4:02 p.m.
I can't remember if this topic was ever resolved, but in an EE 6 container,
will it be necessary to include the FacesServlet declaration in web.xml? I
would assume that if the declaration is absent, but WEB-INF/faces-config.xml
present, the declaration would be implicit and the mapping default to *.jsf.
Is this something that could get resolved before EE 6 platform goes final?
Think about it, when was the last time you declared the JspServlet?
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
11:45 p.m.
I would hope we take this in. However, we would probably have to
specify a default-name for the FacesServlet then, so that people can
override.
Also, is the postfix-mapping the way to go right now? With resource
handling, prefix mapping might be easier to pre-configure?
regards,
Martin
On 11/3/09, Dan Allen <dan.j.allen(a)gmail.com> wrote:
I can't remember if this topic was ever resolved, but in an EE 6
container,
will it be necessary to include the FacesServlet declaration in web.xml? I
would assume that if the declaration is absent, but WEB-INF/faces-config.xml
present, the declaration would be implicit and the mapping default to *.jsf.
Is this something that could get resolved before EE 6 platform goes final?
Think about it, when was the last time you declared the JspServlet?
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
--
http://www.irian.at
Your JSF powerhouse -
JSF Consulting, Development and
Courses in English and German
Professional Support for Apache MyFaces
12:05 a.m.
On Wed, Nov 4, 2009 at 12:45 AM, Martin Marinschek
<mmarinschek(a)apache.org>wrote:
I would hope we take this in. However, we would probably have to
specify a default-name for the FacesServlet then, so that people can
override.
Precisely.
Also, is the postfix-mapping the way to go right now? With resource
handling, prefix mapping might be easier to pre-configure?
I'm fine with either one. I just always use suffix mapping out of habit.
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
9:28 a.m.
Our Servlet EG rep informed us in Servlet3 we can use
http://www.ermalaev.spb.ru/javaee/6/docs/api/javax/servlet/ServletContain...
to do this.
Perhaps something for the MR?
On 4 Nov 2009, at 06:05, Dan Allen wrote:
On Wed, Nov 4, 2009 at 12:45 AM, Martin Marinschek
<mmarinschek(a)apache.org
> wrote:
I would hope we take this in. However, we would probably have to
specify a default-name for the FacesServlet then, so that people can
override.
Precisely.
Also, is the postfix-mapping the way to go right now? With resource
handling, prefix mapping might be easier to pre-configure?
I'm fine with either one. I just always use suffix mapping out of
habit.
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
12:38 p.m.
On Wed, Nov 4, 2009 at 10:28 AM, Pete Muir <pmuir(a)redhat.com> wrote:
Our Servlet EG rep informed us in Servlet3 we can use
http://www.ermalaev.spb.ru/javaee/6/docs/api/javax/servlet/ServletContain...
to
do this.
Are you saying this is how Mojarra could register the servlet? Just
clarifying.
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
5:01 p.m.
On 4 Nov 2009, at 18:38, Dan Allen wrote:
On Wed, Nov 4, 2009 at 10:28 AM, Pete Muir <pmuir(a)redhat.com>
wrote:
Our Servlet EG rep informed us in Servlet3 we can use
http://www.ermalaev.spb.ru/javaee/6/docs/api/javax/servlet/ServletContain...
to do this.
Are you saying this is how Mojarra could register the servlet? Just
clarifying.
No, I was saying we could add a line to the JSF spec:
"If the JSF container is used in a Servlet 3 (or newer) environment,
it should register an implementation of ServletContainerInitializer
(using Service Providers from the JAR file specification). The
ServletContainerInitializer should register the FacesServlet, with a
name of Faces Servlet, if an existing servlet is not yet registered
with this name. Furthermore, a suffix mapping of *.jsf [and/or a
prefix mapping of /faces/*] should be added."
7:19 p.m.
"If the JSF container is used in a Servlet 3 (or newer) environment, it
should register an implementation of ServletContainerInitializer (using
Service Providers from the JAR file specification). The
ServletContainerInitializer should register the FacesServlet, with a name of
Faces Servlet, if an existing servlet is not yet registered with this name.
Furthermore, a suffix mapping of *.jsf [and/or a prefix mapping of /faces/*]
should be added."
Well said. +1 from me.
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
12:41 a.m.
+1 as well!
regards,
Martin
On 11/5/09, Dan Allen <dan.j.allen(a)gmail.com> wrote:
>
> "If the JSF container is used in a Servlet 3 (or newer) environment, it
> should register an implementation of ServletContainerInitializer (using
> Service Providers from the JAR file specification). The
> ServletContainerInitializer should register the FacesServlet, with a name
> of
> Faces Servlet, if an existing servlet is not yet registered with this
> name.
> Furthermore, a suffix mapping of *.jsf [and/or a prefix mapping of
> /faces/*]
> should be added."
>
Well said. +1 from me.
