Browser Kitty
日本語
data

Arrow Viewer

Inspect Apache Arrow IPC File / Stream schema, record batches, dictionaries, metadata, and data locally.

View on GitHub
Local processing
About your filesFiles and content selected in this tool are processed in your browser and are not uploaded to a Browser Kitty backend.
Operation demo See the basic workflow in a short video.

About this tool

Arrow Viewer is a read-only browser tool for inspecting Apache Arrow IPC File / Stream structure and data. It reads .arrow / .arrows / .ipc files, detects the IPC form, and shows schema, metadata version, endianness, and footer / header details.

You can inspect row counts and metadata/body sizes for record batches plus dictionary information, while data browsing reads only batches needed by the current page. Primitive, timestamp, list/struct, and dictionary-encoded values are supported where implemented.

Good for

  • Determine whether an Arrow IPC input is File or Stream and inspect schema / record-batch layout
  • Inspect dictionary encoding and metadata in the browser
  • Preview part of a larger Arrow IPC input without expanding everything and save the current page as CSV

What it can do

Key capabilities available in this tool.

How to use

  1. Add one or more .arrow / .arrows / .ipc files.
  2. Review the detected format, row count, record batches, dictionaries, and metadata.
  3. Inspect the schema in Tree / Raw views.
  4. Browse records by page and use Table / Record or Cell Inspector as needed.
  5. Copy or save the current page as CSV when needed.

Supported

  • Input: Arrow IPC File / Stream in .arrow / .arrows / .ipc files.
  • Structure: Schema Tree / Raw, metadata, metadata version, endianness, footer / header, record batches, and dictionaries.
  • Values: primitive, timestamp, list / struct, dictionary-encoded values, and other supported Arrow values.
  • Viewing/output: read only batches needed for the current page, with Table / Record, Cell Inspector, and current-page CSV.

Limitations and notes

  • Read-only; Arrow IPC files cannot be edited or regenerated.
  • Feather is not advertised as an officially supported input format.
  • IPC body buffers compressed with LZ4 / ZSTD are not supported in v1.0.0.
  • CSV export covers the current page only, and this is not a SQL / query engine.

Frequently asked questions

Are Arrow IPC files uploaded to a server?

No. Selected Arrow IPC File / Stream data is read in the browser and is not uploaded to Browser Kitty.

Can it open both IPC File and Stream formats?

Yes. It reads .arrow / .arrows / .ipc and detects IPC File / Stream inputs.

Can it open LZ4- or ZSTD-compressed Arrow IPC bodies?

Not in v1.0.0. IPC body buffers compressed with LZ4 / ZSTD are treated as unsupported.

Can it convert the entire Arrow file to CSV?

CSV export covers the current page only; it is not a whole-file conversion tool.

Offline version and source code

Save the single-HTML build to inspect Arrow IPC File / Stream schema, batches, dictionaries, and data locally. No runtime external API or CDN is required.

View source on GitHub

Use together

Useful next steps and companion tools, organized around the workflow.

Inspect Parquet Parquet Viewer Use Parquet Viewer for Parquet files, including schema, row groups, and compression. Inspect ORC ORC Viewer Use ORC Viewer for ORC datasets, including stripes, column statistics, and metadata.