excellent, thanks this will cover it nicely.
Cameron Braid wrote:
When an exception occurs, either a http 404 not found response is sent
(when the action isn't found), or a http 500 internal server error
response is sent (both using http headers). For 500 errors, the
exception is also stored in
If you're looking for acceptance testing frameworks - I suggest you look at
Jameleon (http://jameleon.sourceforge.net/).
It has a really, really cool OO way of looking at web acceptance testing -
which looks like it is really neat for creating reusable chunks of tests.
(Oh - and it uses Jelly so y
A nice feature would be to convert values with a converter inheritance
model.
To convert java.sql.date -> String, XWorkConverter looks up a java.sql.date
converter, and if it is not present, XWorkConverter could try to load a
java.util.date converter, and so on... It could minimize the entries in
x
Also.. Pester them about upgrading to have better eclipse cvs support,
since that won't happen til Nov... :(
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Pat Lightbody
Sent: Tuesday, 7 October 2003 4:08 PM
To: [EMAIL PROTECTED]
Subject: Re: [OS-webwork]
Alright alright, don't let Hani bait you. I pretty much agree with you, I
suggest you complain to java.net. While you're at it, can you ask them WTF
happened to our clickstream and sitemesh projects? :)
-Pat
- Original Message -
From: "Drew McAuliffe" <[EMAIL PROTECTED]>
To: <[EMAIL PROT
At my company (Spoke Software), I just finished creating an acceptance test
framework. So for unit tests, I just construct the Action and run execute()
by hand. All other stuff is caught in the acceptance tests.
The way the framework works is: we use JWebUnit and provide a SpokeTestCase
base class
Title: Message
Well, if you want to go ahead and put it in place
(and don't have any bugs to fix ;) then go for it. In the meantime, I'm going to
continue to focus on backwards compatibility and docs. The release happens when
those two things happen.
- Original Message -
From:
Push does the model to the stack as for ModelDrivenInterceptor before
(Bexecuting the action. Afterwards, the action is executed. The model is
(Breflected in the processing of the action and when the reference is
(Breplaced (For instance, data is acquired from DB), the content is not
(Breflecte
Look, that's the way java.net works. It's not something I, or any other
admin of opensymphony can control. All I'm saying is that it's a minor
inconvenience at best. The alternative would be to use sf.net, which as
well as being a day late for anon access, is now back to its old habits
of refus
I disagree with that pretty strongly. I've been getting the latest from
CVS without any hassle from sourceforge for a while now. This is the
first time I've had to work with the latest again, and the repository
has changed. If that repository change means new rules for CVS access,
this is a perfect
That would be cool. I don't have much control over what the JDBC driver
returns as an implementation of Date, but I would think that Timestamp
would be pretty common. This is on Oracle 9i, by the way, if it matters.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Be
The string conversion problem I was having was my own problem, not with
webwork or xwork. But it still doesn't look like conversion is being
used on property gets via $ognl.findValue or $stack.findValue.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Drew
When an exception occurs, either a http 404 not found response is sent
(when the action isn't found), or a http 500 internal server error
response is sent (both using http headers). For 500 errors, the
exception is also stored in the the request scope, according to the
servlet and jsp specs, to a
The params interceptor which sets the action properties uses the
valuestack to set the properties.
Since you have the md interceptor, your action and the model will be on
the stack.
Therefore, when stack.setValue(name, value) gets called, the objects in
the stack are checked from top down to find
Yes, your Action properties will still be available (although they will
be blocked by properties on your Model if they have the same name)
> -Original Message-
> From: Drew McAuliffe [mailto:[EMAIL PROTECTED]
> Sent: Monday, October 06, 2003 9:03 PM
> To: [EMAIL PROTECTED]
> Subject: [OS-
If I have a model driven action (the action uses the md interceptor and
implements the interface), are my other action properties still
available for setting using the parameters interceptor?
For example, I have an action like this:
public class MyAction extends MyBaseAction implements ModelDrive
I believe it's another module next to webwork on dev.java.net CVS... I
haven't checked it out yet, though...
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On
> Behalf Of Eric Pugh
> Sent: Monday, October 06, 2003 2:46 PM
> To: [EMAIL PROTECTED]
> Subject: RE:
You could just set up a error page in your web.xml... Then you get your
pretty error page for all errors, not just errors inside Actions...
> -Original Message-
> From: Francisco Hernandez [mailto:[EMAIL PROTECTED]
> Sent: Monday, October 06, 2003 8:44 PM
> To: [EMAIL PROTECTED]
> Subject
It's probably because it's a timestamp, and not a Date. While a
timestamp "is a" date, the type conversion is using the class name for
looking up converters.
We could add this to the XworkBasicConverter and have it treat it the
same as a date?
Jason
> -Original Message-
> From: Drew McA
You go to java.net, sign up, and apply for the observer role at the
webwork project pages.
