[Geoserver-devel] Last minor feature for 2.4.x: exposing coverage bands in the GUI

GeoServer GeoServer Developer
1

Hi,
as a last minute addition to GeoServer we would like to add in the layer configuration GUI the list of bands that do compose a raster layer, with their name, range, and eventual null values, pretty much in a similar way to how feature attributes are exposed for feature types.

In addition to that, we would like to add a new field to CoverageDimensionInfo, uom, which would be the unit of measure of the band, and that one alone, would be editable in the GUI.

The reason for this is that many single banded coverages are representing various physical phenomenons, but the geotiff reader, to make an example, does not know about their nature, so it always ends up setting “radiance” as the unit of measure, which is in turn exposed via WCS DescribeCoverage (at least in WCS 2.0) and, as said, it ends up often being wrong on images that are not the usual RGB ones.

Soo… opinions?

Cheers
Andrea

–

==
Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information.

Ing. Andrea Aime

@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 1660272
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it

2

Fine with me, sounds like a good change.

···

On Mon, Jul 15, 2013 at 11:15 AM, Andrea Aime <andrea.aime@anonymised.com> wrote:

Hi,
as a last minute addition to GeoServer we would like to add in the layer configuration GUI the list of bands that do compose a raster layer, with their name, range, and eventual null values, pretty much in a similar way to how feature attributes are exposed for feature types.

In addition to that, we would like to add a new field to CoverageDimensionInfo, uom, which would be the unit of measure of the band, and that one alone, would be editable in the GUI.

The reason for this is that many single banded coverages are representing various physical phenomenons, but the geotiff reader, to make an example, does not know about their nature, so it always ends up setting “radiance” as the unit of measure, which is in turn exposed via WCS DescribeCoverage (at least in WCS 2.0) and, as said, it ends up often being wrong on images that are not the usual RGB ones.

Soo… opinions?

Cheers
Andrea

–

==
Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information.

Ing. Andrea Aime

@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 1660272
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it

See everything from the browser to the database with AppDynamics
Get end-to-end visibility with application monitoring from AppDynamics
Isolate bottlenecks and diagnose root cause in seconds.
Start your free trial of AppDynamics Pro today!
http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk

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.

3

+1, excellent feature!

···

On Mon, Jul 15, 2013 at 11:15 AM, Andrea Aime <andrea.aime@anonymised.com> wrote:

Hi,
as a last minute addition to GeoServer we would like to add in the layer configuration GUI the list of bands that do compose a raster layer, with their name, range, and eventual null values, pretty much in a similar way to how feature attributes are exposed for feature types.

In addition to that, we would like to add a new field to CoverageDimensionInfo, uom, which would be the unit of measure of the band, and that one alone, would be editable in the GUI.

The reason for this is that many single banded coverages are representing various physical phenomenons, but the geotiff reader, to make an example, does not know about their nature, so it always ends up setting “radiance” as the unit of measure, which is in turn exposed via WCS DescribeCoverage (at least in WCS 2.0) and, as said, it ends up often being wrong on images that are not the usual RGB ones.

Soo… opinions?

Cheers
Andrea

–

==
Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information.

Ing. Andrea Aime

@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 1660272
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it

See everything from the browser to the database with AppDynamics
Get end-to-end visibility with application monitoring from AppDynamics
Isolate bottlenecks and diagnose root cause in seconds.
Start your free trial of AppDynamics Pro today!
http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk

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.

4

On Mon, Jul 15, 2013 at 7:15 PM, Andrea Aime
<andrea.aime@anonymised.com>wrote:

Hi,
as a last minute addition to GeoServer we would like to add in the layer
configuration GUI the list of bands that do compose a raster layer, with
their name, range, and eventual null values, pretty much in a similar way
to how feature attributes are exposed for feature types.

And here is how it looks, if you open an already configured single banded
layer (I've also added the data type information, it can be quite useful to
understand rendering issues):

[image: Inline image 1]

See that "reload band definitions" button? It forces GS to go down into the
reader and fetch the band informations again.
In case of newly configured layers, the data type will show up
automatically.

