Registry / data / hyparquet-writer

hyparquet-writer

library0.15.6jsnpmunverified

Hyparquet Writer is a JavaScript library for writing Apache Parquet files, focused on being lightweight and fast. Current stable version is 0.15.6, with active development on GitHub under the hyparam organization. It is a companion to the hyparquet reader library. Key differentiators: pure JavaScript with no native dependencies, supports a wide range of Parquet logical types (STRING, JSON, TIMESTAMP, UUID, FLOAT16, GEOMETRY, etc.), automatic type inference, and optional SNAPPY compression. It outputs Parquet files compatible with the Parquet specification.

data devops

npm install hyparquet-writer

github ↗package ↗homepage ↗

INSTALL

IMPORT

SIG · HYPARQUET-WRITER

hyparquet-writer

datajavascriptv0.15.6

harness data pending

Install & Compatibility

Where this runs

No compatibility data collected yet for this library.

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

parquetWriteBuffer

✓ import { parquetWriteBuffer } from 'hyparquet-writer'

✗ const { parquetWriteBuffer } = require('hyparquet-writer')

Package is ESM-only; CommonJS require() will fail unless using dynamic import()

parquetWriteFile

✓ import { parquetWriteFile } from 'hyparquet-writer'

✗ import { parquetWriteFile } from 'hyparquet-writer/src'

Only top-level exports are supported; don't use subpath imports

ParquetWriteOptions

✓ import { ParquetWriteOptions } from 'hyparquet-writer'

TypeScript type; can be imported even if not used at runtime

Demonstrates writing a Parquet file to an ArrayBuffer with two typed columns using parquetWriteBuffer.

import { parquetWriteBuffer } from 'hyparquet-writer'; // Create an ArrayBuffer containing a Parquet file with two columns const arrayBuffer = parquetWriteBuffer({ columnData: [ { name: 'name', data: ['Alice', 'Bob', 'Charlie'], type: 'STRING' }, { name: 'age', data: [25, 30, 35], type: 'INT32' }, ], }); // The buffer can be written to disk or sent over the network const uint8Array = new Uint8Array(arrayBuffer);

Debug

Known issues

breakingIn version 0.15.0, the API changed from exporting a default function to named exports. Previously you could import parquetWriteBuffer as default.

fix

Use named imports: import { parquetWriteBuffer } from 'hyparquet-writer' instead of import parquetWriteBuffer from 'hyparquet-writer'.

affects: <0.15.0

gotchaThe package is ESM-only and cannot be require()'d directly. Attempting const { parquetWriteBuffer } = require('hyparquet-writer') throws an error.

fix

Use import or dynamic import: const { parquetWriteBuffer } = await import('hyparquet-writer')

affects: >=0.15.0

gotchaType inference may guess wrong types if data contains mixed types or is empty. Always provide explicit type to avoid runtime errors.

fix

Specify the 'type' property for each column in columnData. Supported types: BOOLEAN, INT32, INT64, FLOAT, DOUBLE, BYTE_ARRAY, STRING, JSON, TIMESTAMP, UUID, FLOAT16, GEOMETRY, GEOGRAPHY, VARIANT.

affects: >=0.1.0

deprecatedThe 'version' option in ParquetWriteOptions is deprecated and may be removed in future versions.

fix

Remove the 'version' option from ParquetWriteOptions.

affects: >=0.14.0

Errors

Common errors & fixes

TypeError: hyparquet_writer_1.default is not a function

Incorrect default import when using an older version or outdated examples.

fix

Use a named import: import { parquetWriteBuffer } from 'hyparquet-writer'

Error [ERR_REQUIRE_ESM]: require() of ES Module

Using CommonJS require() on an ESM package.

fix

Use dynamic import: const { parquetWriteBuffer } = await import('hyparquet-writer')

RangeError: Invalid typed array length

Passing data that cannot be serialized to the given type (e.g., string in INT32 column).

fix

Ensure data type matches the specified or inferred type. Use JSON.stringify for JSON columns.

Upgrade

Version history

0.15.6latest on npm

Audit

Dependencies

snappyjsrequiredSNAPPY compression codec (included as dependency)

Agent activity

5 hits · last 30 days

node

Resources

githubhttps://github.com/hyparam/hyparquet-writer ↗

packagehyparquet-writer ↗

homepagehttps://hyperparam.app ↗