Dev: Fixes runtime crash when slicing a large array so it becomes small (< 32 elements)

[Skinney]

Includes test to reproduce issue.

The first problem was that depth calculation when slicing down to a one element array (slice 0 1 array) would return NaN, which is a setup for failure.

The second problem was the assumption that hoistTree always would be called with a tree with a size of > 1.

[process-bot]

Thanks for the pull request! Make sure it satisfies this checklist. My human colleagues will appreciate it!

Here is what to expect next, and if anyone wants to comment, keep these things in mind.

[process-bot]

Thanks for the pull request! Make sure it satisfies this checklist. My human colleagues will appreciate it!

Here is what to expect next, and if anyone wants to comment, keep these things in mind.

[Skinney]

Hmm. That second commit was supposed to be a different PR. It's a harmless change though.

It doubles performance of _JsArray_push in Chrome and Safari (no change in Firefox). A similar change of the _JsArray_set function also had some performance increase in Chrome and Safari, but had the opposite effect (way worse performance) in Firefox.

[Skinney]

Hmm. That second commit was supposed to be a different PR. It's a harmless change though.

It doubles performance of _JsArray_push in Chrome and Safari (no change in Firefox). A similar change of the _JsArray_set function also had some performance increase in Chrome and Safari, but had the opposite effect (way worse performance) in Firefox.

[evancz]

Thanks! :)

[evancz]

Thanks! :)