# Version 0.4.0

Sifted through the project to clean up some items that have gotten away from me. Removed/retired some older code and tests, strict adherence to code quality steps, and updated dependencies.

# Patch Notes

## Improvements
- Code Cleanup and re-implementation of stricter code standards (!131, !132, !133)
- Started addressing some earlier poor practice with testing fixtures (!134)

## Changes
- Updated some long overdue dependencies
  - pandas (1.5.1 -> 2.2.2, !131)
  - numpy (1.23.4 -> 126.4, !136)
  - geopandas (0.12.0 -> 0.14.3, !131)
  - pyarrow (8.0 -> 15.0.2, !131)
  - scipy (1.9.3 -> 1.13.0, !136)
  - frictionless (5.15.10 -> 5.16.1, !136)
  - scikit-learn (1.1.2 -> 1.4.2, !136)
  - plotly (!5.13.1 -> 5.20.0, !136)
- Updated some dev dependencies:
  - flake8 (7.0.0 -> 8.0.0, !135)
  - pylint (2.17.3 -> 3.1.0, !135)
  - vulture (2.7 -> 2.11, !135)
  - ipykernel (6.28.0 -> 6.29.4, !135)

- New Features
  - Total rework of the HTML generation (!125)

- New Features
  - Can now include references to Ontologies and Terms in the TiB Terminilogy Service from a Schema file, which will then be incporated when validating data files matching that schema. (!110)
  - Added a new script `run_check.py` for calling quality or validation checks independently. (!120)

- Improvements
  - Improved background object generation (!114)
  - Better collapsing of test cases when many with identical text are generated (!115)
  - Improved organization of the CI/CD scripts (!116)
  - Made it easier to create Datafile objects from local files (!117)
  - Made it easier to add new Quality Checks regardless of data type (!118)
  - Added in better automatic checking of S3 sources (!119)

- Fixes
  - Added missing dependency for allowing Frictionless to interrogate locally available parquet files (!112)
  - Fixed a case where we incorrectly validated an empty schema file (!110)
  - Fixed case where validation status could be missing in the dictionary output of the schema object (!121)

Usage:
- Updated layout of `settings.toml` to allow file by file settings.

Improvements:
- Added support for csv files with different delimiters
- Data field descriptions will now correctly calculate, even if the source data contains incorrect data points.
- Added additional descriptive fields for geodata.

Bugfixes:
- Fixed missing error when S3 credentials were provided, but the user did not include a bucket name.
- Fixed error that could occur when trying to assign schema files that did not yet exist on disk.
- Fixed incorrect calculation of geodata centroid points.
- Fixed various serialization errors related to writing summary data to JSON.
- Fixed display issue related to Categorical values when the categories were numerical.
- Fixed broken links to data file pages
- Fixed broken image links on nav bar