Package dev.langchain4j.store.embedding.pgvector

Interface MetadataStorageConfig

All Known Implementing Classes:

DefaultMetadataStorageConfig

public interface MetadataStorageConfig

Metadata configuration.

Method Summary

Modifier and Type	Method	Description
List<String>	columnDefinitions()	SQL definition of metadata field(s) list.
List<String>	indexes()	Metadata Indexes, list of fields to use as index.
String	indexType()	Index Type: BTREE (default) GIN ...
MetadataStorageMode	storageMode()	Metadata storage mode COMBINED_JSON: For dynamic metadata, when you don't know the list of metadata that will be used.

Method Details

storageMode

MetadataStorageMode storageMode()

Metadata storage mode

• COMBINED_JSON: For dynamic metadata, when you don't know the list of metadata that will be used.
• COMBINED_JSONB: Same as JSON, but stored in a binary way. Optimized for query on large dataset.
• COLUMN_PER_KEY: for static metadata, when you know in advance the list of metadata

Returns:

Metadata storage mode

columnDefinitions

List<String> columnDefinitions()

SQL definition of metadata field(s) list. Example:

• COMBINED_JSON: Collections.singletonList("metadata JSON NULL")
• COMBINED_JSONB: Collections.singletonList("metadata JSONB NULL")
• COLUMN_PER_KEY: Arrays.asList("condominium_id uuid null", "user uuid null")

Returns:

list of column definitions

indexes

List<String> indexes()

Metadata Indexes, list of fields to use as index. Example:

• COMBINED_JSON: Collections.singletonList("metadata") or Arrays.asList("(metadata->'key')", "(metadata->'name')", "(metadata->'age')")
• COMBINED_JSONB: Collections.singletonList("metadata") or Arrays.asList("(metadata->'key')", "(metadata->'name')", "(metadata->'age')")
• COLUMN_PER_KEY: Arrays.asList("key", "name", "age")

Returns:

Metadata Indexes list

indexType

String indexType()

Index Type:

• BTREE (default)
• GIN
• ... postgres indexes

Returns:

Index Type