Slashdot Mirror

Haxe

I use Haxe http://haxe.org/.

I've been looking through the Haxe documentation, but I can't see how it does memory management. Does anyone know? Is it garbage collected?

Haxe is a dream for cross platform development and has been paying the bills for me for a good 6 years now.

http://www.haxe.org/
https://github.com/HaxeFoundat...

It takes 2 clicks on Haxe's site to see it can be used with lots of different kinds of client and server code. Flash is mentioned as an "also, haxe can make swfs" http://haxe.org/use-cases/web/ (despite Flash being a huge part of Haxe's maturing development) http://en.wikipedia.org/wiki/H...

Flash development and ActionScript as a language were never "shit". It certainly was abused and mismanaged, but technologically Flash/AS was amazingly useful -- especially in tying animation to code.

If you ever are willing to challenge your own beliefs you should take some time and checkout Haxe, and Apache Flex. Try keeping an open mind to technologies that greatly shaped the web we have today. A lot of ECMAScript was based on lessons learned from ActionScript. A lot of web games and comics were brought to you by Flash. YouTube, Twitch, Hulu, Yahoo Maps (formerly), and thousands of games, all were built on the backs of Flash. Firefox's original JIT was based on Flash 9 and donated by Adobe and is the second largest open source code donation ever to Mozilla.

Does Flash have problems? Definitely.

Should you dismiss a huge part of the web out of hand? Only if you want to make yourself look like a fool.

It's not required, no. At least not anymore, not sure if it used to be. The self-contained installer is sufficient to develop with.

The IDE with the name "Flash" (or "Flash CC" in it's current version) is by far the best 2d animation tool in the industry. That said, despite an ever increasing IDE set of feature, it's horrendous for coding and debugging. The OpenSource project "Flash Develop" ( http://flashdevelop.org/ ) made AS3 usable by the many hobbyists writing games, as well as the AAA's doing UI work via Scaleform's Flash player.

For those not on the Flash/AS3 scene: there was the meme "Flash is Dead" that started about 3-5 years ago. It's not dead, as-in not at 0% usage, but for about two years it hasn't been a viable tech for most indies to use. (Flash via Adobe's AIR technology does work great on mobile but for some reason, perhaps due to the need of "Flash Builder", this doesn't have as great as a traction amongst indie game devs.) Most indie/AAA devs who really did a stellar job leveraging the low-level bits of Flash, ended up going to HTML5/Javascript or C#/Unity ( http://jacksondunstan.com/ ) . A few did jump over to HaXe ( http://haxe.org/ ), and the award winning "Papers Please" game showed HaXe is viable for indie commercial projects... but it's unproven for larger scale projects and the smaller size of the dev team working on HaXe, has some companies hesitant to explore it.

So it's great Adobe is adding these hooks to allow OpenFL / HaXe to become more accessible, and thereby help out both the Flash community and their own communities. ... but what about "Flash Builder"? The other Flash IDE, built upon Eclipse that is so broken that if you delete a local project through the Finder, it prevents the whole IDE from even starting up? Is Adobe dropping it? Are they adding the functionality to it? Are they going to make it as friendly to use as FlashDevelop? (I'd love to not have to boot Parallels, just to use a Windows-only IDE.)

Half of the (former-)Flash blogs I follow, sound as-if Adobe is transitioning away from Flash, putting resources into HTML/Javascript tools instead. And then occasionally, I hear about some new (usually game industry-related) features Adobe is installing in their Flash tools. But even when 100's of indie developers were making a full-time living, selling Flash games, there wasn't a single year at the Game Developer's Conference (GDC) that Adobe had a Flash presence and talked about games with their technology (with the exception of one year showing off "Adobe Director".)

Depending on if/how the sale of Unity goes to Google, or Microsoft, or whoever... this may be the one opportunity where Adobe can enamor game programmers with a Flash-based development environment (maybe other business sectors as well.)

It will be interesting to see how this plays out.

What constitutes indie is one questions (and AAA is even harder to come to a consensus, even among my work peers) but that said...

As a child of the 80's, who adamantly played video games (e.g., Apple ][, arcade, 2600, NES, etc...) and got into professional game development over 10 years ago (I work for a AAA studio and my have my own gig for nights/weekends) I'd agree with those who say now, 2014, is the best time for indie game development.

Powerful engines and Middleware tools are accessible with licenses that fit indie budgets (e.g., Unity3d, Unreal4, etc...) as well as a swatch of free software for development. (e.g. Phaser: http://phaser.io/ Blender http://www.blender.org/ Love https://love2d.org/ Flixel http://flixel.org/ Haxe http://haxe.org/ )

The internet, as-is, provides indies with a way for
- distance-collaboration (Skype, E-mail, Groups, etc...)
- community building (Twitter, CMSs, Facebook, etc...)
- fundraising (IndieGogo, Kickstarter, HumbleBundle, Paypal, custom web-based donation system, etc...)
- advertising (game communities, news outlets, etc...)

