iel P.
-Ursprüngliche Nachricht-
Von: Poggenpohl, Daniel [mailto:daniel.poggenp...@isst.fraunhofer.de]
Gesendet: Donnerstag, 3. September 2015 21:01
An: users@tapestry.apache.org
Betreff: Tapestry IoC - Two Implementations of the same Service class - Symbol
Providers
Hello everyone,
I h
Hello everyone,
I have a FileSymbolProvider. It looks for a file in the webapp classpath and
reads the properties from the file.
I have defined two services based on the FileSymbolProvider, FileXProvider and
FileYProvider.
One of them is configured to generate default values for properties with
Thanks. We'll explore that path.
The intention is to get rid of the servlet api dependency. The
approach has been: take something that works, and make as few changes
as possible to get something else that works, then go from there.
Ultimately, we want tapestry-ioc-spring. As a matter of
that you need
the servlet api on your classpath for a non-web app. Perhaps
tapestry-spring needs to be split into tapestry-ioc-spring and
tapestry-spring.
At the moment, there is a dummy servlet context being provided. I
suggested keeping the code as close as possible to the original until
we had things working.
That being said, we have created our own SpringModuleDef in
anticipation of additional changes.
I'm accustomed to doing distributed contr
It seems to me that SpringModuleDef needs to be refactored to support
non-web projects, possibly by the use of an ApplicationContextProvider or
similar service.
On 21 Oct 2014 17:34, "Lance Java" wrote:
>
> Considering this is a non-web project, I assume that
WebApplicationContext is not on the c
Considering this is a non-web project, I assume that WebApplicationContext
is not on the classpath. It seems that SpringModuleDef is not suitable for
you since it references ServletContext which I also assume is not on your
classpath.
Hi all,
We have created a stripped-down "tapestry-spring" project to allow us
to use Spring from within a Tapestry-ioc based non-web project. One
objective is to be able to drop in jars and not only have services and
configuration information from Tapestry modules picked up, but also
On Tue, 07 Oct 2014 09:45:21 -0300, Chris Mylonas
wrote:
Hi All Tapestry Ppl,
Hi!
aaa) What are the basics in firing it up? Say from a main() method,
*unmanaged* i know...i read the IoC overview :)
http://tapestry.apache.org/starting-the-ioc-registry.html. ;)
--
Thiago H. de Paula Fi
Hi All Tapestry Ppl,
IoC caught my eye when I first touched tapestry some years back and I've
been fortunate to have a good dig at tapestry for web development a few
times this year again!
I'd like to use Tapestry-IoC in my old-pet open source project for
control
be
>> spun off into its own self-contained project with a release cycle
>> independent of Tapestry core?
>>
>
> We have no plans of spinning it off nor having a independent release cycle
> because, so far, we haven't felt the need for that. Tapestry-IoC is mature
> and
cycle
because, so far, we haven't felt the need for that. Tapestry-IoC is mature
and very flexible, so almost all new features you could think won't need
changes to T-IoC code itself, so they can be implemented in separate
projects. In the last years, the only major changes were
ans to carve out the IOC portion of
>> Tapestry and release that as a separate project? Since I’m moving
>> towards client-side frameworks (such as angular and ember) for future
>> projects, I don’t have much need for TML, page, and component-related
>> logic. But
have much need for TML, page, and component-related
logic. But, I’m quite find of Tapestry IOC. I suspect this approach
will become more common over time.
Tapestry-IoC is a separate project (as in separate source code tree in SVN
then Git) and JAR since it was born, around 7.5 years ago
Tapestry IOC is a separate module and can be used standalone.
On 2 Oct 2014 06:38, "raygarsta...@gmail.com" wrote:
> Just curious - are there any plans to carve out the IOC portion of
> Tapestry and release that as a separate project? Since I’m moving towards
> client-side
quite find of
Tapestry IOC. I suspect this approach will become more common over time.
—
Sent from Mailbox
Hi,
I had this exact same question and problem, and found changing the JDK
to use 7 instead of 8 fixed it.
However, this seems to do it even when I download a snapshot of
Tapestry and run using JDK 8, is there any documentation anywhere with
any requirements and instructions on how to use JDK 8?
It's caused by asm.ClassReader. Which jvm version and tapestry version are
you using? Tapestry has only recently added support for java 8 (asm5).
This was caused by a JDK version conflict!
- Original Message -
From: John
To: users@tapestry.apache.org
Sent: Tuesday, September 02, 2014 11:08 AM
Subject: Tapestry ioc/registry startup exception?
