I'd use the latest version of Seam and seam-gen to generate a new seam project,
then migrate src and resource files. In the future you should be able to then
update the seam-gen generate project when new versions of seam come out, as
seam-gen has an update capability for automatic update
Put an h:messages on your page. Skipping DOB is likely causing a conversion
error, which returns you to the same page. But since you aren't displaying any
error messages it can be a little confusing.
View the original post :
I agree, an upgrade guide from release to release would be useful. However I
doubt you'll get an upgrade guide that describes how to upgrade apps based on
examples. I think your approach of taking the new hibernate2 example as a base
and then patching in your code/config is the way to go
Can't wait for JB AS 4.2.GA. :)
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4040230#4040230
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=4040230
___
jboss-user mailing list
What Interceptors class are you importing, the Seam or EJB 3? The EJB 3
annotation needs to be run within an EJB 3 context.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4040240#4040240
Reply to the post :
Why do you need to get rid of this cid param? I wouldn't think that RESTful
apps would care about it. The server shouldn't mind its absence, it's likely
just a temporary conversation that is being exposed by the conversation filter.
I suppose you could remove that filter, but again, why would
So naturally this
| @Printout
| public String register(.)
| .
|
method is part of an EJB 3 session bean right?
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4040259#4040259
Reply to the post :
If they say yes, remember you'll still need a tomahawk.taglib.xml file for
Facelets. You can find one on the Facelets home page.
I'm using MyFaces 1.1.5, Tomahawk 1.1.5 and the Tomahawk treeTable with Seam
1.2.1.GA. This setup works. Please report back if the MyFaces guys say there
are
Try using JRocket. Permgen issues are tricky and are a real problem for apps
like yours that load and reload a zillion classes. Someone else can probably
go into the details but JBoss isn't alone, for a quick fix grab the latest
R27.2 JRocket release. JRocket handles permgen differently than
EntityHome was designed to be able to be used by JSF CRUD pages without needing
additional java code. That said, it's generally preferred to access Seam
components via @In than Component.getInstance().
View the original post :
You might be able to use an observer on a seam contextual event. Unfortunately
I don't see a seam fully initialized event. Maybe postCreate, or if your app
requires authentication, preAuthenticate. I'm just looking at the docs, there
may be additional events that have been added. It
Hmm, PMs don't seem to be activated. Anyway, please take a look at the placing
of the Events.instance().raiseEvent(org.jboss.seam.postInitialization); line.
I'm guessing that this should come at the end of the method, or at least after
the Lifecycle.endInitialization(); since
[EMAIL PROTECTED] wrote : I prefer exactly the opposite, with a Search
conversation scoped component that holds my prototype entity during the search
conversation. Gives me a lot of control over what the user can select on the
search screen and how I handle this in the backend:
I really like
You might be able to set the default Facelet or JSP flush size to something
large enough to keep everything in memory until your entire paged had finished
rendering. However you apparently have large pages that take a long time to
render, so it's probably not desirable to do this.
If you're
It's late, but I'll give this a go. Christian or Pete feel free to jump in.
It sounds like your mixing your retrieval and persistence mechanisms.
@Factory - Method called when a JSF page tries to reference a contextual
variable that doesn't exist, but has a named factory method. So if
Re EJB3 Entities:
anonymous wrote : Not quite, they are only components if the have @Name (in
which case they would be CONVERSATION scoped). The preferred way to do this
is to use Home objects to manage an Entity, rather than expose it directly with
@Name, and uses pages.xml for wiring. We
anonymous wrote : hey you live in san mateo near by to where i live
dublin/pleasanton :)
Small world. Maybe I'll see ya at the Seam JavaOne get together.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4039567#4039567
Reply to the post :
My apps simply have their own secure login pages so I don't have a best
practice, but I'll help hash something out. I suppose my first question is.
