Announcing Play 2.3-M1

[Christopher Hunt]

After 5 months of having far too much fun the Play team are proud to announce the first release of Play 2.3-M1! This release introduces the following new and exciting functionality:

* The largest new feature for Play 2.3 is the introduction of sbt-web [1]. In summary sbt-web factors out Html, CSS and JavaScript functionality from Play's core into a family of pure sbt plugins. The benefit to you is a whole suite of new functionality that is on par with the best JS tooling out there.

* Play no longer distributed as a zip file download. As a consequence the `play` command as it is entirely integrated with Typesafe Activator [2] and regular sbt. So, instead of "play" to get to Play's console you may now use "activator" or "sbt".
* sbt 0.13.5-M2 is now used by Play. This version brings a new feature named "auto plugins" which permit a large reduction in settings-oriented code for your build files.
* In Play 2.2, a number of result types were deprecated and to facilitate migration to the new results structure some new types introduced.  Play 2.3 finishes this restructuring.
* The WS module has now been factored out into its own library. If you use WS then you must manually add the dependency.

* Java 8 integration including API improvements for lambda compatibility and various new Activator templates.

* A new way to handle WebSockets using actors.

* Various bug fixes.

[3] for more details of what has been incorporated in Play 2.3.

* * *

The new version can be used by:

1. Changing your project/plugins.sbt as follows:

addSbtPlugin("com.typesafe.play" % "sbt-plugin" % "2.3-M1")

2. Update your project/build.properties as follows:

sbt.version=0.13.5-M2

3. Reading the migration guide in full as there are breaking changes: http://www.playframework.com/documentation/2.3.x/Migration23

Enjoy!

For and behalf of the Typesafe Play team.

-- 
Christopher Hunt
Senior Engineer
Typesafe – Build Reactive Apps on the JVM!

Twitter: @huntchr

[1] https://github.com/sbt/sbt-web#sbt-web

[2] https://typesafe.com/activator

[3] https://github.com/playframework/playframework/issues?milestone=3&page=1&state=closed

[Andrew Gaydenko]

Great, thanks to the team!

Earlier I have named play wrapper around sbt as the main drawback. Well, you have eliminated it.
Now the time to find something else... say, long dependencies list.. or... Well, I'll be back later... :)

[Alberto Souza]

Cool!

[TheLovealien .]

Hi 

I will miss "* Play no longer distributed as a zip file download." I like the zip and the play command :(

I don't like Activator!

Thanks

D

--
You received this message because you are subscribed to the Google Groups "play-framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to play-framewor...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Filippo De Luca]

But you can use sbt directly has the same behavior of play cmd.

[Yann Simon]

2014-04-08 9:50 GMT+02:00 Filippo De Luca <m...@filippodeluca.com>:
> But you can use sbt directly has the same behavior of play cmd.

And if you don't want to install sbt, you can write a sbt wrapper,
like that one: https://github.com/yanns/scala-workshop/blob/master/sbt
(win32 version:
https://github.com/yanns/scala-workshop/blob/master/sbt.bat)

[Peter Empen]

Christopher:

Dropping ZIP download is really bad news for us, too. Did you make sure that enterprise customers are happy with an Activator/sbt-only policy?

My experience so far in brief:

Activator's 250 GB download size, compared with Play's 108 GB, is a real obstacle since the company virus scanner takes more than an hour to complete. Often it even aborts so I need to download Activator elsewhere.

I haven't yet managed to run Activator behind the company's firewall. When testing with 1.0.8, Typesafe wasn't able to help me either (issue 2559). I'm sure the cause of the issue is at our side but what my case shows is that a ZIP download is much more preferable from an enterprise point of view. Now I'm about to download Activator 1.1.0 to be able to report about the current state...

Also:
Fortunately, I can download SBT or Scala as a zip. Or opt for installing it. So why does Play need to break best practicies?

Peter

[Julien Richard-Foy]

You don't *need* activator to develop a Play application. sbt is
enough. And sbt installation is very lightweight.

I actually use activator just for the templates and, indeed, a 250 MB
app that essentially performs a git checkout is a bit expensive.

[Bing Ran]

 It's about time!

Can anyone please elaborate on the JDK8 support in this release? Can we use the full java 8 feature set in the controllers? What are the integration patterns available? A sample app is welcome. 

Thanks!

Bing

[Franz]

So basically one must download Activator to create a new Play app. And from there I can go on using sbt, as I would have with the "play" command?

[naturallog]

You could just clone the corresponding activator template from github and then run sbt.

[Maxime]

You can also use Activator on the command line the exact same way you'd use the play command.

There is a Minimal Activator distribution, that weights 1MB: http://downloads.typesafe.com/typesafe-activator/1.1.0/typesafe-activator-1.1.0-minimal.zip

By the release, the skeletons will be back in activator in order to use activator new to create Play projects.

[Peter Empen]

Julien:
Starting with SBT means that all the Play dependencies will be downloaded on demand into the local ivy repo, right?

