withLock {
cachedConfiguration = configuration
state = .ready
}
Self.log("custom LM ready at \(preparedDir.path)")
OSGDiag.log(
"clm.prepare.done path=\(preparedDir.lastPathComponent) \(OSGDiag.memoryTag())",
category: "asr"
)
return configuration
}
// MARK: - DictationTranscriber factory (iOS host app)
#if os(iOS)
public static func makeDictationTranscriber(
locale: Locale,
lmConfiguration: SFSpeechLanguageModel.Configuration?
) -> DictationTranscriber {
let preset = DictationTranscriber.Preset.progressiveLongDictation
guard let lmConfiguration, isChineseLocale(locale) else {
return DictationTranscriber(locale: locale, preset: preset)
}
let contentHints = preset.contentHints.union([
.customizedLanguage(modelConfiguration: lmConfiguration)
])
return DictationTranscriber(
locale: locale,
contentHints: contentHints,
transcriptionOptions: preset.transcriptionOptions,
reportingOptions: preset.reportingOptions,
attributeOptions: preset.attributeOptions
)
}
#endif
// MARK: - Legacy Speech request (macOS Apple Speech fallback)
/// Up to 100 short phrases for `SFSpeechRecognitionRequest.contextualStrings`.
public static func contextualStringsForRecognition(
bias: LocalASRBiasPayload?,
maxCount: Int = 100
) -> [String] {
guard let bias, !bias.hardHotwords.isEmpty else { return [] }
return Array(bias.hardHotwords.prefix(max(1, maxCount)))
}
/// Applies bundled CLM + optional contextual strings to a legacy on-device request.
public static func applyCustomLanguageModel(
to request: SFSpeechURLRecognitionRequest,
locale: Locale,
bias: LocalASRBiasPayload?
) {
request.requiresOnDeviceRecognition = true
if let configuration = shared.configurationForTranscription(locale: locale) {
request.customizedLanguageModel = configuration
}
let phrases = contextualStringsForRecognition(bias: bias)
if !phrases.isEmpty {
request.contextualStrings = phrases
}
}
// MARK: - Bundle / disk helpers
private static var resourceBundle: Bundle {
// CLM assets ship in the host app bundle, so the keyboard process
// never mmaps the training bin.
Bundle.main
}
static func bundledTrainingAssetURL() -> URL? {
// Prefer flat Bundle.main (iOS HostCLM copy), then legacy subdirs.
let candidates: [URL?] = [
resourceBundle.url(forResource: "OSGKeyboardCLM", withExtension: "bin"),
resourceBundle.url(
forResource: "OSGKeyboardCLM",
withExtension: "bin",
subdirectory: "HostCLM/v1"
),
resourceBundle.url(
forResource: "OSGKeyboardCLM",
withExtension: "bin",
subdirectory: Storage.subdirectory
)
]
return candidates.compactMap { $0 }.first
}
static func bundledManifest() -> BundledManifest? {
let candidates: [URL?] = [
resourceBundle.url(forResource: "compiled-manifest", withExtension: "json"),
resourceBundle.url(
forResource: "compiled-manifest",
withExtension: "json",
subdirectory: "HostCLM/v1"
),
resourceBundle.url(
forResource: "compiled-manifest",
withExtension: "json",
subdirectory: Storage.subdirectory
)
]
guard let manifestURL = candidates.compactMap({ $0 }).first,
let data = try? Data(contentsOf: manifestURL),
let manifest = try? JSONDecoder().decode(BundledManifest.self, from: data)
else {
return nil
}
return manifest
}
static func preparedDirectoryURL() -> URL? {
if let container = FileManager.default.containerURL(
forSecurityApplicationGroupIdentifier: AppGroup.identifier
) {
let directory = container.appendingPathComponent(Storage.subdirectory, isDirectory: true)
try? FileManager.default.createDirectory(at: directory, withIntermediateDirectories: true)
return directory
}
#if os(macOS)
guard let appSupport = FileManager.default.urls(
for: .applicationSupportDirectory,
in: .userDomainMask
).first else {
return nil
}
let directory = appSupport
.appendingPathComponent("OSGKeyboard", isDirectory: true)
.appendingPathComponent(Storage.subdirectory, isDirectory: true)