Organizations, such as the International Game Developer's Association (IGDA, http://igda.org/ ) and events like the Global Game Jam, PAX (IndieMegabooth), and MAGFest also contribute to the community of indie game developers.

It is a great time to be an indie game developer in terms of accessibility and ability to achieve a sustainable income.

There isn't an automated way (that I know of), but it's not hard. Most of the syntax is pretty much the same, and part of the point of OpenFL is that the library calls are all the same.

The one time I tried porting a complex AS3 class over to Haxe/OpenFL, I started by copy+pasting the AS3 source. Then, I spent most of my time changing things like Flash Vectors/Arrays to typed arrays, minor changes to for loop syntax, different type names (e.g. Boolean->Bool), and other differences.

Actually, I think there's a really good summary out there...here we go:
http://haxe.org/doc/start/flas...

You can see quite a lot is the same. And I heard there are some even trying to create auto-converters. But it wasn't too hard to convert in my experience. More tedious than difficult.

That's why we need more people using languages like this: http://haxe.org/

It seems as if have seen a hundred new languages, all allegedly solving the problem of programming, and none of them succeeding. That's because the difficult problems are not at the level of syntax.

That's why we need more people using languages like this: http://haxe.org/

I discovered it a couple of months ago. It has its quirks, but not having to worry about rewriting your entire app for another platform is a blessing. And no messy VMs needed.

You know, I like C# and Visual Studio - if I could easily write code that would run across not just all the Windows platforms, but Android and IOS too - and with a UI that looks native on each platform, like QT does - that would be a wonderful thing.

Recently I fell in love with this, which seems to do exactly what you describe: haxe

The consistent UI seems to be the only thing missing; everything else is there.

The weird scope rules and lack of proper object/class support drive me up the wall when working on projects with ~40,000 lines of code.

HAXE has has classes, static typing and all the things you probably want. Then you can just transpile it into javascript (among other languages). There are other efforts along similar lines. Increasingly it seems, due to it's unsuitability for large projects, Javascript is being treated as a sort of assembly language you compile to as a last step.

http://haxe.org/

Multiplatform, statically typed, open source...outputs to desktop and mobile platforms as well.

Any thoughts on this one??

That Mozilla link is along the lines of what I was thinking of for dynamic audio. Too bad it's not supported by all browsers, but it would be a start. With some proper architecture, it should be easy enough to add support for other browsers when they support a similar feature.

Thanks!

I should probably start looking at what it would take to port my Flash NES emulator to JavaScript. I wrote it in Haxe with the goal of doing a JS version at some point. However, at the time, only Chrome could even come close to running the JS fast enough, but now most modern browsers should be fast enough (I'll have to do some performance tests). It should just require minor tweaking to replace the flash calls with JS/HTML5 calls without having to re-do and re-work the bulk of the logic.

Flash (the authoring tool) can now produce HTML 5 Canvas + JavaScript. Flash Player (the runtime) was produced for years because there was no other viable option for running the output of Flash.

The output right now pretty much sucks, but they're working on it. The Flash IDE is pretty nice, so if people who have invested time in learning it can eventually put the output onto a standard web page without requiring a plugin on the client, I'm sure that's what most people will be smart enough to do.

Also, Adobe has done a lot of work on standardizing the SWF format and even the save format used by the Flash IDE. Macromedia's versions used to just dump a memory image to disk to save a Flash project. Now you can save it as an XML file that can be worked on with a node editor, text editor, XSLT, or whatever. The SWF format targeted at the Flash Player is even published so that other players can be written to the exact spec, although HTML 5 + JavaScript will hopefully be the dominant output from the IDE soon.

Now, I don't see the multi-hundred dollar Flash development system itself becoming open source any time soon. Adobe does have some tools they've put in the open realm, though. The quality of clones both open and closed of the IDE is improving. There are scores open source tools that output to SWF now that could also output to HTML 5. One programming language I've worked in (haXe even targets SWF, JavaScript+HTML DOM, or the Neko VM selectably (but with different libraries and some differences in capability for each).

No, it isn't. Clearly you haven't heard of haXe or Swish, and those are just two examples; there are many other applications out there that are capable of creating Flash files.

I normally try and defend Apple, but this one has me feeling like... well, any true nerd will get the Blackadder reference in the title.