[Maxime]

Or you can use activator in the command line.

[Andrew Gaydenko]

On Tuesday, April 8, 2014 12:46:36 PM UTC+4, Peter Empen wrote:

Also:
Fortunately, I can download SBT or Scala as a zip. Or opt for installing it. So why does Play need to break best practicies?

To my (rather old) taste the best practice is to avoid flooding software ecosphere with not-needed matters, abstractions, and so on. Reducing demands to just sbt is a big step in this direction.

As for Activator, I don't see any needing to it in everyday programming. I would treat it as just a great handy game to comfortably become familiar with Play :)

[Franz]

Awesome! 

[Markus Jura]

The new Activator version 1.1 has inspect functionalities. You can analyse your Play request and you Akka actors during development time. For me this is quite useful, e.g. to check the workflow of my Akka messages.

[Anton Kulaga]

Great news!

What about Scala 2.11 support is it already there? Some of Scala 2.11 features like faster incremental compilation, faster immutable collections, better reflection look rather promising for webdevelopment

[Andrew Gaydenko]

On Tuesday, April 8, 2014 2:34:34 PM UTC+4, Anton Kulaga wrote:

Great news!

What about Scala 2.11 support is it already there? Some of Scala 2.11 features like faster incremental compilation, faster immutable collections, better reflection look rather promising for webdevelopment

No, there isn't, Just some hours ago: https://github.com/playframework/playframework/commit/9bb358f30b16799f04f8e795a14f5e0ca14448b6

[Andrew Gaydenko]

On Tuesday, April 8, 2014 2:22:24 PM UTC+4, Markus Jura wrote:

The new Activator version 1.1 has inspect functionalities. You can analyse your Play request and you Akka actors during development time. For me this is quite useful, e.g. to check the workflow of my Akka messages.

Of course, I guess the best way is to separate monitoring into dedicated application (and just use it as an additional tool in Activator) :) OK, have you probably tried (and is it possible) to monitor already existing applications running Play 2.3.x?

[Bernhard Berger]

Congrats to the Play team!

Is there anywhere a minimal Play 2.3-M1 Java8 example?


I don't understand how I can migrate my Play 2.2-Java-App to use the
"Auto plugins" as described in
http://www.playframework.com/documentation/2.3-M1/Migration23 and want
to have a look into an example.

The examples from the Typesafe Activator 1.1 don't use these feature.
(Btw, the "Java 8"-filter in the Activator Templates don't work as
expected.)

I maybe need this, because I cannot get my WS-java-calls get running
with 2.3-M1, maybe I'm using a wrong dependency for WS/Java (I use:
`"com.typesafe.play" % "play-ws_2.10" % "2.3-M1"`). So I want to add the
WS dependency just like in the actual documentation.

When I just add the line
lazy val root = (project in file(".")).addPlugins(PlayJava)
to project/plugins.sbt I get compile errors.

[Julien Richard-Foy]

> Is there anywhere a minimal Play 2.3-M1 Java8 example?

Actually I'm just using jdk8 and it works. There is no additional
configuration to set up.

[James Roper]

Hi all,

There's been a number of questions asked here, and a few of them are following a few themes, so rather than answer each individually, I thought I'd some up some of the answers here.

First of all, those asking about more details about the new features, we should have in the next few days an activator template published that is a feature tour of the new features in Play.  In it there will be a tutorial that explains some of the coolest new things in Play.  I seriously encourage everyone, whether you're a beginner or an advanced user, to download that template and read through the tutorial.

A number of you are asking about activator and why we got rid of the play command.  Play has always provided an ultra productive development environment, both suited well to newcomers and advanced users.  At Typesafe we wanted to take this developer experience further, providing a library of community contributed templates, providing tutorials and resources for getting started, providing additional developer tools and an even more productive development environment.  But we didn't want to just limit it to Play.  So we took what Play had, and evolved it to something that was bigger and better.  Activator is the next evolution of the Play command.

And it's far more similar to Play than you may think - the Play command was basically a wrapper around SBT that provided a few extra things like the ability to create a new project.  Activator too is a wrapper around SBT that provides a few extra things like the ability to create a new project.  You can use Activator on the CLI exactly the same way that you use Play.  If you really liked the Play command so much, you can have it back, just put the following in your ~/.bashrc:

alias play=activator

You won't notice a difference.  I recently updated all the Play getting started docs, basically all I had to do was replace every reference to the play command with activator:

http://www.playframework.com/documentation/2.3.x/NewApplication

On the topic of download size.  Yes Activator is bigger than Play 2.2.  But it's not much bigger than Play 2.1.  Play 2.2 was kind of lucky, all the stars aligned, and by stars, I mean versions components that were bundled with Play, it contained the latest of every component in the Typesafe stack, and SBT for example was on the same version of Scala that Play was.  But that was special for Play 2.2, this wasn't the case for Play 2.1, and its download was close to 200mb, due to having two versions of Scala, and multiple versions of all the SBT dependencies.  It also wouldn't have been the case for Play 2.3 if it still had a distribution.  Activator of course also contains more than just Play, and has at least 3 different versions of Scala in it.

