lumen-ai/com.subhashh.lumen.ai/EmbeddingStore

EmbeddingStore

class EmbeddingStore : SQLiteOpenHelper(source)

Smart Search's index for one model: which library items are done (at which version) and their vectors. A photo has one vector; a video one per kept frame, with the frame's time. One file per model, in app-private storage that isn't backed up, since vectors from different models can't be mixed and are cheap to rebuild.

Types

object Companion

Properties

open val databaseName: String?

readableDatabase

open val readableDatabase: SQLiteDatabase?

var version: Int

Bumped on every write, so a search cache knows its copy is stale.

writableDatabase

open val writableDatabase: SQLiteDatabase?

Functions

open fun close()

fun count(): Int

fun indexed(): Map<Long, Long>

Item id → the modified it was indexed at.

fun loadAll(): Embeddings

Everything, for searching.

open fun onConfigure(db: SQLiteDatabase?)

open override fun onCreate(db: SQLiteDatabase)

open fun onDowngrade(db: SQLiteDatabase?, oldVersion: Int, newVersion: Int)

open fun onOpen(db: SQLiteDatabase?)

open override fun onUpgrade(db: SQLiteDatabase, oldVersion: Int, newVersion: Int)

fun put(mediaId: Long, modified: Long, frames: List<Pair<Long, FloatArray>>)

Replaces whatever mediaId had.

fun remove(mediaIds: Collection<Long>)

setIdleConnectionTimeout

open fun ~~setIdleConnectionTimeout~~(idleConnectionTimeoutMs: Long)

setLookasideConfig

open fun setLookasideConfig(slotSize: Int, slotCount: Int)

open fun setOpenParams(openParams: SQLiteDatabase.OpenParams)

setWriteAheadLoggingEnabled

open fun setWriteAheadLoggingEnabled(enabled: Boolean)

fun vectorsOf(mediaId: Long): List<FloatArray>

One item's vectors: a photo's one, or a video's frames.