Clojurians Log v2

Cheers! Helpful intro to route->url .

thanks tony for the quick answer.

I would nudge you to use the function/macros, because I personally think they are better, but in an objective sense there’s not really a ton of difference that can’t be overridden by personal preference. The objective things are: • The fulcro macros can, in dev, add source lines to the HTML so you can find the code that emitted them. This is a huge plus in large projects. • The macros, when used properly, emit low-level js with NO runtime overhead, so they are as fast as you can possibly go. • There are both functions and macros (seamlessly chosen for you by the compiler). This isn’t an advantage, just a note that you can use the dom elements as values. • The :classes pseudoprop on DOM function/macros makes composing dynamic classes a little cleaner. • The :.classname#id support in the macros does the transform at compile time. The syntactic convenience has no cost.

(div :.x {} "Hello") 

[:div.x {} "Hello"]

and the former in Fulcro results in

<div data-fulcro-source="namespace:44" class="x">Hello</div>

In shadow-cljs builds you just include this option to enable the source annotations:

:dev {
   :compiler-options {
     :external-config {
        :fulcro     {:html-source-annotations? true}}}}

I went to a lot of trouble to make all of that work super well in a first-class sense…Also, the INPUT types in React need special treatment, which hiccup processors probably do in a different way. I assume they will work right, but I have not used them.

(the symptom of a broken React input is that if you try to edit the middle of the value when using it in an async controlled sense will cause cursor jumps)

I am sold. Thanks for this details. Will start with the macros then.

The performance of the dom function/macros in Fulcro is actually pretty much native js performance. You probably won’t have any issue with using hiccup if you really want to, as the performance difference really isn’t significant in most cases.