Thanks Jason. I guess this will give us another option to consider.
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Jason
Carreira
Sent: Thursday, March 04, 2004 1:26 AM
To: [EMAIL PROTECTED]
Subject: RE: [OS-webwork] Webwork 1.4.1?
This can
re if there is some demand for it (along with an example), as it stands it's an add on that isn't checked in anywhere, but people are free to copy etc.On Mar 3, 2004, at 2:20 PM, James Cook wrote:> Should we be expecting a point release soon with Ha
Should we be expecting a point release soon with Hani's addition of a
SpringActionFactory?
---
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with
a free DVD software kit from IBM. Clic
> -Original Message-
> Behalf Of Patrick Lightbody
> Sent: Thursday, January 08, 2004 5:46 PM
> I was thinking that a huge performance boost could come from
> remembering the index for certain properties, but I admit I
> haven't fully thought this through and there might be more
> than
Title: Message
Cameron,
Can
you contribute the code to pass the ValueStack to FreeMarker, or is that in CVS
now?
The sessionMap may not be available in your constructor. It should be
available in nearly any other method that the framework will invoke.
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On
> Behalf Of Patrick Holzmann
> Sent: Tuesday, November 25, 2003 11:46 A
What about reloading the configuration contexts? For example, if a
change to xwork.xml or components/xml also triggers a reload?
> -Original Message-
> Behalf Of Hani Suleiman
>
> Not all containers support reloading of random classes. Resin does,
> Orion will only reload modified servle
Can I specify an array of strings when using the tag within an
tag?
---
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and
> -Original Message-
> Behalf Of Joseph Ottinger
> What I'd thought of was adding an extra method to Action and
> ActionSupport. The Action method's signature might look like this:
> ...
> Thoughts? Insults? Points?
I thought of a pointless insult. ;-)
We used this pattern in a large app
I had thought the components.xml file would be a natural integration
point for Spring factory beans. To this end, I have written my own
WebWork enabler interface and exposed our Spring factories this way. It
seems lighter weight and a more natural WebWork integration point than
coupling the web-tie
> -Original Message-
> Behalf Of Jason Carreira
> You also get number format messages for non int values?
> That's a bug... It should just add a field error...
Here is the exception for a non-numeric value. How can this ever be a
field error if OGNL throws this before the calidation inter
I have the following field in my Action:
Integer _personIDNumber;
And the following input field in my JSP:
My validation requirements are:
You must enter a value for person ID
number.
When the code executes and I supply no value in the input
Does anyone have any good ideas for determining the presence of field or
action errors using only JSTL?
Doesn't work, because a new HashMap is created on the fly.
Doesn't work, because it would expect a getSize() function on the Map,
when the function is size().
So, I have to resort to som
Actually in WW1.3 it is/was still execute(). The ActionSupport class
that you extended implemented the execute() method for you, and this
method called your doExecute() method.
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On
> Behalf Of Lars Fischer
> Sent:
WHERE DOES THE MADNESS END!!!
:-)
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On
> Behalf Of Matt Ho
> Sent: Wednesday, October 01, 2003 6:38 PM
> To: [EMAIL PROTECTED]
> Subject: Re: [OS-webwork] interceptors as components?
>
>
> Currently, no. However
BTW, the filter is on the Wiki at:
http://wiki.opensymphony.com/space/Using+JSTL+seamlessly+with+WebWork
> -Original Message-
> From: Cameron Braid
> Though for JSTL to access your action properties you need to
> do one of either
>
> A) set it into the page context (someone correct me i
Thanks for the details on this approach. Can you post some code to the
Wiki? Just one question regarding transactions in such a scheme.
If my action chain looks like this:
ActionA --> redirect --> ActionB
Will your solution easily allow a transaction to be ACID if I do
database work in both Acti
In most cases, it would be easier to be able to indicate the "allowed"
parameters, versus the disallowed. Of course, both are necessary
depending on the situation.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Cameron Braid
Sent: Thursday, September 18,
+1 on removing legacy dependency on Action interface. Let's make Xwork
truly a "generic command framework".
-Original Message-
Sorry that I didn't post this comment to the mailing list to begin with.
It is also on the Wiki, probably not the best place for a meaningful
technical discussion
Multiple requests does not necessarily mean a long-lived transaction. I
was thinking of a general case when one may want to use a redirect thus
creating a second request.
Basically, I was asking Pat if the book's framework used Hibernate in a
way that supports transactions especially in a componen
Does the book document the "Yet Another Petstore" as a web-tier only, or
as a web-tier with an EJB (SLSB) backend?
