PreferencesKeys class - preferences_keys library

Properties

hashCode → int: The hash code for this object.
no setterinherited
runtimeType → Type: A representation of the runtime type of the object.
no setterinherited

Methods

noSuchMethod(Invocation invocation) → dynamic: Invoked when a nonexistent method or property is accessed.
inherited
toString() → String: A string representation of this object.
inherited

Operators

operator ==(Object other) → bool: The equality operator.
inherited

Static Methods

bundledPath(String filename) → String: Get key for bundled file path mapping Format: 'bundled_path_{filename}'
externalPath(String filename) → String: Get key for external file path mapping Format: 'external_path_{filename}'

Constants

activeEmbeddingFilename → const String: Filename of the currently active embedding model.
activeEmbeddingSource → const String: Same encoding as activeInferenceSource, for the embedding model file.
activeEmbeddingTokenizerFilename → const String: Filename of the currently active embedding tokenizer.
activeEmbeddingTokenizerSource → const String: Same encoding as activeInferenceSource, for the embedding tokenizer.
activeInferenceFilename → const String: Filename of the currently active inference model. Required because the "installed" filename key (installedModelFileName) is legacy-only — the new multi-model system tracks installs through installedModels, not a single filename, so we need a dedicated active-pointer.
activeInferenceFileType → const String: ModelFileType.name of the currently active inference model.
activeInferenceModelType → const String: ModelType.name of the currently active inference model. Read on FlutterGemma.initialize() together with activeInferenceFileType and installedModelFileName to rehydrate _activeInferenceModel.
activeInferenceSource → const String: Encoded source for the active inference model. Format: <kind>|<value> where kind ∈ {network,asset,bundled} and value is the URL / asset path / bundle resource name. file is not encoded — Mobile uses a resolved FileSource(filePath) reconstructed from activeInferenceFilename directly.
embeddingModelFile → const String: Legacy: Single embedding model filename
embeddingTokenizerFile → const String: Legacy: Single tokenizer filename
installedEmbeddingModels → const String: List<String> of installed embedding model files
installedLoraFileName → const String: Legacy: Single LoRA filename
installedLoras → const String: List<String> of installed LoRA files
installedModelFileName → const String: Legacy: Single inference model filename
installedModels → const String: List<String> of installed inference model files
installedTokenizers → const String: List<String> of installed tokenizer files
webCacheLastCleanup → const String: Last cache cleanup timestamp
webCacheMetadataPrefix → const String: Prefix for web cache metadata keys Format: 'web_cache_{url.hashCode}'
webCachePersistentGranted → const String: Whether persistent storage was granted