But as others have pointed out, if you are particularly adverse to large downloads, you can get a minimal version of Activator that will download all its dependencies on demand here:

http://downloads.typesafe.com/typesafe-activator/1.1.0/typesafe-activator-1.1.0-minimal.zip

Personally, I wouldn't download that, I have a 35mb internet connection at home, but because I'm in Australia, there's a large latency between myself and the ivy/maven repositories in the US, and due to the way ivy works, it ends up taking 5-10 seconds per dependency to download them all.  It is far faster for me to download one 250mb zip than to let ivy download each dependency individually.  I suspect this might be the case even in the US.

Those of you asking for Java 8 samples, just read our docs, eg:

http://www.playframework.com/documentation/2.3.x/JavaAsync

Every Java code snippet that can be expressed in a nicer fashion in Java 8 has a Java and a Java 8 tab.

I think that addresses most of the issues raised so far.

Cheers!

James

--
You received this message because you are subscribed to the Google Groups "play-framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to play-framewor...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

James Roper
Software Engineer

Typesafe – Build reactive apps!
Twitter: @jroper

JOIN US. REGISTER TODAY!

Scala

Days

June 16th-18th,

Berlin

[Bing Ran]

I'm trying the activator and my new test app gets stuck at

[info] Resolving com.typesafe.play#play-exceptions;2.3-2014-03-25-58325fd-SNAPSHOT

Not sure if it's due to my slow connection to typesafe though. It has been staying there forever. 

 

[Bernhard Berger]

Thanks, after looking in `playframework/templates/play-java/` I have
figured it out and my app works again :).

[TonyB]

Thanks for the clarification James.

In the interest of checking out Activator, as I have been using Play command line up until now, I downloaded it and tried running my Play 2.2.2 app in it.  I'm experiencing the VerifierError problem from the Javassist lib that is well known.  I was using -XX:-UseSplitVerifier in my JVM parameters via the _JAVA_OPTS env var before starting play.  For the life of me I can't figure out how to apply custom JVM paramters to my application as it runs within Activator.  I tried editing the activator.bat script directly but I have a feeling that its for the activator JVM and maybe not the underlying application JVM as it doesn't seem to work.  Any help would be appreciated, I'm been googling for "typesafe activator application jvm parameters" but can't find anything.

-Tony

[Kevin Bowling]

I think it would be a good move to integrate play-slick, in terms of showing direction and unification of the Typesafe stack, even if as a "Technology Preview"

Regards,

The information in this message may be confidential.  It is intended solely for

the addressee(s).  If you are not the intended recipient, any disclosure,

copying or distribution of the message, or any action or omission taken by you

in reliance on it, is prohibited and may be unlawful.  Please immediately

contact the sender if you have received this message in error.

[Christopher Hunt]

On Wednesday, 9 April 2014 04:30:55 UTC+10, Kevin Bowling wrote:

I think it would be a good move to integrate play-slick, in terms of showing direction and unification of the Typesafe stack, even if as a "Technology Preview"

Hi Kevin,

We're deferring the integration of play-slick for 2.3 given that there is some more foundational work on our side that needs to be done in the area of Play and its approach to dependency injection.

Kind regards,

Christopher

[Joe Zulli]

Awesome work everyone!

Quick question. Is the Forms / JSON API harmonization work still slated for 2.3? I notice that the ticket no longer has a milestone :(. I'm excited to use it after seeing it demo'd at the Ping conference. 

Joe

[Drew Kutcharian]

Julien/Pascal,

Is 2.3 going to include the new unified validation framework? https://github.com/playframework/playframework/pull/1904 is still open.

Cheers,

Drew

[James Roper]

Hi Tony,

I'm not sure how to add JVM parameters when using the Activator UI, but if you use the activator command (which is just the same thing as the play command, run it with no arguments, you get the SBT console like in Play, or you can run "activator run" etc) then JAVA_OPTS should work.

Cheers,

James

--
You received this message because you are subscribed to the Google Groups "play-framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to play-framewor...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[TheLovealien .]

Great news on the release :) Is M1 stable for production or will it change a lot ?

[Christopher Hunt]

Thanks. It isn't yet ready for production given its milestone status.

Kind regards

Christopher

You received this message because you are subscribed to a topic in the Google Groups "play-framework" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/play-framework/-Wuopt6NAxg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to play-framewor...@googlegroups.com.

[Julien Tournay]

That's a question for James actually.

[Julien Richard-Foy]

James, what are the arguments to not include the new validation API?

It would be great to unify form and json validation...

>> Typesafe - Build Reactive Apps on the JVM!

[sc]

the biggest issue of activator is it dont have option for debug. This was solved in this version?