Are you using Facelets? My knee jerk solution to this would be to write a
simple my:loginForm tag with Facelets that used a regular html form
Borrow whatever you like Pete. Anything for the cause.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4039575#4039575
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=4039575
___
One down side to using EntityHome for generic crud is lack of built in
security. One needs to be careful when using Homes for crud operations that
allow or require RequestParameters. You need to ensure the operation on this
ID is valid. You don't want to expose information you shouldn't and
Take a look at the documentation on Factory and manager components. They will
help you do what you want. Also the blog and messages examples show how you
can initialize a contextual variable with DB data when a page is viewed.
The messages example makes use of a DataTable binder, but you can
I like to keep my entities clean. Just get/set/toString/equals/hashcode. So
I'd put the @Factory on a method in your controller/action/manager Bean. If
you need some heavyweight resources that you'd like to clean up, you can use
the manager pattern and the @Unwrap annotation. In the case of
...and on scope, go as narrow as possible while still being functional. :)
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4039405#4039405
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=4039405
If that's correct, then great. I'm wrong and everything is safe. I've been
cleaning out a lot of select ... where name like +name+% from a codebase I
inherited. So I may just have been spooked by seeing #{expression} + %.
That said, I don't recall reading anywhere that Seam does what you
I guess I'm just expecting the values to be run through Seam's interpolator,
which just does a string replace. I don't expect that Seam will generate a
prepared statement, replace the EL with bind variables, etc. If it does, more
power to Gavin.
View the original post :
Yep, +1 Seam crew. I should have known better. :)
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4038452#4038452
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=4038452
___
The framework schema seems to support the following form.
framework:entity-query
|framework:hints
| key/keyvalue/value
|/framework:hints
| /framework:entity-query
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4038616#4038616
Reply to the
Looks like a problem to me. As a work around try something like:
@Name(myQuery)
| public class MyQuery extends EntityQuery
| {
| protected String getCountEjbql()
| {
| return select count(*) without fetch;
| }
|
| public String getEjbql()
| {
|
Oops, PersonQuery class in the last code block should be MyQuery.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4037846#4037846
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=4037846
Doing a lot of hot deploy? I usually find when things are screwy like this
there's a permgen memory error at the very bottom of that stack trace. A
restart of the app server causes the problem to go away.
I've recently migrated to JRocket R27.2 and haven't had any trouble so far.
JRocket
Lol, should really not drink and make forum posts. :) A combination of
dsmith's solution and my components.xml should give you what you're looking
for. You only need 1 small class that you reuse in components.xml. You could
even write your own namespace and have your own
Try creating a build.properties file next to build.xml and include
profile = prod
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4038003#4038003
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=4038003
or simply pass as a command line arg to ant
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4038004#4038004
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=4038004
___
jboss-user
exception?
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4038020#4038020
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=4038020
___
jboss-user mailing list
In Bean B try sharing with
@In @Out DataModel sections;
If you're trying to manipulate the DataModel in B. Otherwise your
investigation seems to show that Seam isolates the @DataModel data binding on a
per bean basis. Not sure why. Gavin is around here somewhere, try adding a
post to the
ui:repeat is the preferred way to go. If you're having problems with it I'd
try and resolve the issue on the facelets mailing list. Or you could try an
alternate implementation. I believe that RichFaces has a repeat tag of their
own as well.
View the original post :
anonymous wrote : ...it doesn't do a subsequent injection
Really? Although I only outject in my code, I thought DataBinders performed
bidirectional binding. Why else implement a getWrappedData() DataBinder
method?
View the original post :
So you're asking for something like @Begin(timeout=ms)?
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4038097#4038097
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=4038097
___
This just seems begging for a SQL injection problem. Be careful when using
this feature.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4038215#4038215
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=4038215
This is really a Hibernate/JPA question, you might get more help from that
forum. What I'd expect is happening is that due to your cascade level, refresh
is attempting to refresh the database state of your detail entities. Since
they don't exist, you get an error.
Perhaps you should
You appear to be using an EJB3 SFSB and a JPA extended persistence context.
Only Hibernate supports a manual flush mode. What I'd assume is happening here
is that your @Begin(flushMode=...) declaration is being ignored and that EJB3
is defaulting to its standard transaction/flush
Take a look at the message attribute of the Hibernate Validator annotations,
they allow per field overrides of the default message. NotNull is still
handled by JSF's required=true mechanism, but you have greater flexibility for
other field constraints.
