Create 0.4.0 release

[veenstrajelmer]

The code quality, test coverage and user friendliness was improved in 0.2.0 and the output format was improved in 0.3.0. Several steps are required for the results to be official (e.g. updates in measurement data and automatic retrieval of HYDRA statistics). These dependencies might only be resolved in 2025.

Todo:

• compute_expected_counts gives incorrect values for HANSWT ext #88
• correct measurements amount #147
• avoid RuntimeError due to nans in overschrijding #139
• expand physical_break_dict #150
• fix testcases related to physical_break #152
• Move to linear fit for slotgemiddelden #156
• expose yearly tidal components #158
• update pandas version to use diff #160
• neater timezone handling and time corrections in calc_havengetallen #162
• reconsider 20 minute correction in culm_addtime for calc_havengetallen #164
• pass complete measurement series throughout code #166
• deprecate debug argument for calc_gemiddeldgetij #169
• align code with min_coverage and modular newyear clipping #167
• apply slotgemiddelden to havengetallen and gemiddeldgetij #171
• Expose mean/max/min for spring/neap tide #137
• check size of resulting netcdf files with engine='h5netcdf' (which is often used after fixing 125 add windows tests Deltares/ddlpy#126) >> nothing changed in kw env, so no changes expected
• only convert back to cm in kenmerkendewaarden if unit==m (after release of hatyan with Align units in ddlpy_to_hatyan Deltares/hatyan#363)
• consider enforcing nyears in kw.crop_timeseries_last_nyears() instead of warning, requires quite some updates to gemiddelgetij and havengetallen tests and a df_ext_12 (of 10y)
• some stations failed because of almost-duplicate times: Fix numbering of extremes for havengetallen #101 (waiting on update of extremes data in DDL by TvL, will be somewhere after 4 nov 2024)
• add urls in changelog again
• release and supply data to RWS? (slotgemiddelden fit is different, still won't include units)

Later (2025):

• Optional data corrections #29
• Improve overschrijdingsfrequenties #80
• Improve havengetallen #84
• Improve slotgemiddelden #85
• Improve gemiddeldgetij #86
• Add additional indicators/kengetallen to kenmerkendewaarden scope #25
• update to new RWS WaterWebservices Deltares/ddlpy#113
• use complete and updated timeseries: DDL improvements (data and waterwebservices) #4
• release
• derive (now official?) results for all stations and for 2011/2021 (include in deliverable: retrieve+process script and station figure/csv)

Todo not related to methods:

• check warnings in pytest workflow
• move to Deltares Github organization?
• decide on dutch/english in documentation, docstrings, function/argument/key/variable names
• generate DOI and add badge to readme
• GPL or LGPL license? Default of Deltares-research is MIT, not for Deltares
• code is formatted with black, do this automatically via .pre_commit_config.yml?
• add contributing docs (including release workflow)
• publish to database instead of csv's? Makes it possible to add metadata like unit, station x/y/code/name, vertref, kenmerkendewaarden-package-version etc. Temporarily add to Deltares database, or wait for RWS database to be consolidated?

Reporting:

• rapportage evt met Quarto: iets online via bookdown R of opvolger, een webpage of pdf. Mooie methode voor technische rapporten, kan ook met Deltares LaTeX template: https://www.testsysteemrapportage.nl/wadden/_main.pdf Of gewoon sphinx docs op github?
• docstrings do/will contain the technical user information about input/output of functions and what they do
• the description of the kenmerkende waarden can be in sphinx documentation (what does it represent, why this method, interaction)
• additionally a pdf report with description of historic methods and motivation of improvements/simplification? Maybe including comparison of methods if still reproducible This information is not all relevant for the online documentation, since not all users will be interested in this. But it is good to document motivation for changes somewhere.
• which language where?
• the RWS dashboard will show the results for all stations, so here we can focus on one example station (or some)
• also document which output (csv files, figures, boi timeseries) is relevant for which process (including contact person)

Products:

• github toolbox
• final results
• report?