Hi,
Any pointers on why my app won't start pls?
John
ioc.Reg
Hi,
Any pointers on why my app won't start pls?
John
ioc.Registry Error building service proxy for service 'RegistryStartup' (at
org.apache.tapestry5.ioc.internal.services.RegistryStartup(Logger, List) (at
RegistryStartup.java:36) via
org.apache.tapestry5.ioc.services.TapestryIOCModule.bind
He still needs to manually contribute the interface-implementing services
to this single service. Still, it sounds like the closes solution.
*-*
*Muhammad Gelbana*
http://www.linkedin.com/in/mgelbana
On Sat, Feb 22, 2014 at 11:18 AM, Lance Java wrote:
> I think the "tapestry
I think the "tapestry way" of doing what you want is to make multiple
contributions to a single service and inject the single service.
On Fri, 21 Feb 2014 12:58:26 -0300, Michael Gagauz
wrote:
Is it possible in T5 to inject multiple services which implemented same
interface into array field?
I.e:
@Inject
private SomeService[] services;
No.
--
Thiago H. de Paula Figueiredo
Tapestry, Java and Hibernate consultant and deve
Is it possible in T5 to inject multiple services which implemented same
interface into array field?
I.e:
@Inject
private SomeService[] services;
-
To unsubscribe, e-mail: users-unsubscr...@tapestry.apache.org
For additional com
On Tue, 11 Feb 2014 18:16:01 -0200, Lance Java
wrote:
you must bind every actor and make sure it uses the correct scope
You could do something like this:
public void bind(ServiceBinder binder) {
Class[] actorClasses = ...;
for (Class actorClass : actorClasses) {
bin
ts you instantiate yourself. Just
> create a buildXXX() method. Of course, you'll need to create one for each
> actor.
>
>
> The solution with autobuild() is much more elegant,
>>
>
> Have you tried using the proxy() method instead of autobuild? I'm not sure
but that's something I'd try. After all, the whole
Tapestry-IoC implementation of decoration and aspects are based on proxies.
--
Thiago H. de Paula Figueiredo
Tapestry, Java and Hibernate consultant and developer
http://machina.com.br
-
Thiago, every time you call ObjectLocator.getService() you get a singleton.
Every time you call ObjectLocator.autobuild() you get a new object.
ks for the input!
Thilo
From: Thiago H de Paula Figueiredo
Sent: Tuesday, February 11, 2014 17:30
To: Tapestry users
Subject: Re: Tapestry IOC / Autobuild / Advise Question
On Tue, 11 Feb 2014 13:51:13 -0200, Thilo Tanner
wrote:
> Hi Lance, h
On Tue, 11 Feb 2014 13:51:13 -0200, Thilo Tanner
wrote:
Hi Lance, hi Thiago,
Hi!
thanks a lot for your feedback. To be honest, I'm still a bit confused:
Creating a bean or prototype scope seem to be controversial, according
to older discussions:
e.g. http://web.archiveorange.com/arch
On Tue, 11 Feb 2014 13:21:20 -0200, Lance Java
wrote:
Thiago, perhaps my wording was a little ambiguous. I consider perthread
scoped services are still singletons. The proxy is a singleton (the
underlying target may not be a singleton).
I would never say that. The proxy itself may be a sing
ossible to transform an object
created by autobuild()?
http://tapestry.apache.org/5.3/apidocs/org/apache/tapestry5/plastic/PlasticClass.html
Thanks again and best,
Thilo
From: Lance Java
Sent: Tuesday, February 11, 2014 16:21
To: Tapestry users
Subject: Re: Ta
Thiago, perhaps my wording was a little ambiguous. I consider perthread
scoped services are still singletons. The proxy is a singleton (the
underlying target may not be a singleton). Using autobuild creates a new
instance.
You might also want to consider implementing a custom scope (eg
"prototype") by contributing a ServiceLifecycle to ServiceLifecycleSource.
You could then mark your services with @Scope("prototype").
Take a look at PerThreadServiceLifecycle for inspiration.
defined with an interface. I've been using it on
perthread services a lot without any issues. I think the solution is to
create a new Tapestry-IoC scope, similar to Spring's prototype one
(basically, create a new object everytime injection is done) and turn the
actor objects into se
Decoration is really for singleton services whereas you are using a scope
which can be compared to spring's prototype scope.
