[Geoserver-devel] Extension point for GetCapabilities links in home page

Hi all,

As per http://jira.codehaus.org/browse/GEOS-4162, I worked out a patch
that incorporates an extension point for contributing to the
getcapabilities links in the home page.
You can see the diff at
<https://github.com/groldan/geoserver_trunk/compare/master...getcaps_ui_integration&gt;

There's a default implementation of this extension point that grabs the
available ServiceInfo objects and creates the standard ows getcaps links
as GeoServerHomePage used to do, and an implementation in the GWC module
that contributes the links for the WMS-C, WMTS and TMS services.

If there's no oppositions I'd like to merge this into trunk by this
week.

Best regards,
Gabriel

--
Gabriel Roldan
groldan@anonymised.com
Expert service straight from the developers

On Sun, Jan 16, 2011 at 8:56 AM, Gabriel Roldán <groldan@anonymised.com> wrote:

Hi all,

As per http://jira.codehaus.org/browse/GEOS-4162, I worked out a patch
that incorporates an extension point for contributing to the
getcapabilities links in the home page.
You can see the diff at
<https://github.com/groldan/geoserver_trunk/compare/master...getcaps_ui_integration&gt;

There's a default implementation of this extension point that grabs the
available ServiceInfo objects and creates the standard ows getcaps links
as GeoServerHomePage used to do, and an implementation in the GWC module
that contributes the links for the WMS-C, WMTS and TMS services.

If there's no oppositions I'd like to merge this into trunk by this
week.

Sound good to me. Just a question, why passing around the GeoServerApplication
class?
That is made available as a singleton to all pages
by using GeoServerApplication.get().
Using a model indirection seems un-necessary extra code to me.

Cheers
Andrea

--
Ing. Andrea Aime
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy

phone: +39 0584962313
fax: +39 0584962313

http://www.geo-solutions.it
http://geo-solutions.blogspot.com/
http://www.linkedin.com/in/andreaaime
http://twitter.com/geowolf

-----------------------------------------------------

On Sun, 2011-01-16 at 10:53 +0100, Andrea Aime wrote:

On Sun, Jan 16, 2011 at 8:56 AM, Gabriel Roldán <groldan@anonymised.com> wrote:
> Hi all,
>
> As per http://jira.codehaus.org/browse/GEOS-4162, I worked out a patch
> that incorporates an extension point for contributing to the
> getcapabilities links in the home page.
> You can see the diff at
> <https://github.com/groldan/geoserver_trunk/compare/master...getcaps_ui_integration&gt;
>
> There's a default implementation of this extension point that grabs the
> available ServiceInfo objects and creates the standard ows getcaps links
> as GeoServerHomePage used to do, and an implementation in the GWC module
> that contributes the links for the WMS-C, WMTS and TMS services.
>
> If there's no oppositions I'd like to merge this into trunk by this
> week.

Sound good to me. Just a question, why passing around the GeoServerApplication
class?
That is made available as a singleton to all pages
by using GeoServerApplication.get().
Using a model indirection seems un-necessary extra code to me.

Good catch, I removed it as argument and accessed as singleton.
Committed now. Thanks for the feedback.

Gabriel.

Cheers
Andrea

--
Gabriel Roldan
groldan@anonymised.com
Expert service straight from the developers

Is trunk actually a real trunk yet? Looking at svn it looks like there’s a RC1 tag, but not a 2.1.x branch yet. I think this is a great patch for trunk/2.2.x, to be backported for 2.1.0, but I was hoping RC1 would be a real Release Candidate - become 2.1.0 if there were no major bugs.

On Sun, Jan 16, 2011 at 5:32 PM, Gabriel Roldán <groldan@anonymised.com> wrote:

On Sun, 2011-01-16 at 10:53 +0100, Andrea Aime wrote:

On Sun, Jan 16, 2011 at 8:56 AM, Gabriel Roldán <groldan@anonymised.com> wrote:

Hi all,

As per http://jira.codehaus.org/browse/GEOS-4162, I worked out a patch
that incorporates an extension point for contributing to the
getcapabilities links in the home page.
You can see the diff at
<https://github.com/groldan/geoserver_trunk/compare/master…getcaps_ui_integration>

There’s a default implementation of this extension point that grabs the
available ServiceInfo objects and creates the standard ows getcaps links
as GeoServerHomePage used to do, and an implementation in the GWC module
that contributes the links for the WMS-C, WMTS and TMS services.

