+1 to start from ViewRoot
It would be possible to overload the method and pass the current component,
which is available via #{component}
e.g.
<h:commandButton onclick="doSomething(#{f:clientId("componentId”,
component)})" />
But AFAIR it's not possible to overload a EL function? Maybe have to choose
another name for the overloaded method.
2015-03-31 23:53 GMT+02:00 Cagatay Civici <cagatay.civici_at_gmail.com>:
> @Arjan,
>
> Considering;
>
> #{f:clientId("componentId”)}
>
> Where you do you suggest we start search, UIViewRoot? The
> #p:component(‘componentId’) does it.
>
> Also how to customize where to start search? .e.g
>
> Regards,
>
> Cagatay Civici
> PrimeFaces Lead
> PrimeTek Informatics
> www.primefaces.org
>
> On Monday 16 March 2015 at 16:22, Josh Juneau wrote:
>
> Hi Kito-
>
> Thanks for the response. I can definitely see your point regarding the
> collapse of namespace using widgetVar. I typically do not run into issues
> with duplicate widgetVar names because I try to be mindful of the names
> I've chosen...but I could see how it may be problematic in some cases.
>
> My main concern is referencing the component in JavaScript...just
> attempting to make this easier for the user. In my opinion, a standardized
> version of something like the PrimeFaces widgetVar would resolve this issue
> to some degree. It is much nicer to reference a single widgetVar or unique
> component identifier than to ensure that the component is being referenced
> correctly by walking the DOM to build the correct component ID.
>
> Josh Juneau
> juneau001_at_gmail.com
> http://jj-blogger.blogspot.com
> https://www.apress.com/index.php/author/author/view/id/1866
>
>
> On Mon, Mar 16, 2015 at 8:38 AM, Kito Mann <kito.mann_at_virtua.com> wrote:
>
> Hello Josh,
>
> I like the widgetVar, but it collapses the namespace; preserving the
> namespace is the whole reason we have naming containers to begin with.
> Moreover, not every component has a widget (at least that's true for the
> standard component). In other words, two components with the same name can
> still exist within different naming containers; this isn't true with the
> widgetVar naming approach. This can be problematic if you have composite
> components or Facelet includes that have assigned a component the same
> widgetVar name you have on your main page.
>
> Are you mainly concerned with referencing the component in JavaScript or
> on the server?
>
> ___
>
> Kito D. Mann | @kito99 | Author, JSF in Action
> Virtua, Inc. | http://www.virtua.com | JSF/Java EE training and consulting
> http://www.JSFCentral.com | @jsfcentral
> +1 203-998-0403
>
> * Listen to the Enterprise Java Newscast: *http://
> <http://blogs.jsfcentral.com/JSFNewscast/>enterprisejavanews.com
> <http://ww.enterprisejavanews.com>*
> * JSFCentral Interviews Podcast:
> http://www.jsfcentral.com/resources/jsfcentralpodcasts/
> * Sign up for the JSFCentral Newsletter:
> http://oi.vresp.com/?fid=ac048d0e17
>
> On Sat, Mar 7, 2015 at 10:52 AM, Josh Juneau <juneau001_at_gmail.com> wrote:
>
> Hi everyone-
>
> From a user perspective, it is cumbersome to reference components within
> JavaScript using the :form:container:component notation. It isn't bad when
> you have a basic form within a view, but when container components, such as
> TabViews or Panels come into play, then it can become a minor headache.
>
> In the PrimeFaces world, components contain an attribute "widgetVar",
> which is basically a pseudo ID for the component, and it is used when
> referencing the component using the PrimeFaces JavaScript API. It is easy
> to use, and less error prone/time consuming then walking the DOM tree to
> determine the correct component ID for complex views...where oftentimes the
> user has to view the sources to determine the correct ID of a component.
> This is the functionality that I'd like to see standardized...and I think
> it would make life easier for component developers as well in some cases.
>
> Hope this helps.
>
> Josh Juneau
> juneau001_at_gmail.com
> http://jj-blogger.blogspot.com
> https://www.apress.com/index.php/author/author/view/id/1866
>
>
> On Fri, Mar 6, 2015 at 10:42 AM, Bauke Scholtz <balusc_at_gmail.com> wrote:
>
> Hi,
>
> I'd rather first lookup in current NamingContainer. If there's nothing,
> then look in NamingContainer children, working up-bottom in the tree. If
> there's nothing, then look in NamingContainer parents, working bottom-up
> until UIViewRoot. That would be more intuitive in case the same ID is used
> in multiple NamingContainers.
