do you think that would be a good enhancement?
# tornadofxc
carlw
09/04/2018, 5:50 PMdo you think that would be a good enhancement?
r
Ruckus
09/04/2018, 5:51 PMBy "that", do you mean changing JavaFX to support structural pseudo-classes? If so, I personally don't think so, but that may have more to do with my style than anything.
c
carlw
09/04/2018, 6:07 PMno, enhancing CSSSelectors so that you can do styleclass.child(3) for the 3rd child
r
Ruckus
09/04/2018, 6:22 PMHow would that work?
Ruckus
09/04/2018, 6:25 PMJavaFX has no support for structural pseudo-classes, so there's no way we could have
styleclass.child(3)c
carlw
09/04/2018, 7:10 PMthinking more of a css preprocessor
r
Ruckus
09/04/2018, 7:11 PMI'm still not following...
c
carlw
09/04/2018, 7:11 PMapply a generated style for example
carlw
09/04/2018, 7:11 PMrun a tornadofx preprocess on the style objects and add new objects based on these extension functions
carlw
09/04/2018, 7:12 PMsomething like
vbox {
nthChild(3) {
background = RED
results in a new style object to be applied to the vbox's child
carlw
09/04/2018, 7:12 PMw. a listener for changes to vbox
r
Ruckus
09/04/2018, 7:14 PMAh, I see, though I don't think it would work. If I add a new
.vboxRuckus
09/04/2018, 7:15 PMThe CSS DSL just generates CSS text, it ultimately has nothing to do with the running application itself.
Ruckus
09/04/2018, 7:17 PM(Much like Sass, Less, or any CSS preprocessor, you should treat it as statically compiled AoT. By the time the app sees it, it's just plain CSS.)
c
carlw
09/04/2018, 7:21 PMoptional AOP on the builder?
carlw
09/04/2018, 7:22 PMwhich results in an aspect putting the builder's object through the preprocessor
r
Ruckus
09/04/2018, 7:26 PMThat seems extremely fragile, and would pretty much break interop with any JavaFX libraries (or at the very least not work with them).
3 Views