If there’s no oppositions I’d like to merge this into trunk by this
week.

Sound good to me. Just a question, why passing around the GeoServerApplication
class?
That is made available as a singleton to all pages
by using GeoServerApplication.get().
Using a model indirection seems un-necessary extra code to me.

Good catch, I removed it as argument and accessed as singleton.
Committed now. Thanks for the feedback.

Gabriel.

Cheers
Andrea


Gabriel Roldan
groldan@anonymised.com
Expert service straight from the developers


Protect Your Site and Customers from Malware Attacks
Learn about various malware tactics and how to avoid them. Understand
malware threats, the impact they can have on your business, and how you
can protect your company and customers by using code signing.
http://p.sf.net/sfu/oracle-sfdevnl


Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel

Good question, I'm not sure though.
I checked if the pom version was updated before committed and since it
was not I just assumed it was ok. Let's see what Justin says.

Gabriel

On Sun, 2011-01-16 at 22:12 -0500, Chris Holmes wrote:

Is trunk actually a real trunk yet? Looking at svn it looks like
there's a RC1 tag, but not a 2.1.x branch yet. I think this is a
great patch for trunk/2.2.x, to be backported for 2.1.0, but I was
hoping RC1 would be a real Release Candidate - become 2.1.0 if there
were no major bugs.

On Sun, Jan 16, 2011 at 5:32 PM, Gabriel Roldán <groldan@anonymised.com>
wrote:
        On Sun, 2011-01-16 at 10:53 +0100, Andrea Aime wrote:
        > On Sun, Jan 16, 2011 at 8:56 AM, Gabriel Roldán
        <groldan@anonymised.com> wrote:
        > > Hi all,
        > >
        > > As per http://jira.codehaus.org/browse/GEOS-4162, I worked
        out a patch
        > > that incorporates an extension point for contributing to
        the
        > > getcapabilities links in the home page.
        > > You can see the diff at
        > >
        <https://github.com/groldan/geoserver_trunk/compare/master...getcaps_ui_integration&gt;
        > >
        > > There's a default implementation of this extension point
        that grabs the
        > > available ServiceInfo objects and creates the standard ows
        getcaps links
        > > as GeoServerHomePage used to do, and an implementation in
        the GWC module
        > > that contributes the links for the WMS-C, WMTS and TMS
        services.
        > >
        > > If there's no oppositions I'd like to merge this into
        trunk by this
        > > week.
        >
        > Sound good to me. Just a question, why passing around the
        GeoServerApplication
        > class?
        > That is made available as a singleton to all pages
        > by using GeoServerApplication.get().
        > Using a model indirection seems un-necessary extra code to
        me.
        
        Good catch, I removed it as argument and accessed as
        singleton.
        Committed now. Thanks for the feedback.
        
        Gabriel.
        >
        > Cheers
        > Andrea
        >
        
        --
        Gabriel Roldan
        groldan@anonymised.com
        Expert service straight from the developers
        
        ------------------------------------------------------------------------------
        
        Protect Your Site and Customers from Malware Attacks
        Learn about various malware tactics and how to avoid them.
        Understand
        malware threats, the impact they can have on your business,
        and how you
        can protect your company and customers by using code signing.
        http://p.sf.net/sfu/oracle-sfdevnl
        _______________________________________________
        Geoserver-devel mailing list
        Geoserver-devel@lists.sourceforge.net
        https://lists.sourceforge.net/lists/listinfo/geoserver-devel
        
--
Gabriel Roldan
groldan@anonymised.com
Expert service straight from the developers

I have not created the 2.1.x branch, i was waiting for some more general consensus.

If people can agree it should be created now i can create the branch. Two options:

(a) grab trunk as of now and branch it
(b) grab trunk as of the revision for 2.1-RC1 and create the branch from it

It sounds like Chris would prefer (b). Last week we more or less agreed that “stable development” meant bug fixes and modest features generally considered low risk. This would tend to jive with what we have done in the past as well on the stable branch although at times we have been pretty liberal with the term stable development.

