Can I see the serverside code for your edit page?
I get the feeling that onPassivate() is being called on the edit page and
returning an id persisted in the session.
Hello,
I'm using the AlertManager service along with the Alerts component to display
messages. For non ajax requests it behaves like I prefer, but for ajax requests
messages are appended to the alerts list and old ones are only cleared when
the whole page is refreshed.
This is of course
package org.pacificcis.piko.pages.person;
import org.hibernate.Session;
import org.pacificcis.piko.entities.Person;
import org.pacificcis.piko.pages.Persons;
import org.apache.tapestry5.annotations.InjectPage;
import org.apache.tapestry5.annotations.PageActivationContext;
import
Commenting out onPassivate does not help. Thanks
--
View this message in context:
http://tapestry.1045711.n5.nabble.com/Pagelink-above-grid-picks-up-context-from-last-pagelink-in-grid-tp5660049p5661333.html
Sent from the Tapestry - User mailing list archive at Nabble.com.
Neither does this work...
Object onPassivate() { return new Person(); }
Only context=0.
--
View this message in context:
http://tapestry.1045711.n5.nabble.com/Pagelink-above-grid-picks-up-context-from-last-pagelink-in-grid-tp5660049p5661353.html
Sent from the Tapestry - User mailing
For some reason, HTTP request to edit/create page of the entity results in
attempted coercion of the word css into the Id parameter (type Long) of
the entity -- application workflow seems normal though. Just plenty of
vomit in jetty console log...shown below.
S T A C K T R A C E
I think you want flash persistence rather than the longer lived default
session persistence
@Persist(PersistanceConstants.FLASH)
I have not used @PageActivationContext but the javadoc states that it
performs onPassivate() automatically
I don't think jdo matters as well.
Just trying using POJOs and annotations are ignored as well.
However, I did discover that annotations are only ignored when I ask for
application/json, for application/xml everything seems to be working.
Am I mistaken to think that the @Xml* annotations would
I think you might be having a name clash with a page name and a CSS file
name in the same folder and tapestry is thinking it's an event URL to the
page instead of a CSS path.
Solution: name the CSS file differently or move it to a different folder
On Tue, 24 Apr 2012 06:45:03 -0300, Lance Java lance.j...@googlemail.com
wrote:
I think you might be having a name clash with a page name and a CSS file
name in the same folder and tapestry is thinking it's an event URL to the
page instead of a CSS path.
Solution: name the CSS file
You could overwrite the FieldValidatorSource and the
FieldValidatorImpl, which is responsible for creating validator
instances, so that it .
Or you use this hack in your Validator. It uses the message object of
the current field:
public void validate(Field aField, Integer aConstraintValue,
Tapestry has already passed an instance of MessageFormatter to you in the
Validate method.
Use this instead of the Messages directly as the actual message key is
dependent on the field being validated.
Download the tapestry source and take a look at an existing Validator (eg
Required)
You need to use an additional constructor in the Custom validator class.
public MyPasswordValidator() {
super(null, String.class, msg-key);
}
Where msg-key is the key for the validation message displayed on the the
field. You can put this key in ValidationMessages.properties file.
--
Most likely, your template has a relative path to it ... you've done:
link rel=stylesheet src=css/site.css/
That works great inside a static page, where you know exactly what the
browser URL is going to be.
In Tapestry, once you add page activation contexts, various folders,
and other mappings
Sorry I thought this was a very basic question. Anyone?
?? 2012/4/23 17:33, Rural Hunter :
Hi,
We put parameters in context parameters or servlet parameters web.xml
for normal java web application. But where is the best place to put
application or page parameters in tapestry?
is it a
Hello!
I just read Howard's response in regards to using Tapestry's
${context:css/blah.css} stuff so T5 handles the stuff as an Asset.
Is this a concept that people generally do not use because they know of other
ways of handling assets (i.e. insert relative paths), or is it new in T5?
It
There is no silver bullet solution for this and it's a matter of taste.
Tapestry has a number of built in SymbolProviders. You can specify a
combination of system properties, init-params to the tapestry filter,
factory defaults and application defaults
http://tapestry.apache.org/symbols.html
If
Hi all,
Just a quick question, how do I do to make dinamic content in an
accordion component.
Thanks!
--
View this message in context:
http://tapestry.1045711.n5.nabble.com/Tapestry-jQuery-tp5662365p5662365.html
Sent from the Tapestry - User mailing list archive at Nabble.com.