View the original post :
Yep. It makes sense, it would be nice if you could override the validation
message at the s:validate level. This has long been a complaint of JSF, and
one that I believe will be fixed in JSF 2.0. JBoss will start using the 1.2 RI
in 4.2, and that may allow some jsf-ext work to be used,
The is the typical @In vs @PersistenceContext question. The answer is summed
up nicely in this thread.
http://jboss.com/index.html?module=bbop=viewtopict=106019
Basically to answer your questions:
1. What the core:... is doing for you is configuring a Seam managed
persistence context. You
Hmm, what I was seeing on JBoss 4.0.5, which should be MyFaces on JSF 1.1, is
that my converter wasn't being called at all in either direction when I tried
to override forClass=String.class.
What I've written is a quick converter to translate empty Strings into null.
This makes my DB happier
Double check that jndi-name with the pattern in components.properties. All of
my lookups are of the form #{ejbName}/local no domain.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4037278#4037278
Reply to the post :
I've written a Seam component JSF Converter. It works great if I specify an
f:converter tag on my input. However if I specify
@org.jboss.seam.annotations.jsf.Converter(forClass=String.class)
and leave off the f:converter, the converter isn't called even if there is a
String value binding.
anonymous wrote : Caused by: org.jboss.tm.JBossRollbackException: Unable to
commit, tx=TransactionImpl:XidImpl[FormatI
| d=257, GlobalId=kostja-mobil/468, BranchQual=, localId=468]
status=STATUS_NO_TRANSACTION; - nested t
| hrowable: (javax.persistence.EntityNotFoundException: deleted entity
How about the second part? Have you done something like this?
1. Lookup entity.
2. Remove entity.
3. Try and persist removed managed entity.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4036844#4036844
Reply to the post :
anonymous wrote : 1. Lookup entity.
| 2. Remove entity.
| 3. Try and persist removed managed entity.
The above is what not to do. So if you're doing that, that's likely your
problem. Sry, your code is just too long to sift through.
View the original post :
In this case you never try and re-persist kfz1. If you did you'd probably get
the same error here as well.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4036875#4036875
Reply to the post :
Question regarding the semantics of @Convert(forClass=...)
http://www.jboss.com/index.html?module=bbop=viewtopict=106354
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4036880#4036880
Reply to the post :
Up for drinks +1. Do you really want everyone to E-mail? Easier filing system
than forum?
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4036882#4036882
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=4036882
Same here Rob. It shouldn't be a problem though, there are plenty of hotels
and bars nearby as well as good cab service. I'm not even sure Moscone allows
alcohol.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=4036910#4036910
Reply to the post :
@RequestParameter is for HTTP request params.
http://host/page.jsf?username=Steve
@RequestParameter(username)
String username;
assert username == Steve
I'd guess the problem here is that you don't actually have a request parameter
with this data or that Seam has no idea how translate from
anonymous wrote : I assumed that because the init method has been annotated
with a Factory annotation and the testVar component is initially null - that
this would cause the init method (i.e. method annotated with the Factory
annotation) to be invoked whenever one invokes any business method in
The difference between @Factory and @Unwrap is that @Unwrap only works on
components. @Factory can work on contextual variables as well, like the String
example given.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3973923#3973923
Reply to the post :
anonymous wrote : One alternative approach is instead of passing the entity
object around, just pass the id in a request parameter and use something like
the ManagedEntity component (in Seam CVS) to auto-load the entity instance and
expose it via a context variable.
Since request parameters
I'm pretty sure this is a classloader issue. I think there's a Seam FAQ out
somewhere that addresses this. I'm on the run, but poke around a little more.
I'm pretty sure this has been answered.
JBoss has a number of classloader options, it also ships with a JSF
implementation so your
I'm pretty sure we were both using JSTL correctly, but had the wrong namespace.
