Scala.JS 0.6.17 released

https://www.scala-js.org/news/2017/06/04/announcing-scalajs-0.6.17/

by scalatohaskell • 9 years ago

89 59 9 years ago

scala-js.org

partycoder• 9 years ago

Sad that Scala.js will not use WebAssembly, since WebAssembly doesn't provide GC.

sjrd16• 9 years ago

WebAssembly will most likely eventually provide a GC, at which point we'll do what we can to support compiling to wasm.

themihai• 9 years ago

They don't seem be on Chrome roadmap. We'll all grow full beards by the time DOM and GC become available. https://wasmdash.appspot.com

user12413203• 9 years ago

My two cents: to me WebAssembly seems like something Scala Native might address, more than Scala.js. Or even better, both project can sort of blend for this purpose as on one hand (SN) addresses LLVM compilation and the other (SJS) is perfect JS interop (not so sure how WebAssembly does that).

sjrd16• 9 years ago

not so sure how WebAssembly does that

Basically: badly. That's the number one reason we're not trying to compile Scala.js to wasm at the moment. For wasm to get the level of JS interop that Scala.js has, it needs to integrate with JS objects, which means with the JS GC, which means basically have a GC of its own. Hence we're waiting for the wasm GC spec to take life :)

Matthias247• 9 years ago

What do you exactly expect from targetting WASM instead of JS? Higher performance? Smaller bundles and faster load times? Better interoperability?

sjrd16• 9 years ago

Better interoperability, definitely not. Scala.js already has the best interoperability out there, excluding languages with the same run-time semantics as JS (such as TypeScript). In fact, it is the lack of good interoperability in wasm that currently prevents us from compiling Scala.js to that target.

We could hope for smaller bundles and faster load times, yes. But mostly we should be able to get faster "cruising" execution. Scala.js being statically typed, there is a lot of static guarantees that we lose when compiling to JS. JS engines are very good and can "recover" a lot of it through JIT'ing, but it's still 3x slower than Scala/JVM on average (after warmup of both), so clearly we lose something non-negligible. By compiling to wasm, we would lose the JIT, but keep the static types. It is unclear how fast we would manage to be, but I would hope we can get within 2x of the JVM instead of 3x-ish.

There is a lot of misconception on the Internet that "wasm makes everything faster". For example, a lot of people want to "compile JS down to wasm" to make it faster. That's bullshit, because your language does not have static types to begin with, so you lose the JIT and gain nothing in exchange, which means you'll run much slower. But if you compile a statically typed language down to wasm, you should be able to gain some significant speed.

jaimex2• 9 years ago

Does anyone use this?

wbillingsley• 9 years ago

Yes, for quite a few projects.

For a number of them I used scala-js-react, though very recently (as of about two weeks ago) I've been playing around with my own (smaller, less opinionated) virtual dom library. http://www.wbillingsley.com/veautiful/

continuational• 9 years ago

We use Scala.JS in production, and so far we haven't had a single issue with it. Cross-JS/JVM projects have hairy .sbt files though.

slovic• 9 years ago

yes, I've one big project in it. Currently working with: ui -> scalatags, serialisation -> upickle.

lutzh• 9 years ago

Well, for one, the world's 3rd biggest software company, SAP. See http://event.scaladays.org/scaladays-cph-2017#!#schedulePopu... - quote: "The application development stack we are building embraces Scala all over the place – including the development of user interfaces for which Scala.js is utilized to provide a type-safe interface to SAP's own JavaScript UI library SAP UI5."

virtualwhys3• 9 years ago

Sure, replacing a legacy Coffeescript + jQuery app with pure Scala.js. Implementing core functionality of various jQuery plugins (Datatables, Validation, FancyBox, Bootstrap Date/Time pickers, etc.) in Scala.js. Payload is less than 50% of previous version's app bundle. You can of course pull in client-side libs as you like, but for this project it's been really nice to strip out the bloat.

Having Scala's type system on both client and server is pretty surreal, not to mention full IDE support -- the entire world of javascript just appears as you type ;-)

keth• 9 years ago

What's the generated file size after gzipping for the project compared to coffeescript + jquery if I may ask?

Also, can scala.js be used for progressive enhancement on static sites, or is it limited to single page apps due to the need to map scala semantics to js?

scalatohaskell2• 9 years ago

yes, several personal stuff, some internal facing business apps , and few public facing I did as contractor.

It's a whole different world than anything else in frontend. The libraries from scala are immense.

atemerev4• 9 years ago

Yes, we do, and we love it.

If you know Scala, you can mostly forget about Javascript and just code everything in Scala both on server and client side, and everything just works magically (using the same data types is PURE PINK UNICORN AND RAINBOW BLISS). If you don't know Scala, your mileage may vary, but I'd learn a bit of Scala just to experience that.

sjrd16• 9 years ago

Scala.js author here. Ask me anything ;)

atemerev4• 9 years ago

Do your know that you have done the impossible and single-handedly revived Scala? It was slowly dying two years ago, with all that Dotty uncertainty and Typesafe->Lightbend transition.

Then you came and showed us that Scala can be productively used beyond JVM. A new hope appeared, Scala-native was inspired by your example, and Dotty was upgraded from a purely academic exercise to a viable pragmatic way to Scala 3. The attitude was "if Scala.js guys did it, so can we".

