- Unlock pandas and pyarrow versions in setup.cfg. Thanks @DManowitz for the reminder
- Clean up some accidentally commit egg files
- Update pyarrow to <11.0.0
- Change urlparse import to be only Python 3 compatible
- Drop support for Python 2
- Update pyarrow to <6.0.0, pandas to <2.0.0 and numpy to >=1.14.0. Thanks to @sweco for the PR!
- Update pyarrow to <2.0.0 and pandas to <2.0.0. Thanks to @zmjjmz for the PR!
- Update pyarrow to 0.0.14 and pandas to 0.24.2. Thanks to @rudaporto for the PR!
- Update pyarrow to 0.0.12 to fix zlib issues and gain improved memory perf with lots of strings. Thanks @nimish and @Madhu1512
- Add field_aliases kwarg to loading methods to allow mapping a JSON column name to a different parquet column name. Thanks to @sojovi for the idea.
- Add write_parquet_dataset available as a default export, thanks to @gregburek
- Bump pyarrow, lower lock on Numpy to >=1.14
- Bump pyarrow, numpy and Pandas versions
- Bump pyarrow and Pandas versions
- Don't lock ciso8601 version.
- Add support for DATE fields. h/t to Spectrify for the implementation
- Properly handle boolean columns with None.
- Allow schema to be an optional argument to convert_json
- Bring write_parquet_dataset to a top level import
- Properly convert Boolean fields passed as numbers to PyArrow booleans.
- Add support for custom datetime formatting (thanks @Madhu1512)
- Add support for writing partitioned datasets (thanks @mthota15)
- Stop silencing Redshift errors.
- Fix decimal type for newer pyarrow versions
- Allow casting of int64 -> int32
- Bump PyArrow and allow int32 data
- Allow passing partition columns when getting a Redshift schema, so they can be skipped
- Fix conversion of timestamp columns again
- Fix conversion of timestamp columns
- Force converted Timestamps to max out at pandas.Timestamp.max if they exceed the resolution of datetime[ns]
- Add automatic downcasting for Python
float
tofloat32
via pandas when schema specifiespa.float32()
- Fix conversion of float types to be size specific
- Fix ingestion of timestamp data with ns resolution
- Add pandas dependency
- Add proper ingestion of timestamp data using Pandas
to_datetime
- Fix formatting of README so it displays on PyPI
- Initial release
- JSON/data writing support
- Redshift Schema reading support