jsr372-experts@javaserverfaces-spec-public.java.net

[jsr372-experts] Re: Report Issues of JSF 2.3 Public Review

From: Bauke Scholtz <balusc_at_gmail.com>
Date: Tue, 31 Jan 2017 22:24:13 +0100

Hi,

> I suspect Bauke did that for a reason, but I would like to ask him to
make it be the same as Ajax, if possible. Bauke?

Not a technical reason really. It was just a coding style which advocates
encapsulation. It's possible to revert it to global style. I only wonder if
I should create a spec or impl issue on this.

> Bauke, can we remove this? I can factually state it is not used anywhere.

Indeed, it was a leftover of the initial proof of concept. It can safely be
removed.

Cheers, B


On Tue, Jan 31, 2017 at 9:59 PM, Edward Burns <edward.burns_at_oracle.com>
wrote:

> >>>>> On Tue, 24 Jan 2017 19:26:04 -0500, Leonardo Uribe <
> leonardo.uribe_at_irian.at> said:
>
> LU> I have found two issues on the javascript documentation:
>
> LU> - javascript doc of jsf.push.init(...), jsf.push.open() and
> LU> jsf.push.close() adds "self". Ex: jsf.push.self.init(...). This is
> LU> clear a bug in the documentation generation, not something related
> LU> to the API.
>
> This is more than just an API documentation artifact. The way the Push
> functions are exposed uses a different JavaScript idiom than the way the
> Ajax functions are exposed. I suspect Bauke did that for a reason, but
> I would like to ask him to make it be the same as Ajax, if possible.
> Bauke?
>
> LU> - There is a variable here called jsf.contextpath . Javascript files
> must
> LU> be static, no contextual information must be sent there because that
> cause
> LU> problems on resource caching (remember JSF jars can be shared among
> LU> applications in the same container). I guess this was added on early
> LU> implementations of f:websocket, but it MUST be removed. If you need to
> LU> pass contextual information to the page, there are other ways like an
> LU> inline script, or a query parameter.
>
> Yes, I agree. However, given that this is Bauke's stuff I'll defer to
> him. Bauke, can we remove this? I can factually state it is not used
> anywhere.
>
> Ed
>
>
> --
> | edward.burns_at_oracle.com | office: +1 407 458 0017
> | 25 business days until planned start of JSF 2.3 Final Approval Ballot
> | 15 business days until DevNexus 2017
> | 40 business days until JavaLand 2017
>