Does the framework also support an ACID transaction across multiple
requests, against more than one database connection, or across more than
one VM.
I'll be first in line to buy the
I was replying to his statement that he:
> currently in my daos i create a session object for every
> dao operation like dao.save dao.load, etc
I was assuming that he was also doing a flush after each of these
operations.
-Original Message-
From: Jason Carreira
> This is not how Freerol
Yes, the order of, not related, elements is constrained in XML.
-Original Message-
The problem was that the order of these elements was not the same as in
the error message above. Is this how it should be? (I don't know very
much about xml and parsing but it seems pretty stupid if the or
> currently in my daos i create a session object for every
> dao operation like dao.save dao.load, etc
I think this is exactly what caused the freeroller application (using
Hibernate also) to suffer some extreme scalability failures.
As far as best practices go, Jason has started Conductor that
Shouldn't there be a class with a psvm to execute the Xwork environment?
I am writing some Xwork documentation and an example, and I was
surprised that there isn't a way to run XWork as a standalone command
processor from the command line.
---
Sorry that I didn't post this comment to the mailing list to begin with.
It is also on the Wiki, probably not the best place for a meaningful
technical discussion.
September 15, 2003 pazu, 9 hours, 27 minutes ago.
--
Just one thing bothers me: if
PROTECTED]
> Subject: Re: [OS-webwork] DefaultComponentManager (fallback?)
>
>
> Yes, there is a use case. Look at the lifecycle listeners,
> they set up fallbacks.
>
> -Pat
>
> - Original Message -
> From: "James Cook" <[EMAIL PROTECTED]>
> To
t; To: [EMAIL PROTECTED]
> Subject: RE: [OS-webwork] DefaultComponentManager (fallback?)
>
>
> I believe that allows the hierarchy of component managers
> (request -> session -> application)
>
> > -Original Message-
> > From: James Cook [mailto:[EMAIL P
Is
private DefaultComponentManager fallback;
going to be deprecated/removed? What was the intent?
---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
I just ran the Xwork tests with no problems. I had to add the clover jar
to my /lib directory. (Junit jar was already there). What is your
error?
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On
> Behalf Of Cameron Braid
> Sent: Sunday, August 31, 2003 10:19
Conceptually, XSLT has always seemed pretty cool. I would be interested
in how your application fairs when you are done. (Seriously, please
report back.) We experienced unusable performance issues and some very
complicated XSLT once we started moving beyond trivial examples. This
was some time ago,
+1
We have our own getSession method that sync's WebWork's map. I thought the
lack of syncing was raised as a bug a couple years ago!
---
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals
I would think that this behavior would be desireable for other presentations
and not just Sitemesh. You may want to give the filter a generic name that
describes the cleanup.
Also, any reason why this filter should not be enabled by default?
> -Original Message-
> From: [EMAIL PROTECTED]
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] Behalf Of
> Mike Cannon-Brookes
> What is wrong with the filter/servlet combination? It works
> perfectly, will
> allow SiteMesh usage with WW tags in the decorator (that's a 'nice'
> combination - no interdependenc
> [mailto:[EMAIL PROTECTED] Behalf Of
> James Pan
> Sent: Monday, July 28, 2003 9:55 AM
> What I want to is actually very simple: to allow some kind of
> link with parameters, like
>
> .../mylink?parm1=hello&parm2=world
>
> but to allow for omitting parameters as I see fit... so after
> that call a
Dudley, I can't speak for the core development team, but I suspect 1.3 will
get some attention to fix any major bugs if they are found, but I would
expect (and hope) that most of their attention is focused on WW2.
One of the original goals was stated that WW2 will be backward portable with
WW 1.3
Well, even though the original WebWork was touted as a general-purpose
command architecture, 99% of those who participated on the lists were using
it as a web framework. In this current generation, the distinction is made
even more clear. XWork gives you a basic command infrastructure, and
WebWork2
You could clean up the ActionContext in the dispatcher (or a filter)
earlier, rather than later. For example, the first thing the dispatcher
could do is look for this context in the session/request and null it at that
time. It should acomplish the same effect, right?
> Jason Carreira
> Sent: Tuesd
If the filter cleans up, shouldn't it also be the place where the action
context is added to the request/session? Perhaps this is not possible given
the particular wrapping needs, but it just seems slightly kludgey to add the
context in the dispatcher servlet and remove it in a filter.
> -Orig
I noticed in Mike's TSSS write-up that WebWork2 now has FreeMarker
integration. Is that true? Did someone twist someone's arm, or did the
FreeMarker developers pony up? ;-)
---
This SF.net email is sponsored by: VM Ware
With VMware you can run
We run a large webwork (1.3) application using velocity as well on
production on AIX.