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
--
http://www.irian.at
Your JSF powerhouse -
JSF Consulting, Development and
Courses in English and German
Professional Support for Apache MyFaces
12:44 a.m.
+1. Nicely written
Sent from my Verizon Wireless BlackBerry
-----Original Message-----
From: Martin Marinschek <mmarinschek(a)apache.org>
Date: Thu, 5 Nov 2009 07:41:50
To: <jsr-314-open(a)jcp.org>
Subject: Re: [jsr-314-open] FacesServlet declaration in Java EE 6
+1 as well!
regards,
Martin
On 11/5/09, Dan Allen <dan.j.allen(a)gmail.com> wrote:
>
> "If the JSF container is used in a Servlet 3 (or newer) environment, it
> should register an implementation of ServletContainerInitializer (using
> Service Providers from the JAR file specification). The
> ServletContainerInitializer should register the FacesServlet, with a name
> of
> Faces Servlet, if an existing servlet is not yet registered with this
> name.
> Furthermore, a suffix mapping of *.jsf [and/or a prefix mapping of
> /faces/*]
> should be added."
>
Well said. +1 from me.
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
--
http://www.irian.at
Your JSF powerhouse -
JSF Consulting, Development and
Courses in English and German
Professional Support for Apache MyFaces
11:19 a.m.
+1 from me as well, but shouldn't "should" be "must"?
2009/11/4 Dan Allen <dan.j.allen(a)gmail.com>
> "If the JSF container is used in a Servlet 3 (or newer) environment, it
> should register an implementation of ServletContainerInitializer (using
> Service Providers from the JAR file specification). The
> ServletContainerInitializer should register the FacesServlet, with a name of
> Faces Servlet, if an existing servlet is not yet registered with this name.
> Furthermore, a suffix mapping of *.jsf [and/or a prefix mapping of /faces/*]
> should be added."
>
Well said. +1 from me.
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
11:31 a.m.
On Thu, Nov 5, 2009 at 12:19 PM, David Geary <clarity.training(a)gmail.com>wrote:
+1 from me as well, but shouldn't "should" be
"must"?
+1 for me for both mappings, and I agree with David -- "must" is preferred.
2009/11/4 Dan Allen <dan.j.allen(a)gmail.com>
>
>> "If the JSF container is used in a Servlet 3 (or newer) environment, it
>> should register an implementation of ServletContainerInitializer (using
>> Service Providers from the JAR file specification). The
>> ServletContainerInitializer should register the FacesServlet, with a name of
>> Faces Servlet, if an existing servlet is not yet registered with this name.
>> Furthermore, a suffix mapping of *.jsf [and/or a prefix mapping of /faces/*]
>> should be added."
>>
>
> Well said. +1 from me.
>
> -Dan
>
> --
> Dan Allen
> Senior Software Engineer, Red Hat | Author of Seam in Action
> Registered Linux User #231597
>
> http://mojavelinux.com
> http://mojavelinux.com/seaminaction
> http://www.google.com/profiles/dan.j.allen
>
12:16 p.m.
Now that I think more about it, there really is no harm, and plenty of
benefits, in having both mapped, so +1 to both.
-Dan
p.s. In a way it's like having .jsp and .jspx mapped in JspServlet.
On Thu, Nov 5, 2009 at 12:31 PM, Kito Mann <kito.mann(a)virtua.com> wrote:
On Thu, Nov 5, 2009 at 12:19 PM, David Geary <clarity.training(a)gmail.com>wrote:
> +1 from me as well, but shouldn't "should" be "must"?
>
+1 for me for both mappings, and I agree with David -- "must" is preferred.
>
> 2009/11/4 Dan Allen <dan.j.allen(a)gmail.com>
>
>
>>
>>> "If the JSF container is used in a Servlet 3 (or newer) environment, it
>>> should register an implementation of ServletContainerInitializer (using
>>> Service Providers from the JAR file specification). The
>>> ServletContainerInitializer should register the FacesServlet, with a name of
>>> Faces Servlet, if an existing servlet is not yet registered with this name.
>>> Furthermore, a suffix mapping of *.jsf [and/or a prefix mapping of /faces/*]
>>> should be added."
>>>
>>
>> Well said. +1 from me.
>>
>> -Dan
>>
>> --
>> Dan Allen
>> Senior Software Engineer, Red Hat | Author of Seam in Action
>> Registered Linux User #231597
>>
>> http://mojavelinux.com
>> http://mojavelinux.com/seaminaction
>> http://www.google.com/profiles/dan.j.allen
>>
>
>
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
5:15 p.m.
+1 for both mappings.
Ted.
On 2009-11-05, at 11:16 AM, Dan Allen wrote:
Now that I think more about it, there really is no harm, and plenty
of benefits, in having both mapped, so +1 to both.
-Dan
p.s. In a way it's like having .jsp and .jspx mapped in JspServlet.