I think you could transform the class (via PlasticManager /
PlasticClassTransformer) prior to calling
ObjectLocator.autobuild(transformedClass)
Hi all,
I have a quick question concerning Tapestry IOC. I'm currently integrating Akka
into one of my applications. In order to use it with a DI container, I created
a IndirectActorProducer which uses the autobuild() method of ObjectLocator, in
order to create a new actor object whe
Thanks Dmitry, works perfect :).
On 06.02.2014 16:11, Dmitry Gusev wrote:
You can try anjlab-tapestry-quartz from here:
https://github.com/anjlab/anjlab-tapestry-commons/tree/master/anjlab-tapestry-quartz
On Thu, Feb 6, 2014 at 2:46 PM, Matthias wrote:
Hi, I need to write a service where jo
You can try anjlab-tapestry-quartz from here:
https://github.com/anjlab/anjlab-tapestry-commons/tree/master/anjlab-tapestry-quartz
On Thu, Feb 6, 2014 at 2:46 PM, Matthias wrote:
> Hi, I need to write a service where jobs can be added and should executed
> after a given delay. After execution t
Hi, I need to write a service where jobs can be added and should
executed after a given delay. After execution they are finished, so
periodical execution is not required. I prefer to user the
ScheduledExecutorService for this task.
The question is now, how do I use my @Inject objects in the th
Hi, I need to write a service where jobs can be added and should
executed after a given delay. After execution they are finished, so
periodical execution is not required. I prefer to user the
ScheduledExecutorService for this task.
The question is now, how do I use my @Inject objects in the th
I see what you mean Denis, thank you.
On Fri, May 31, 2013 at 10:21 AM, Denis Stepanov
wrote:
> I have tried to push exactly same idea but wasn't successful.
>
> https://issues.apache.org/jira/browse/TAP5-1842
>
> Denis
>
> May 31, 2013 v 4:16 PM, Eli Doran :
>
> > There's already a @ServiceId
I have tried to push exactly same idea but wasn't successful.
https://issues.apache.org/jira/browse/TAP5-1842
Denis
May 31, 2013 v 4:16 PM, Eli Doran :
> There's already a @ServiceId annotation to specify the ID. This isn't a
> "service" exactly, it's close enough.
>
>
> On Fri, May 31, 2013
If you derive the id from the method name, this might break backwards
compatability. I'm guessing there are a lot of startup methods called
startup()
Perhaps a new annotation is required
@OrderedStartup(id="doEarly", constraints="before:*")
public static void doStuff() { … }
@OrderedStartup(id="doStuff")
public static void doStuff() { … }
@OrderedStartup(id="doMoreStuff", constraints="after:doStuff")
public static void doMoreStuff() { … }
On
There's already a @ServiceId annotation to specify the ID. This isn't a
"service" exactly, it's close enough.
On Fri, May 31, 2013 at 10:12 AM, Lance Java wrote:
> Perhaps a new annotation is required
>
> @OrderedStartup(id="doEarly", constraints="before:*")
> public static void doStuff() { … }
Yeah, like with builder methods which use the name after "build" as the
Service ID.
For example, to have the ID "SomeTask":
> @Startup("before:SomeOtherOne")
> public void startupSomeTask() { }
On Fri, May 31, 2013 at 9:50 AM, Lance Java wrote:
> I'm not sure this would work. OrderedConfigur
I'm not sure this would work. OrderedConfiguration requires that every
entry has a unique id and @Startup does not require an id. I guess you
could use ModuleClass.methodName for the default id but it's messy.
You can always post a JIRA asking for that.
On Thu, May 30, 2013 at 12:56 PM, Eli Doran wrote:
> I am not using Tapestry web framework (which means I'd have everything in
> my App's module, and its submodules, to keep it all together.) This means
> the "other startup methods" are in other modul
I am not using Tapestry web framework (which means I'd have everything in
my App's module, and its submodules, to keep it all together.) This means
the "other startup methods" are in other modules.
I work with a lot of modules which run together in different combinations.
They contribute to the st
on when it isn't needed.
I use Tapestry IoC for a modular application framework (not the tapestry
web framework). Maintaining many modules used in a variety of combinations
requires lots of defaults management and overriding. This is just one of
those things which I'd like to customize and
Hi Eli,
This would be clever, but you can always have a single @Startup method that
then calls all of your other startup methods in the order you want. In
some ways, I think this is easier to read/understand. The startup order is
clearly defined in a single spot instead of having to go decipher
+1
On Mon, May 27, 2013 at 7:50 PM, Eli Doran wrote:
> I like the simplicity of using the @Startup annotation compared to the
> startup contribution method.