try? FileManager.default.createDirectory(at: directory, withIntermediateDirectories: true)
return directory
#else
return nil
#endif
}
static func loadCachedConfigurationFromDisk() -> SFSpeechLanguageModel.Configuration? {
guard let manifest = bundledManifest(),
storedFingerprint() == fingerprint(for: manifest),
let preparedDir = preparedDirectoryURL()
else {
return nil
}
let languageModelURL = preparedDir.appendingPathComponent("LM")
let vocabularyURL = preparedDir.appendingPathComponent("Vocab")
let fm = FileManager.default
guard fm.fileExists(atPath: languageModelURL.path),
fm.fileExists(atPath: vocabularyURL.path) else {
return nil
}
return SFSpeechLanguageModel.Configuration(
languageModel: languageModelURL,
vocabulary: vocabularyURL
)
}
static func isChineseLocale(_ locale: Locale) -> Bool {
locale.identifier(.bcp47).lowercased().hasPrefix("zh")
}
private static func fingerprint(for manifest: BundledManifest) -> String {
"\(manifest.identifier)|\(manifest.version)|\(manifest.binBytes)"
}
private static func storedFingerprint() -> String? {
persistenceDefaults.string(forKey: Storage.fingerprintKey)
}
private static func removeItemIfExists(at url: URL) throws {
let fm = FileManager.default
if fm.fileExists(atPath: url.path) {
try fm.removeItem(at: url)
}
}
private static func prepareLanguageModel(
assetURL: URL,
configuration: SFSpeechLanguageModel.Configuration
) async throws {
try await withCheckedThrowingContinuation { (continuation: CheckedContinuation<Void, Error>) in
SFSpeechLanguageModel.prepareCustomLanguageModel(
for: assetURL,
configuration: configuration
) { error in
if let error {
continuation.resume(throwing: error)
} else {
continuation.resume()
}
}
}
}
// MARK: - Retry / backoff
private static var persistenceDefaults: UserDefaults {
AppGroup.defaultsIfAvailable ?? .standard
}
private static func storedAttemptCount() -> Int {
persistenceDefaults.integer(forKey: Storage.attemptCountKey)
}
private static func storedLastFailureAt() -> TimeInterval? {
let value = persistenceDefaults.double(forKey: Storage.lastFailureAtKey)
return value > 0 ? value : nil
}
private static func recordFailure() {
let nextAttempt = storedAttemptCount() + 1
persistenceDefaults.set(nextAttempt, forKey: Storage.attemptCountKey)
persistenceDefaults.set(Date().timeIntervalSince1970, forKey: Storage.lastFailureAtKey)
}
private static func clearRetryState() {
persistenceDefaults.removeObject(forKey: Storage.attemptCountKey)
persistenceDefaults.removeObject(forKey: Storage.lastFailureAtKey)
}
/// Returns false when retry budget is exhausted or backoff has not elapsed.
private static func canRetryAfterFailure() -> Bool {
let attempts = storedAttemptCount()
guard attempts > 0 else { return true }
guard attempts <= Storage.maxRetryAttempts else { return false }
guard let lastFailureAt = storedLastFailureAt() else { return true }
let backoffIndex = min(attempts - 1, Storage.backoffIntervals.count - 1)
let requiredDelay = Storage.backoffIntervals[backoffIndex]
let elapsed = Date().timeIntervalSince1970 - lastFailureAt
return elapsed >= requiredDelay
}
private static func log(_ message: String) {
OSGLog.clm.info("\(message, privacy: .public)")
}
enum PrepareError: LocalizedError {
case missingManifest
case missingTrainingAsset
case missingAppGroupContainer
case missingPreparedArtifacts
case retryBudgetExhausted
var errorDescription: String? {
switch self {
case .missingManifest:
return SharedL10n.string("error.clm.missingManifest")
case .missingTrainingAsset:
return SharedL10n.string("error.clm.missingTrainingAsset")
case .missingAppGroupContainer:
return SharedL10n.string("error.clm.missingAppGroupContainer")
case .missingPreparedArtifacts:
return SharedL10n.string("error.clm.missingPreparedArtifacts")
case .retryBudgetExhausted:
return SharedL10n.string("error.clm.retryBudgetExhausted")
}
}
}
}