Chris B

I just looked at the source code and see line uses `agg=any()` by default, while points uses `agg=count()` by default. So just a note to confirm that the findings above...

Yes, I guess the title should have been more like, 'Is line the expected amount slower than points?', or maybe, 'Can line be made faster?'.

I think examples/environment.yml is probably out of date - the dependencies are quite different from the current ones (see https://github.com/pyviz/datashader/blob/master/setup.py#L32, which is the basis for the conda package). In the...

Just an aside: I see above in a few places using `pip install -e .` in conda environments. Maybe that's ok now with the latest conda, or if you know...

Before I can get to the actual fix, I'm having some problems with the test framework :) 1. Where is `--nbval-lax` tested? I.e. I wasn't sure where to put my...

Hmm, re. problem 2 above: travis does not have the problem. So I guess it's related to my environment; I'll try a new environment and see what happens.

I had been using conda. I tried in a newly created python virtualenv (using my system python 3) - same problem :( I suppose I should compare the packages here...

It's not the dependencies. Locally I get * Same correct result as on travis if I run `py.test -v tests --nbval --current-env --sanitize-with tests/sanitize_defaults.cfg --ignore tests/ipynb-test-samples` * The weird `traitlets.traitlets.TraitError:...

I could try to narrow down, but I don't suppose you want to take a guess @vidartf ?

Ok, the test results are now as I expect, i.e. they reflect the bug I'll fix in this PR: `%this_magic_does_not_exist` passes silently (which is wrong, hence causing the tests added...