There are several ways of doing this and I don't think there is any
consensus.
Personally I don't like storing things in text/xml files because you can't
refactor them.
Symbol providers are final and Strings if that works for you (but are run
thru Type Coercion)
I prefer to use Environmental
Hi,
Using this demo: http://jqueryui.com/demos/accordion/
I would take a guess to put an ${accordionPage1} within the first p tags in
the tml below.
and in the java file have a getAccordionPage1(){ return Some text goes here ;}
You can then take it further and get down into your lower layers
ok, thanks a lot. I don't care if it has to be in my own properties file
or not. I just want it easy to edit/chage. Now I work this way with your
hint:
1. define the parameter as context-param in web.xml, eg. myParam
2. inject it in my page class:
@Inject
@Value(${myParam})
String myParam;
Hi, I actually did in this way an it works, but I want to use the jquery
accordion provided on Tapestry-Jquery...
div id=accordion
t:loop t:source=nodes t:value=node
# ${node.name}
div
ul
I prefer to narrow things down to one system property: the Tapestry
execution mode:
http://tapestry.apache.org/configuration.html#Configuration-SettingExecutionModes
This determines which Tapestry IoC modules(s) will be loaded.
The modules make contributions to the @ApplicationDefaults and
That last email was all tapestry5-jquery. It was a copy of the jqueryui.com
demo as an example.
All you have to do is substitute where h3 begins with your loop.
HTH :D
On 25/04/2012, at 1:43 AM, arterzatij wrote:
Hi, I actually did in this way an it works, but I want to use the jquery
For modules you really have to use Assets because you need to serve them from
the classpath. For an app you could just hard code paths but it's harder
than using Assets because you have to deal with the context plus they handle
compression, expires etc. Lastly you can override the url lookup part
Thanks for the info Howard. Actually my requirement is like this: my
application generates reporting files in doc/pdf format. I need to set a
path where to store the generated reporting files. It maybe different
when the app is deployed on different servers. So I think the execution
mode way
That kind of config I prefer to do it by db and handle them by page config,
Just an opinion :)
--
View this message in context:
http://tapestry.1045711.n5.nabble.com/Best-place-to-put-application-page-parameters-tp5658805p5662525.html
Sent from the Tapestry - User mailing list archive at
That's another advantage of configuration in Java code ... you can put
more complex decisions about these things in code, such as looking at
the hostname to decide symbol defaults.
On Tue, Apr 24, 2012 at 9:04 AM, Rural Hunter ruralhun...@gmail.com wrote:
Thanks for the info Howard. Actually my
Sound reasoning to do it the right way :)
On your last point re: CDN delivery - I only found this
http://tapestry.1045711.n5.nabble.com/Referencing-external-assets-td4346215.html
There's no further mention of CDN/Asset overrides in the tapestry documentation
for Assets
There's a JIRA here
While it seems like a good idea (and believe me I've tried) to have one page
handle create/update it turns out that many things conspire to make this
more difficult than it seems and I've just returned to having a PersonNew
and a PersonEdit.
The first problem is if you use @Persist like you are
You can override AssetPathConverter
--
View this message in context:
http://tapestry.1045711.n5.nabble.com/Tapestry-s-css-js-coming-from-other-languages-frameworks-tp5662291p5662599.html
Sent from the Tapestry - User mailing list archive at Nabble.com.
awesome quote: ... This is the hook needed to make use of a Content Delivery
Network...
http://tapestry.apache.org/tapestry5/apidocs/org/apache/tapestry5/services/AssetPathConverter.html
Thanks
On 25/04/2012, at 2:34 AM, trsvax wrote:
You can override AssetPathConverter
--
View this
Did you try flash persistence as I suggested earlier?
Flash persistence only stores in the session momentarily so that state can
be maintained through a redirect after post.
Flash will change it but I don't think it will fix it. For example
if you go to
editPerson/5
and then save, Person 5 will be in the session and you return to Persons
Now press new and you go to
editPerson
The first read will return Person 5 but what you want is a blank form. It's
possible to
https://code.google.com/p/corner/source/browse/corner3/trunk/src/main/java/corner/asset/services/impl/CDNAssetPathConverterImpl.java
Excuse my lack of knowledge up in these levels of tapestry stuff, but...
