Version History¶
Tafra: a minimalist dataframe
Copyright (c) 2020 Derrick W. Turk and David S. Fulford
1.0.7¶
Handle missing or NULL values in
read_csv()
.Cast empty elements to None when updating dtypes to avoid failure of
np.astype()
.Update some typing, minor refactoring for performance
1.0.6¶
Additional validations in constructor, primary to evaluate Iterables of values
Split
col_map
tocol_map
andkey_map
as the original function’s return signature depending upon an argument.Fix some documentation typos
1.0.5¶
Add
tuple_map
methodRefactor all iterators and
..._map
functions to improve performanceUnpack
np.ndarray
if given as keys to constructorAdd
validate=False
in__post_init__
if inputs are known to be valid to improve performance
1.0.4¶
Add
read_csv
,to_csv
Various refactoring and improvement in data validation
Add
typing_extensions
to dependenciesChange method of
dtype
storage, extractstr
representation fromnp.dtype()
1.0.3¶
Add
read_sql
andread_sql_chunks
Add
to_tuple
andto_pandas
Cleanup constructor data validation
1.0.2¶
Add object_formatter to expose user formatting for dtype=object
Improvements to indexing and slicing
1.0.1¶
Add iter functions
Add map functions
Various constructor improvements
1.0.0¶
Initial Release