div(derive(() => vs derive(() => div: Which is "better"?

[bobinspace]

const A = () => div(derive(() => state.val ? 1  : 0));
const B = () => derive(() => div(state.val ? 1 : 0));

Although both components are visually identical, is there some "best practice" that favours a particular implementation?

[bleistivt]

A will only change the Text Node contained in the div, when the state changes, B will recreate the div, so A is preferable. It could also just be rewritten as:

const A = () => div(() => state.val ? 1 : 0);

Edit: In general, you will not need to use derive very much when dealing with the tags functions of VanJS. When a function appears in attribute or content position, it will automatically be wrapped in a derive.

[bobinspace]

When a function appears in attribute or content position, it will automatically be wrapped in a derive. <-- Thanks, I didn't know that.

[Tao-VanJS]

You can also refer to Minimize the scope of DOM updates section in Advanced Topics page.