So thank you, from me and all Scala hackers out there. Martin owes you a beer. :)

sjrd16• 9 years ago

Thank you for your kind words :) It's always nice to see that one's work is appreciated!

adriaanm3• 9 years ago

Typesafe->Lightbend transition

The renaming never affected the staffing of the Scala team at Lightbend -- we've been doing about 2/3 of core Scala development (major thank you to the community for the other third!) since 2.10 (when I went from post-doc in Martin's lab to the Scala team lead position).

In my opinion, it was a good thing that feature development slowed down in Scala 2.x (while Martin pushed the research frontier forward in Dotty), so we could focus on (in no particular order) compiler performance, a new back-end and optimizer, Java 8 support, modularization,...

virtualwhys3• 9 years ago

The in-process Scala collections library overhaul, what impact will that have on Scala.js when Scala 2.13 lands?

From what I've seen so far there hasn't been much in the way of removals. Ideally the compiler and standard library would be as small as possible in order to help Scala.js streamline generated application binaries.

sjrd16• 9 years ago

Hopefully the new design means less interdependencies between all the collections. That would mean that the reachability analysis of Scala.js could be more aggressive in its dead code elimination.

AzzieElbab2• 9 years ago

Is there any way or anything in the works that would allow dealing with JSON without relying on libs that originated on JVM? They always feels like quite a bit of an overkill considering that underlying platform is JS.

sjrd16• 9 years ago

Sure, you can just use `js.Dynamic`, or write `trait`s describing the "schema" of your JSON, and of you go. Let's say you have have an array of Persons with a name and age, you can type it as:

    @ScalaJSDefined
    trait Person extends js.Object {
      var name: String
      var age: Int
    }

    type Persons = js.Array[Person]

Done deal. With the new option `-P:scalajs:sjsDefinedByDefault` of this release, you don't even need the `@ScalaJSDefined`.

sraquo2• 9 years ago

Scala.js has been an amazing tool for my personal projects. I literally hit euphoria sometimes writing stuff in it. I am yet to truly grasp the benefits of full stack Scala (shared models, Autowire, etc.) but even for purely front-end development it's great, it beats Typescript and Flow out of the water in features, soundness and stability.

I wish I could use it at work since we (Hootsuite) already use Scala heavily on the backend, but I am reluctant in part because Scala.js does not quite have financial support of Lightbend. Or so it seems, it's a bit hard to tell where Lightbend ends and the non-profit Scala Center begins. The latter did pay to get some features implemented but reading their advisory board minutes, I'm not sure if they would have enough funding to pay for the majority of Scala.js development, which if I understand correctly happens for free as part of a PhD right now (note: my information might be wrong/outdated!)

So, if anyone involved with Scala.js is reading this and has better insights on the situation, it would be nice to know.

But I will keep using it regardless. It's marvellous.

sjrd16• 9 years ago

You are mostly accurate wrt. the "financial" situation of Scala.js. Except that the Scala Center has an explicit Recommendation to "ensure the continuity of Scala.js" [1]. At least that means that, should I (the Ph.D. student doing Scala.js right now) stop working on it, the Scala Center has a mission to ensure someone keeps working on it. As with all Scala Center Recommendations, this is not a guarantee. But given the popularity of Scala.js within the Scala eco-system, I am confident that solutions will be found! Read more about the Scala Center and how it works at [2]

[1] https://github.com/scalacenter/advisoryboard/blob/master/min... [2] https://scala.epfl.ch/

sraquo2• 9 years ago

Thanks! You know, if it comes to that, Scala.js community is a passionate bunch, we could probably crowdfund it :)

lomnakkus2• 9 years ago

the Scala Center has a mission to ensure someone keeps working on it.

Heh, I hope it's you! :)

Seriously, scala.js has been amazing for us! Thank you for all that hard work!

EDIT: Of course you might still be beaten out by scala.native -> WebAssembly -> Browser... nah, not really. :)

virtualwhys3• 9 years ago

It would be shocking if they hired anyone but the creator of Scala.js -- hopefully after you finish your doctorate this winter Lightbend and/or Scala Center will step up and make it happen.

lomnakkus2• 9 years ago

(Too late to edit my comment.)

Were you at ScalaDays 2017 CPH? I don't think I saw you there, but you deserve a round of applause, you Magnificent Bastard, you!

adriaanm3• 9 years ago

I am reluctant in part because Scala.js does not quite have financial support of Lightbend. Or so it seems, it's a bit hard to tell where Lightbend ends and the non-profit Scala Center begins.

As the Scala team lead at Lightbend, I'd love to have a few members of my team focus on scala.js and scala-native. We do financially support their development (most recently, as part of our funding of the Scala Center).

As a business, it's a bit of a chicken-and-egg problem: our customers usually indicate they are hesitant to switch from JS for their frontend work. It would be great to have more customers provide feedback like yours (via our customer surveys)!

user12413203• 9 years ago

If I was Lightbend (whose work I use/enjoy/thank for) I think I would put Scala.js inside the Play Framework (as opposed to say Coffeescript)

scalatohaskell2• 9 years ago

I am exactly in same situation and experiences with Scala.JS as you... my #1 choice atm.

galfarragem• 9 years ago

Scala is still growing popular:

https://insights.stackoverflow.com/trends?tags=scala%2Cgo%2C...

brianwawok2• 9 years ago

Looks like it has peaked to me..

todd8• 9 years ago

This looks pretty good, but Google trends doesn't show any growth for functional languages over the years :(

https://trends.google.com/trends/explore?q=scala%20programmi...

Scala.JS 0.6.17 released

Related Stories