> -Original Message-
> Behalf Of Butt, Dudley
> Sent: Wednesday, July 16, 2003 11:18 AM
>
> will WW run on AIX?
>
---
This SF.net email is sponsore
> I agree with what you are saying, it should be kept pure. However, from
> a practical aspect I am looking to test my actions, the stuff they call,
> and the many descriptors (validators, xwork.xml, etc) that I have setup.
We are considering the following approach to testing.
Like you, testing
> Jason Carreira
> Sent: Tuesday, June 17, 2003 9:55 AM
>
> So you don't think there should be a HttpHeaderResult to enable
> you to return special Http header codes?
I like the idea, but should it be a simple map (ala session)?
---
This SF.N
Thanks Frank, I will try to check it out this weekend.
Just one suggestions:
How about using @webwork.action for *both* class decalrations and methods
declarations? Can the XDT tell the difference between the same tagname used
for class and method?
thanks,
jim
> -Original Message-
> Fr
> I'm getting pretty good at this XDoclet stuff, I can probalby
> write a template to generate the actions.xml file if need be.
What does your xdoclet tags look like. I didn't think that XDoclet supported
command driven actions yet. Jason had a good proposal a few weeks ago:
http://sf.net/mailarc
On Thursday, March 20, 2003, at 12:24 AM, Snehal K.Gandhi wrote:
javax.xml.parsers.ParserConfigurationException: AElfred parser is
non-validating
at
com.icl.saxon.aelfred.SAXParserFactoryImpl.newSAXParser(SAXParserFactor
yImpl.java:34)
at
org.apache.commons.digester.Digester.g
I am using xdoclet to generate my views.properties file, although I am
leaning towards not going this route. In the back of my head I feel that
I am coupling my views to tightly with my action. I suppose there has to
be a coupling somewhere (views.properties), but it is just so damn
convenient.
Th
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On
> Behalf Of Jason Carreira
> > And eventually SLSB -> Action -> Web Tier?
> >
>
> Well, one would hope not. Again, it's dependent on how you're
> using you Actions.
Gotcha, I just thought that was his int
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On
> Behalf Of Anders Engström
> The idea is to make the actions "mobile" and move their place
> of execution to the EJB-container. This way you minimize the
> amount of network use - instead of letting each act
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On
> Behalf Of Jason Carreira
> I think his suggestion makes sense. It's a valid pattern, IMHO.
Maybe I'm missing it, but moving action execution to the EJB-tier is a
good idea why? Why not just code some POJO's
gt; -----Original Message-
> > From: James Cook [mailto:[EMAIL PROTECTED]
> >
> > Sorry but this doesn't make much sense to me. When you say
> Actions, I
> > guess you are not referring to WW actions. If you are, I
> > think you may
> > need to re
On Tuesday, February 25, 2003, at 02:54 PM, Anders Engström wrote:
We want each HTTP-request to run under the scope of one transaction in
the business layer. This is where I need some advice - how do we best
utilize WW to achieve this?
Most of the transactions in our application are initiated on t
Title: Message
I
don't think it belongs in WebWork proper. I suppose you just want this
class?
class
ActionForward implements Action {
public String execute() {return
SUCCESS;}
}
If
that is it, then it would take 2 secs to add it to your personal project. I
imagine it wouldn't be f
I think the comments are very valid from a theoretical pov, and I was
opposed to the Command-driven pattern as a means of convenience. However, as
our project progressed, the number of classes increased tremendously because
of the inheritence model for shared data. The structure made it difficult
f
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of
> Jason Carreira
> How could Xwork know to use it? With Constructors, at least the purpose
> of the method is clear when you're doing runtime reflection.
I see...
The AccountManager class is basically
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of
> Patrick Lightbody
> Yes, I'm working on getting this feature in place. Currently it
> works -only-
> if getAccount() doesn't return null. My goal is to have Ognl realize this
> and create a new Accoun
I'm surprised it is working when the view is rendered. Does your 'value'
attribute actually get set somehow? You are trying to make a call to
getToolItem(String). Does this method exist in your action, and if it does
you would usually have to expose the action using a method like getModel()
{ retur
Do you guys bundle the Hibernate package in your web-app, or do you
deploy it to the app server standalone? I am confused why their
examples show Hibernate deployed to jBoss as a service, yet all the
configuration files reference app-specific classpaths and mapping files.
It would probably be a service to all to make the underlying Map that
represents the Session threadsafe using the Collections threadsafe wrapper.
