Class PrismaVectorStore<TModel, TModelName, TSelectModel, TFilterModel>

A specific implementation of the VectorStore class that is designed to work with Prisma. It provides methods for adding models, documents, and vectors, as well as for performing similarity searches.

Type Parameters

TModel extends Record<string, unknown>
TModelName extends string
TSelectModel extends ModelColumns<TModel>
TFilterModel extends PrismaSqlFilter<TModel>

Hierarchy

VectorStore
- PrismaVectorStore

Constructors

constructor

new PrismaVectorStore<TModel, TModelName, TSelectModel, TFilterModel>(embeddings, config): PrismaVectorStore<TModel, TModelName, TSelectModel, TFilterModel>
Type Parameters
- TModel extends Record<string, unknown>
- TModelName extends string
- TSelectModel extends ModelColumns<TModel>
- TFilterModel extends PrismaSqlFilter<TModel>
Parameters
- embeddings: EmbeddingsInterface<number[]>
- config: {
      columns: TSelectModel;
      db: PrismaClient;
      filter?: TFilterModel;
      prisma: PrismaNamespace;
      tableName: TModelName;
      vectorColumnName: string;
  }
  - columns: TSelectModel
  - db: PrismaClient
  - Optionalfilter?: TFilterModel
  - prisma: PrismaNamespace
  - tableName: TModelName
  - vectorColumnName: string
Returns PrismaVectorStore<TModel, TModelName, TSelectModel, TFilterModel>
Overrides VectorStore.constructor
- Defined in libs/langchain-community/src/vectorstores/prisma.ts:141

Properties

FilterType

FilterType: TFilterModel

`Protected`Prisma

Prisma: PrismaNamespace

contentColumn

contentColumn: keyof TModel & string

`Protected`db

db: PrismaClient

embeddings

embeddings: EmbeddingsInterface<number[]>

Embeddings interface for generating vector embeddings from text queries, enabling vector-based similarity searches.

`Optional`filter

filter?: TFilterModel

idColumn

idColumn: keyof TModel & string

`Protected`selectColumns

selectColumns: string[]

`Protected`tableName

tableName: string

`Protected`vectorColumnName

vectorColumnName: string

`Static`ContentColumn

ContentColumn: typeof ContentColumnSymbol = ContentColumnSymbol

`Static`IdColumn

IdColumn: typeof IdColumnSymbol = IdColumnSymbol

Methods

addDocuments

addDocuments(documents): Promise<void>
Adds the specified documents to the store.
Parameters
- documents: Document<TModel>[]
  The documents to add.
Returns Promise<void>
A promise that resolves when the documents have been added.
Overrides VectorStore.addDocuments
- Defined in libs/langchain-community/src/vectorstores/prisma.ts:287

addModels

addModels(models): Promise<void>
Adds the specified models to the store.
Parameters
- models: TModel[]
  The models to add.
Returns Promise<void>
A promise that resolves when the models have been added.
- Defined in libs/langchain-community/src/vectorstores/prisma.ts:271

addVectors

addVectors(vectors, documents): Promise<void>
Adds the specified vectors to the store.
Parameters
- vectors: number[][]
  The vectors to add.
- documents: Document<TModel>[]
  The documents associated with the vectors.
Returns Promise<void>
A promise that resolves when the vectors have been added.
Overrides VectorStore.addVectors
- Defined in libs/langchain-community/src/vectorstores/prisma.ts:301