People who run cvs should definitely NOT be users who are too lazy to
go through those incredibly simple steps!
A project is pretty screwed if its growth depends on
instant-gratification anonymous cvs ac
what do you guys think about returning a result like
invalid.action.result or something along those lines when the action
executed throws NPE? this way I can just map an global-result to show a
pretty screen to the users
currently if an action throws NPE i get 500 Internal Server Error
-
How does one go about doing that? And whatever happened to anonymous
access? I would think that anonymous access would allow for a much wider
distribution of the codebase. If everyone has to apply for observer
access, the project isn't going to grow very fast.
-Original Message-
From: [EMA
I'm having trouble with OGNL outputting date values from a database
read. Date fields in my object get populated with Timestamp objects (a
sql specific subclass of Date). If I try to output field "dateLogged" on
object "myObject" using the following call,
$!{ognl.findValue("dateLogged", $myObject)}
this is what i use, i just hit enter when it asks me for a password
cvs -d :pserver:[EMAIL PROTECTED]:/cvs login
cvs -d :pserver:[EMAIL PROTECTED]:/cvs checkout webwork
Drew McAuliffe wrote:
I'm having problems getting the latest from the new CVS repository.
Neither the anoncvs or the guest/
Just apply for an observer role in the webwork project.
On Monday, October 6, 2003, at 08:20 PM, Drew McAuliffe wrote:
I'm having problems getting the latest from the new CVS repository.
Neither the anoncvs or the guest/guest logins work. Does anyone have
the
valid guest account name, and can th
I'm having problems getting the latest from the new CVS repository.
Neither the anoncvs or the guest/guest logins work. Does anyone have the
valid guest account name, and can the java.net site be updated with it?
Thanks,
Drew
---
This sf.net
addFieldError is the proper way. I should probably make the
getFieldError return an unmodifiable list...
> -Original Message-
> From: Francisco Hernandez [mailto:[EMAIL PROTECTED]
> Sent: Monday, October 06, 2003 7:25 PM
> To: [EMAIL PROTECTED]
> Subject: [OS-webwork] whats the proper way
has anyone taken a look at this issue lately?
http://jira.opensymphony.com/secure/ViewIssue.jspa?key=WW-261
basically id like to know what should be way to add field errors,
addFieldError("field","field Error") // this adds the field error to the
List of fieldErrors for field
or
getFieldErrors.
Of course you realise the irony here. If there were such people, they would
never read your message? ;)
I subscribe to both lists, so I'm really on the fence. I don't mind either
way, it makes little difference!
M
On 7/10/03 1:39 AM, "Hani Suleiman" ([EMAIL PROTECTED]) penned the words:
> So, i
Sure - just create a filter which shows all issues created in the last 7
days, and then subscribe the list to it every 7 days :)
M
On 7/10/03 4:16 AM, "Jason Carreira" ([EMAIL PROTECTED]) penned the
words:
> I don't think so
>
>> -Original Message-
>> From: Robert Douglass [mailto:[EMAI
I picked it up reading some email archive. Was the work to integrate
PicoContainer committed to CVS anywhere? Is webwork-extensions an
OpenSymphony project? That might give me the pointer I need.. Although it
seems straightforward..
Eric Pugh
> -Original Message-
> From: [EMAIL PROTE
On 10/6/03 2:16 PM, "Jason Carreira" <[EMAIL PROTECTED]> wrote:
> I don't think so
Easy to do by having the messages bounced to mailman.
---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_
i think issue creation emails are good ok to be sent to
[EMAIL PROTECTED] list and the other emails like comments, updates, etc
to issues sent to the cvs list is fine too
Jason Carreira wrote:
I reset it back to the Xwork/WebWork notification scheme (it had been
turned off altogether)... I thin
the purpose of CoolUrlTag would be to generate the Urls in the 'Cool'
style just like the existing UrlTag creates 'boring' Urls now, but now
im thinking that it should be a config paramter so that I can switch the
whole webapp between Cool and Boring Urls with a single change to an
option somew
Is anyone using that issue tracker? I would hope not...
I'll change it to just send to the CVS list
> -Original Message-
> From: Patrick Lightbody [mailto:[EMAIL PROTECTED]
> Sent: Monday, October 06, 2003 1:58 PM
> To: [EMAIL PROTECTED]
> Subject: RE: [OS-webwork] [JIRA-OS] Created: (WW
I don't think so
> -Original Message-
> From: Robert Douglass [mailto:[EMAIL PROTECTED]
> Sent: Monday, October 06, 2003 2:17 PM
> To: [EMAIL PROTECTED]
> Subject: RE: [OS-webwork] [JIRA-OS] Created: (WW-336)
> Iterator tags value attribute required
>
>
> Is there a way to have this ty
Is there a way to have this type of notification collected in a digest, and
sent to this list once a day|week?