[sc]

I almost forgot the problemof lock PID file when you force close(close running tab of browser), this issue dont have in play cmd. I like play cmd because it is simple and fast, the last time I used activator was miles away from that. BUT...I can be wrong because was some time ago.

[Drew Kutcharian]

+1 on integrating the new validation API. It’s really an elegant solution.

- Drew

[James Ward]

Right now you can only launch the app in debug mode using the CLI. This
is on the todo list for UI improvements.

-James

> <https://groups.google.com/d/optout>.

> >>
> >>
> > --
> > You received this message because you are subscribed to the
> Google Groups
> > "play-framework" group.
> > To unsubscribe from this group and stop receiving emails from it,
> send an

> > email to play-framewor...@googlegroups.com <javascript:>.

> > For more options, visit https://groups.google.com/d/optout

> <https://groups.google.com/d/optout>.

>
> --
> You received this message because you are subscribed to the Google
> Groups "play-framework" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to play-framewor...@googlegroups.com

> <mailto:play-framewor...@googlegroups.com>.

[James Roper]

Just to further what James Ward said, please read the documentation on how to debug applications:

http://www.playframework.com/documentation/2.3.x/PlayConsole

an email to play-framework+unsubscribe@googlegroups.com
<mailto:play-framework+unsub...@googlegroups.com>.

For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "play-framework" group.

To unsubscribe from this group and stop receiving emails from it, send an email to play-framework+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[sc]

Reading documentation, I see this change isn't a problem, just a minor change. 

Em segunda-feira, 7 de abril de 2014 21h10min26s UTC-3, Christopher Hunt escreveu:

After 5 months of having far too much fun the Play team are proud to announce the first release of Play 2.3-M1! This release introduces the following new and exciting functionality:

* The largest new feature for Play 2.3 is the introduction of sbt-web [1]. In summary sbt-web factors out Html, CSS and JavaScript functionality from Play's core into a family of pure sbt plugins. The benefit to you is a whole suite of new functionality that is on par with the best JS tooling out there.

* Play no longer distributed as a zip file download. As a consequence the `play` command as it is entirely integrated with Typesafe Activator [2] and regular sbt. So, instead of "play" to get to Play's console you may now use "activator" or "sbt".
* sbt 0.13.5-M2 is now used by Play. This version brings a new feature named "auto plugins" which permit a large reduction in settings-oriented code for your build files.
* In Play 2.2, a number of result types were deprecated and to facilitate migration to the new results structure some new types introduced.  Play 2.3 finishes this restructuring.
* The WS module has now been factored out into its own library. If you use WS then you must manually add the dependency.

* Java 8 integration including API improvements for lambda compatibility and various new Activator templates.

* A new way to handle WebSockets using actors.

* Various bug fixes.

[3] for more details of what has been incorporated in Play 2.3.

* * *

The new version can be used by:

1. Changing your project/plugins.sbt as follows:

addSbtPlugin("com.typesafe.play" % "sbt-plugin" % "2.3-M1")

2. Update your project/build.properties as follows:

sbt.version=0.13.5-M2

3. Reading the migration guide in full as there are breaking changes: http://www.playframework.com/documentation/2.3.x/Migration23

Enjoy!

For and behalf of the Typesafe Play team.

-- 
Christopher Hunt
Senior Engineer

Typesafe – Build Reactive Apps on the JVM!

Twitter: @huntchr

[1] https://github.com/sbt/sbt-web#sbt-web

[Juan Miguel Cejuela]

Congratulations and thanks for this new milestone!

I for one welcome replacing the "play" command by the "sbt" command. One less thing to remember.

[Christopher Hunt]

One small update if you're stuck on jdk6. You'll need to add the following to your plugins.sbt:

    libraryDependencies += "io.apigee.trireme" % "trireme-core" % "0.7.3-play23M1"

This issue will be resolved for the next milestone.

Kind regards,

Christopher

[Fabio Tiriticco]

Great work guys, thank you. The improvements to the WebSockets system look awesome. Can't wait for the following release!

[Miuler]

mmm .. really is a shame, I would have really liked to have slick once in the core of Play.

[Yoel R GARCIA DIAZ]

the name choices are most strange... now play is not called play anymore! it's called activator or sbt (both are weird choices to replace play - sbt is something else and not simple and activator is poorly descriptive!)...

I think that is really great that we have a 2.3-M1 out so quickly but better watch out you don't loose your identity altogether

[Christopher Hunt]

So our identity is entirely associated with the command's name...? :-P

--
You received this message because you are subscribed to a topic in the Google Groups "play-framework" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/play-framework/-Wuopt6NAxg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to play-framewor...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Yoel R GARCIA DIAZ]

might not at all... just seems odd to me ^^

thanks a lot for the new additions... I will probably wait before embarking upgrading to 2.3 but I am very happy to hear about it

[Peter Empen]

Certainly not. But it's really odd: originally, the name 'Activator' applied to a different action, more or less to the process of creating a new app based on templates, at least in my understanding. So what Yoel is asking is whether it's good to call the Play development console/ui something not related (related only in a technical sense). Think of an API - changing the underlying representation won't mean changing the API itself.

