THE FILE FIELD GUIDE

.feather

Feather Data

Data & Science

Feather stores typed column data. V2 uses the Arrow IPC file format; legacy V1 differs. Establish writer/version instead of inferring it from the suffix. New Arrow documentation deprecates V1 support, so review compatibility for long-term use.

What it is used for: Storage and exchange of typed tabular data
What the file contains: Binary data

All sections, questions and sources

Content reviewed：2026-10-12 · Editor：Kai

Opening steps

Choose your environment or purpose to see its steps.

Inspect Feather version and types

V1 and V2 have different requirements.

Identify the writer’s Arrow/library version.
Inspect column types and rows with a compatible reader.
Compare representative strings, times and missing values.

Read into Python or R

Verify support in the reading environment.

Use supported Feather or V2 IPC reading.
Read required columns as a table.
Compare order, types and rows with source metadata.

Prepare an archival or interchange copy

Inspect legacy V1 compatibility first.

Record schema/counts in an environment that reads the source.
Export separately to supported V2/IPC or another format.
Reopen at the destination and compare types, values and nulls.

Menu labels and support vary by app, OS and version. Preserve the original before making changes.

Troubleshooting by symptom

Old reader fails: check format version and compression support.
Legacy V1 unsupported: preserve originals and rewrite using a trusted compatible reader.
Changed timestamps/categories: validate time zones and dictionary-type round trips.
Legacy file fails: inspect V1/V2 and reader-version support.
Compression error: inspect support for the file’s compression.
Types change in a data frame: compare conversion and missing-value semantics.

Feather → Parquet / CSV / Arrow IPC

Read with a compatible tool after identifying version/schema, then export.
Compare dimensions, types, nulls, categories and times.
Retain legacy source and compare rows, types and missing values after reopening.

Renaming V1 does not make IPC; CSV requires separate type/metadata documentation.

Distinguish legacy V1 and V2

Current Arrow documentation deprecates V1; verify a working reader before migrating a copy.

Source for these steps：Apache Software Foundation · Feather File Format / Apache Software Foundation · Streaming, Serialization, and IPC

Inspect data-frame conversion differences

Visible columns do not prove identical types; compare dates, categories, nulls and integer precision.

Record required counts/types and validate by reopening at the destination.

Source for these steps：Apache Software Foundation · Feather File Format / Apache Software Foundation · Streaming, Serialization, and IPC

Related-format differences

Related-format differences
Format	Main role	What to check
Feather V1	Legacy Feather format	Reader version
Feather V2	Arrow IPC file format	Compression/types
Parquet	Other columnar storage	Types/dataset scope

Handling and security

Document format/library versions; avoid unknown converters and review personal data.

Questions and answers

Is all Feather IPC?

V2 uses IPC files; V1 differs.

Is it always compressed?

Depends on version/options.

Is it Python-only?

Compatible tools in R and other ecosystems can use it.

When should I migrate legacy V1?

Use a working reader to migrate a copy, validate counts/types/values and retain provenance.

Does renaming upgrade V1 to V2?

Read V1 with compatible software and rewrite it; renaming does not upgrade structure.

Format facts: name, steward and MIME type

Formal name: Feather V1 / Feather V2 (Arrow IPC)
Steward / specification: Apache Software Foundation
MIME type: Depends on the producer or server; an extension alone does not establish a MIME type.

Primary and official sources

Project documentationFeather File Format
Explains Feather V2, Arrow IPC, compression and legacy format.
Apache Software Foundation · Link checked：2026-10-12
Project documentationStreaming, Serialization, and IPC
Distinguishes IPC stream and file readers.
Apache Software Foundation · Link checked：2026-10-12