I kinda understand the "innocent explanation" for this - the Mac platform has had its share of half-assed "ports" that don't really follow Mac UI standards as well as development tools to produce such ports - but generally users and developers have eventually voted with their feet (anybody remember when Microsoft produced a Visual C++/MFC cross-compiler for Mac?) - something Adobe are risking with their current Mac offerings.

However, this could be achieved by setting UI and performance standards for apps, not the blunt instrument of dictating which programming language could be used and making it very difficult for developers to target multi-platform.

Of course - this is all Job's risk: even in the phone/tablet market, Apple only "dominates" certain narrowly defined sectors and doesn't enjoy the sort of near-total monopoly that MS enjoyed in its prime. If he decimates the iProduct developer community then its still possible for the market to show him the error of his ways.

Meanwhile, I'll repeat what I've said before: the future in mobile "Apps" might just be browser-based apps hosted by the "cloud" or on your home server which completely bypass the App store and these rules - so although the end result has to be in Javascript, you are free to use cross-compilers like this or this.

Right now, mobile internet is barely up to snuff for this, but once that improves it makes so much more sense in a world where you may have multiple mobile gadgets.

The other 10% - games - it is great at that but native platforms are stronger.

Provided that the native platform is even open to development. For example, the only legit way for individuals and small companies to make and publish games for Wii is through Internet Channel. (Otherwise, you have to have a corporation or LLC, a dedicated office, and one or more prior published commercial titles.) The iPhone/iPod Touch devkit is far easier to get, but it still costs $600. Tools for JavaScript and SWF development, on the other hand, run on the commodity PC that you are more likely to already have.

The Flash plugin is closed source but Adobe has published the SWF specification. There are third party tools which produce SWF output which will run in the Flash plugin. My favourite is haXe, a Java like language which compiles to a number of targets including SWF.

The last time I looked into haXe, it mentioned loading assets from a .swf file created with swfmill or Sam Haxe. But unless I have Flash, I can't make vector animations; I can only import PNG and JPEG. If I'm limited to raster assets, I might as well code for JavaScript Canvas to run in Chrome Frame; at least that'll run on an iPhone.

Don't forget about sound.

To which model were you referring?

There are a few other means of doing it as well, such as writing as few portions of the code as possible in native code and doing the rest as scripted code. Generalize the platform-specific calls as much as possible so that you have a very small library of graphics, sound and input functions which require platform-specific code. You're going to have to fix bugs no matter how good your cross-platform libraries are - nobody said game programming was easy, after all.

again, C# translates fairly readily to Objective-C.

Can it be done automatically, so that any changes I make in the C# are reflected in the Objective-C?

Would you trust a tool which claimed to do it automatically without bugs? Again, game development isn't easy. Everybody ends up putting up with the same crap in their development process; the decision as to whether the effort is worth the potential gains is yours.

hAxe is a fairly robust Actionscript compiler which can produce fully funtioning SWF movies.

The last time I looked into haXe, it mentioned loading assets from a .swf file created with swfmill or Sam Haxe. But unless I have Flash, I can't make vector animations; I can only import PNG and JPEG. If I'm limited to raster assets, I might as well code for JavaScript Canvas to run in Chrome Frame; at least that'll run on an iPhone.

Apparently C# XNA code isn't too bad to port across to Objective C and the iPhone SDK.

There is more than one model of developing an application for multiple platforms. First, there's the waterfall model, where a program is completed on one platform and then translated line by line into the language of the other platform. Translation errors between languages can and do happen, and changes to the old version don't propagate automatically to the new version.

But then there's the front-and-back-end model related to MVC, where the "back end" (gross physics, AI, and arguably map loading) is implemented once and linked into each version, while the "front end" (graphics, audio, input) is implemented for each platform, and the two call into each other. An advantage of the front-and-back-end model is that if a bug is fixed in the back end on one platform The big disadvantage of the front-and-back-end model is that all platforms have to support the language in which the back-end is written, and that's more difficult if you have to deploy the same app on XNA, SWF, Java, and iPhone, all of which use different languages.

To which model were you referring?

As far as physics goes, apparently Bullet has both iPhone and XNA flavors and uses the zlib license

By "physics", I mean it in a generalized sense, including all rules of a game such as how much damage an attack does.

again, C# translates fairly readily to Objective-C.

Can it be done automatically, so that any changes I make in the C# are reflected in the Objective-C? Or does it leave open an opportunity for errors in manual translation if I try to keep codebases in sync?

Well, haXe is a Java-like language which compiles to Javascript among other languages. It's definitely general purpose enough to use in the client as well as the back-end, and it has nice libraries which make communication between them easy. It happens to compile to Flash, too. I'd imagine that any reasonably complex project could be built just about entirely in haXe.

