Great! Only comment is that it could be useful to have a STF-to-FORT90ish converter, for legacy users.

Maybe it would be better to just have a standalone executable. I doubt anyone relying on this uses SixTrack 5.x anyway. I assume the code can be extracted from pospr or read90?

For what it's worth, the converter was written and can be seen in:
vkbo#4

It works, but needs to be integrated with the build and test system. It should probably also accept a flag to write the old-style fort.91-i filenames.

The PR to Veronica's branch is now ready to be merged.
One thing that should be changed (and I think e.g. Veronica is better situated to do it) is to use a dedicated singletrackfile.dat for testing. I basically picked a random one, however it should ideally be one that (a) has > 32 pairs, and (b) has a few 100 turns.

It should probably also be tested with ntwin=1...

Thanks for merging that.

Only thing that one should look at is that the test for the conversion tool is currently referencing the singletrackfile.dat.canonical in the test dump7_fma. It should really have it's own little file, ideally with > 64 particles and ~100 turns (losses are cool), but I feel that Veronica @vkbo would be better placed to choose or generate such a test file; given the churn of these files I did not want to add one.