Change AbstractCoordinatePoint to not be a FieldVector#489
Conversation
4ad755a to
c787b66
Compare
Implement proper math for affine frames, Refactor code base where necessary.
fhagemann
left a comment
There was a problem hiding this comment.
Can we PLEASE first make a release with the current main before merging this in case this messes some things up that are not covered by the tests?
Sure, registering now: |
Also scale default marker size automatically. Need to plot points with units in samplesurface mode now too.
Previous versions require pos in DetectorHit to be an AbstractVector
Many recent changes that might need re-reviewing
|
I'm deactivating the |
e0bda75 to
ac08c04
Compare
|
For the release notes for 0.11.0: |
This reverts commit 3425c2d.
ac08c04 to
27c8b07
Compare
|
|
||
| # LH5 can't handle CartesianPoint, turn positions into CartesianVectors which will be saved as SVectors | ||
| pos_vec = VectorOfVectors([p .- Ref(cartesian_zero) for p in mcevents_sub.pos]) | ||
| pos_vec = VectorOfVectors([[CartesianPoint(p...) - cartesian_zero for p in ps] for ps in mcevents_sub.pos]) |
There was a problem hiding this comment.
Instead of defining a new cartesian_zero, I changed this commit to convert whatever p is to a CartesianPoint first, to then subtract the original cartesian_zero to get a CartesianVector that can be saved to a file by LegendHDF5IO.
3188c33 to
087cb82
Compare
087cb82 to
b6cdbf6
Compare
fhagemann
left a comment
There was a problem hiding this comment.
I would suggest merging this into main and then making an issue with all the things that still have to be done for the 0.11.0 release in a separate issue (and separate PRs), including drafting a release note with breaking changes.
Sounds good to me. |
Corresponds to mathematically cleaner "affine vs. linear" language.