We also tend to go through a number of release candidates before the official .0 release. Not sure we have every done a single RC and then called it good if no issues come up. Not against it by any means, it just never really turned out that way.

Anyways, like I said something I would like others to comment on. I would be good with branching 2.1.x from current trunk since nothing to core and considered risk has taken place. And then just keep with our definition of stable for what gets backported to 2.1.x branch.

On Sun, Jan 16, 2011 at 10:04 PM, Gabriel Roldán <groldan@anonymised.com> wrote:

Good question, I’m not sure though.
I checked if the pom version was updated before committed and since it
was not I just assumed it was ok. Let’s see what Justin says.

Gabriel

On Sun, 2011-01-16 at 22:12 -0500, Chris Holmes wrote:

Is trunk actually a real trunk yet? Looking at svn it looks like
there’s a RC1 tag, but not a 2.1.x branch yet. I think this is a
great patch for trunk/2.2.x, to be backported for 2.1.0, but I was
hoping RC1 would be a real Release Candidate - become 2.1.0 if there
were no major bugs.

On Sun, Jan 16, 2011 at 5:32 PM, Gabriel Roldán <groldan@anonymised.com>
wrote:
On Sun, 2011-01-16 at 10:53 +0100, Andrea Aime wrote:

On Sun, Jan 16, 2011 at 8:56 AM, Gabriel Roldán
<groldan@anonymised.com501…> wrote:

Hi all,

As per http://jira.codehaus.org/browse/GEOS-4162, I worked
out a patch
that incorporates an extension point for contributing to
the
getcapabilities links in the home page.
You can see the diff at

<https://github.com/groldan/geoserver_trunk/compare/master…getcaps_ui_integration>

There’s a default implementation of this extension point
that grabs the
available ServiceInfo objects and creates the standard ows
getcaps links
as GeoServerHomePage used to do, and an implementation in
the GWC module
that contributes the links for the WMS-C, WMTS and TMS
services.

If there’s no oppositions I’d like to merge this into
trunk by this
week.

Sound good to me. Just a question, why passing around the
GeoServerApplication
class?
That is made available as a singleton to all pages
by using GeoServerApplication.get().
Using a model indirection seems un-necessary extra code to
me.

Good catch, I removed it as argument and accessed as
singleton.
Committed now. Thanks for the feedback.

Gabriel.

Cheers
Andrea


Gabriel Roldan
groldan@anonymised.com…
Expert service straight from the developers


Protect Your Site and Customers from Malware Attacks
Learn about various malware tactics and how to avoid them.
Understand
malware threats, the impact they can have on your business,
and how you
can protect your company and customers by using code signing.
http://p.sf.net/sfu/oracle-sfdevnl


Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel

Gabriel Roldan
groldan@anonymised.com
Expert service straight from the developers


Protect Your Site and Customers from Malware Attacks
Learn about various malware tactics and how to avoid them. Understand
malware threats, the impact they can have on your business, and how you
can protect your company and customers by using code signing.
http://p.sf.net/sfu/oracle-sfdevnl


Geoserver-devel mailing list
Geoserver-devel@anonymised.comsts.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel


Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.

I think the one thing why I would go for branching off the 2.1-RC1
revision is the geotools version. So that only trunk keeps on gt-trunk
and the branch on gt-2.7-beta1.

Then I can backport the front page capabilities stuff as it's low risk,
but looks cleaner to branch as of the tagged release version?

Cheers,
Gabriel
On Mon, 2011-01-17 at 08:34 -0700, Justin Deoliveira wrote:

I have not created the 2.1.x branch, i was waiting for some more
general consensus.

If people can agree it should be created now i can create the branch.
Two options:

(a) grab trunk as of now and branch it
(b) grab trunk as of the revision for 2.1-RC1 and create the branch
from it

It sounds like Chris would prefer (b). Last week we more or less
agreed that "stable development" meant bug fixes and modest features
generally considered low risk. This would tend to jive with what we
have done in the past as well on the stable branch although at times
we have been pretty liberal with the term stable development.