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of
> Jason Carreira
> Sent: Thursday, February 13, 2003 10
Not that I really care, but for some strange reason I feel like the
only adult in the room. Can't you leave this offline? I am sure there
is a "He is saying bad things about me" mailing list somewhere,
although it is probably dominated by 10 year olds.
On Thursday, January 30, 2003, at 02:25 A
The one thing that may prevent FreeMarker from being "drop-in"
replacement for the WebWork JSP is the expression language supported by
the WebWork taglibs. It also affects JSTL integration and Velocity
integration. It concerns the hierarchical nature of the ValueStack.
Most of these third-party
We use webwork 1.2 as well on a project, but sidestep webwork tags in
favor of Velocity. That being said, Velocity makes callbacks to our
actions in much the same way that the webwork ui tags do, however we
have no performance problem on large select lists. I would assume that
the bottleneck is in
[EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]] On
> Behalf Of James Cook
> Sent: Sunday, January 12, 2003 11:55 AM
> To: [EMAIL PROTECTED]
> Subject: [OS-webwork] broken in RC1?
>
>
> Is anyone having success with the when the value
> is an action? Sure it executes the
age-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]] On
> Behalf Of James Cook
> Sent: Thursday, January 09, 2003 9:18 PM
> To: [EMAIL PROTECTED]
> Subject: [OS-webwork] Inclusion of a webwork call invoked by JSTL
>
>
> I have a web page coded as a JSP and utiliz
Is JasperReports already integrated as a view technology in 1.3? I saw some
documentation regarding its use in WebWork.
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of
> Michael Blake Day
> Sent: Thursday, January 09, 2003 7:02 PM
> To: [EMAIL PROTEC
Quick clarification...
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]] On
> Behalf Of James Cook
> Sent: Thursday, January 09, 2003 9:18 PM
> The invokes the webwork action and the view is
> returned, however the view is returned a
I have a web page coded as a JSP and utilizing JSTL tags. On this page,
I have a dynamic report that is generated by a webwork call. I am using
the tag to reference the webwork action, and it is mapped to
a JSP view for rendering.
Time for a visual?
index.jsp
+--+
|
No, it should be able to do this on its own. Try adding a method to your
action with the following signature:
public String[] getTitles();
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of
> Rick Salsa
> Sent: Friday, December 20, 2002 10:00 AM
> To:
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of
> boxed
> I find this rather confusing myself. Why not just do name="com.example.weblog.actions.entry.CreateEntry"
> alias="CreateEntry"> if
> you want to avoid any ambiguitiy? This seems like a lot bet
I agree with the filter approach, however an unsophisticated technique that
can be applied to just those pages that you are interested in is to embed a
callback to your action in the web page. We use this approach to time how
long it takes for an action to complete, including the client-side
callba
I have chosen to test actions by substituting another views.properties file.
This "testing_view.properties" file includes the same type of calls the
normal view would execute, however the results of these views produce an XML
stream based on their outcome.
Success
At the risk of further embarassment, I'm not a newbie.
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:opensymphony-webwork-admin@;lists.sourceforge.net] On
> Behalf Of Maurice Parker
> But Anders, you didn't really help him any did you? One of the best
> ways to earn the re
absolutely correct
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:opensymphony-webwork-admin@;lists.sourceforge.net] On
> Behalf Of boxed
> Sent: Tuesday, November 05, 2002 9:41 AM
> To: [EMAIL PROTECTED]
> Subject: Re: [OS-webwork] User-Agent Variations
>
>
> > I am working
I am working on a website that will have a completely different set of
actions based on the user-agent. Think of IE6 and Pocket PC. What is the
best technique in a WebWork framework to serve up this site?
---
This sf.net email is sponsored by:
I don't particularly like actions.xml, and I really like the simplicity of
views.properties. I find it much easier to read.
I'm ambivalent about deprecating views.properties only because it is a bitch
to maintain in a multi-user environment. At early stages in a project, it is
a heavily edited fil
Has always worked for me, although I use views.properties.
nosession = /common/no_session.vm
noauthorization = /common/no_authorization.vm
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:opensymphony-webwork-admin@;lists.sourceforge.net]On Behalf Of
> Mike Cannon-Brookes
> Sent:
Friend of mine threw together this filter that exposes methods on the
ValueStack (hence your action) to the JSTL expression language. Now I can
finally do a:
and have it hit the getName() method in my action. Very cool.
Just take the class below and add it as a filter. Use filter mapping of
*.a
I think we should finally separate core Webwork servlet dispatching from the
view technologies (and taglibs).
---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_
78 matches
Mail list logo