[James Ward]

The activator and play commands should work the same way. So if you
still want to type `play` then setup a symlink or rename the activator
script.

-James

On 04/13/2014 02:18 AM, Peter Empen wrote:
> Certainly not. But it's really odd: originally, the name 'Activator'
> applied to a different action, more or less to the process of creating a
> new app based on templates, at least in my understanding. So what Yoel
> is asking is whether it's good to call the Play development console/ui
> something not related (related only in a technical sense). Think of an
> API - changing the underlying representation won't mean changing the API
> itself.
>
> Am Sonntag, 13. April 2014 07:19:20 UTC+2 schrieb Christopher Hunt:
>
> So our identity is entirely associated with the command's name...? :-P
>
>
> On 13 Apr 2014, at 1:53 pm, Yoel R GARCIA DIAZ <yoe...@gmail.com

> <javascript:>> wrote:
>
>> the name choices are most strange... now play is not called play
>> anymore! it's called activator or sbt (both are weird choices to
>> replace play - sbt is something else and not simple and activator
>> is poorly descriptive!)...
>>

> --
> You received this message because you are subscribed to the Google
> Groups "play-framework" group.

> To unsubscribe from this group and stop receiving emails from it, send
> an email to play-framewor...@googlegroups.com

> <mailto:play-framewor...@googlegroups.com>.

[Marius Soutier]

Hopefully Homebrew will install it that way. Otherwise I agree with the rest, typing activator to start Play feels odd (and is more to type), and having to setup one more thing is not user friendly.

[Anton Kulaga]

What about making a symbolic link "play" for activator executable so the process will be smoother for those who loves shorter "play command"?

Overall I think that a decision to move people to activator is good because it is superior (for those who love minimalism there is still sbt command) and developers are often lazy to learn new things unless they have to=)

an email to play-framework+unsubscribe@googlegroups.com
<mailto:play-framework+unsub...@googlegroups.com>.

For more options, visit https://groups.google.com/d/optout.

--

You received this message because you are subscribed to a topic in the Google Groups "play-framework" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/play-framework/-Wuopt6NAxg/unsubscribe.

To unsubscribe from this group and all its topics, send an email to play-framework+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--
Best regards,
Anton Kulaga

[Peter Empen]

I thought the Play team could add a synonym in the Activator distribution easily. Why let all users bother with it?

[Christopher Hunt]

Is this naming concern boiling down to just having more characters to type ie "activator" instead of "play"?

On 14 Apr 2014, at 12:35 am, Marius Soutier <mps...@gmail.com> wrote:

Hopefully Homebrew will install it that way. Otherwise I agree with the rest, typing activator to start Play feels odd (and is more to type), and having to setup one more thing is not user friendly.

On 13 Apr 2014, at 16:14, James Ward <james...@typesafe.com> wrote:

The activator and play commands should work the same way.  So if you still want to type `play` then setup a symlink or rename the activator script.

-James

--
You received this message because you are subscribed to a topic in the Google Groups "play-framework" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/play-framework/-Wuopt6NAxg/unsubscribe.

To unsubscribe from this group and all its topics, send an email to play-framewor...@googlegroups.com.

[virtualeyes]

pla[tab] vs. act[tab]ivat[tab]

creating an alias to `play` is pretty trivial but I agree that `activator` is an odd way to start play.

[James Ward]

On my system it is just act[tab] --- is there something on Mac that
makes that not the case?

On 04/13/2014 02:09 PM, virtualeyes wrote:
> |
> pla[tab]vs.act[tab]ivat[tab]

> |
>
> creating an alias to `play` is pretty trivial but I agree that
> `activator` is an odd way to start play.
>
>
> On Sunday, April 13, 2014 9:11:26 PM UTC+2, Christopher Hunt wrote:
>
> Is this naming concern boiling down to just having more characters
> to type ie "activator" instead of "play"?
>
>
> On 14 Apr 2014, at 12:35 am, Marius Soutier <mps...@gmail.com

> <javascript:>> wrote:
>
>> Hopefully Homebrew will install it that way. Otherwise I agree
>> with the rest, typing activator to start Play feels odd (and is
>> more to type), and having to setup one more thing is not user
>> friendly.
>>
>> On 13 Apr 2014, at 16:14, James Ward <james...@typesafe.com

>> <javascript:>> wrote:
>>
>>> The activator and play commands should work the same way. So if
>>> you still want to type `play` then setup a symlink or rename the
>>> activator script.
>>>
>>> -James
>>
>> --
>> You received this message because you are subscribed to a topic in
>> the Google Groups "play-framework" group.
>> To unsubscribe from this topic, visit
>> https://groups.google.com/d/topic/play-framework/-Wuopt6NAxg/unsubscribe

>> <https://groups.google.com/d/topic/play-framework/-Wuopt6NAxg/unsubscribe>.