If you don't specify the proper namespace facelets will happily just spit out
your tag as text instead of performing the appropriate taglib processing. This
is the problem we both had, and it's solved by using the
Writing a filter that blocks direct .xhtml access shouldn't be too hard. I
wouldn't expect facelets to perform a web request to access the .xhtml files,
so you should be able to 404 any direct HTTP access.
I'm not 100%, but this should be easy enough to test. Either just write the
filter, or
You could also try setting your Faces Servlet to match on *.xhtml instead of
*.seam. I'm not sure if that causes a problem or not. I wish I were closer to
a dev environment. These are all quick tests.
For some reason I thought that facelets was still like JSP in that if you screw
up and
You can also define components, including JavaBeans, in components.xml. If you
define them here, you don't need the @Name annotation.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3973104#3973104
Reply to the post :
You could also look into using a dedicated security framework like Acegi that
would have richer options.
http://acegisecurity.org/
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3973152#3973152
Reply to the post :
This is the expected behaviour. The @SelectItems annotation applies a data
binder and outjects the wrapped object as a contextual variable (regions).
Direct access to the method selector.getRegions() will return the method's
original values. Useful for calling the method from other
home.iface doesn't appear to be handled by the Faces Servlet. Have you tried
home.seam?
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3972767#3972767
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3972767
The thing about this is that null or not found is a valid use case. Many ppl
have pages that rely on this functionality to not display something when a
variable isn't in context and then to come alive once an action puts this
variable into context. In this case you wouldn't want a log message
Thanks Pete. This clears things up for me.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3971143#3971143
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3971143
___
jboss-user
What I figured. Thanks. Any plans to publish a protocol spec?
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3970660#3970660
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3970660
Take a look at the Seam wiki. There are a couple third-party selectItems
DataBinders there. These will let you take a list returned from your DB and
use it within your JSF menu.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3970806#3970806
Reply to the post
Take a look at the RESTful section in the documentation and the blog example
for tips on your mostly stateless apps.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3970807#3970807
Reply to the post :
The given Seam examples and documentation isn't PhD work, or even that complex.
It does try and provide a complete example (non-trivial) and it does assume a
certain foundation.
In the case of your team, can I suggest some training. At least on JSF and
Hibernate/JPA and then sprinkle in some
Agreed, docs can always be better. Although as far as OS projects go, these
are pretty good. I did a quick search for PAGE scope and found that the issues
example uses it. In particular the ProjectFinderBean.
The usage and design pattern seem to fit my guess from a previous post. PAGE
It isn't clear how page scoped resources are cleaned up. They appear to
survive multiple requests, however can be out of date and require a refresh()
as indicated in the issues example ProjectFinderBean class.
Is it possible to set a page scope timout so that resources can be cleaned up
in
If I recall the JNDI lookup for Tomcat/embedded EJB3 is different than regular
JBoss. I don't have time right now for a full doc/forum search, but try that
first. The root cause appears to be a naming problem.
View the original post :
Also, make sure you've updated your Seam's build.properties and generated the
example via deploy.tomcat?
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3970640#3970640
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3970640
Not to knock the work done by the remoting team, but is there a reason a custom
protocol was chosen vs a Seam RemotingEndpoint that would handle remoting via
some version of SOAP?
I would have a thought the benefits of a standard transport protocol would
outweigh the down side. Since
This way idle conversations are timed out, yet the conversation that you're
currently in will last as long as your session lasts or until you finish the
conversation.
You don't want the foreground work you're currently doing to be ripped away
from you. But you do want a way for the resources
I agree with Gavin. I've always been ok with complex validation in the action
method (or control flow). If you need something a little more declarative, it
shouldn't be too hard to use Seam plus EJB interceptors to write your thin
framework.
If you think about it this is what JSF is doing
@Out's scope defaults to the scope of the enclosing controller, in this case
the SLSB. SLSB have a stateless scope. I think that Seam may actually promote
the outjection of context variables of stateless components to event scope for
you (but I'm not positive).
Event scope is like JSP
Make sure there is a value for your @In tradeDetail.userPrincipal in scope. If
not you'll get this error.