asRetriever

asRetriever(kOrFields?, filter?, callbacks?, tags?, metadata?, verbose?): VectorStoreRetriever<PrismaVectorStore<TModel, TModelName, TSelectModel, TFilterModel>>
Creates a VectorStoreRetriever instance with flexible configuration options.
Parameters
- OptionalkOrFields: number | Partial<VectorStoreRetrieverInput<PrismaVectorStore<TModel, TModelName, TSelectModel, TFilterModel>>>
  If a number is provided, it sets the k parameter (number of items to retrieve).
  - If an object is provided, it should contain various configuration options.
- Optionalfilter: TFilterModel
  Optional filter criteria to limit the items retrieved based on the specified filter type.
- Optionalcallbacks: Callbacks
  Optional callbacks that may be triggered at specific stages of the retrieval process.
- Optionaltags: string[]
  Tags to categorize or label the VectorStoreRetriever. Defaults to an empty array if not provided.
- Optionalmetadata: Record<string, unknown>
  Additional metadata as key-value pairs to add contextual information for the retrieval process.
- Optionalverbose: boolean
  If true, enables detailed logging for the retrieval process. Defaults to false.
Returns VectorStoreRetriever<PrismaVectorStore<TModel, TModelName, TSelectModel, TFilterModel>>
- A configured VectorStoreRetriever instance based on the provided parameters.
Example
Basic usage with a k value:
```
const retriever = myVectorStore.asRetriever(5);
```
Usage with a configuration object:
```
const retriever = myVectorStore.asRetriever({
  k: 10,
  filter: myFilter,
  tags: ['example', 'test'],
  verbose: true,
  searchType: 'mmr',
  searchKwargs: { alpha: 0.5 },
});
```
Inherited from VectorStore.asRetriever
- Defined in langchain-core/dist/vectorstores.d.ts:596

buildSqlFilterStr

buildSqlFilterStr(filter?): null | Sql
Parameters
- Optionalfilter: TFilterModel
Returns null | Sql
- Defined in libs/langchain-community/src/vectorstores/prisma.ts:417

delete

delete(_params?): Promise<void>
Deletes documents from the vector store based on the specified parameters.
Parameters
- Optional_params: Record<string, any>
  Flexible key-value pairs defining conditions for document deletion.
Returns Promise<void>
A promise that resolves once the deletion is complete.
Inherited from VectorStore.delete
- Defined in langchain-core/dist/vectorstores.d.ts:477

`Optional`maxMarginalRelevanceSearch

maxMarginalRelevanceSearch(query, options, _callbacks): Promise<DocumentInterface<Record<string, any>>[]>
Return documents selected using the maximal marginal relevance. Maximal marginal relevance optimizes for similarity to the query AND diversity among selected documents.
Parameters
- query: string
  Text to look up documents similar to.
- options: MaxMarginalRelevanceSearchOptions<TFilterModel>
- _callbacks: undefined | Callbacks
Returns Promise<DocumentInterface<Record<string, any>>[]>
- List of documents selected by maximal marginal relevance.
Inherited from VectorStore.maxMarginalRelevanceSearch
- Defined in langchain-core/dist/vectorstores.d.ts:527

similaritySearch

similaritySearch(query, k?, filter?): Promise<Document<SimilarityModel<TModel, TSelectModel>>[]>
Performs a similarity search with the specified query.
Parameters
- query: string
  The query to use for the similarity search.
- k: number = 4
  The number of results to return.
- filter: undefined | TFilterModel = undefined
Returns Promise<Document<SimilarityModel<TModel, TSelectModel>>[]>
A promise that resolves with the search results.
Overrides VectorStore.similaritySearch
- Defined in libs/langchain-community/src/vectorstores/prisma.ts:328

similaritySearchVectorWithScore

similaritySearchVectorWithScore(query, k, filter?): Promise<[Document<SimilarityModel<TModel, TSelectModel>>, number][]>
Performs a similarity search with the specified vector and returns the results along with their scores.
Parameters
- query: number[]
  The vector to use for the similarity search.
- k: number
  The number of results to return.
- Optionalfilter: TFilterModel
  The filter to apply to the results.
Returns Promise<[Document<SimilarityModel<TModel, TSelectModel>>, number][]>
A promise that resolves with the search results and their scores.
Overrides VectorStore.similaritySearchVectorWithScore
- Defined in libs/langchain-community/src/vectorstores/prisma.ts:367

similaritySearchWithScore

similaritySearchWithScore(query, k?, filter?): Promise<[DocumentInterface<Record<string, any>>, number][]>
Performs a similarity search with the specified query and returns the results along with their scores.
Parameters
- query: string
  The query to use for the similarity search.
- Optionalk: number
  The number of results to return.
- Optionalfilter: TFilterModel
  The filter to apply to the results.
Returns Promise<[DocumentInterface<Record<string, any>>, number][]>
A promise that resolves with the search results and their scores.
Overrides VectorStore.similaritySearchWithScore
- Defined in libs/langchain-community/src/vectorstores/prisma.ts:351