We also tend to go through a number of release candidates before the
official .0 release. Not sure we have every done a single RC and then
called it good if no issues come up. Not against it by any means, it
just never really turned out that way.

Anyways, like I said something I would like others to comment on. I
would be good with branching 2.1.x from current trunk since nothing to
core and considered risk has taken place. And then just keep with our
definition of stable for what gets backported to 2.1.x branch.

On Sun, Jan 16, 2011 at 10:04 PM, Gabriel Roldán <groldan@anonymised.com>
wrote:
        Good question, I'm not sure though.
        I checked if the pom version was updated before committed and
        since it
        was not I just assumed it was ok. Let's see what Justin says.
        
        Gabriel
        
        On Sun, 2011-01-16 at 22:12 -0500, Chris Holmes wrote:
        > Is trunk actually a real trunk yet? Looking at svn it looks
        like
        > there's a RC1 tag, but not a 2.1.x branch yet. I think this
        is a
        > great patch for trunk/2.2.x, to be backported for 2.1.0, but
        I was
        > hoping RC1 would be a real Release Candidate - become 2.1.0
        if there
        > were no major bugs.
        >
        > On Sun, Jan 16, 2011 at 5:32 PM, Gabriel Roldán
        <groldan@anonymised.com>
        > wrote:
        > On Sun, 2011-01-16 at 10:53 +0100, Andrea Aime
        wrote:
        > > On Sun, Jan 16, 2011 at 8:56 AM, Gabriel Roldán
        > <groldan@anonymised.com> wrote:
        > > > Hi all,
        > > >
        > > > As per
        http://jira.codehaus.org/browse/GEOS-4162, I worked
        > out a patch
        > > > that incorporates an extension point for
        contributing to
        > the
        > > > getcapabilities links in the home page.
        > > > You can see the diff at
        > > >
        >
        <https://github.com/groldan/geoserver_trunk/compare/master...getcaps_ui_integration&gt;
        > > >
        > > > There's a default implementation of this
        extension point
        > that grabs the
        > > > available ServiceInfo objects and creates the
        standard ows
        > getcaps links
        > > > as GeoServerHomePage used to do, and an
        implementation in
        > the GWC module
        > > > that contributes the links for the WMS-C, WMTS
        and TMS
        > services.
        > > >
        > > > If there's no oppositions I'd like to merge this
        into
        > trunk by this
        > > > week.
        > >
        > > Sound good to me. Just a question, why passing
        around the
        > GeoServerApplication
        > > class?
        > > That is made available as a singleton to all pages
        > > by using GeoServerApplication.get().
        > > Using a model indirection seems un-necessary extra
        code to
        > me.
        >
        >
        > Good catch, I removed it as argument and accessed as
        > singleton.
        > Committed now. Thanks for the feedback.
        >
        > Gabriel.
        > >
        > > Cheers
        > > Andrea
        > >
        >
        > --
        > Gabriel Roldan
        > groldan@anonymised.com
        > Expert service straight from the developers
        >
        >
        >
        ------------------------------------------------------------------------------
        >
        >
        > Protect Your Site and Customers from Malware Attacks
        > Learn about various malware tactics and how to avoid
        them.
        > Understand
        > malware threats, the impact they can have on your
        business,
        > and how you
        > can protect your company and customers by using code
        signing.
        > http://p.sf.net/sfu/oracle-sfdevnl
        > _______________________________________________
        > Geoserver-devel mailing list
        > Geoserver-devel@lists.sourceforge.net
        >
        https://lists.sourceforge.net/lists/listinfo/geoserver-devel
        >
        >
        
        --
        
        Gabriel Roldan
        groldan@anonymised.com
        Expert service straight from the developers
        
        ------------------------------------------------------------------------------
        Protect Your Site and Customers from Malware Attacks
        Learn about various malware tactics and how to avoid them.
        Understand
        malware threats, the impact they can have on your business,
        and how you
        can protect your company and customers by using code signing.
        http://p.sf.net/sfu/oracle-sfdevnl
        _______________________________________________
        Geoserver-devel mailing list
        Geoserver-devel@lists.sourceforge.net
        https://lists.sourceforge.net/lists/listinfo/geoserver-devel
        
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.