>
> However, it doesn't have the ability to order its execution as in the
> contribution method.
>
> It seems it would be a simple thing t
s this something the flexibility of Tap IoC would allow me to override
> some processors or decorate current services to alter in my own
> applications?
>
> Is this something anyone else is interested in having in Tapestry IoC?
>
>
> A part of my motivation for this is it seems a nuis
Having an optional default value on @Symbol seems useful to me.
On Mon, May 27, 2013 at 4:42 PM, Thiago H de Paula Figueiredo <
thiag...@gmail.com> wrote:
> On Mon, 27 May 2013 17:23:36 -0300, Eli Doran wrote:
>
> I've used @Symbols a lot. I appreciate its flexibility and coercion
>> ability.
>
On Mon, 27 May 2013 17:23:36 -0300, Eli Doran wrote:
I've used @Symbols a lot. I appreciate its flexibility and coercion
ability.
I started programming around it though and instead accessing SymbolSource
and TypeCoercer directly because it is missing the ability to specify a
default value, a
me to override
some processors or decorate current services to alter in my own
applications?
Is this something anyone else is interested in having in Tapestry IoC?
A part of my motivation for this is it seems a nuisance being required to
put a default symbol value in the FactoryDefaults/Applicat
I like the simplicity of using the @Startup annotation compared to the
startup contribution method.
However, it doesn't have the ability to order its execution as in the
contribution method.
It seems it would be a simple thing to accept the ordering constraint in
the annotation like:
> @Startup(
his :) As much as I would love to, I
>> cannot afford this :)
>
> Same for me (except I don't see the need for replacing Tapestry-IoC, but
> sometimes we write code just because "why not?")). :) On the other hand, you
> could use some free time and do
On Wed, 22 May 2013 16:30:04 -0300, Lenny Primak
wrote:
I don't see myself getting paid for this :) As much as I would love to,
I cannot afford this :)
Same for me (except I don't see the need for replacing Tapestry-IoC, but
sometimes we write code just because "why not
On May 22, 2013, at 3:27 PM, Thiago H de Paula Figueiredo wrote:
> On Wed, 22 May 2013 15:00:41 -0300, Lenny Primak
> wrote:
>
>>> You're really interested in removing Tapestry-IoC of Tapestry. I see your
>>> good intentions there even if I disagree. I
On Wed, 22 May 2013 15:00:41 -0300, Lenny Primak
wrote:
You're really interested in removing Tapestry-IoC of Tapestry. I see
your good intentions there even if I disagree. I suggest you something
which I'd love to see in this discussion: Tapestry is open-source, so
what
iguration of services/beans, and
> services/beans are the core of IoC.
Not in my view. Beans can ** use ** configuration service, it doesn't need to
be tied into IoC
>
>> The only way it is related is because its baked into tapestry IOC.
>> These ought to be 2 separ
because its baked into tapestry IOC.
These ought to be 2 separate modules.
If, indeed there is a dire need to distributed configuration (I don't
believe there is such an integral need)
You keep saying that and it makes me think you don't know Tapestry well,
but you do. The mind boggles.
You guys keep talking about distributed configuration.
How is this related to IOC anyway?
The only way it is related is because its baked into tapestry IOC.
These ought to be 2 separate modules.
If, indeed there is a dire need to distributed configuration (I don't believe
there is su
cumentation and it's not
comparable to Tapestry's distributed configuration. Multibindings doesn't
seem to be have any ordering constraints, something which is crucial to
Tapestry-the-web-framework. Tapestry-IoC wasn't created out of NIH, but
based on Tapestry-the-web-fr
ns I've seen to date just don't provide the flexibility and
extensibility you get with "distributed configuration", and those concepts are
critical for Tapestry.
Robert
On May 21, 2013, at 5/216:28 PM , Lenny Primak wrote:
> You are missing my point.
> This is not ab
On Tue, May 21, 2013 at 7:12 PM, Thiago H de Paula Figueiredo <
thiag...@gmail.com> wrote:
> On Tue, 21 May 2013 23:09:29 -0300, Kalle Korhonen <
> kalle.o.korho...@gmail.com> wrote:
>
>> Lance has not participated in this thread even with a single message.
>>
> Thanks for correcting me, Kalle! I
On Tue, 21 May 2013 23:09:29 -0300, Kalle Korhonen
wrote:
Lance has not participated in this thread even with a single message.
Thanks for correcting me, Kalle! I was talking about Lenny, not Lance.