toJSON

toJSON(): Serialized
Returns Serialized
Inherited from VectorStore.toJSON
- Defined in langchain-core/dist/load/serializable.d.ts:74

toJSONNotImplemented

toJSONNotImplemented(): SerializedNotImplemented
Returns SerializedNotImplemented
Inherited from VectorStore.toJSONNotImplemented
- Defined in langchain-core/dist/load/serializable.d.ts:75

`Static`fromDocuments

fromDocuments(docs, embeddings, dbConfig): Promise<DefaultPrismaVectorStore>
Creates a new PrismaVectorStore from the specified documents.
Parameters
- docs: Document<Record<string, any>>[]
  The documents to use to create the store.
- embeddings: EmbeddingsInterface<number[]>
  The embeddings to use.
- dbConfig: {
      columns: ModelColumns<Record<string, unknown>>;
      db: PrismaClient;
      prisma: PrismaNamespace;
      tableName: string;
      vectorColumnName: string;
  }
  The database configuration.
  - columns: ModelColumns<Record<string, unknown>>
  - db: PrismaClient
  - prisma: PrismaNamespace
  - tableName: string
  - vectorColumnName: string
Returns Promise<DefaultPrismaVectorStore>
A promise that resolves with the new PrismaVectorStore.
Overrides VectorStore.fromDocuments
- Defined in libs/langchain-community/src/vectorstores/prisma.ts:507

`Static`fromTexts

fromTexts(texts, metadatas, embeddings, dbConfig): Promise<DefaultPrismaVectorStore>
Creates a new PrismaVectorStore from the specified texts.
Parameters
- texts: string[]
  The texts to use to create the store.
- metadatas: object[]
  The metadata for the texts.
- embeddings: EmbeddingsInterface<number[]>
  The embeddings to use.
- dbConfig: {
      columns: ModelColumns<Record<string, unknown>>;
      db: PrismaClient;
      prisma: PrismaNamespace;
      tableName: string;
      vectorColumnName: string;
  }
  The database configuration.
  - columns: ModelColumns<Record<string, unknown>>
  - db: PrismaClient
  - prisma: PrismaNamespace
  - tableName: string
  - vectorColumnName: string
Returns Promise<DefaultPrismaVectorStore>
A promise that resolves with the new PrismaVectorStore.
Overrides VectorStore.fromTexts
- Defined in libs/langchain-community/src/vectorstores/prisma.ts:475

