@parqui/core

Framework-agnostic TypeScript core for reading Apache Parquet files and building fast parquet viewers, parquet file explorers, and data inspection tools in browser or desktop apps.

Why @parqui/core

• Fast Apache Parquet metadata and row access in JavaScript/TypeScript
• Built on hyparquet with lazy/random-access reading
• Pipeline utilities for sorting, filtering, grouping, and index building
• Reusable across React, Vue, Angular, desktop, and custom frameworks
• Designed for large datasets and virtualized data grid workflows

Installation

npm install @parqui/core

Quick Start

import { sourceFromUrl, readParquetMetadata, readParquetData } from "@parqui/core";

const source = sourceFromUrl("https://example.com/data/events.parquet");

const metadata = await readParquetMetadata(source);
console.log(metadata.columns.map((c) => c.name));

const data = await readParquetData(source, { offset: 0, limit: 1000 });
console.log(data.rows.length);

Main Exports

• Readers: readParquetMetadata, readParquetData, readColumnValues, readRowsByIndices
• Sources: sourceFromFile, sourceFromBuffer, sourceFromUrl
• Pipeline: buildSortIndex, buildFilterIndex, buildGroups, collectUniqueValues, createEmptyPipeline
• Types: ParquetSource, ParquetMetadata, ParquetData, ParquetRow, ReadOptions

Package Ecosystem

• React component: @parqui/react
• Vue component: @parqui/vue
• Angular component: @parqui/angular

Links

• Website: parqui.dev

License

Business Source License 1.1 (BUSL-1.1). See LICENSE.