On Thu, Nov 5, 2009 at 12:31 PM, Kito Mann <kito.mann(a)virtua.com>
wrote:
On Thu, Nov 5, 2009 at 12:19 PM, David Geary <clarity.training(a)gmail.com
> wrote:
+1 from me as well, but shouldn't "should" be "must"?
+1 for me for both mappings, and I agree with David -- "must" is
preferred.
2009/11/4 Dan Allen <dan.j.allen(a)gmail.com>
"If the JSF container is used in a Servlet 3 (or newer) environment,
it should register an implementation of ServletContainerInitializer
(using Service Providers from the JAR file specification). The
ServletContainerInitializer should register the FacesServlet, with a
name of Faces Servlet, if an existing servlet is not yet registered
with this name. Furthermore, a suffix mapping of *.jsf [and/or a
prefix mapping of /faces/*] should be added."
Well said. +1 from me.
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
1:18 p.m.
I'd like to resurrect this thread, since we got +1 votes from just about
everyone who replies on a a regular basis.
"If the JSF container is used in a Servlet 3 (or newer)
environment, it
should register an implementation of ServletContainerInitializer (using
Service Providers from the JAR file specification). The
ServletContainerInitializer should register the FacesServlet, with a name of
Faces Servlet, if an existing servlet is not yet registered with this name.
Furthermore, a suffix mapping of *.jsf and a prefix mapping of /faces/*
should be added."
Is this something that can be added to the Accepted Changes for the JSF MR
(and is it possible to require this behavior from the JSF spec, or must it
be in the platform spec)?
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
12:37 p.m.
On Wed, Nov 4, 2009 at 1:05 AM, Dan Allen <dan.j.allen(a)gmail.com> wrote:
On Wed, Nov 4, 2009 at 12:45 AM, Martin Marinschek
<mmarinschek(a)apache.org
> wrote:
> I would hope we take this in. However, we would probably have to
> specify a default-name for the FacesServlet then, so that people can
> override.
>
Precisely.
From what I can tell, the JSP spec does not specify a standard servlet
name
for the JspServlet. It just talks about *.jsp being the default extension
(or *.jspx if <isxml> is true). Catalina names this servlet "jsp". There
are
proprietary overrides to configure the servlet in the <jsp-config> element
of course.
I think we should go with the name "Faces Servlet" since that appears to be
the most common reference. If the space is a problem, we can drop it. Here
are alternatives:
Faces Servlet
FacesServlet
faces
Faces
jsf
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
1:41 a.m.
On 11/03/2009 10:05 PM, Dan Allen wrote:
On Wed, Nov 4, 2009 at 12:45 AM, Martin Marinschek
<mmarinschek(a)apache.org <mailto:mmarinschek@apache.org>> wrote:
I would hope we take this in. However, we would probably have to
specify a default-name for the FacesServlet then, so that people can
override.
Precisely.
Also, is the postfix-mapping the way to go right now? With resource
handling, prefix mapping might be easier to pre-configure?
I'm fine with either one. I just always use suffix mapping out of habit.
Prefix
mapping would be even better. Framework can believe on prefix to
load resources that resolves relative references form styles and
scripts, but user still can append its own prefixes and suffixes if he
wants.
The only problem is portlet environment where Faces Servlet is not
necessary and it even had better to not include servlet configuration at
all to prevent direct access to JSF pages.
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
11:40 p.m.
Btw, this is already in GlassFish V3.
http://weblogs.java.net/blog/cayhorstmann/archive/2009/10/19/another-smal...
..JBoss AS 5.2 will most likely have it as well. We just need to get it
inked into the specification (and/or platform) at this point.
-Dan
On Fri, Nov 6, 2009 at 2:41 AM, Alaxander Smirnov <asmirnov(a)exadel.com>wrote:
On 11/03/2009 10:05 PM, Dan Allen wrote:
On Wed, Nov 4, 2009 at 12:45 AM, Martin Marinschek <mmarinschek(a)apache.org
> wrote:
> I would hope we take this in. However, we would probably have to
> specify a default-name for the FacesServlet then, so that people can
> override.
>
Precisely.
>
> Also, is the postfix-mapping the way to go right now? With resource
> handling, prefix mapping might be easier to pre-configure?
>
I'm fine with either one. I just always use suffix mapping out of habit.
Prefix mapping would be even better. Framework can believe on prefix to
load resources that resolves relative references form styles and scripts,
but user still can append its own prefixes and suffixes if he wants.
The only problem is portlet environment where Faces Servlet is not
necessary and it even had better to not include servlet configuration at all
to prevent direct access to JSF pages.
-Dan
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen
5497
days inactive
5536
days old
jsr-314-open-mirror@lists.jboss.org
16 comments
8 participants
participants (8)
-
Alaxander Smirnov -
Dan Allen -
David Geary -
Kito Mann -
lincolnbaxter＠gmail.com -
Martin Marinschek -
Pete Muir -
Ted Goddard