sjrd on 0.6.x
Remove dead code: specific coll… Adapt the signature of `js.Arra… Merge pull request #3554 from s… (compare)
sjrd on master
Fix analyzer cycle detection to… Add toString methods to analyze… Do not provide linked ClassInfo… and 1 more (compare)
sjrd on master
Remove Logger.success It is un… Make level helpers final Clean-up ScalaConsoleLogger cod… and 1 more (compare)
@lewisjkl as for not using the bundler – it was quite a while ago, I can barely remember… maybe I couldn’t customize something I wanted to, or something like that, and found it easier back then to handle webpack manually.
Also, as of now I like it better this way: running ~fastLinkJS
in one terminal and webpack serve
in another :) (thinking to try the snowpack)
I’ve made this g8 template a while ago (it’s a bit outdated – it's on my TODO, but the tailwind stuff has barely changed since then): https://github.com/yurique/scala-js-laminar-starter.g8
I'm running in to some issues when upgrading FS2 to sjs 1.4.0. Some are self inflicted due to use of .isInstanceOf[Int]
, which I've fixed. However now I'm seeing runtime exceptions like this:
Exception: scala.scalajs.js.JavaScriptException: TypeError: dest.u.set is not a function
Looking in the compiled js, I see:
$ac_I.prototype.copyTo = (function(srcPos, dest, destPos, length) {
$arraycopyCheckBounds(this.u.length, srcPos, dest.u.length, destPos, length);
dest.u.set(this.u.subarray(srcPos, ((srcPos + length) | 0)), destPos)
});
System.arraycopy
calls. It seems dest
might not be an Array value.
System.arraycopy
with a 3rd parameter than is not an Array value.
def copyToArray[O2 >: O](xs: Array[O2], start: Int): Unit =
if (xs.isInstanceOf[Array[AnyRef]])
System.arraycopy(values, offset, xs, start, length)
else {
values.iterator.slice(offset, offset + length).copyToArray(xs, start)
()
}
Build.scala
, it might need an explicit import of the autoImport._
members.
Hi there :wave: I have a couple of open pull requests for different repositories around Scala.js. Ideally, I would love to get feedbacks & be able to participate in the community. (feel free to discard this comment if it is not in the correct conversation)
scalajs-bundler:
Anyway, thank you to all the people involved in this project and for the time spent!
js.Any
I believe)js.Dynamic.literal()
to produce a crude codec, which is boring but ok, but I'm wondering about the allocate costs of such a thing (?) and thought I'd see if there was a simpler/better solution? I would be doing this 60 times per second...
we don't use web workers, but for lots of small objects we do use literals, but our data structures are small and not recursive, so hand cranking those codecs is simple.
In a different usecase with a huge recursive case class I'm using upickle, but the size of the result doesn't please me - so I want to give boopickle a go, which should improve the serialisation roundtrip. I wonder if it can benefit from 1.4.0 typed array improvements?