Fail output comparison on unexpected output

[takluyver]

Closes #81.

[vidartf]

Regarding the failure in jupyter/notebook#3212, I'm guessing the reason we didn't check for added outputs previously was because it didn't allow for keeping un-run notebooks for testing. Maybe a solution is to add a check for execution_count if output was added (unless the addition is an unexpected exception)?

[vidartf]

Or, formulated a little better: Cells without an execution_count should implicitly be evaluated in lax mode.

[takluyver]

That makes sense. Do you want to code it up?

[vidartf]

I'll submit a PR tomorrow.

[takluyver]

Thanks!

…

On 15 Jan 2018 10:25 p.m., "Vidar Tonaas Fauske" ***@***.***> wrote: I'll submit a PR tomorrow. — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#82 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAUA9RqSzvfMYUVxkmgNc9AFtpKvPFHuks5tK9BSgaJpZM4RdNc2> .