>> To unsubscribe from this group and all its topics, send an email

>> to play-framewor...@googlegroups.com <javascript:>.

>> For more options, visit https://groups.google.com/d/optout

>> <https://groups.google.com/d/optout>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "play-framework" group.

> To unsubscribe from this group and stop receiving emails from it, send

> an email to play-framewor...@googlegroups.com
> <mailto:play-framewor...@googlegroups.com>.

[Łukasz Śliwiński]

I also agree that you changed something that was good. Now we have have to type activator to start play which is odd and also word is longer. Unnecessary change that causing confusion. I don't see pros.

[virtualeyes]

Ah yes, I dropped my McMac dependency in 2009, on LtU (Linux the Ultimate) here ;-)

> <mailto:play-framework+unsub...@googlegroups.com>.

[James Roper]

You start a django app by running "python".

You start an express app by running "node".

You start a PHP app by running "apache".

If PHP developers are smart enough to cope with having a command name that bears no relation whatsoever to PHP, then I'm sure Scala/Java developers can cope with it.

--
You received this message because you are subscribed to the Google Groups "play-framework" group.

To unsubscribe from this group and stop receiving emails from it, send an email to play-framewor...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Marius Soutier]

No, it just depends on what else you have installed. act-[tab]-[space]-run is not as elegant as play-[space]-run, but if the sbt command still works it’s really not an issue.

Renaming activator to play won’t work because you’d have to do it with every update (on the Mac you typically use the homebrew package manager).

[Hector Miuler Malpica Gallegos]

I use sbt run, sbt test, sbt xx... 

Use activator seems heavy

--
You received this message because you are subscribed to a topic in the Google Groups "play-framework" group.

To unsubscribe from this topic, visit https://groups.google.com/d/topic/play-framework/-Wuopt6NAxg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to play-framewor...@googlegroups.com.

[James Ward]

Activator from the command line is the same thing as sbt and play. They
are all just sbt launchers. The only differences are that Activator has
an optional UI, templates, and a different command name.

-James

On 04/14/2014 08:25 AM, Hector Miuler Malpica Gallegos wrote:
> I use sbt run, sbt test, sbt xx...
>
> Use activator seems heavy
>
>
> On Mon, Apr 14, 2014 at 7:50 AM, Marius Soutier <mps...@gmail.com
> <mailto:mps...@gmail.com>> wrote:
>
> No, it just depends on what else you have installed.
> act-[tab]-[space]-run is not as elegant as play-[space]-run, but if
> the sbt command still works it's really not an issue.
>
> Renaming activator to play won't work because you'd have to do it
> with every update (on the Mac you typically use the homebrew package
> manager).
>
>
> On 13.04.2014, at 22:12, James Ward <james...@typesafe.com

> <mailto:james...@typesafe.com>> wrote:
>
> > On my system it is just act[tab] --- is there something on Mac
> that makes that not the case?
>
> --
> You received this message because you are subscribed to a topic in
> the Google Groups "play-framework" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/play-framework/-Wuopt6NAxg/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> play-framewor...@googlegroups.com

> <mailto:play-framework%2Bunsu...@googlegroups.com>.

> For more options, visit https://groups.google.com/d/optout.
>
>

> --
> You received this message because you are subscribed to the Google
> Groups "play-framework" group.

> To unsubscribe from this group and stop receiving emails from it, send
> an email to play-framewor...@googlegroups.com
> <mailto:play-framewor...@googlegroups.com>.

[Hector Miuler Malpica Gallegos]

It would be great to be able to do that since the sbt as  maven archetypes

On Mon, Apr 14, 2014 at 9:27 AM, James Ward <james...@typesafe.com> wrote:

Activator from the command line is the same thing as sbt and play.  They are all just sbt launchers.  The only differences are that Activator has an optional UI, templates, and a different command name.

-James


On 04/14/2014 08:25 AM, Hector Miuler Malpica Gallegos wrote:

I use sbt run, sbt test, sbt xx...

Use activator seems heavy


On Mon, Apr 14, 2014 at 7:50 AM, Marius Soutier <mps...@gmail.com
<mailto:mps...@gmail.com>> wrote:

    No, it just depends on what else you have installed.
    act-[tab]-[space]-run is not as elegant as play-[space]-run, but if
    the sbt command still works it's really not an issue.

    Renaming activator to play won't work because you'd have to do it
    with every update (on the Mac you typically use the homebrew package
    manager).


    On 13.04.2014, at 22:12, James Ward <james...@typesafe.com

[James Ward]

Where'd you hear that Play 2.3 is not supporting Java? That is
definitely wrong. Play 2.3 actually has significantly better support
for Java, especially Java 8.