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of
Patrick Lightbody
Sent: Monday, October 06, 2003 6:58 PM
To: [EMAIL PROTECTED]
Subject: RE: [OS-webw
I haven't heard of any effort to integrate Avalon and XW/WW2... Some
months ago, there was some work on integrating PicoContainer, but that
has moved into webwork-extensions. There's also been some talk of
integrating Spring's BeanFactory...
> -Original Message-
> From: Eric Pugh [mailto:[
I don't mind getting them.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of
Hani Suleiman
Sent: Monday, October 06, 2003 4:40 PM
To: [EMAIL PROTECTED]
Subject: RE: [OS-webwork] [JIRA-OS] Created: (WW-336) Iterator tags
value attribute required
So, is there a
For me, I like keeping this list for purely communication only. Four
reasons:
1) I check the CVS list for automated stuff and this list for discussions.
Having stuff sent here blurs that line
2) People sometimes reply to the creation emails, but that discussion is
never logged in to JIRA -- the
Hi all,
I am looking into integrating some existing components that were written to
use the Avalon IoC scheme. I saw posts a couple months back about
integrating an Avalon container for the sample app, but they didn't seem to
go anywhere. The way I understand it, WW2 uses XWork as the command pa
From: "Jason Carreira" <[EMAIL PROTECTED]>
> I reset it back to the Xwork/WebWork notification scheme (it had been
> turned off altogether)... I think it's valuable for new issues to come
> here, so everyone can see what the issues are and might see that they're
> not the only one seeing this... On
So, is there anyone who doesn't automatically hit delete when they see
that [JIRA-OS] subject on this mailing list?
Jason Carreira said:
> I reset it back to the Xwork/WebWork notification scheme (it had been
> turned off altogether)... I think it's valuable for new issues to come
> here, so every
What are you trying to do? What do you need static methods for?
ActionContext.getContext() is a static method.
> -Original Message-
> From: Samuel Mota [mailto:[EMAIL PROTECTED]
> Sent: Monday, October 06, 2003 11:08 AM
> To: [EMAIL PROTECTED]
> Subject: [OS-webwork] ValueStack
>
>
> H
I reset it back to the Xwork/WebWork notification scheme (it had been
turned off altogether)... I think it's valuable for new issues to come
here, so everyone can see what the issues are and might see that they're
not the only one seeing this... Only issue creation comes here now, more
notification
Hi,
I've read some posts and docs but I'm still not able to use the ValueStack at JSP
blocks ...
I've tried: OgnlValueStack stack = ActionContext.getContext().getStack();
but ActionContext has no static methods (I've imported xwork.*)...
I'd like to test some labels (strings) and do some flow
Title: RES: [OS-webwork] Advanced URL mapping?
I think we could use a CoolURI tag to render tags pretty much in the same way the URL tag does today (maybe even patch the url tag so that it generates both cool and "boring" urls?)
-cv
-Mensagem original-
De: Jérôme BERNARD [mailto:[EM
Francisco Hernandez <[EMAIL PROTECTED]>:
> I think a CoolUrlTag would still be usefull, just so that we can have
> CoolUrl's used throughout the site, I used a shopping cart written in
> PHP that had these kind of Urls and there was a config option to use
> CoolUrls or regular Urls, it worked p
ARGH!
We've been over this before. Please do NOT have every new issue sent
here, this is supposed to go to a separate cvs list for those who care.
On Monday, October 6, 2003, at 05:29 AM, [EMAIL PROTECTED]
wrote:
Message:
A new issue has been created in JIRA.
--
Title: Message
I'm
wrong again.. (and now I am speaking to myself! )
There
is a difference between the way OgnlUtil.setProperties(Map props, Object o, Map
context) works and how the params interceptor works.
I have
added a unit test SetPropertiesTest into xwork. To show the same thing be
Title: Message
well... looks like I have proven myself wrong.. my unit test
passes.
I'll
investigate deeper.
-Original Message-From:
[EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Cameron BraidSent: Monday, 6 October 2003 8:54 PMTo:
[EMAIL PROTECTED]Subjec
Title: Message
Nah..
it has
void
setId(Long);
Long
getId();
I'll
write a unit test to show that it is happening.
-Original Message-From:
[EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Jason CarreiraSent: Monday, 6 October 2003 7:39
PMTo: [EMAIL PROTEC
ooh thanks for letting me know about dbunit, and now that you mention
ant, another tool i need to get deeper into, where will the madness stop!?
Cameron Braid wrote:
I don't think that you need to run tests against your live data.
We do the following
A) provide ant tasks to (re)create the data
This is a cool article that describes how you can avoid having to set up a
test database for unit testing. I don't use this method because I haven't
yet had the guts to start deploying stuff with AspectJ code, but it's a nice
read.
http://www-106.ibm.com/developerworks/library/j-mocktest.html
---
Title: Message
Does
the class you're using have a get method for the same property name as a String?