Sorry, Lance! Damn similar names . . . :P
--
Thiago H. de Paula Figueiredo
On Tue, May 21, 2013 at 6:53 PM, Thiago H de Paula Figueiredo <
thiag...@gmail.com> wrote:
> On Tue, 21 May 2013 20:28:11 -0300, Lenny Primak
> wrote:
>
If Tapestry replaces T-IoC with something else, we would cause such a huge
> backward compatibility problem that most people would abandon Tapes
On Tue, 21 May 2013 20:28:11 -0300, Lenny Primak
wrote:
You are missing my point.
This is not about how bad / great tapestry-ioc is.
This is about having to learn yet another DI system
before you can truly use tapestry to its full potential.
You still need to learn one anyway. And, after
"Well, yes, your screwdriver is great I guess, but I already know how to
use a hammer."
On Wed, May 22, 2013 at 12:28 AM, Lenny Primak wrote:
> You are missing my point.
> This is not about how bad / great tapestry-ioc is.
> This is about having to learn yet another DI syst
You are missing my point.
This is not about how bad / great tapestry-ioc is.
This is about having to learn yet another DI system
before you can truly use tapestry to its full potential.
If it used an existing IOC, the barrier to entry would be lower.
On May 21, 2013, at 6:01 AM, Inge Solvoll
Couldn't agree more with Inge
I have worked with tapestry-Guice & tapestry-Spring IOC and I think one of the
merits of Tapestry IOC is how easily you can integrate it with any IOC.
Any web framework needs some build-in IOC, It may be a couple of Java classes
but it is there. In Tap
I love Tapestry IOC. When used in a very basic way, it's almost
indistinguishable from Guice. Actually it's less intrusive since you don't
need annotations for injection.
Tapestry is very powerful when you do more advanced stuff, and I just love
that the power's there even t
of it. I would do the same thing.
There were legitimate reasons at the time to make decisions that were made,
at least most of them.
But, the landscape changed. Now there are better, simpler, more popular choices
in IOC than Tapestry-IOC, and I would prefer a world without Tapestry-IOC.
I can a
And I forgot another huge reason for not replacing Tapestry-IoC in
Tapestry: backward compatibility. Unfortunately, we just can't break
compatibility in such a large way. Many people still void Tapestry due to
its past history of completely non-backward compatible changes and the
Tap
On Wed, May 15, 2013 at 10:23 PM, hantsy wrote:
> Tapestry should embrace the existed and mature specs, such JSR330, Bean
> Validation, Managed Bean, etc, Spring has supported them in 3.0 natively.
>
I don't really agree with this logic. It leaves no room for innovation.
Everyone would just us
I'm not getting what are you trying to say. Is it "lets replace
tapestry-ioc with some other ioc"?
Or "lets implement proper CDI support"?
> If you are implying that this is all so important, why isn't every
project on the planet using Tapestry-IOC?
>
> As I said in another thread, you're suggesting replacing Tapestry-IoC
> with CDI. If that was done, people would still learn one IoC framework
> in order to learn Tapestry. CDI has a broader reach (in termos of
> concepts and features) than T-IoC. Not much people use CDI now
On Wed, 15 May 2013 18:57:55 -0300, Lenny Primak
wrote:
I agree that distributed configuration is great.
But, it's not equivalent to tapestry-ioc.
Agreed.
There are lots of ready-made solutions for distributed configurations
already.
You can also easily build one on top of ready
I agree that distributed configuration is great.
But, it's not equivalent to tapestry-ioc. There are lots of ready-made
solutions for distributed configurations already.
You can also easily build one on top of ready-made CDI implementation, if you
really want to (I don't think there
ou want to have flexible code.
For background on why Tapestry-IoC was created, read
http://tapestry.apache.org/ioc.html, specially the "Why not X?" sections.
If you are implying that this is all so important, why isn't every
project on the planet using Tapestry-IOC?
Few peop
If you are implying that this is all so important, why isn't every project on
the planet using Tapestry-IOC?
I would be very happy using the Web Framework without Tapestry-IOC, using just
plain beans for configuration,
or even using CDI events to gather configuration.
Given the history of Tape
blic TransformedItem getTransformedItem() {
return doSomeTransformation(item);
}
}
In this example, "transformedItem" can be cached until "item" changes.
--
View this message in context:
http://tapestry.1045711.n5.nabble.com/implementing-a-Tapestry-IoC-annotation-to-cache-method-resu
thanks, that could do the job - but is there any example of using the watch
parameter?