And if you don't want to directly code in Javascript, some great alternative languages can actually compile Javascript code :

- haXe : http://www.haxe.org/
- Fan : http://www.fandev.org/

haxe looks promising, heres the link :o

http://haxe.org/

if only someone ported html to xhtml :(

This is good news for the haXe community.

And don't forget about haXe as well, which has been designed to be able to output JS, but also PHP, SWF, Neko, and more recently C++

ask, and ye shall receive:

http://haxe.org/

Does your compiler produce typed code for Flash and untyped for Javascript? Note that typical web code is untyped, and on untyped code the Flash Javascript interpreter is pretty slow, since ActionScript is typically typed.

I've no idea - I didn't write it, just playing around.

AIUI, haXe itself is strong, statically typed with type inference. However once all the types have been inferred and it is generating JS or Flash in the backend then it's anyone's guess what happens to the type information. Guess you'd have to read the compiler source or the generated code to find out ...

Rich.

Safari, FireFox, and Opera (in that order) have been showing marked improvements in Javascript performance. To the point where Javascript performance is a major point of competition.

I was playing with haXe which is a programming language that targets both Flash and Javascript back ends. You can (within some constraints) compile the same program to both targets, so that's what I did yesterday:

• Flash version (requires Adobe plugin)
• Javascript version (requires Firefox 3 or Safari with the canvas widget)

Use keys 1-9 to change demos, and click with the mouse to fire blocks.

I found that Adobe's Flash plugin beat everything, Safari+Javascript was pretty slow, and Firefox+Javascript was in the middle (but still pretty slow compared to Flash). Gnash doesn't run this demo at all, although it can run other haXe code.

Rich.

Next step is that we shall be able to see a server-side compiled JavaScript

If you're talking about servers sending precompiled bytecode in place of javascript, no that isn't the next step or even on the agenda. Unless that is you have a VM in javascript.

In that case we will be able to have language-agnostic browsers since the compiled code won't necessarily have to reflect which language that was used to write the script.

You don't need bytecode for that.

• http://haxe.org/doc/intro
• http://www.openlaszlo.org/
• http://code.google.com/p/pyjamas/
• http://code.google.com/webtoolkit/

Let's spell it out, some want to use the CLR for web applications (I'm never installing Mono). Some don't like the fact that web client apps in javascript are inherently open source. Seems to me that there are more compelling reasons not to standardize on a common bytecode.

Flash apps must be developed in ActionScript, and must developed in IDEs specifically designed for Flash - of which there are about 3.

No, some alternatives exists such as:

• haXe which is a compiler that target the Flash VM (among others).
• OpenLaszlo

I'd love to see Haxe ( http://www.haxe.org/ ) supersede Ecmascript.

Besides, ANYTHING is better than javascript.

You may want to take a look at haXe.

There's already Java (with GWT) and haXe

I've been making SWFs on Linux for years. Swfmill is quite capable (the svn version has very good SVG support and works well with Inkscape), there is a fine language and compiler called haXe that can even compile for other targets as well (the Neko and generated Javascript, with PHP support in the works), among other tools.

Also, the Flex SDK is already open and works on Linux (it's Java). Finally, their (proprietary) Flexbuilder for Linux is currently a public alpha.

If by supporting you mean "have thrown an alpha or two over the wall for 32-bit x86 processors back in December", then yes, Adobe supports Linux with Flex.

Personally, I'm very happy about them releasing alphas. It's already quite usable.
Also, there's another commercial IDE, the SDK itself is under the MPL, and there are alternative (non-Adobe) tools as well.
Anyway, I highly recommend haXe, it's a fine language that you can also use to generate JavaScript, with a great type system.

These new features are nice and all, but what I really want as a Web developer is for a Javascript standard thorough and widespread enough that I can write scripts that work on most browsers without a bunch of hacks to make sure that each browser gets the right code. Anyone have a prognosis on this?

This submition is breaking new ground: Proprietary MS plattform lockin now combined with the crappyness of pointless fanboy analysis in the style of über-crappy layouted GNU Open Source web posts. Great.

No thanks people. I'm staying the heck away from MS, .Net, 'SilverShite' and whatnot. For anything. But especially for web stuff. And because nearly 20 years of very good reasons too.

There are a bazillion webkits out there, one better than the next and *all* of them open. (I just ran across haxe again the other day - yet another totally awesome OSS solution). No need to even waste a minute considering MS. Unless you're being paid tripple for your time, waste of brainpower and for learning stuff that corporations have no interest in keeping around for long, because they want to sell it again and again.

