Dialog: Allow for subpixel calculation errors in tests #1737
Conversation
|
@scottgonzalez, thanks for your PR! By analyzing the annotation information on this pull request, we identified @jzaefferer, @mikesherov and @arschmitz to be potential reviewers |
Tests were failing in IE 10-11 with values that were off by 0.01 pixels.
95b4b79 to
6bf10e8
Compare
| actualDH = heightAfter - heightBefore; | ||
| actualDW = widthAfter - widthBefore; | ||
|
|
||
| assert.ok( Math.abs(actualDH - dh) <= 1 && Math.abs(actualDW - dw) <= 1, "resized[50, 50] " + msg ); |
There was a problem hiding this comment.
How about using two assert.close() calls instead? It already does the Math.abs and comparison stuff.
Something like this (didn't test):
assert.close(actualDH, dh, 1, "...");
assert.close(actualDW, dw, 1, "...");
Would provide much more useful output for failures.
There was a problem hiding this comment.
I was going to, and I can, but it means updating every test that uses this helper to adjust the expected count.
There was a problem hiding this comment.
Also, if we're going to do that, we should update the helper above, and probably many other places. Didn't seem worth it right now.
There was a problem hiding this comment.
I see. Let's add a TODO comment then, for whenever we end up having to adjust all the tests anyway.
There was a problem hiding this comment.
I see. Let's add a TODO comment then, for whenever we end up having to adjust all the tests anyway.
Done.
|
👍 |
Tests were failing in IE 10-11 with values that were off by 0.01 pixels.