- Original Message -
From: Lance Java
To: users@tapestry.apache.org
Sent: Monday, December 31, 2012 8:49 AM
Subject: Re: implementing a Tapestry IoC annotation to cache method results
/apidocs/org/apache/tapestry5/internal/transform/CachedWorker.html
--
View this message in context:
http://tapestry.1045711.n5.nabble.com/implementing-a-Tapestry-IoC-annotation-to-cache-method-results-tp5719068p5719069.html
Sent from the Tapestry - User mailing list archive at Nabble.com
On Wed, Sep 12, 2012 at 6:02 PM, Thiago H de Paula Figueiredo
wrote:
> On Wed, 12 Sep 2012 17:05:51 -0300, Muhammad Gelbana
> wrote:
>
>>>
>>> HiveMind just makes me think of StarCraft. ;-)
>>
>>
>> Exactly eheheh :D
>
>
> No one thought of the Borg hive mind in Star Trek?
I very much did, when
On Wed, 12 Sep 2012 17:05:51 -0300, Muhammad Gelbana
wrote:
HiveMind just makes me think of StarCraft. ;-)
Exactly eheheh :D
No one thought of the Borg hive mind in Star Trek?
--
Thiago H. de Paula Figueiredo
-
To unsu
On Wed, 12 Sep 2012 13:26:09 -0300, Howard Lewis Ship
wrote:
Maybe we could resurrect the HiveMind name? I always loved that name!
I've just had a very bad idea: Unicorn IoC! It matches the Tapestry logo
and searching for 'unicorn ioc' in Google has no relevant results . . .
--
Thiago
>
> HiveMind just makes me think of StarCraft. ;-)
Exactly eheheh :D
On Wed, Sep 12, 2012 at 8:11 PM, Lenny Primak wrote:
> HiveMind always had negative connotations in my mind as well.
> I say de-emphasize Tapestry in Tapestry-IoC.
> Just call it IoC.
>
> I hate Spring
HiveMind always had negative connotations in my mind as well.
I say de-emphasize Tapestry in Tapestry-IoC.
Just call it IoC.
I hate Spring myself, and anything that gets people off it is a good thing IMHO
On Sep 12, 2012, at 1:52 PM, Daniel Jue wrote:
> I'd say keep the same name but ma
at 12:26 PM, Howard Lewis Ship wrote:
> On Wed, Sep 12, 2012 at 2:59 AM, Lance Java
> wrote:
> > I've often had a conversation like this in the workplace:
> >
> > colleague: Let's start a new spring project
> > me: I hate spring, how about we use Tap
On Wed, Sep 12, 2012 at 2:59 AM, Lance Java wrote:
> I've often had a conversation like this in the workplace:
>
> colleague: Let's start a new spring project
> me: I hate spring, how about we use Tapestry IOC instead?
> colleague: Tapestry, isn
I've often had a conversation like this in the workplace:
colleague: Let's start a new spring project
me: I hate spring, how about we use Tapestry IOC instead?
colleague: Tapestry, isn't that a web framework?
me: Well... yes, but Tapestry (the web framework) is
I'll post that as a separate thread to make it easier to find later.
Norman Franke
Answering Service for Directors, Inc.
www.myasd.com
On Aug 9, 2012, at 2:08 PM, Dmitry Gusev wrote:
> Hello, Norman!
>
> Sure, I'd like to see your approach.
>
> On Thu, Aug 9, 2012 at 10:02 PM, Norman Franke
Hello, Robert!
This looks like what I need!
Now I can remove bind() calls and just do:
@Contribute(WorkerManager.class)
public static void defineWorkers(Configuration conf) {
conf.addInstance(GitPullWorker.class);
conf.addInstance(GitCloneWorker.class);
}
And I e
Hello, Norman!
Sure, I'd like to see your approach.
On Thu, Aug 9, 2012 at 10:02 PM, Norman Franke wrote:
> If all the workers are in the same package you could make/Google a
> PackageEnumerator class that uses a ClassLoader to get all of the classes
> in the package, and then register those. I
Hi Dimitry.
What you need is to contribute your workers to the worker manager. Something
like:
contributeWorkerManager(Configuration conf) {
conf.add(...)
conf.add(...)
}
Now, the useful thing about this is that any module that is used in your app
that has a "contributeWorkerManager" will
If all the workers are in the same package you could make/Google a
PackageEnumerator class that uses a ClassLoader to get all of the classes in
the package, and then register those. I do this with my DAOs interface and
implementations to keep me from having to remember to add them to the AppModu
1 - 100 of 324 matches
Mail list logo
