Objective-C Runtime Overhead about flexlayout

[lucdion]

Hi @flagello, and thanks for the nice comment! 👍

Thanks for pointing me to this yoga's comment. You're right that it would probably reduce the overhead. But I think that before optimizing something (often, its time consuming and it adds complexity to codes) you must be sure what is the real gain in code execution time. So before optimizing you must measure it. In that case, a function call will always be very efficient and compare to the computation required to layout a view and to update the screen, it will be negligible.

Also, using FlexLayout, most code is executed while initializing views, so the cost of calling multiple times objc_msgSend() its done only once.

I will add this to my TODO list and check it when I have the time.

If you have any other concerns or ideas, they are always welcomed.

I do agree — and although it’s indeed a one-time operation I’d be curious to see some tests. Should I decide to use FlexLayout I’ll be sure to test it out myself as it is an interesting optimization case.

[lucdion]

You'right. Did you have the time to checks FlexLayout/Yoga performance in this benchmark: https://github.com/layoutBox/LayoutFrameworkBenchmark

[lucdion]

Hi @flagello, I've thought about your comment, and it comes it was pretty easy to fix it, simply by keeping a reference to the yoga's interface. See #42