In this example - does it mean that all assets will be delivered from CDN, or
that can be
I think that example serves everything from a CDN but you can do pretty much
anything you like. I have one that checks S3 and if the Asset is there is
serves it via cloudfront. If it's not there it schedules a copy just serves
the Asset locally. This way you just deploy and all your Assets
cross that bridge when i need to i guess :)
sounds pretty cool, thanks for clarifying stuff
cheers
chris
On 25/04/2012, at 3:02 AM, trsvax wrote:
I think that example serves everything from a CDN but you can do pretty much
anything you like. I have one that checks S3 and if the Asset is there
Yes I was. Jackson has @Json* annotations for json output.
On Apr 24, 2012, at 10:43 AM, Paulo Andrade wrote:
Am I mistaken to think that the @Xml* annotations would also influence the
JSON output?
On Tue, 24 Apr 2012 11:52:46 -0300, Rural Hunter ruralhun...@gmail.com
wrote:
Sorry I thought this was a very basic question. Anyone?
I'd say configuration symbols.
于 2012/4/23 17:33, Rural Hunter 写道:
Hi,
We put parameters in context parameters or servlet parameters web.xml
for
the best solution I found in order to point on external folders (generally
on J2ee, not Tapestry specific) without a fixed or a common base path is a
deployment plan in Weblogic. Basically, you put in web.xml one(or more)
env-entry with a fixed key and your path(or url, or every config value) as
Tried several pre-compilation tricks as well, and it still renders things out
as text, and I think that's a result of the CDATA being XML whereas that
data is rendered and read differently in the browser.
I have read Mr. Shipp's blog about JS and the 5.4 release. I would love to
know more about
So...been experimenting more...and from the fine folks at Stack
Overflow...this solution for pulling HTML out of CDATA. Fine point, it has
to be CDATA and not the JS/CSS commented CDATA.
You will then have to unescape the content before compiling with
Handlebars.compile.
var
I'd create a component so you can do something like this
t:handlebars id=entry-template
p:template
/p:template
/t:handlebars
Then use something like this to write out the script tag
On Tue, 24 Apr 2012 08:34:56 -0300, membersound memberso...@web.de wrote:
I would like to create a temporary input field, and use it for
validating of a password.
Check the add parameter of BeanEditForm (and BeanEditor) in its
documentation. Regarding validation, check the validation
On Tue, 24 Apr 2012 17:26:09 -0300, trsvax trs...@gmail.com wrote:
You still need the CDATA but I'd say the benefits of tml files being
valid XML easily out weight the drawbacks.
Agreed. In addition, to support other formats besides XML, we would need
to write a parser for it. Not easy,
Something like this was on my TODO list. I wanted to pass scripts via
p:script
/p:script
So here it is.
public class Handlebars {
@Parameter(required = true)
private RenderCommand template;
@Parameter(defaultPrefix = literal)
private String id;
Thanks. I'll give it a shot and report back.
I still think there has to be a better way to handle client side templating,
and one that should be baked into the system. If 5.4 is going to support
more MVC style JavaScript, this is going to get a lot of use.
--
View this message in context:
What's the recommended method to reference assets in a .css file - for
example background images?
Currently I use a hard-coded relative URL, but this has the drawback
of causing the browser to cache the resource after it's changed
leading to browsers caching old resources unless I change the name
Hi Dave,
Good question - I don't know about from within a CSS file.
However within a template it offers this:
img src=${asset:context:image/tapestry_banner.gif} alt=Banner/
from this page: https://tapestry.apache.org/assets.html
image/tapestry_banner.gif lives in
If you're using Less CSS (or maybe sass possibly), you can set some basic
variables at the top of your file that can be used as a cache buster in your
paths.
http://lesscss.org/#-variables
--
View this message in context:
On Tue, Apr 24, 2012 at 4:04 PM, Chris Mylonas ch...@opencsta.org wrote:
Good question - I don't know about from within a CSS file.
...
Looking into the tapestry source, specifically
src/main/resources/org/apache/tapestry5/tapestry-console.css
it uses relative links to the pics in
Thanks, guys.
LANCE: @Persist(PersistenceConstants.FLASH) makes nullifies Person in
Persons page (listing page), thereby breaking it...
THIAGO: return person.getId() breaks the Edit Page completely with Render
queue error in BeginRender...
Here is probably the clincher that this is a bug or
Aha. Thanks, Howard. Makes sense. That is probably what was going on.
I am, however, using the context: in all my templates and corresponding
java classes, as intended.
And, for the life of me, I am unable to trap the issue any longer. The logs
seem to be free of it last time I checked
54 matches
Mail list logo