Here is a rain layer, reloaded and with the unit of measure manually filled
in:

[image: Inline image 2]

And here is a RGB layer instead, sporting the default uom for rgb bands,
which is the radiance unit of measure in SI
terms (http://en.wikipedia.org/wiki/Radiance):

[image: Inline image 3]

Now... looking at the GUI one would be very tempted to make the band name
and null values editable... which is doable,
but not immediate, we'd have to wrap the readers and getting out of
resource pool so that the reader returns
GridCoverage objects that have the desired metadata. It's doable, not
extremely hard, but I'll leave that for another day :wink:

Cheers
Andrea

--

Our support, Your Success! Visit http://opensdi.geo-solutions.it for more
information.

Ing. Andrea Aime
@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 1660272
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it

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

5

I am afraid you images did not come through, if you can advise when this has been added to master I will wait for it.
The GeoTools release has run into a couple of snags (my own fault), and we have some leeway while we wait for GWC to be released.

And +1 for the functionality.

–
Jody Garnett

On Wednesday, 17 July 2013 at 11:22 PM, Andrea Aime wrote:

On Mon, Jul 15, 2013 at 7:15 PM, Andrea Aime <andrea.aime@anonymised.com> wrote:

Hi,
as a last minute addition to GeoServer we would like to add in the layer configuration GUI the list of bands that do compose a raster layer, with their name, range, and eventual null values, pretty much in a similar way to how feature attributes are exposed for feature types.

And here is how it looks, if you open an already configured single banded layer (I’ve also added the data type information, it can be quite useful to understand rendering issues):

Inline image 1

See that “reload band definitions” button? It forces GS to go down into the reader and fetch the band informations again.
In case of newly configured layers, the data type will show up automatically.

Here is a rain layer, reloaded and with the unit of measure manually filled in:

Inline image 2

And here is a RGB layer instead, sporting the default uom for rgb bands, which is the radiance unit of measure in SI
terms (http://en.wikipedia.org/wiki/Radiance):

Inline image 3

Now… looking at the GUI one would be very tempted to make the band name and null values editable… which is doable,
but not immediate, we’d have to wrap the readers and getting out of resource pool so that the reader returns
GridCoverage objects that have the desired metadata. It’s doable, not extremely hard, but I’ll leave that for another day :wink:

Cheers
Andrea

–

==
Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information.

Ing. Andrea Aime

@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 1660272
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it

See everything from the browser to the database with AppDynamics
Get end-to-end visibility with application monitoring from AppDynamics
Isolate bottlenecks and diagnose root cause in seconds.
Start your free trial of AppDynamics Pro today!
http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk

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

6

On Thu, Jul 18, 2013 at 1:50 AM, Jody Garnett <jody.garnett@anonymised.com>wrote:

I am afraid you images did not come through, if you can advise when this
has been added to master I will wait for it.

Gah... not sure why? I've added them as attachments this time. But the
commit is in master already

Cheers
Andrae

--

Our support, Your Success! Visit http://opensdi.geo-solutions.it for more
information.

Ing. Andrea Aime
@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 1660272
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it

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

(attachments)

customized-nq8.png
orignal-nq8.png
rbg-nq8.png

7

Thanks Andrea, Kevin has made a GWC 1.5.0-beta so I think I am ready to go.

–
Jody Garnett

On Thursday, 18 July 2013 at 4:00 PM, Andrea Aime wrote:

On Thu, Jul 18, 2013 at 1:50 AM, Jody Garnett <jody.garnett@anonymised.com> wrote:

I am afraid you images did not come through, if you can advise when this has been added to master I will wait for it.

Gah… not sure why? I’ve added them as attachments this time. But the commit is in master already

Cheers
Andrae

–

==
Our support, Your Success! Visit http://opensdi.geo-solutions.it for more information.

Ing. Andrea Aime

@geowolf
Technical Lead

GeoSolutions S.A.S.
Via Poggio alle Viti 1187
55054 Massarosa (LU)
Italy
phone: +39 0584 962313
fax: +39 0584 1660272
mob: +39 339 8844549

http://www.geo-solutions.it
http://twitter.com/geosolutions_it

Attachments:

  • customized-nq8.png

  • orignal-nq8.png

  • rbg-nq8.png