I'm just wondering if it's getting confused...
-Original Message-From: Cameron Braid
[mailto:[EMAIL PROTECTED] Sent: Monday, October 06, 2003 4:01
AMTo: [EMAIL PROTECTED]Sub
Message:
A new issue has been created in JIRA.
-
View the issue:
http://jira.opensymphony.com/secure/ViewIssue.jspa?key=WW-337
Here is an overview of the issue:
-
If you separate your persistence code into some kind of
PersistenceManager with an Interface, you can use Mock objects to test
that your Action interacts with your PM correctly, then have unit tests
for your PM to make sure it works correctly.
> -Original Message-
> From: Francisco Hernand
Title: Message
This
sounds like it would be ok to add in... provided you add the doc to the
Wiki
-Original Message-From: Cameron Braid
[mailto:[EMAIL PROTECTED] Sent: Monday, October 06, 2003 3:46
AMTo: [EMAIL PROTECTED]Subject:
RE: [OS-webwork] Idea to enhance the veloc
I think it's #params['home']
> -Original Message-
> From: Frederick N. Brier [mailto:[EMAIL PROTECTED]
> Sent: Monday, October 06, 2003 3:07 AM
> To: [EMAIL PROTECTED]
> Subject: RE: [OS-webwork] Using the PushTag
>
>
> That seems like a pretty cool idea. And the filter could
> null o
Message:
A new issue has been created in JIRA.
-
View the issue:
http://jira.opensymphony.com/secure/ViewIssue.jspa?key=WW-336
Here is an overview of the issue:
-
Message:
A new issue has been created in JIRA.
-
View the issue:
http://jira.opensymphony.com/secure/ViewIssue.jspa?key=XW-103
Here is an overview of the issue:
-
Message:
A new issue has been created in JIRA.
-
View the issue:
http://jira.opensymphony.com/secure/ViewIssue.jspa?key=XW-102
Here is an overview of the issue:
-
Pat Lightbody <[EMAIL PROTECTED]>:
> Jerome,
> I think that at this point it's a nice feature but we probably won't have
> the time to get this in 2.0. I'd like for all the developers to focus on
> getting 2.0 out the door with 1.3 compatibility and all the open bugs
> resolved.
Pat,
I agree wit
Message:
A new issue has been created in JIRA.
-
View the issue:
http://jira.opensymphony.com/secure/ViewIssue.jspa?key=WW-335
Here is an overview of the issue:
-
Title: Message
You're
right, code is re-entrant.
It's
strange enough because ognl/xwork seems to work differently when the value
is null, it looks for the accessor with the input class
argument...
-Message d'origine-De :
[EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]De la part
I don't think that you need to run tests against your live data.
We do the following
A) provide ant tasks to (re)create the database
since we are using hibernate this is easy
B) provide ant tasks to populate the database with the test data
we use DBUNIT to insert the data from a
Title: Message
nope..
the code that is there works fine, because the convertValue(...) is re-entrant
for the case where the value is an array.
so
result does get set to OK_NULL_RESULT, and then returned as
null.
What I
think is happening internally with ognl is that since the value is b
That seems like a pretty cool idea. And the filter could null out the
ActionContext ThreadLocal on exit allowing it to be garbage collected and
not hanging around. BTW, what is the syntax for accessing a request
parameter off of the value stack? Is it "$req.home"? Assuming home is the
key o
Title: Message
yeah..
its optional..
That
is the idea behind this mod - to allow re-use of the existing
vm's
-Original Message-From:
[EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Robert DouglassSent: Monday, 6 October 2003 5:17
PMTo: [EMAIL PROTECTED]Sub
Title: Message
Obviously, it doesnt work, and the error comes from a bad test in the
convertValue(...) method.
The
code :
if(
"".equald(value)) {
result = (String.class.equals(toType)) ? "" :
OK_NULL_RESULT;
return result;
}
has to
be changed for :
if(
value instanceof Strin
so i need to write tests to for successfull conditions and then also
failure conditions, I was planning on having the tests all work against
a test database so i dont touch my live data.
Robert Douglass wrote:
I make sure to write tests that break the code too. So the first test I run
is said ac
I make sure to write tests that break the code too. So the first test I run
is said action without any input parameters, for example. Then with bad
parameters. The bad tests have to all fail before I test the correct case.
And, by the way, make sure you have worked out whether you're going change
s
Title: Message
It's
an optional parameter, right? I think it's a great idea. In particular, I need
to rewrite a proprietary widget toolkit in WW and now I can just provide custom
header and footer templates instead of limiting myself to a single view
technology by writing a taglib or macro l
73 matches
Mail list logo
