New matrices support in sympy.physics.matrices #140
Conversation
rotation about a single axis to sympy.physics.matrices. Tests for these were added to tests_physics_matrices.py.
| st = sin(theta) | ||
| mat = ((1,0,0), | ||
| (0,ct,st), | ||
| (-st,ct,0)) |
There was a problem hiding this comment.
check this line, cell with 1, must not be intersected with cos and sin cells.
…ean up the associated tests which were needlessly verbose.
| r2_minus = rotAxis1(-pi/3) | ||
| assert r3_minus*r3_plus*eye(3)*r3_plus.transpose()*r3_minus.transpose() == eye(3) | ||
| assert r2_minus*r2_plus*eye(3)*r2_plus.transpose()*r2_minus.transpose() == eye(3) | ||
| assert r1_minus*r1_plus*eye(3)*r1_plus.transpose()*r1_minus.transpose() == eye(3) | ||
|
|
There was a problem hiding this comment.
Also it would be useful if you create test saying that trace of rotation matrix is equal 1 + 2*cos(theta)
|
I'm not sure how to properly call these from the python prompt though, but all the tests pass when run separately. Try To be sure that yours test are connected properly you can play with them (made them incorrect) and see reaction in this case (letters DO NOT COMMIT). |
|
I want to know how to run just one set of tests, not the whole set. Running ./bin/test runs the whole set. |
|
Taking into consideration that rotation can happen not only in physics, but in mathematics and geomentry too I think that this methods can be placed in matrices module. What about but at the same time you must be sure that yours tests are connected with './bin/test' properly too. |
| st = sin(theta) | ||
| mat = ((ct,0,st), | ||
| (0,1,0), | ||
| (st,0,-ct)) |
There was a problem hiding this comment.
Also, check this line too.
So can be added another test: that when theta==0 then rotation is equal to identity matrix).
|
I've fixed this. All tests pass, but I'm going to move the rotation matrices to the matrices module and test again. |
|
I've moved the rotation matrices over to matrices.py in the matrices module and removed them from physics.matrices. The pat_matrix function for parallel axis theorem is still in the physics.matrices module and running ./bin/test everything seems fine so I'll commit the new version. |
…to the matrices module, made minor corrections and added new tests. The parallel axis theorem matrix remains in the physics.matrices module. All tests now pass and the functions have been re-named to match PEP8 expectations.
Just add an arbitrary string you want to be present in the test's name. In your case for instance: |
| @@ -2203,3 +2204,31 @@ def symarray(prefix, shape): | |||
| arr[index] = Symbol('%s_%s' % (prefix, '_'.join(map(str, index)))) | |||
| return arr | |||
|
|
|||
| def rot_axis3(theta): | |||
| """Returns a rotation matrix for a rotation of theta about the 3-axis.""" | |||
|
Otherwise it looks good, thanks! |
|
I wrote them as separate tests, I was trying to follow the style that I saw in the sympy.physics.matrices. I could move a test or two for pat_matrix so they're doctests, but I think the other tests for the rot_axis functions should stay with the matrix tests. I guess I could add an example for each of the rot_axis functions. I'm guessing there really should be doctests for the other functions in sympy.physics.matrices. |
…doctests for pat_matrix in sympy/physics/matrices.py.
|
This needs to be rebased and a test run on sympy-bot would be good. |
|
Test results html report: http://pastehtml.com/view/axlivyuqc.html Automatic review by sympy-bot. There are merge conflicts at the moment. |
|
I have cleaned up this pull request - rebased, then squashed the commits into one commit because most of the stuff was implemented in first commit and then moved around and rewritten in later commits. I also made some extra cleanups and bunch of fixes (like wrong exception names) in another commit. https://github.com/plaes/sympy/tree/pull-140 So, should I open a new pull request and this can be closed? |
|
Yes, please open a new pull request. |
|
Ok, new one is @ #515 |
|
Superseded by pull request #515. |
I've added support in sympy.physics.matrices for calculating the parallel axis theorem matrix for translating the inertia matrix. Currently, it uses individual parameters for the distances dx, dy, dz, but that could be changed if desired.
I've also added rotation matrices for rotation about individual axes. While I use them to rotate the inertia matrix, they could be used elsewhere. I wasn't sure if I should put them in sympy.physics.matrices or directly in matrices.
Tests for these are added to the tests_physics_matrices.py file in the tests subdirectory of physics. I'm not sure how to properly call these from the python prompt though, but all the tests pass when run separately.