Jody,

 

Even though I am not entitled to vote, I’d like to point out that experiences 
from Testbed 17 clearly show that it is not wise to use GPKG as output format 
to synchronous WMS and WFS requests issued by users. 

 

Each request binds resources until the GeoPackage is complete. Cancellation is 
not possible. If the GeoPKG production takes too long, the user may “click” 
again and again and again and thereby cause binding of resources for completing 
the same request that perhaps is still in progress (but defunct). The fact that 
the client is no longer able to consume the response (either connection timeout 
or user cancelled the request) is found out only after the GeoPackage is fully 
produced and to be send to the client (IO Exception as socket is closed). Until 
then, all necessary resources of the thread (CPU, Memory, diskspace) are bound 
to produce a GeoPackage for nothing…

 

In addition we found the GeoPackage axis order issue (as you point out in #3) 
and reported that as bug: https://osgeo-org.atlassian.net/browse/GEOT-7011 

 

Best

Andreas

 

From: Jody Garnett <jody.garn...@gmail.com>
Date: Wednesday, 5. January 2022 at 03:16
To: Simone Giannecchini <simone.giannecch...@geosolutionsgroup.com>
Cc: Geoserver Devel <geoserver-devel@lists.sourceforge.net>
Subject: Re: [Geoserver-devel] GSIP-206 Promote GeoPackage to extension

 

Simone, this is the definition of slow moving and not urgent. I first asked if 
this was a good idea around 2 years ago as a customer is interested. 
Procurement moves slowly, funding has now come through to do this activity. The 
customer has been using geopackage WFS output for at least three years. Indeed 
this funding is as a result of a review of the community modules the customer 
is using in production, and asking if they would be interested in helping it be 
cleaned up and made into an official extension.

 

While I do not have a sizable number of clients, once this functionality has 
been cleaned up I will be adding it to the geocat enterprise products for all 
our customers (presently it is an option by request).

 

Let's review the checklist:

 

1. The module has at least a “handful” of users.

 

I have three users of this functionality, only one I would consider as using it 
in production.

 

2. The module has a designated and active maintainer

 

I offered to join Andrea on this (on the assumption the functionality can be 
cleaned up and be included in geocat enterprise products).

 

3. The module is considered “stable” by the majority of the PSC

 

I checked in with this on november/decemeber expecting wfs and wps to be stable 
and wms output format to be removed. Andrea asked me to revise the proposal to 
include gs-wms.

 

So far my personal testing has been mixed, EPSG:4326 output is coming out in 
Y/X order which does not match up with the specification:

 

The axis order in WKB stored in a GeoPackage follows the de facto standard for 
axis order in WKB and is therefore always (x,y{,z}{,m}) where x is easting or 
longitude, y is northing or latitude, z is optional elevation, and m is 
optional measure. This ordering explicitly overrides the axis order as 
specified in the SRS metadata, applying Case 4 from OGC 08-038r7, Revision to 
Axis Order Policy and Recommendations[K11]. This was done to maintain 
consistency with previous implementations of WKB that predated the OGC policy.

 

The above indicates additional QA is needed. You can also see the proposal 
where I noted frustration with a couple design decisions.

 

4. The module maintains 40% test coverage

 

The coverage looks good, but I have not measured it.

 

5. The module has no IP violations.

 

So far everything seems to be original work, with links to OGC where 
appropriate. If something comes up during the activity I will let you know.

 

6. The module has a page in the user manual

 

Not directly useful as the documentation will be distributed across several 
pages:

* https://docs.geoserver.org/stable/en/user/community/geopkg/index.html

 

I also note geosolutions has training materials 
https://docs.geoserver.geo-solutions.it/edu/en/wps/geopackage_output.html

 

7. The maintainer has signed the GeoServer Contributor Agreement

 

OSGeo has a signed CLA from both myself and GeoCat BV.

 

 

--

Jody Garnett

 

 

On Tue, 4 Jan 2022 at 14:23, Simone Giannecchini 
<simone.giannecch...@geosolutionsgroup.com> wrote:

Good Morning Jody,

I am not too inclined on having the gpkg output jump from community to core for 
WMS and WFS.

The process we have in place is there to exactly  prevent something like this 
from happening because "someone needs it urgently".

I mean, have you been using them in production enough to be confident with 
them? Do you already have a sizable number of clients using the extensions so 
we can trust them? I guess not given what you said above...

 

For the moment my vote is a -1 on this, but I am happy to hear your thoughts on 
my points above.


Regards,
Simone Giannecchini
==
GeoServer Professional Services from the experts!
Visit http://bit.ly/gs-services for more information.
==
Ing. Simone Giannecchini
@simogeo
Founder/Director GeoSolutions Italy

President GeoSolutions USA

phone: +39 0584 962313
fax:     +39 0584 1660272
mob:   +39  333 8128928

 

http://www.geosolutionsgroup.com
http://twitter.com/geosolutions_it

-------------------------------------------------------
This email is intended only for the person or entity to which it is addressed 
and may contain information that is privileged, confidential or otherwise 
protected from disclosure. We remind that - as provided by European Regulation 
2016/679 “GDPR” - copying, dissemination or use of this e-mail or the 
information herein by anyone other than the intended recipient is prohibited. 
If you have received this email by mistake, please notify us immediately by 
telephone or e-mail.

