change_log.md

ver 0.0.1 create tpxo9.zarr

ver 0.0.2 u,v = U/(D/100) from transport (API not yet)/n1

ver 0.0.3 u,v = small U/(D/100) problem: use pyTMD extract constant (not yet)/n2

ver 0.0.4 small U/(D/100) problem: use shallowLimit or without (not yet)/n3

ver 0.0.5 re-fill NA in extract_ATLAS_pytmd both extract_constants and interpolate methods (not yet)/n4

ver 0.0.6 first draft-version of tide API, re-fill NA by 3*1 neighbors (not yet)/n5

-- add /tide, /tide/const to load each harmonic constituent
-- refine project structure before rerun zarr dataset https://github.com/tsutterley/pyTMD/discussions/241

ver 0.0.7 Rescale u,v amplitude by model.scale=1e-4 according to pyTMD/pull/243, fill NA/n7

-- small fix TypeError: expected unicode string, found 15 problem in constituents, fill NA/n11

ver 0.0.8 Add nearest method to locate points by Zarr; Validate api results with Jen's paper

-- change to use pipenv for package version management

ver 0.0.9 Use vectorized/nearest method to solve mult-point constituents/fix datetime, mode None problems

-- change wide dataframe format is default. Specify mode=long if long format is preferred
-- /tide/const example to plot tidal ellipses. Parallel re-fill-NA savings
-- fix Parallel re-fill-NA savings deadlock problem by Dask version
-- add FastAPI pydantic response_model
-- fix bounding box convert to (0,360) bug
-- fill 2x2 neighbor NA. Note that trouble to mix tpxo9.zarr/_fillna.zarr should be avoided
-- small package upgrade/fill_NA trials(15th --> 18th)

ver 0.1.0 Add API Readme.md, disclaimer, fastapi lifespan for package upgrade

-- various package upgrade/test(18->25th), dev files small update

ver 0.1.1 Let constituents oscillation contribute 0 not mask as NA

-- small update in dev files for testing/README.md move swagger doc to Ocean APIverse
-- various package upgrade/n2/test(->35th), dev files small update

ver 0.1.2 Start validation with observation(CWA)/model(Marea API) data

-- Add validation with NOAA tide observation data api.tidesandcurrents.noaa.gov/
-- Modify testbench (for CWA, NOAA) bugs e.g. local time (NOAA) to UTC
-- small update testbench, tpxo9.zarr(->36th), dev files

ver 0.2.0 Fix get_tide_series u,v unit conversion bug, upgrade pyTMD 2.1.4 breaking change

-- a temporary solution to make xarry internal index error when lon=0 but actually at -4.06e-6 (floating error)

ver 0.2.1 Fix shared dask worker leak by unique naming scheme for Dask tasks/major package upgrade (numpy v2)

ver 0.2.2 Compare Tide API with CMEMS cmems_mod_glo_phy_anfc_merged-uv_PT1H-i

-- pacakge upgrade/testing files update (->47th)
-- refine the plots that compare Tide API with CMEMS