Hi, if it is part of the tlddocs, it is considered to be part of the
spec. Hence you are right. Can you open a JIRA issue with this? (and
even better, provide a patch if you have the time?).
Thank you!
Bruno
On 18/01/2008, Philip Newton <[EMAIL PROTECTED]> wrote:
> Are there plans for MyFaces' implementation of the onclick attribute
> of h:commandLink to change?
>
> Currently, the behaviour seems to be to render onclick code as <a
> onclick="USERCODE; JSFCODE;" href="#">, which doesn't interact well
> with user onclick attributes such as "return validate()" or "return
> confirm('Are you sure?')".
>
> Is this behaviour compatible with the spec?
>
> http://java.sun.com/javaee/javaserverfaces/1.2/docs/tlddocs/h/commandLink.html
> says that "If an "onclick" attribute was specified by the user, render
> this JavaScript in a function, and render the user's JavaScript in a
> function. Render both functions in a choice function [...] such that
> if the user's JavaScript returns true, then the rendered JavaScript
> will also execute."
>
> I'm not sure whether that page documents Sun's implementation of JSF
> 1.2 or whether it documents the JSF 1.2 spec (i.e. whether the
> behaviour is something that users may rely on with any JSF 1.2
> compatible implementation); the page title is "commandLink
> (jsf-ri-1.2)" which implies that it documents the RI, but that page is
> part of the frameset
> http://java.sun.com/javaee/javaserverfaces/1.2/docs/tlddocs/index.html
> , which is linked to from
> http://java.sun.com/javaee/javaserverfaces/reference/api/index.html
> under the "JavaServer Faces Technology 1.2 Reference Documentation"
> section.
>
> Apparently, this change was made in the RI response to
> https://javaserverfaces.dev.java.net/issues/show_bug.cgi?id=71 .
>
> So, I suppose I have two questions. 1) Is the bit I quoted part of the
> JSF 1.2 specification for h:commandLink? 2) If so, does MyFaces plan
> to implement that portion of the specification at some point?
>
> (And if that's not part of the spec but only part of the JSF 1.2 spec
> - does someone know whether it's documented anywhere what the effect
> of an "onclick" attribute with an h:commandLink does, or whether every
> implementation is free to pick its own behaviour?)
>
> I've seen a couple of workarounds (including an explicit onclick="if
> (! myfunc()) return false" rather than an unconditional
> onclick="return myfunc()", and using onsubmit instead of onclick), but
> after having read that portion of the documentation of h:commandLink,
> I thought I could rely on the behaviour I had read there.
>
> Cheers,
> --
> Philip Newton <[EMAIL PROTECTED]>
>
