tab-s3-converter

Migrate a Tableau workbook or data source from the legacy S3 connector (class='s3') to the modern Amazon S3 connector (class='amazon-s3').

The tool has no runtime dependencies — it uses only the Python standard library. Dev/test tooling (pytest, ruff) is managed with uv.

Usage

python -m tab_s3_converter <input> [-o output] [--force] [--debug]

• <input> — a .twb, .tds, .twbx, or .tdsx file.
• -o/--output — output path (default: <name>.converted.<ext>).
• --force — overwrite an existing output file.
• --debug — print a traceback instead of a one-line error.

The original file is never modified. .twbx/.tdsx archives are rewritten with all supporting files preserved byte-for-byte; only the inner document is converted.

What it does

For each s3 connection it: parses bucket/region out of the JSON connectionData, rewrites the connection to the amazon-s3 attribute set (auth auth-user-pass, blank credentials — Tableau prompts on open), renames the s3.<id> named-connection to amazon-s3.<id> and updates all references, and replaces the relation's column list with CSV parse options. Per-column <metadata-records> are preserved unchanged, so ordinal, local-name, and local-type carry forward exactly.

Development

uv run pytest          # run the test suite
uvx ruff check .       # lint
uvx ruff format .      # format