Port the SeisSol Preprocessing folder here (now with its history)

[davschneller]

Add the SeisSol preprocessing folder (or what's left of it after some recent purge of very old files).

This time, it includes the commit history of everything from there up to date (done using git-filter-repo). ... The question would be—do we (really) want that, or not? Otherwise, it would just be a "normal" merge.

Also, it should make rconv more SeisSol-independent.

Probably a point that's still open is the folder structure of the repo now.

Once this one's merged, we can close the preprocessing folder on the SeisSol.

[vikaskurapati]

Add the SeisSol preprocessing folder (or what's left of it after some recent purge of very old files).

This time, it includes the commit history of everything from there up to date (done using git-filter-repo). ... The question would be—do we (really) want that, or not? Otherwise, it would just be a "normal" merge.

I think it is a nice idea to have to the commit history, just in case some changes need to be tracked, or any kind of results diverged with time.

Also, it should make rconv more SeisSol-independent.

Probably a point that's still open is the folder structure of the repo now.

Once this one's merged, we can close the preprocessing folder on the SeisSol.

[Thomas-Ulrich]

I also think it is a good idea to export the full history (but may clutter the history of this repo?).
Alternatively, you can add a line in each file saying.

"This file has been transferred from SeisSol/SeisSol. The full history can be found by checking out SeisSol/SeisSol at commit xxxx."
(simpler and cleaner?).

Apart from that, I would suggest removing:

• gmsh_examples
• gmsh2gambit
• everything directly in science folder, except generating_ASAGI_file.py, read_ini_fault_parameters.py (now updated in Update easi python script SeisSol#1544) and ViscoelasticModComp.m. The rest are outdated MATLAB scripts and Python 2 scripts.

I would also consider removing asagiconv (I have the impression that it could be re-implemented with a few lines of Python and the easi module).

[davschneller]

I also think it is a good idea to export the full history (but may clutter the history of this repo?). Alternatively, you can add a line in each file saying.

"This file has been transferred from SeisSol/SeisSol. The full history can be found by checking out SeisSol/SeisSol at commit xxxx." (simpler and cleaner?).

Apart from that, I would suggest removing:

* gmsh_examples

* gmsh2gambit

* everything directly in science folder, except generating_ASAGI_file.py, read_ini_fault_parameters.py (now updated in [Update easi python script SeisSol#1544](https://github.com/SeisSol/SeisSol/pull/1544)) and ViscoelasticModComp.m. The rest are outdated MATLAB scripts and Python 2 scripts.

I would also consider removing asagiconv (I have the impression that it could be re-implemented with a few lines of Python and the easi module).

That sounds great as well. ... Maybe, as suggestion: how about we delete these files in the SeisSol main repo first; then we can see further? Might help with deciding further about this PR or its follow-up version.

(I can prepare a branch, if you'd want me to (but anyone else, feel free to volunteer); though I'd probably keep asagiconv until we have the Python script ready ... Probably we could also spin off rconv to a Python version somewhat similarly)

[vikaskurapati]

I also think it is a good idea to export the full history (but may clutter the history of this repo?). Alternatively, you can add a line in each file saying.
"This file has been transferred from SeisSol/SeisSol. The full history can be found by checking out SeisSol/SeisSol at commit xxxx." (simpler and cleaner?).
Apart from that, I would suggest removing:

* gmsh_examples

* gmsh2gambit

* everything directly in science folder, except generating_ASAGI_file.py, read_ini_fault_parameters.py (now updated in [Update easi python script SeisSol#1544](https://github.com/SeisSol/SeisSol/pull/1544)) and ViscoelasticModComp.m. The rest are outdated MATLAB scripts and Python 2 scripts.

I would also consider removing asagiconv (I have the impression that it could be re-implemented with a few lines of Python and the easi module).

That sounds great as well. ... Maybe, as suggestion: how about we delete these files in the SeisSol main repo first; then we can see further? Might help with deciding further about this PR or its follow-up

Can we move them somewhere? Or just collect them in a common place where all the legacy scripts are stored, in a case someone wants to find or look for them later on.

(I can prepare a branch, if you'd want me to (but anyone else, feel free to volunteer); though I'd probably keep asagiconv until we have the Python script ready ... Probably we could also spin off rconv to a Python version somewhat similarly)

You can go ahead with it, I guess given that you already did this.

[Thomas-Ulrich]

No need for a legacy scripts folder, IMO. They can still be found in the GitHub history, even if the files have been removed (you need to know they existed to look for them).

[davschneller]

Ok, done. Have SeisSol/SeisSol#1545 ; I've also thrown out cube_c, as we have a PUML version of that now in this repository here (cf. #49 ).

W.r.t. old files: there's also the "old" tags, like e.g. v1.0 which should have preprocessing/postprocessing dirs that contain all old files.
... which might impose the question: will we need to somehow also tag the meshing and visualization repositories when releasing a new SeisSol version? (and maybe retrospectively releasing past versions).

[davschneller]

Well then; the remaining "mass" of scripts and tools is rather small (see here for the current state). Personally, I wouldn't really care on how we exactly proceed. We can either:

• keep this whole history of all scripts and tools
• try to reduce the history to all files we port now (maybe much less "mass")
• just copy in a single commit with a notice as Thomas suggested.

Open for all three options. :)

[vikaskurapati]

Well then; the remaining "mass" of scripts and tools is rather small (see here for the current state). Personally, I wouldn't really care on how we exactly proceed. We can either:

• keep this whole history of all scripts and tools
• try to reduce the history to all files we port now (maybe much less "mass")
• just copy in a single commit with a notice as Thomas suggested.

Open for all three options. :)

Same, open for all three, but as Thomas already prefers the third, we can go with that.