>
> Cheers, B
>
> On Fri, Mar 6, 2015 at 5:08 PM, arjan tijms <arjan.tijms_at_gmail.com> wrote:
>
> Hi
>
> On Fri, Mar 6, 2015 at 12:14 AM, Cagatay Civici
> <cagatay.civici_at_gmail.com> wrote:
> > Hi,
> >
> > Sorry for the late reply, due to ongoing workload I was unable to look
> into
> > it.
> >
> > Arjan if you think you can allocate some time instead, I’d really
> appreciate
> > your help.
>
> Sure for general (serverside) ID lookup/find issues I'm always willing
> to help out. If I understood correctly, there's no javascript really
> involved here right? It's just an EL function?
>
> E.g. that EL function itself will become something like:
>
> #{f:clientId("componentId")}
>
> In OmniFaces we have a couple of find component utility methods, that
> basically do relative lookup. My first stab at this would be to use a
> method like the following:
>
>
> https://github.com/omnifaces/omnifaces/blob/master/src/main/java/org/omnifaces/util/Components.java#L203
>
> And then feed
> UIComponent.getCurrentComponent(FacesContext.getCurrentInstance())
> as the first parameter.
>
> This will, starting from the "current component", first look in the
> naming container parents and if it fails to find the component there
> in the children from the view root, working down.
>
> What do you think?
>
> Kind regards,
> Arjan
>
>
> >
> > Regards,
> >
> > Cagatay Civici
> > PrimeFaces Lead
> > PrimeTek Informatics
> > www.primefaces.org
> >
> > On Wednesday 25 February 2015 at 20:32, arjan tijms wrote:
> >
> > Hi Cagatay,
> >
> > Just curious, but any progress to report here?
> >
> > Thanks!
> >
> > Kind regards,
> > Arjan
> >
> > On Monday, January 26, 2015, manfred riem <manfred.riem_at_oracle.com>
> wrote:
> >
> > Hi Cagatay,
> >
> > I have assigned it to you.
> >
> >
> > Thanks!
> > Manfred
> >
> > On 1/26/15, 6:33 AM, Cagatay Civici wrote:
> >
> > Hi,
> >
> > It is cagatay_civici.
> >
> > Regards,
> >
> > Cagatay Civici
> > PrimeFaces Lead
> > PrimeTek Informatics
> > www.primefaces.org
> >
> > On Thursday 22 January 2015 at 20:23, manfred riem wrote:
> >
> > Whoops,
> >
> > Can you let me know what your java.net id is?
> >
> > Thanks!
> > Manfred
> >
> > On 1/22/15, 12:19 PM, manfred riem wrote:
> >
> > Hi Cagatay,
> >
> > Thanks! I'll assign to you it and please let us know how it goes :)
> >
> > Regards,
> > Manfred
> >
> > On 1/22/15, 12:15 PM, Cagatay Civici wrote:
> >
> > Hi,
> >
> > ACTION: Cagatay, may we assign [49-jsClientId] [1] to you? As part of
> > owning this, can you please create a text document of the specification
> > for this. It should probably reside inside the f:tag library. The
> > specification must be clean-room implementable, of course.
> >
> > Sorry for the late reply, it has been a busy week. Sure please assign it
> to
> > me and I’ll work on it.
> >
> > Regards,
> >
> > Cagatay Civici
> > PrimeFaces Lead
> > PrimeTek Informatics
> > www.primefaces.org
> >
> > On Tuesday 20 January 2015 at 21:15, Edward Burns wrote:
> >
> > On Wed, 14 Jan 2015 18:58:16 +0100, Bauke Scholtz <balusc_at_gmail.com>
> said:
> >
> >
> > BalusC> PrimeFaces #{p:component('componentId')} already does exactly
> > BalusC> this:
> > BalusC>
> >
> http://www.primefaces.org/docs/vdl/5.0/core/primefaces-p/component.fn.html
> >
> > On Thu, 15 Jan 2015 08:01:53 -0600, manfred riem <
> manfred.riem_at_oracle.com>
> > said:
> >
> >
> > MR> So Cagatay are you willing and able to push this issue further
> > MR> along?
> >
> > ACTION: Cagatay, may we assign [49-jsClientId] [1] to you? As part of
> > owning this, can you please create a text document of the specification
> > for this. It should probably reside inside the f:tag library. The
> > specification must be clean-room implementable, of course.
> >
> > Ed
> >
> > --
> > | edward.burns_at_oracle.com | office: +1 407 458 0017
> > | 33 days til DevNexus 2015
> > | 43 days til JavaLand 2015
> > | 53 days til CONFESS 2015
> >
> > [1] https://java.net/jira/browse/JAVASERVERFACES_SPEC_PUBLIC-49
> >
> >
> >
> >
> >
> >
> >
> >
>
>
>
>
>
>
>