addDocument method

@override Future<void> addDocument({ required String id, required String content, required List<double> embedding, String? metadata, })

Add document with embedding to vector store

Auto-dimension detection:

• First document sets the dimension (e.g., 768D)
• Subsequent documents must match that dimension

INSERT OR REPLACE behavior:

• If id exists, document is updated (replace)
• If id is new, document is inserted

Parameters:

• id: Unique document identifier (PRIMARY KEY)
• content: Document text content
• embedding: Vector embedding (dimension auto-detected from first add)
• metadata: Optional JSON string metadata

Throws:

• StateError if not initialized
• ArgumentError if embedding dimension doesn't match existing documents Throws VectorStoreException if the store is initialized but holds on-disk data it could not read — an implementation must refuse rather than report an unreadable corpus as an empty one.

Implementation

@override
Future<void> addDocument({
  required String id,
  required String content,
  required List<double> embedding,
  String? metadata,
}) async {
  throw UnimplementedError(
    'WebSqliteVectorStore is only available on web; use SqliteVectorStore',
  );
}