Release April 2024 #35
Conversation
…907eb0dc10555a7072e2f74b10a5655 commit
…it (we love VSCode sometimes)
Revamped Adapter to Work for SU2 v7.5.0-v7.5.1, Both CHT + FSI
…wrapper (#34) * Nondimensional mode implementation * Fix compilation w/ OMP bug * Specify problem dimension as argument * Clarify meaning of Monitor * Save time and iter for implicit coupling * Fix CHT initial read/write (was misleading) (read_data --> write_data) * Add disclaimer/note on restarting simulations
|
This new version now still has a small part that acts as a patch to the SU2 sources, right? Does this mean that we likely need to update it for every new SU2 version, similar to e.g. the CalculiX adapter? Then, maybe, a versioning scheme following the SU2 versions could be helpful? |
|
Good point about the SU2 replacement files. That might still change soon. Let's just keep this as-is for now (commit id), and decide later. |
* Refactor checkpointing and output. * Revert unrelated whitespace changes. * Move stopCalc after final output
|
I updated the documentation as well, which is now sourced from this repository (develop branch): https://precice.org/adapter-su2-overview.html You might need to refresh each page. I have also integrated the recent changed of @BenjaminRodenberg and changed the default branch to This is now ready to merge from my side. |
MakisH
requested review from
uekerman and
BenjaminRodenberg
and removed request for
uekerman
There was a problem hiding this comment.
Thanks a lot @MakisH 🙏
I looked at the docs and the diff. All looks reasonable. I did not run anything.
@MakisH Are you sure that v7.5.0 is still supported? The replacement file is apparently from v7.5.1. Could be that there are simply no changes in that file, but could be safer to simply remove the statement that it also supports v7.5.0.
This is a big update, which:
This state is compatible with the tutorials PR precice/tutorials#324.
Still needs updates for the documentation on the website (or make the website source them from here).
No versioning system is established at the moment, but this is now pointing towards v1.0 @uekerman (assuming 0.x would have been the previous state). We could still now start with v0.2 to be conservative, or keep using untagged releases. But I think we should take the opportunity to mark this with a clear version.
BIG thanks to @j-signorelli for his excellent contribution and his patience getting that reviewed and merged.