On 04/17/2014 08:37 AM, Cong Nguyen wrote:
> I'm new to Play, just last week :), and my choice is 2.2 because of it's
> stability. Although I have learned Play for 2 weeks, I must say that I
> love zip archive rather than install + sbt or Activator. But it is not
> terrible, it just a problem of time to learn again. The most terrible
> thing in my thought is that Play will not support Java (Hope that this
> won't be the truth, otherwise I must learn Scala, the most weird
> language I've seen :( ).
> Come back to 2.3, I must say that sbt-web is really really attractive

[Will Sargent]

I just use zsh and "alias play=activator" :-)

You received this message because you are subscribed to the Google Groups "play-framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to play-framewor...@googlegroups.com.

[Chanan Braunstein]

Cong,

I started using Play 2.3 and I got to tell you, the Java 8 support is amazing! They integrated lambda's into Play and Akka. It is really cuts down on the amount of ceremonial code (to borrow a phrase from Venkat) that you need to write in Java.

[Will Sargent]

On Thu, Apr 17, 2014 at 8:15 AM, Cong Nguyen <congn...@gmail.com> wrote:

I'm sorry, my English is not good. It's just my "negative thought", because I've seen some softwares which are written in Scala, at first they supported Java, but then they focus on only Scala

No, Play is very serious about providing both idiomatic Scala and Java APIs.

Will. 

[Łukasz Śliwiński]

Hi,

I didn't find quick answer so I want to ask - Will Play 2.3 support also java in new validation API that is unified for json and forms?

Best regards,

Lukas

[Julien Tournay]

Hi Łukasz,

The unified validation API does not support Java. 

Some of the features required for implementing and using it are not available in Java.

Julien.

[Andy Polack]

Peter,

I echo your concerns.  Having the .zip distributable for Play was really very helpful in an enterprise environment.

-Andy

On Tuesday, April 8, 2014 4:46:36 AM UTC-4, Peter Empen wrote:

Christopher:

Dropping ZIP download is really bad news for us, too. Did you make sure that enterprise customers are happy with an Activator/sbt-only policy?

My experience so far in brief:

Activator's 250 GB download size, compared with Play's 108 GB, is a real obstacle since the company virus scanner takes more than an hour to complete. Often it even aborts so I need to download Activator elsewhere.

I haven't yet managed to run Activator behind the company's firewall. When testing with 1.0.8, Typesafe wasn't able to help me either (issue 2559). I'm sure the cause of the issue is at our side but what my case shows is that a ZIP download is much more preferable from an enterprise point of view. Now I'm about to download Activator 1.1.0 to be able to report about the current state...

Also:
Fortunately, I can download SBT or Scala as a zip. Or opt for installing it. So why does Play need to break best practicies?

Peter

[Andy Polack]

On the subject of DI - we just completed a large refactor of a medium-sized Play app to use the scalaz.Reader monad to perform our DI as inspired by this article: http://blog.originate.com/blog/2013/10/21/reader-monad-for-dependency-injection/ and with lots of gracious help from the ScalaZ community.

We're very happy with the results.  I humbly commend it to the Play team as a consideration if they're looking to overhaul any Play native DI capabilities. 

Kind Regards,
-Andy

On Tuesday, April 8, 2014 8:16:24 PM UTC-4, Christopher Hunt wrote:

On Wednesday, 9 April 2014 04:30:55 UTC+10, Kevin Bowling wrote:

I think it would be a good move to integrate play-slick, in terms of showing direction and unification of the Typesafe stack, even if as a "Technology Preview"

Hi Kevin,

We're deferring the integration of play-slick for 2.3 given that there is some more foundational work on our side that needs to be done in the area of Play and its approach to dependency injection.

Kind regards,

Christopher

[Andrew Gaydenko]

On Friday, April 18, 2014 5:16:55 PM UTC+4, Andy Polack wrote:

Having the .zip distributable for Play was really very helpful in an enterprise environment.

So, you deploy/run your application with Play instead of deploying zip resulted in dist, don't you?

[Peter Empen]

Andy, don't worry. I was tought that instad of Play.zip we should download Activator.zip. Then starting activator is the same as stating play previously. So it's just misleading naming. In addition we may also start activator ui.

[Peter Empen]

As for me, it's not deploying the application but deployeing the Play development environment for Play developers.

[Łukasz Śliwiński]

Julien Tournay,

Are there any plans to create different solution for Java to unify validation api?

[Cédric Chantepie]

Anorm shipped with Play 2.3-M1 will also includes various fixes, and some new features (String interpolation, multivalued parameter, positional column parser).

--

Github: https://github.com/cchantep/

Twitter: @cchantep

[Rémy Darcel]

I have similar issues with proxys.

I guess it would not be to complex to build a zip including sbt & an ivy directory containing all the dependencies of play through continous integration.

Sometimes it's very complex to push people to adopt new technologies such as play (especially people who are used to work since 10 years on same old technologies) and the only way to convince them is simplicity. And simplicity is a sum of little things.