`Static`withModel

withModel<TModel>(db): {
    create: (<TPrisma, TColumns, TFilters>(embeddings: EmbeddingsInterface<number[]>, config: {
        columns: TColumns;
        filter?: TFilters;
        prisma: TPrisma;
        tableName: keyof TPrisma["ModelName"] & string;
        vectorColumnName: string;
    }) => PrismaVectorStore<TModel, ModelName, TColumns, TFilters>);
    fromDocuments: (<TPrisma, TColumns, TFilters>(docs: Document<TModel>[], embeddings: EmbeddingsInterface<number[]>, dbConfig: {
        columns: TColumns;
        prisma: TPrisma;
        tableName: keyof TPrisma["ModelName"] & string;
        vectorColumnName: string;
    }) => Promise<PrismaVectorStore<TModel, ModelName, TColumns, TFilters>>);
    fromTexts: (<TPrisma, TColumns>(texts: string[], metadatas: TModel[], embeddings: EmbeddingsInterface<number[]>, dbConfig: {
        columns: TColumns;
        prisma: TPrisma;
        tableName: keyof TPrisma["ModelName"] & string;
        vectorColumnName: string;
    }) => Promise<DefaultPrismaVectorStore>);
}
Creates a new PrismaVectorStore with the specified model.
Type Parameters
- TModel extends Record<string, unknown>
Parameters
- db: PrismaClient
  The PrismaClient instance.
Returns {
    create: (<TPrisma, TColumns, TFilters>(embeddings: EmbeddingsInterface<number[]>, config: {
        columns: TColumns;
        filter?: TFilters;
        prisma: TPrisma;
        tableName: keyof TPrisma["ModelName"] & string;
        vectorColumnName: string;
    }) => PrismaVectorStore<TModel, ModelName, TColumns, TFilters>);
    fromDocuments: (<TPrisma, TColumns, TFilters>(docs: Document<TModel>[], embeddings: EmbeddingsInterface<number[]>, dbConfig: {
        columns: TColumns;
        prisma: TPrisma;
        tableName: keyof TPrisma["ModelName"] & string;
        vectorColumnName: string;
    }) => Promise<PrismaVectorStore<TModel, ModelName, TColumns, TFilters>>);
    fromTexts: (<TPrisma, TColumns>(texts: string[], metadatas: TModel[], embeddings: EmbeddingsInterface<number[]>, dbConfig: {
        columns: TColumns;
        prisma: TPrisma;
        tableName: keyof TPrisma["ModelName"] & string;
        vectorColumnName: string;
    }) => Promise<DefaultPrismaVectorStore>);
}
An object with create, fromTexts, and fromDocuments methods.
- create: (<TPrisma, TColumns, TFilters>(embeddings: EmbeddingsInterface<number[]>, config: {
      columns: TColumns;
      filter?: TFilters;
      prisma: TPrisma;
      tableName: keyof TPrisma["ModelName"] & string;
      vectorColumnName: string;
  }) => PrismaVectorStore<TModel, ModelName, TColumns, TFilters>)
  - - <TPrisma, TColumns, TFilters>(embeddings, config): PrismaVectorStore<TModel, ModelName, TColumns, TFilters>
    - Type Parameters
      TPrisma extends PrismaNamespace
      TColumns extends ModelColumns<TModel>
      TFilters extends PrismaSqlFilter<TModel>
      Parameters
      embeddings: EmbeddingsInterface<number[]>
      config: {
          columns: TColumns;
          filter?: TFilters;
          prisma: TPrisma;
          tableName: keyof TPrisma["ModelName"] & string;
          vectorColumnName: string;
      }
      columns: TColumns
      Optionalfilter?: TFilters
      prisma: TPrisma
      tableName: keyof TPrisma["ModelName"] & string
      vectorColumnName: string
      Returns PrismaVectorStore<TModel, ModelName, TColumns, TFilters>
- fromDocuments: (<TPrisma, TColumns, TFilters>(docs: Document<TModel>[], embeddings: EmbeddingsInterface<number[]>, dbConfig: {
      columns: TColumns;
      prisma: TPrisma;
      tableName: keyof TPrisma["ModelName"] & string;
      vectorColumnName: string;
  }) => Promise<PrismaVectorStore<TModel, ModelName, TColumns, TFilters>>)
  - - <TPrisma, TColumns, TFilters>(docs, embeddings, dbConfig): Promise<PrismaVectorStore<TModel, ModelName, TColumns, TFilters>>
    - Type Parameters
      TPrisma extends PrismaNamespace
      TColumns extends ModelColumns<TModel>
      TFilters extends PrismaSqlFilter<TModel>
      Parameters
      docs: Document<TModel>[]
      embeddings: EmbeddingsInterface<number[]>
      dbConfig: {
          columns: TColumns;
          prisma: TPrisma;
          tableName: keyof TPrisma["ModelName"] & string;
          vectorColumnName: string;
      }
      columns: TColumns
      prisma: TPrisma
      tableName: keyof TPrisma["ModelName"] & string
      vectorColumnName: string
      Returns Promise<PrismaVectorStore<TModel, ModelName, TColumns, TFilters>>
- fromTexts: (<TPrisma, TColumns>(texts: string[], metadatas: TModel[], embeddings: EmbeddingsInterface<number[]>, dbConfig: {
      columns: TColumns;
      prisma: TPrisma;
      tableName: keyof TPrisma["ModelName"] & string;
      vectorColumnName: string;
  }) => Promise<DefaultPrismaVectorStore>)
  - - <TPrisma, TColumns>(texts, metadatas, embeddings, dbConfig): Promise<DefaultPrismaVectorStore>
    - Type Parameters
      TPrisma extends PrismaNamespace
      TColumns extends ModelColumns<TModel>
      Parameters
      texts: string[]
      metadatas: TModel[]
      embeddings: EmbeddingsInterface<number[]>
      dbConfig: {
          columns: TColumns;
          prisma: TPrisma;
          tableName: keyof TPrisma["ModelName"] & string;
          vectorColumnName: string;
      }
      columns: TColumns
      prisma: TPrisma
      tableName: keyof TPrisma["ModelName"] & string
      vectorColumnName: string
      Returns Promise<DefaultPrismaVectorStore>
- Defined in libs/langchain-community/src/vectorstores/prisma.ts:186

