Summary
The PyPI release databow==0.1.0 was built before commit c49f597 (fix: render timestamp with time zone data, PR #14) landed on main. As a result, the published wheel cannot render arrow arrays whose elements are timestamps with a named IANA timezone — running any query that returns such a column from the CLI fails with:
Failed to print batches: Parser error: Invalid timezone "Etc/UTC": only offset based timezones supported without chrono-tz feature
The fix on main (enabling the chrono-tz feature on the arrow umbrella crate) already resolves this; building from main produces a working binary. Only the published artifact is affected.
Reproduction
Against any driver that returns a Timestamp(_, Some("Etc/UTC")) array (e.g. DuckDB after SET TimeZone = 'Etc/UTC'):
$ uv tool install databow # 0.1.0
$ databow --driver duckdb --query "SET TimeZone = 'Etc/UTC'; SELECT now()"
Failed to print batches: Parser error: Invalid timezone "Etc/UTC": only offset based timezones supported without chrono-tz feature
Ask
Cut a new release (e.g. 0.1.1) so PyPI consumers stop hitting this. No code change required — main already contains the fix.
Workarounds
- Install from source:
cargo install --git https://github.com/columnar-tech/databow --locked
- Pin to a future release once published.
Summary
The PyPI release
databow==0.1.0was built before commit c49f597 (fix: render timestamp with time zone data, PR #14) landed onmain. As a result, the published wheel cannot render arrow arrays whose elements are timestamps with a named IANA timezone — running any query that returns such a column from the CLI fails with:The fix on
main(enabling thechrono-tzfeature on thearrowumbrella crate) already resolves this; building frommainproduces a working binary. Only the published artifact is affected.Reproduction
Against any driver that returns a
Timestamp(_, Some("Etc/UTC"))array (e.g. DuckDB afterSET TimeZone = 'Etc/UTC'):Ask
Cut a new release (e.g.
0.1.1) so PyPI consumers stop hitting this. No code change required —mainalready contains the fix.Workarounds
cargo install --git https://github.com/columnar-tech/databow --locked