Relocatable Ocean Modelling in PYthon (rompy) combines templated cookie-cutter model configuration with various xarray extensions to assist in the setup and evaluation of coastal ocean model
@tomdurrant the timestap-based buffering has been added back in cdf3d83fef33296e5adbe75b278535a1cc77aaae. It now uses the new coordinates property of the source field to avoid accessing the actual data
A few changes to fix the datamesh source objects. One change concerns schism @tomdurrant which is the removal of the time offset logic from the _filter_time() method (https://github.com/rom-py/rompy/compare/raf-data?expand=1#diff-1ea32bb20ed105e939a30996d7a70cd9dfa420c2328c750a2035be817fda18d9R372-L393). In order to achieve the same result now, the time buffer now needs to be specified as the number of seconds, instead of a multiplier for the timestep calculated from the data (https://github.com/rom-py/rompy/compare/raf-data?expand=1#diff-1ea32bb20ed105e939a30996d7a70cd9dfa420c2328c750a2035be817fda18d9L365-L368). Ultimately, the data filtering option probably needs some rethinking. Ideally we should be able to be more flexible and deal with longitude conventions, non-matching slices, etc.