Class PrismaVectorStore<TModel, TModelName, TSelectModel, TFilterModel>

Type Parameters

Hierarchy

Index

Constructors

Properties

Methods

Constructors

constructor

Type Parameters

Parameters

columns: TSelectModel

db: PrismaClient

Optionalfilter?: TFilterModel

prisma: PrismaNamespace

tableName: TModelName

vectorColumnName: string

Returns PrismaVectorStore<TModel, TModelName, TSelectModel, TFilterModel>

Properties

FilterType

ProtectedPrisma

contentColumn

Protecteddb

embeddings

Optionalfilter

idColumn

ProtectedselectColumns

ProtectedtableName

ProtectedvectorColumnName

StaticContentColumn

StaticIdColumn

Methods

addDocuments

Parameters

Returns Promise<void>

addModels

Parameters

Returns Promise<void>

addVectors

Parameters

Returns Promise<void>

asRetriever

Parameters

Returns VectorStoreRetriever<PrismaVectorStore<TModel, TModelName, TSelectModel, TFilterModel>>

Example

buildSqlFilterStr

Parameters

Returns null | Sql

delete

Parameters

Returns Promise<void>

OptionalmaxMarginalRelevanceSearch

Parameters

Returns Promise<DocumentInterface<Record<string, any>>[]>

similaritySearch

Parameters

Returns Promise<Document<SimilarityModel<TModel, TSelectModel>>[]>

similaritySearchVectorWithScore

Parameters

Returns Promise<[Document<SimilarityModel<TModel, TSelectModel>>, number][]>

similaritySearchWithScore

Parameters

Returns Promise<[DocumentInterface<Record<string, any>>, number][]>

toJSON

Returns Serialized

toJSONNotImplemented

Returns SerializedNotImplemented

StaticfromDocuments

Parameters

columns: ModelColumns<Record<string, unknown>>

db: PrismaClient

prisma: PrismaNamespace

tableName: string

vectorColumnName: string

Returns Promise<DefaultPrismaVectorStore>

StaticfromTexts

Parameters

columns: ModelColumns<Record<string, unknown>>

db: PrismaClient

prisma: PrismaNamespace

`Optional`filter?: TFilterModel

`Protected`Prisma

`Protected`db

`Optional`filter

`Protected`selectColumns

`Protected`tableName

`Protected`vectorColumnName

`Static`ContentColumn

`Static`IdColumn

`Optional`maxMarginalRelevanceSearch

`Static`fromDocuments

`Static`fromTexts

`Static`withModel

create: (<TPrisma, TColumns, TFilters>(embeddings: EmbeddingsInterface<number[]>, config: {
columns: TColumns;
filter?: TFilters;
prisma: TPrisma;
tableName: keyof TPrisma["ModelName"] & string;
vectorColumnName: string;
}) => PrismaVectorStore<TModel, ModelName, TColumns, TFilters>)

`Optional`filter?: TFilters

fromTexts: (<TPrisma, TColumns>(texts: string[], metadatas: TModel[], embeddings: EmbeddingsInterface<number[]>, dbConfig: {
columns: TColumns;
prisma: TPrisma;
tableName: keyof TPrisma["ModelName"] & string;
vectorColumnName: string;
}) => Promise<DefaultPrismaVectorStore>)