Home/Software/Tad · A CSV and Parquet viewer with pivot tables

Tad · A CSV and Parquet viewer with pivot tables

Inspect CSV and Parquet files locally: filter rows, group data, build hierarchical pivots and export a filtered result. A viewer for analysis, rather than a spreadsheet editor.

Creator: Antony Courtney0.14.0Nov 6, 2026MITSource repository
Get from publisher
WindowsmacOSLinux

Requirements

Windows, macOS or Linux matching an official release asset. Access only files you are authorized to inspect; protect customer exports and preserve the original files. Runtime compatibility and performance have not been tested by Boloop.

Quick start

Download the Windows EXE, macOS Intel/Apple Silicon DMG, or Linux DEB/RPM from the official v0.14.0 release. Open a working copy of your CSV/Parquet file, inspect types and rows, then filter and group. Export a filtered result to a new CSV/Parquet file and check it before replacing any existing workflow.

Inspect CSV and Parquet files locally: filter rows, group data, build hierarchical pivots and export a filtered result. A viewer for analysis, rather than a spreadsheet editor.

When Tad is useful

Tad is an open-source desktop viewer by Antony Courtney. Its official documentation lists CSV, Parquet, SQLite and DuckDB files as inputs. DuckDB handles the data, while the interface provides filtering, sorting, aggregation, column selection and hierarchical pivots. This is useful when an export needs inspection by several dimensions without first building a separate analysis application.

For an order export, for example, you could group by region and then product to examine the structure of the data. That is a suggested workflow, not a Boloop benchmark or a claim that the project ships an order-management system. The official screenshot below uses museum collection data and shows how nested groups expand into individual records.

Official Tad example: museum collection records grouped in a hierarchical pivot. Screenshot by Antony Courtney from the project repository; not a Boloop test or an order-data example.
Official Tad example: museum collection records grouped in a hierarchical pivot. Screenshot by Antony Courtney from the project repository; not a Boloop test or an order-data example. Image source

A practical first pass through a file

  1. Get the release matching your operating system from the official release page. Start with a copy of a non-sensitive or appropriately authorized file.
  2. Open the file and inspect its column names, inferred types and record count. Check dates, identifiers and missing values before trusting a total; a numeric-looking identifier is not necessarily a quantity to sum.
  3. Narrow the rows with a filter, choose the columns you need, and add a pivot dimension. Expand a group to see the underlying records. For an order export, compare a few groups against the source system before relying on an aggregate.
  4. In v0.14.0, the release notes add export of filtered tables to CSV and Parquet. Export to a new file, then inspect that file separately to confirm that the selected rows and columns match your intended result. Keep the original export intact.

The repository also documents opening a file from a terminal, for example tad MetObjects.csv. That command assumes the Tad command is available in your environment; use the desktop file-opening flow if it is not.

What to check before choosing it

Tad is a viewer and analysis interface, not a general spreadsheet cell editor. Prefer an editor if your main job is correcting individual cells. The repository's experimental web and cloud backends are proof-of-concept work; do not assume that they provide supported production connectors in the desktop release. File size, schema and your machine affect the result; this page does not provide a tested maximum file size or performance guarantee.

As checked on November 6, 2026, the latest official release remains v0.14.0, published on June 21, 2024. Its notes include DuckDB 1.0, filtered CSV/Parquet export and a scientific-notation formatting fix. We have checked the documentation, release assets and feedback, but have not installed or runtime-tested Tad for this recommendation.

Independent feedback also matters: in issue #305, user mindoo reports that a failed file-open operation in a second Tad window closes all Tad windows after the error is dismissed. The report does not identify a version and is still open at this check. It is a reason to test your own multi-window workflow and keep work copies, rather than proof that every installation has this behavior.

Sources and attribution

See the official project documentation, v0.14.0 release and downloads, and user report #305. Tad and its official example screenshot are credited to Antony Courtney; the repository is distributed under the MIT license.

Sources and further reading

  1. Tad project documentationOfficial documentation · Antony Courtney
  2. Tad v0.14.0 releaseOfficial documentation · Antony Courtney
  3. File-open error in a second window — issue #305Review · mindoo
  4. Tad MIT licenseOfficial documentation · Antony Courtney