Skip to content

CHANGELOG.md

Latest commit

 

History

History
124 lines (97 loc) · 6.64 KB

CHANGELOG.md

File metadata and controls

124 lines (97 loc) · 6.64 KB
Raw

Changelog is organized by the version of this library, commit date and main points of change

[VERSION] - DATE

  • Important fix
  • Important feature

[Unreleased]

Dependencies

  • Updated snowflake-connector-python dependency: >=3.0.0,<4 → >=3.0.0,<5 (allows v4.x)

2025 January

[8.1.0] - 2025-01-20

Build System Migration: Poetry → uv + Hatch

  • Rewrote pyproject.toml using PEP 621 standard format
  • Replaced Poetry's poetry.lock with uv.lock
  • Switched build backend from poetry.core.masonry.api to hatchling.build
  • Removed legacy configuration files: setup.py, setup.cfg, tox.ini, .coveragerc
  • Removed individual requirements files (requirements.txt, requirements-dev.txt, requirements-extras-*.txt)
  • Consolidated all configuration (coverage, pytest, linting) into pyproject.toml

CI/CD Improvements

  • New unified ci.yml workflow replaces black.yml and pythonpackage.yml
    • Updated to GitHub Actions v5 (checkout@v5, setup-uv@v5)
    • Expanded test matrix: Python 3.9, 3.10, 3.11, 3.12, 3.13, 3.14
    • Coverage reporting runs on Python 3.13
  • New release.yml workflow for automated PyPI publishing using OIDC trusted publishing
  • Switched linting from Black to Ruff (same 120 char line length, quote style preserved)

Pandas 2.x Compatibility Fixes

  • Fixed deprecated DataFrame.applymap() → DataFrame.map() throughout codebase
  • Added future.no_silent_downcasting option context to avoid FutureWarnings
  • Fixed groupby() level parameter to use scalar when single level
  • Added infer_objects(copy=False) after fillna for proper dtype inference
  • Convert DataFrame to object dtype before applying string formatting in pandas widget

PyPika Compatibility

  • Updated pypika dependency: 0.48.2 (pinned) → >=0.50.0,<1.0.0
  • Fixed test assertions for PyPika's NULL keyword formatting change

Other Changes

  • Added pre-commit configuration with Ruff hooks for linting and formatting
  • Relaxed pandas dependency: ==2.0.3 (pinned) → >=2.0.0
  • Relaxed toposort dependency: ==1.6 (pinned) → >=1.6,<2
  • Updated psycopg2-binary dependency: >=2.9.0 → >=2.9.11 (adds Python 3.14 support)
  • Dropped Python 3.8 support; minimum Python version is now >=3.9 (required by psycopg2-binary 2.9.11 and pandas 2.x)
  • Snowflake extra is temporarily unavailable on Python 3.14 (upstream snowflake-connector-python requires cffi<2, which lacks Python 3.14 wheels); Snowflake tests gracefully skip when dependencies are missing
  • Fixed invalid escape sequence in type_engine.py (was SyntaxWarning, will be error in Python 3.16)
  • Removed dead code from Snowflake database module
  • Configured Ruff to ignore intentional patterns (re-exports in __init__.py, star imports in tests)
  • Applied Ruff formatting to codebase
  • Fixed duplicate test method in test_build_filters.py
  • Fixed duplicate import in test_references.py
  • Snowflake: Moved cryptography.hazmat imports inside method for lazy loading
  • Version now read from package metadata using importlib.metadata.version()
  • Fixed filter_nones() to use lambda instead of None.__ne__
  • Added CLAUDE.md with project overview and development commands
  • Added Snowflake and MSSQL database extras to setup documentation
  • Added Snowflake database configuration example to documentation
  • Removed test API documentation files from docs

2023 December

[8.0.5] - 2023-12-18

  • Fix dataframe sort with a pivot with the new pandas v2 API
  • Optimize it by getting rid of reset_index and set_index method calls when sorting

2023 November

[8.0.4] - 2023-11-06

Make removal of date total when generating highcharts more flexible by localizing the datetime when filtering out the totals. This will work with timezone aware datetimes as well not aware ones.

2023 September

[8.0.3] - 2023-09-06

  • Specify extras dependencies correctly

[8.0.2] - 2023-09-06

  • Specify extras names and the library dependencies for them under tool.poetry.extras

[8.0.1] - 2023-09-06

  • Describe all the necessary group dependencies in pyproject.toml as none of other than Vertica and Snowflake libraries have been included in the published library
  • Add newly installed libraries to poetry.lock
  • Update pymssql to 2.2.8

2023 August

[8.0.0] - 2023-08-29

  • Drop support for Python3.7 as pandas v2 does not support it
  • Add Python3.9 support
  • Use the newest pandas (2.0.3 in requirements) and add the minimum version 2.0.0 of pandas to pyproject.toml
  • Use the newest vertica-python (1.3.4 in requirements) and add the minimum version 1.0.0 of vertica-python to pyproject.toml
  • Use the newest snowflake-connector-python (3.0.4 in requirements) and add the minimum version 3.0.0 of snowflake-connector-python to pyproject.toml
  • Use the newest coverage (7.3.0 in requirements) and add the minimum version 7.3.0 of coverage to pyproject.toml
  • Use the newest watchdog (3.0.0 in requirements) and add the minimum version 3.0.0 of watchdiog to pyproject.toml
  • Remove python-dateutil from dependencies as it is part of other libraries' dependencies
  • Bump psycopg-binary==2.9.6 though it seems not needed for the tests
  • Bump pymssql==2.2.7 though it seems not needed for the tests
  • Bump Cython==3.0.0 though it seems not needed for the tests
  • Get rid of SyntaxWarning: "is" with a literal. Did you mean "=="?
  • Get rid of DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated since Python 3.3, and in 3.10 it will stop working
  • Replace Dataframe.append to pd.concat because append does not exist since pandas v2
  • Add group_keys=False to DataFrame.groupby() method because it is no longer ignored since pandas v1.5
  • Fix _apply_share method for Shares with the new libraries.
  • Rename TestDatabase to MockDatabase since it is used only for mocking. This is beneficial also because Python testing method will not delve into it to find methods to run
  • Rename test_connect and test_fetch to mock_connect and mock_fetch as these are mocks. This is beneficial also because Python testing method will not delve into it to find methods to run
  • Rename TestMySQLDatabase to MockMySQLDatabase for the same reason
  • When concatenating DataFrames, use .tail(1) instead of .iloc[-1] as it includes indexes
  • Use static CSVs to get the expected DataFrames in tests instead of applying methods of fireant to a DataFrame to get those expected DataFrames
  • Replace np.float to float since it was deprecated
  • Get rid of None and [] as ascending parameters for Pandas class
  • Replace .iteritems() with .items() as the former method was deprecated