Is PGVector store node only ?

[Stefenyo]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

import { OpenAIEmbeddings } from "@langchain/openai";
import {
  DistanceStrategy,
  PGVectorStore,
} from "@langchain/community/vectorstores/pgvector";
import { PoolConfig } from "pg";

// First, follow set-up instructions at
// https://js.langchain.com/docs/modules/indexes/vector_stores/integrations/pgvector

const config = {
  postgresConnectionOptions: {
    type: "postgres",
    host: "127.0.0.1",
    port: 5433,
    user: "myuser",
    password: "ChangeMe",
    database: "api",
  } as PoolConfig,
  tableName: "testlangchain",
  columns: {
    idColumnName: "id",
    vectorColumnName: "vector",
    contentColumnName: "content",
    metadataColumnName: "metadata",
  },
  // supported distance strategies: cosine (default), innerProduct, or euclidean
  distanceStrategy: "cosine" as DistanceStrategy,
};

const pgvectorStore = await PGVectorStore.initialize(
  new OpenAIEmbeddings(),
  config
);

export {pgvectorStore}

Description

I'm trying to use PGVector as a retriever but my app keeps crashing with "process is not defined" in console. It isn't listed as "node-only" in the docs, but I'm guessing this isn't the case...

System Info

Platform: Mac
Node: v18.20.0

[dosubot[bot]]

The error "process is not defined" suggests that your code, which relies on Node.js-specific features, is being executed in an environment where the Node.js process object is not available, such as a browser. To prevent this error, ensure that any Node.js-specific code is only run in a Node.js environment. You can use a conditional check to safeguard this:

if (typeof process !== "undefined" && process.versions && process.versions.node) {
    // Place your Node.js-specific code here
}

This snippet checks for the presence of the process object and its versions.node property, which confirms that the code is running in a Node.js environment.

Regarding PGVector, it is primarily designed for server-side use with PostgreSQL databases, particularly in Node.js environments. Its functionality is centered around supporting efficient vector operations within these databases. While PGVector is not limited to Node.js and can be used in any server-side environment that can connect to PostgreSQL, its direct use in client-side environments like browsers is not feasible without a server-side intermediary to handle database interactions.

For your specific case, ensure that your application's architecture correctly distinguishes between server-side and client-side code, especially when dealing with database operations and Node.js-specific features.

Details

Sources

• langchain-core/src/utils/env.ts
• libs/langchain-community/src/vectorstores/pgvector.ts
• environment_tests/test-exports-esbuild/src/entrypoints.js
• langchain/package.json
• examples/src/indexes/vector_stores/pgvector_vectorstore/pgvector_pool.ts
• examples/src/indexes/vector_stores/pgvector_vectorstore/pgvector.ts

About Dosu
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

To continue the conversation, mention @dosu.

[MJDeligan]

PGVector uses pg-node to query the DB, so it 's probably Node only. Don't know why it isn't marked as such in the documentation. I'm pretty sure @jacoblee93 has also referenced it being node-only before.

For what it's worth, connecting to a DB in the browser isn't something you'd normally want to do.

[Stefenyo]

Ty!