v3.30.1

What's Changed

• Move _normalize_hive_syntax to DefaultTypeConverter and fix pass stubs by @laughingman7743 in #695
• Always use C engine for CSV parsing and simplify engine selection logic by @laughingman7743 in #697

Bug Fixes

• Fix CSV parsing performance regression — PR #594 incorrectly forced the Python CSV engine for files over 50MB based on fabricated claims about pandas C parser int32 limitations. The C engine is now always used as the default (same as pandas' own default), restoring up to 28% faster CSV parsing. See #696 for details.

Internal

• Refactored type converter: moved _normalize_hive_syntax to DefaultTypeConverter and cleaned up pass stubs.
• Simplified CSV engine selection from 4 methods to 2 by removing dead code and inlining pyarrow compatibility checks.

Affected Versions

All versions from v3.17.0 through v3.30.0 (released 2025-08-09 to 2026-02-28) are affected by this performance regression. The last unaffected version is v3.16.0.

Workaround for older versions

If you cannot upgrade to v3.30.1, explicitly specify the C engine to bypass the incorrect engine selection:

cursor = connection.cursor(PandasCursor, engine="c")

Full Changelog: v3.30.0...v3.30.1