Take it from a professional web-developer: Stear clear. Jump on the Rails bandwagon if you must (only if you really must) but do yourself a favour and stay away from .Net.

My 2 cents.

All you need is a text editor and a text-oriented tool for Flash to get a Flash site going.

There are lots of tools for Flash-compatible SWF files out there besides Flash. Flex is one. HaXe is another. Laszlo Systems has a proprietary product and an open version called Open Laszlo, which IIRC is built on Java. There are probably more I'm forgetting.

HaXe is its own language from the guy who designed the Neko VM. It run on the Neko runtime, and it targets Neko, Javascript browser DOM with its own Ajax libraries, or Flash. I haven't done anything huge with it, but it was pretty quick to pick up for a couple of small projects.

There are also graphical Flash authoring tools besides Flash and Dreamweaver. They range from Swish Max which is meant to be a full Flash replacement for most people down to specialized things like animated banner creators and photo gallery creators. There's also a lot of royalty-free and even some Open Source components you can download and reuse.

Flash isn't as open as JavaScript and HTML, and it is dominated by one company. It's not exactly useful only to people who buy Flash, though.

actually you would want to look at haXe mtasc was AS2.

Free development in Flash has been around for a while. I particularly like working in Haxe.

To me, I had never been interested in Flash development because the dev environments I saw always seemed semi-hostile to something other than timeline-based animation. With haxe, it's just you and your text editor - the way programming should be.

For one sense, the FSF uses the term "freedom" rather than "openness". For another, what is so non-free about haXe and Gnash?

I happen to think that Python and Ruby (fine scripting languages that they are) make poor replacements for PHP. Haxe or Pike are better but not as widely available.

From a hosting perspective, I'd like everyone to target parrot and we can have done with it.

So that raises a few questions:
1) Can reverse-engineering the file format give enough information to make a fully-featured flash decoder/player?
2) Will Adobe try to stop such reverse-engineering efforts?
3) Is it worth it to continue along the Flash route, or should supporters of Open Standards promote an alternate vector-based animation/movie format?

1) What is it you're missing? Google SWF spec and the first hit will be the specs without the restriction. It seems to be a copy of the original specs, though, so better stick with Alexis' reference. You could also read the sources of various tools, or the existing OS player. For the opcodes of the new VM, you could read the sources of a compiler or the VM itself.

2) Apparently they didn't. Neither did they try to stop the OS streaming server.

3) Both would make sense, depending on how you plan to use it. For the web, SWF will stay the king, IMHO. Users don't like installing additional plugins, but that wouldn't be a problem for standalone apps.
The bigger problem is content creation, both for a new format and for SWF. There are very good OS tools for making SWFs already, but they are focussed on programmers. If you want to write code and maybe include some assets like graphics and fonts that you then use with it, I'd say you're better off with the OS tools. But for graphical work like animations or layout, there isn't really a way around Adobe products for professional work (just that you can do animations and layout without them doesn't mean your designer will consider it an efficient, comfortable workflow, and he's right). This is where work needs to be done. Better SVG import for the tools, or direct SWF output for Inkspace or even a specialized app.

http://www.osflash.org/swfmill

also see

http://mtasc.org/ (a better, faster, open-source Actionscript compiler)

and http://haxe.org/ (a better, faster, open-source language for writing flash and other things in)

...but have you looked into using Flash?

There's Red5 as an open source streaming server, as a compiler you could use the open source haXe or the free-as-in-beer Flex 2 SDK.

I bet on neko having AMD64 JIT first.

I wouldn't be so sure, there will be many more working on Tamarin then on the Neko VM.
But who cares which one will be first, haXe can be compiled for both. :)

In my personal tests, Actionscript is over 100 times slower than Quickbasic. Why the hell is that the case? Both are interpreted languages. Actionscript even compiles to bytecode before it's executed, and I think Quickbasic does something similar as well. Does static typing alone really cause a language to run faster? Or is it just what happens when you design interpreters for high vs. low-specification processors?

With which version of the FlashPlayer did you do that test?
Tamarin is the VM introduced for FlashPlayer 9, aka. AVM2. The above sounds like you tested on AVM1, which is included in FP9 for backwards compatibility. AVM2/Tamarin is JIT compiled, and significantly faster than AVM1. If you want to test, you will need to specifically compile for it, either by using Adobe's free as in beer Flex SDK if you like to use ActionScript 3, or haXe for an open source alternative that has some aditional features.

The ActionScript compiler isn't open source (but available for free as in beer), but haXe is. It's not ECMA262 v4, but a relative with some additional goodies, like its type system. It can compile for FlashPlayer 9, among other platforms, which uses the VM now known as Tamarin.