Can you please explain in details “contexts will build up: memory leak” statement?
Using recursive setTimeout guarantees that method which we want to execute on some interval will never show up 2 or more times on the execution queue. Since next call in the recursion will only be initiated if the method is fully executed…
I guess that they got sick of other people using their library in a “wrong” way so they made the new version so that it can be used only as is.
Honestly I think that in modern FE development there is not so much space for such things like Bootstrap. Design for any project is almost never generic, thus the implementation can not be generic as well.
This comment is in conflict with your first comment. You first pointed out that to achieve the true obfuscation of CSS selectors one needs to use that lib which does renaming. Now you state that you use CSS modules, which is legit. As I stated in the article, if you are not comfortable with 4 character long obfuscation just go with 5 or 6 or … also…
Still the way I showed that it can be done is not only for class names. It is applicable to all CSS selectors out of the box. Also, it is very configurable and if you think that you are safe to have 1 character long selectors you are free to do so.
Good point! But if IE completely ignores this feature that is pretty bad in Front-End world. IMHO 90% support is quite bad and 66.32% (which caniuse shows for async functions) is horrible.
Additionally caniuse do not mention await operator, just async function in general. Am curious to know actual support for await.
You are a unique person Stephanie and lots of people enjoy your creations as much as you do. That is the key to success.
Am awaiting conference video of your talk :)