#2300: Update: Add Piecewise Polytrope EoS Support to IllinoisGRMHD, Improved TOV Solver?
| Reporter: | Zach Etienne |
| Status: | open |
| Milestone: | ET_2020_04 |
| Version: | |
| Type: | enhancement |
| Priority: | major |
| Component: |
Comment (by Roland Haas):
Unfortunately even requiring a trivial modification is not acceptable. This is partially due to what is trivial tending to depend to be on the level of experience a user has with the code in question, ranging from the author, for whom everything is trivial, to someone trying to run a gallery parfile or example parfile from the web, for whom nothing is trivial.
An acceptable way would be the following:
ID_converter_ILGRMHD which must default to “false”. Have ID_converter_ILGRMHD use either its own parameter value or ILGRMHD’s values depending on the value of this new parameterTo retire ID_converter_ILGRMHD it also needs to be deprecated for one release. Note that we typically retire code only once it no longer functions, not when it is not longer considered interesting, both to allow very old parfiles to run and also since we never quite know what the ET community is still using.
As far as I can tell there is no way to put the parameters back in and warn a user about them not be supported anymore (b/c one cannot change the default values and must accept runs with the default values for only one of the pair of corresponding parameters).
Policy on retiring functionality is outlined on the wiki: https://docs.einsteintoolkit.org/et-docs/Policies_to_retire_functionality
Thorns and parameters probably should not stay deprecated for very long before being removed (ie don’t just mark everything as deprecated to get a blank cheque on removing anything anytime) but given --
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2300/update-add-piecewise-polytrope-eos-support