create workflow ci #6
Conversation
|
you ok if I do a squash-and-merge, or should I crush these down to a single commit first? |
testing a workflow ci reverting/patching lot the plot: exception pep 8 ing Trying out badge Modified rules, refactored code accordingly Fixing test import paths checking in on imports adding a touch function to use cython imports, horribly had a bad assertion
samgawthrop
force-pushed
the
samgawthrop-gh-ci-1
branch
from
9bd0176
to
3cc29ec
Compare
| def touch(): | ||
| """Touch all imports from cython to alleviate flake8.""" | ||
| if valenti_q_acc == valenti_q_mag: | ||
| print(rotation_vector_to_matrix) | ||
| identity(valenti_q_mag) | ||
| raise Exception("Never run this") |
There was a problem hiding this comment.
I'm not clear on what this does.
There was a problem hiding this comment.
This is an abomination to use the things imported from the _pyquat. I regret it already, but I'm not sure how to do this any better.
Unused imports are bad, but the unused imports here are kinda-sorta necessary to load the context for the other things downrange that use this init as the conduit into the cython.
Cython itself didn't have any particularly clear patterns on how to explicitly import without leaving unused imports, or having unclear context.
| else: | ||
| raise InputError("expected 1- or 2-D array") | ||
| raise Exception("expected 1- or 2-D array") |
There was a problem hiding this comment.
This should be a ValueError.
There was a problem hiding this comment.
fair I couldn't find a deceleration for InputError. I'll change these. I too dislike base Exceptions.
| for i in range(1,ary.shape[1]): | ||
| axes.plot(ary[0,i-1:i+1], ary[1,i-1:i+1], ary[2,i-1:i+1], c = tuple(c[:,i]), **kwargs) | ||
| for i in range(1, ary.shape[1]): | ||
| axes.plot(ary[0, i - 1:i + 1], | ||
| ary[1, i - 1:i + 1], | ||
| ary[2, i - 1:i + 1], | ||
| c=tuple(c[:, i]), **kwargs) |
There was a problem hiding this comment.
I don't love this type of change. It's going to make some code absolutely unreadable. I realize that expressions should be space-separated in general, but these are expressions within bracket notation.
| xh = numpy.zeros((3,2)) | ||
| xh = numpy.zeros((3, 2)) | ||
| yh = numpy.zeros_like(xh) | ||
| zh = numpy.zeros_like(xh) | ||
| xh[0,1] = axis_size | ||
| yh[1,1] = axis_size | ||
| zh[2,1] = axis_size | ||
| xh[0, 1] = axis_size | ||
| yh[1, 1] = axis_size | ||
| zh[2, 1] = axis_size |
There was a problem hiding this comment.
Same comment as with the array indexing. I don't think this change substantially improves readability.
There was a problem hiding this comment.
Okay, I'm going to push back on the second part of this one a bit more because it doesn't actually appear in pep8: https://stackoverflow.com/questions/31741418/pep8-space-after-a-comma/37064013#37064013
I think there should be spaces after commas except within array notation. The examples in pep8 do show spaces after commas, even though this isn't specifically enumerated; but they don't show multi-dimensional array indexing notation.
Thoughts?
| @@ -55,11 +54,9 @@ def rand( | |||
|
|
|||
| """ | |||
| if axis is not None and angle is not None: | |||
| raise StandardError("expected non-fixed angle or non-fixed axis or both") | |||
| raise Exception("expected non-fixed angle or non-fixed axis or both") | |||
There was a problem hiding this comment.
| raise Exception("expected non-fixed angle or non-fixed axis or both") | |
| raise ValueError("expected non-fixed angle or non-fixed axis or both") |
| @@ -58,7 +60,6 @@ def test_davenport_eigenvalues(self): | |||
| [1.0, 0.0], | |||
| [0.0, 1.0]]) | |||
| B = pq_esoq.attitude_profile_matrix(obs = obs, ref = ref) | |||
| irot = pq_esoq.sequential_rotation(B) | |||
There was a problem hiding this comment.
This line is needed. It does an in-place rotation on B.
There was a problem hiding this comment.
so it should just be pq_esoq.sequential_rotation(B)? the irot variable isn't used. Also curious why it passes tests with an essential rotation removed, but that's a differnt question.
There was a problem hiding this comment.
I suggest an additional line that verifies that irot is not None. self.assertNotEqual(irot, None) maybe?
| q = pqv.dq_acc(a0) | ||
| q = pqv.q_acc(a0) |
There was a problem hiding this comment.
IIRC, these are different functions.
There was a problem hiding this comment.
where's dq_acc declared? I see valenti_q_acc imported through the base init , which is laundered through wahabi.valenti to be q_acc.
from pyquat import valenti_q_mag as q_mag
from pyquat import valenti_q_acc as q_acc
but I don't see dq used elsewhere in the code, is that renamed elsewhere in an import?
these are getting hard to follow.
There was a problem hiding this comment.
I think we settled this during our zoom meeting yesterday. Two additional import lines should be added (in addition to the two imports you gave in your comment):
from pyquat import valenti_dq_mag as dq_mag
from pyquat import valenti_dq_acc as dq_acc
and the above change should be reverted.
| dqm = pqv.dq_mag(l) | ||
| dqm = pqv.q_mag(l) |
There was a problem hiding this comment.
Also different functions.
| @@ -47,7 +43,7 @@ def test_q_mag_points_north(self): | |||
| self.assertEqual(dqm.y, 0.0) | |||
|
|
|||
| q_G_to_L = qa * qm | |||
| q_L_to_G = q_G_to_L.conjugated() | |||
| # q_L_to_G = q_G_to_L.conjugated() | |||
There was a problem hiding this comment.
| # q_L_to_G = q_G_to_L.conjugated() |
testing a workflow ci