--
Gabriel Roldan
groldan@anonymised.com
Expert service straight from the developers

On Mon, Jan 17, 2011 at 4:34 PM, Justin Deoliveira <jdeolive@anonymised.com> wrote:

I have not created the 2.1.x branch, i was waiting for some more
general consensus.
If people can agree it should be created now i can create the branch. Two
options:
(a) grab trunk as of now and branch it
(b) grab trunk as of the revision for 2.1-RC1 and create the branch from it
It sounds like Chris would prefer (b). Last week we more or less agreed that
"stable development" meant bug fixes and modest features generally
considered low risk. This would tend to jive with what we have done in the
past as well on the stable branch although at times we have been pretty
liberal with the term stable development.
We also tend to go through a number of release candidates before the
official .0 release. Not sure we have every done a single RC and then called
it good if no issues come up. Not against it by any means, it just never
really turned out that way.
Anyways, like I said something I would like others to comment on. I would be
good with branching 2.1.x from current trunk since nothing to core and
considered risk has taken place. And then just keep with our definition of
stable for what gets backported to 2.1.x branch.

I would go for b) (branch from the RC tag) and then backport bug fixes
and innocuous (think extensions, GUI tweaks) features back.
And yes, just release the last good RC as stable without changes, that's a good
practice that I'd like to see used in GS releases :slight_smile:

I'm wondering though, what about GeoTools? GS is just as locked down as GT
is, makes little sense to be very adamant on GS with two branches and all and
then allow whatever change to get into GT2 (not saying we have to
freeze everything,
just saying the policy in the two should be consistent, two branches
and discussion
before backports on both sides)

Cheers
Andrea

--
Ing. Andrea Aime
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy

phone: +39 0584962313
fax: +39 0584962313

http://www.geo-solutions.it
http://geo-solutions.blogspot.com/
http://www.linkedin.com/in/andreaaime
http://twitter.com/geowolf

-----------------------------------------------------

Right. As geotools approaches 2.7-RC1 it would be good to branch off as well, but a separate discussion and process so there may be a bit of time where we are on branch and have to stick to geotools trunk. Hopefully not too long though.

So I will branch from RC1 and anyone who has committed changes since then please backport as you see fit. In the meantime I will start the discussion going about branching gt to 2.7.x on the upcoming RC1 release.

On Mon, Jan 17, 2011 at 9:40 AM, Andrea Aime <andrea.aime@anonymised.com> wrote:

On Mon, Jan 17, 2011 at 4:34 PM, Justin Deoliveira <jdeolive@anonymised.com> wrote:

I have not created the 2.1.x branch, i was waiting for some more
general consensus.
If people can agree it should be created now i can create the branch. Two
options:
(a) grab trunk as of now and branch it
(b) grab trunk as of the revision for 2.1-RC1 and create the branch from it
It sounds like Chris would prefer (b). Last week we more or less agreed that
“stable development” meant bug fixes and modest features generally
considered low risk. This would tend to jive with what we have done in the
past as well on the stable branch although at times we have been pretty
liberal with the term stable development.
We also tend to go through a number of release candidates before the
official .0 release. Not sure we have every done a single RC and then called
it good if no issues come up. Not against it by any means, it just never
really turned out that way.
Anyways, like I said something I would like others to comment on. I would be
good with branching 2.1.x from current trunk since nothing to core and
considered risk has taken place. And then just keep with our definition of
stable for what gets backported to 2.1.x branch.

I would go for b) (branch from the RC tag) and then backport bug fixes
and innocuous (think extensions, GUI tweaks) features back.
And yes, just release the last good RC as stable without changes, that’s a good
practice that I’d like to see used in GS releases :slight_smile:

I’m wondering though, what about GeoTools? GS is just as locked down as GT
is, makes little sense to be very adamant on GS with two branches and all and
then allow whatever change to get into GT2 (not saying we have to
freeze everything,
just saying the policy in the two should be consistent, two branches
and discussion
before backports on both sides)

Cheers
Andrea


Ing. Andrea Aime
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy

phone: +39 0584962313
fax: +39 0584962313

http://www.geo-solutions.it
http://geo-solutions.blogspot.com/
http://www.linkedin.com/in/andreaaime
http://twitter.com/geowolf



Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.