Tool

Query Pathway vector store for semantic document retrieval

Retrieve nearest-neighbor documents from a Pathway vector store into LlamaIndex.

Works with pathwayllamaindex

74
Spark score
out of 100
Updated 2 days ago
Version 0.14.23

Add to Favorites

Why it matters

Retrieve semantically relevant documents from a Pathway data indexing pipeline by querying the vector store with natural language text and returning the closest matching neighbors.

Outcomes

What it gets done

01

Connect to Pathway server using host and port configuration

02

Query vector store with natural language text to find similar documents

03

Filter results by metadata criteria to narrow search scope

04

Return top-k nearest neighbor documents for downstream processing

Install

Add it to your toolbox

Run in your project directory:

curl -fsSL https://spark.entire.vc/get/li-reader-readers-pathway | bash

Overview

LlamaIndex Readers Integration: Pathway

A LlamaIndex reader that queries a running Pathway server's vector store for the nearest-neighbor documents to a text query. Use when you already run a Pathway indexing pipeline and want its results as LlamaIndex documents.

What it does

The Pathway Reader is a utility class for retrieving documents from the Pathway data indexing pipeline. It queries the Pathway vector store to get the closest neighbors of a given text query, functioning as a bridge between an already-running Pathway indexing pipeline and LlamaIndex's document format.

PathwayReader is initialized with the host and port of a running Pathway server, and load_data takes a query_text (the text to find the closest neighbors of), a k parameter controlling how many results to return, and an optional metadata_filter to narrow the search further.

When to use - and when NOT to

Use it when you already run a Pathway data indexing pipeline and want its nearest-neighbor search results available as LlamaIndex documents - for example, feeding Pathway-indexed content into a broader LlamaIndex retrieval or query workflow. Do not use it without an already-running Pathway server; the reader is a thin client over Pathway's vector store, not a standalone indexing pipeline itself.

Capabilities

load_data queries a running Pathway server's vector store for the k closest neighbors of a text query, optionally filtered by metadata, and returns the results as LlamaIndex documents.

How to install

pip install llama-index-readers-pathway

Requires a reachable, already-running Pathway server (host and port).

Who it's for

Developers who already operate a Pathway indexing pipeline and want its nearest-neighbor search results integrated into a LlamaIndex retrieval or query workflow.

Source README

LlamaIndex Readers Integration: Pathway

Overview

Pathway Reader is a utility class for retrieving documents from the Pathway data indexing pipeline. It queries the Pathway vector store to get the closest neighbors of a given text query.

Installation

You can install Pathway Reader via pip:

pip install llama-index-readers-pathway

Usage

from llama_index.readers.pathway import PathwayReader

### Initialize PathwayReader with the URI and port of the Pathway server
reader = PathwayReader(host="<Pathway Host>", port="<Port>")

### Load data from Pathway
documents = reader.load_data(
    query_text="<Query Text>",  # The text to get the closest neighbors of
    k=4,  # Number of results to return
    metadata_filter="<Metadata Filter>",  # Filter to be applied
)

This loader is designed to be used as a way to load data into
LlamaIndex.

FAQ

Common questions

Discussion

Questions & comments · 0

Sign In Sign in to leave a comment.