Although I can't imagine you'd get this error while the GC is calling your
finalize() method as this would happen outside of the JSF lifecycle and no
bijection should be going on.
View
Well this is kind of core to bijection and contextual components. @Out is a
general concept supported by all components over six searchable contexts.
Where did you think your object was going? I'm not trying to be mean, rather
trying to figure out where any new documentation should go. Is
And required checking happens outside of conversion/validation at yet another
time. Unfortunately, JSF seems to be designed to support providing a
consistent checked state through each phase of the lifecycle and short circuit
phases if there are problems. This almost explicitly designs away
Whoa. What are you trying to do in your finalizer?
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3970207#3970207
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3970207
___
Well EVENT obviously won't work because state is destroyed at the end of the
request, so it won't be available across redirects (which have 2 requests).
PAGE on the other hand I'm not 100% sure about. I'll admit I don't use page
scope. I was under the impression that this was like a
Just curious. What JSF state saving technique are you using? If you change
this to client, does page scope work?
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3970214#3970214
Reply to the post :
Well it just doesn't make sense for a stateless component to outject anything,
because it wouldn't go anywhere. *IF* (I'm still not sure) Seam is nice and
automatically outjects objects from stateless components into event scope, then
the examples were likely written in a time where Seam
CptnKirk wrote : Well it just doesn't make sense for a stateless component to
outject anything, because it wouldn't go anywhere.
And this isn't strictly true either. There are a whole class of problems where
you want to have a stateless component (like a SearchManager) to outject into
another
My guess is the client vs server state saving problem is due to a bug in ealier
versions of MyFaces that JBoss may still ship with.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3970220#3970220
Reply to the post :
...or maybe that was an earlier version of facelets (or both). Try upgrading
those if you need server side state saving.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3970222#3970222
Reply to the post :
Actually I'm pretty sure Java does have limited meta-annotation support. Look
at how Seam implements its DataBinding mechanism.
| @Target(ANNOTATION_TYPE)
| @Retention(RUNTIME)
| @Documented
| public @interface DataBinderClass
| {
|Class? extends DataBinder value();
| }
Then
EJB 3 interceptors also make use of meta-annotations.
http://docs.jboss.com/seam/1.0.0.GA/reference/en/html/concepts.html#d0e2635
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3969611#3969611
Reply to the post :
make sure you have a getter/setter and that if using EJBs that these methods
are exposed via the local/remote interface.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3962944#3962944
Reply to the post :
Later versions of facelets default to UTF-8. If you want to change that I
think you can specify the encoding parameter on the xml declaration. If that
doesn't seem to work for you, you might want to ask the facelets mailing list.
View the original post :
This doesn't appear to be related to Seam. From this error it looks like you
have different versions of your classes on your client and server.
You also appear to be mixing RMI, EJB 2.x and EJB 3. I suggest using pure EJB
3 with Seam. Try looking at an EJB 3 tutorial and/or an EJB 3 book
Your example is just too huge for most ppl on the form to walk through. Try
reproducing your problem in a much smaller cut down.
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3961354#3961354
Reply to the post :
I think the original post is conceptually wrong. ui:repeat is an iterator.
It can iterate over lists of things, including DataModels. However it doesn't
modify the model that is iterates over. For example it will never set a
DataModel's selected row and using @DataModelSelection with it
An exception shouldn't implicitly kill the conversation. In fact you've proven
that it doesn't. You might roll back a transaction on an exception, but
conversations aren't the same as transactions and should survive your given use
case (although if you wanted to, it wouldn't be hard to end a
This is really an EJB3/JPA question, not a Seam one. You may get better
support if you post your question in that forum. It sounds like you want a
ManyToMany relation. If Tasks can have many TaskItems and a TaskItem can be
associated with more than one Task this is a ManyToMany relation, not
How are you making this call? You'll only be able to lookup a Seam context
after Seam has initialized and while you're within some web request.
So calling this method via a startup servlet will probably fail. Calling this
method via an EJB or MDB that is executed via a non-web request (ie
1 - 100 of 121 matches
Mail list logo