 

 

On Fri, Dec 31, 2021 at 1:50 AM Jody Garnett <jody.garn...@gmail.com> wrote:

Proposal is renamed.

 

With respect to gs-wps module I would like to see the matching gt-wps 
unsupported module which forms its foundation cleaned up (finally). Something 
we can discuss in the new year.
--

Jody Garnett

 

 

On Thu, 30 Dec 2021 at 07:40, Jody Garnett <jody.garn...@gmail.com> wrote:

Thanks Andrea, I will rename the proposal.

 

I have capacity to support the wps module on this one (as indeed a customer is 
funding this activity). 

 

Jody

 

On Thu, Dec 30, 2021 at 5:11 AM Andrea Aime <andrea.a...@geosolutionsgroup.com> 
wrote:

Hi Jody,

checking the proposal, I believe the title is misleading, it seems to suggest a 
classic module move from community to extension, as is... which does not match 
the actual proposal.

The actual proposal is:

·         Fold the WFS output format gs-wfs, the WMS output format in gs-wms, 
hence, move these two bits in _core_

·         Fold the WPS process in gs-wps-core

About the move to cover of the WMS/WFS output formats, from a technical point 
of view I'm not concerned:

·         The WFS output format will eventually break for large outputs due to 
HTTP  time outs (needs to be written on disk first, streamed back once 
complete), but the same already happens for zipped shapefiles, so nothing 
really new

·         The WMS output format now honors the rendering time outs, so it 
should be fine

I should however point out that the core of GeoServer is "maintained by the 
PSC" so the rest of the PSC should be comfortable having that code in core, and 
realize the associated obligation.

 

About the move of the WPS process to wps-core, I'm also personally not deeply 
concerned, if the documentation

is very clear about the experimental extensions baked into the process (so doc 
updates are needed).

I'm however noting the code moving also moves its maintainership from "nobody" 
to me (the WPS module maintainer), which I'm not too fond of [1].

 

Since you're making the proposal, I'd like you to step up as co-maintainer of 
the code you're trying to push up. For the core bits, as a PSC member, you're 
taking that

responsibility automatically anyways. Please step up to co-maintain the 
processes once moved in gs-wps-core.

 

Cheers

Andrea

 

[1] Due to both project and business obligations I'm responsible for way too 
many modules already,

something which is too big of a onus on my shoulders, and a big project 
liability in case I get sick

or decide to leave. It's something we'll have to address, possibly sooner 
rather than later.

 

On Thu, Dec 30, 2021 at 5:28 AM Jody Garnett <jody.garn...@gmail.com> wrote:

Please have a look at 

https://github.com/geoserver/geoserver/wiki/GSIP-206 which outlines moving 
different sections of the geopackage community module into the appropriate core 
module: gs-wfs, gs-wms, and gs-wps.

 

The proposal is solid, please make note of the backwards compatibility section 
which proposes calling out wps geopackage supports for not yet finalized 
extensions. While there is nothing wrong with having geoserver specific 
extensions they should be documented as such (and marked as in progress if they 
are chasing a moving target).  If folks feel strongly about documentation not 
being enough warning I can look at leaving these geopackage extensions as an 
optional install.

 

Jody

-- 

--

Jody Garnett

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


 

-- 

Regards,

Andrea Aime

==
GeoServer Professional Services from the experts!

Visit http://bit.ly/gs-services-us for more information.
==

Ing. Andrea Aime 
@geowolf
Technical Lead



GeoSolutions Group
phone: +39 0584 962313

fax:     +39 0584 1660272

mob:   +39  333 8128928

 

https://www.geosolutionsgroup.com/

http://twitter.com/geosolutions_it

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


Con riferimento alla normativa sul trattamento dei dati personali (Reg. UE 
2016/679 - Regolamento generale sulla protezione dei dati “GDPR”), si precisa 
che ogni circostanza inerente alla presente email (il suo contenuto, gli 
eventuali allegati, etc.) è un dato la cui conoscenza è riservata al/i solo/i 
destinatario/i indicati dallo scrivente. Se il messaggio Le è giunto per 
errore, è tenuta/o a cancellarlo, ogni altra operazione è illecita. Le sarei 
comunque grato se potesse darmene notizia.

This email is intended only for the person or entity to which it is addressed 
and may contain information that is privileged, confidential or otherwise 
protected from disclosure. We remind that - as provided by European Regulation 
2016/679 “GDPR” - copying, dissemination or use of this e-mail or the 
information herein by anyone other than the intended recipient is prohibited. 
If you have received this email by mistake, please notify us immediately by 
telephone or e-mail

-- 

--

Jody Garnett

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

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

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

Reply via email to