[neverminder]

Will Play 2.3 work with Scala 2.11 now that it's about to be released? Is that at least in plans or do we have to wait for Play 2.4?

[Andrew Gaydenko]

On Sunday, April 20, 2014 2:59:25 PM UTC+4, neverminder wrote:

Will Play 2.3 work with Scala 2.11 now that it's about to be released? Is that at least in plans or do we have to wait for Play 2.4?

 https://github.com/playframework/playframework/pull/2470

[Christopher Hunt]

A great point. I'll make sure we mention this in our release notes for rc1.

--

[Christopher Hunt]

Yeah, I believe we are going to be cross compiling for 2.10 and 2.11.

--

[Anton Kulaga]

Yeah, Scala 2.11 support is one of the most expected features of play 2.3 as many libs already crosscompile to scala 2.11 and we are looking forward to deal with 2.11 benefits like faster compilation...

[Julien Tournay]

Not that I'm aware of.

[Andy Czerwonka]

+1

[Filippo De Luca]

​Wonderful answer. In addiction I would say, if any framework will create its own build system, it will be a babylon. Using standard tool is the way, if they are good enough.

When the play command tool has born, there were nothing out there that was meet the play needs. Now the play tool is just an sbt launcher, so makes sense to use sbt.​

On 14 April 2014 08:14, James Roper <james...@typesafe.com> wrote:

You start a django app by running "python".

You start an express app by running "node".

You start a PHP app by running "apache".

If PHP developers are smart enough to cope with having a command name that bears no relation whatsoever to PHP, then I'm sure Scala/Java developers can cope with it.

On Mon, Apr 14, 2014 at 4:38 PM, Łukasz Śliwiński <sliwins...@gmail.com> wrote:

I also agree that you changed something that was good. Now we have have to type activator to start play which is odd and also word is longer. Unnecessary change that causing confusion. I don't see pros.

W dniu niedziela, 13 kwietnia 2014 05:53:38 UTC+2 użytkownik Yoel R GARCIA DIAZ napisał:

the name choices are most strange... now play is not called play anymore! it's called activator or sbt (both are weird choices to replace play - sbt is something else and not simple and activator is poorly descriptive!)...

I think that is really great that we have a 2.3-M1 out so quickly but better watch out you don't loose your identity altogether

On Monday, April 7, 2014 8:10:26 PM UTC-4, Christopher Hunt wrote:

After 5 months of having far too much fun the Play team are proud to announce the first release of Play 2.3-M1! This release introduces the following new and exciting functionality:

* The largest new feature for Play 2.3 is the introduction of sbt-web [1]. In summary sbt-web factors out Html, CSS and JavaScript functionality from Play's core into a family of pure sbt plugins. The benefit to you is a whole suite of new functionality that is on par with the best JS tooling out there.

* Play no longer distributed as a zip file download. As a consequence the `play` command as it is entirely integrated with Typesafe Activator [2] and regular sbt. So, instead of "play" to get to Play's console you may now use "activator" or "sbt".
* sbt 0.13.5-M2 is now used by Play. This version brings a new feature named "auto plugins" which permit a large reduction in settings-oriented code for your build files.
* In Play 2.2, a number of result types were deprecated and to facilitate migration to the new results structure some new types introduced.  Play 2.3 finishes this restructuring.
* The WS module has now been factored out into its own library. If you use WS then you must manually add the dependency.

* Java 8 integration including API improvements for lambda compatibility and various new Activator templates.

* A new way to handle WebSockets using actors.

* Various bug fixes.

[3] for more details of what has been incorporated in Play 2.3.

* * *

The new version can be used by:

1. Changing your project/plugins.sbt as follows:

addSbtPlugin("com.typesafe.play" % "sbt-plugin" % "2.3-M1")

2. Update your project/build.properties as follows:

sbt.version=0.13.5-M2

3. Reading the migration guide in full as there are breaking changes: http://www.playframework.com/documentation/2.3.x/Migration23

Enjoy!

For and behalf of the Typesafe Play team.

-- 
Christopher Hunt
Senior Engineer
Typesafe – Build Reactive Apps on the JVM!

Twitter: @huntchr

[1] https://github.com/sbt/sbt-web#sbt-web

[2] https://typesafe.com/activator

[3] https://github.com/playframework/playframework/issues?milestone=3&page=1&state=closed

--

You received this message because you are subscribed to the Google Groups "play-framework" group.

To unsubscribe from this group and stop receiving emails from it, send an email to play-framewor...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

James Roper
Software Engineer

Typesafe – Build reactive apps!
Twitter: @jroper

JOIN US. REGISTER TODAY!

Scala

Days

June 16th-18th,

Berlin

--

You received this message because you are subscribed to the Google Groups "play-framework" group.

To unsubscribe from this group and stop receiving emails from it, send an email to play-framewor...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Filippo De Luca
---------------------

WWW: http://filippodeluca.com

IM:  filosg...@gmail.com