I had a look at this error and can confirm that it's a legit error with the
main branch. So an Issue in Jira would be welcomed.
Carsten provided two different sld. Here is the one I've been using:
There isn't enough of the stack trace provided but this commit is a possible
source of the regression (GeoServer 2.23.3+):
https://github.com/geotools/geotools/commit/d7991fdcd1630ea1c18b5dfad91d4e291b8bb95b
Steve Ikeoka
From: Jody Garnett
Sent: Monday, June 17,
Hi Jody,
I made the same mistake and used layer preview for testing. However, it sends
always WFS 1.0.0 GetFeature, only with different outputformat. I think it is
not the same thing to ask GML 3.2 from WFS 1.0.0 than to use WFS 2.0.0.
-Jukka Rahkonen-
Lähettäjä: Jody Garnett
Lähetetty: maana
Hi
The giant_polygon layer seems to return by default, without touching the
publishing tab, this with GML 3.2:
numberMatched="unknown" numberReturned="0"
however, one feature is returned. Tested with Geoserver 2.25.2.
-Jukka Rahkonen-
Lähettäjä: Andrea Aime
Lähetetty: maanantai 17. kesäkuu
I was testing main, but I guess I mixed GML 3.2 output format with a
different version of WFS because... I cannot reproduce any longer LOL.
For reference, here is a layer where I disabled the count matched, using
the WFS 2.0 default output format:
curl "
https://gs-main.geosolutionsgroup.com/geos
It is more that I am not aware of any change that could cause this
regression.
Were you able to determine when the problem was introduced and report the
issue to the issue tracker?
I am curious to what you will find. Determing which release the problem
started in will help a great deal.
--
Jody
Thanks Andrea, I was getting the specification numbers confused.
I am trying to reproduce your "when setting it, the numberReturn is zero"
with the 2.24.4 release in preflight testing...
- GeoServer 2.24.4 Layer Preview GML3.1: unknown
- GeoServer 2.24.4 Layer Preview GML3.2: unknown
-
Hello GeoServer community,
Wanted to ask if there was any interested parties in the marine formats S57
(and S100 perhaps although it seems to be stalled out at S101 ).
I am going to research what is possible and would like to ask if anyone is
interested in joining in :)
Please just reply if you a
Out of curiosity can you share the SLD file that did not validate? I am
curious if it had anything unusual and/or missing.
I am sorry about the "Entity resolution disallowed for null" message not
being helpful, when I wrote that I assumed every external entity would have
a name - but found out la
Hey folks,
I added the ENTITY_RESOLUTION_ALLOWLIST option for a long time but nobody
was noticing very much! I am glad you found the setting and have been
working though how it works.
You are correct that it is used to mitigate the service side request
forgery attacks. Some software is very susc
Hello Jean-Christophe,
I just upgraded to V2.25.1 and the error is gone, so no more workaround is
necessary.
Regards
Daniel
From: Jean-Christophe Bastin
Sent: Dienstag, 23. April 2024 12:05
To: Calliess Daniel Ing. ; 'geoserver-users'
Subject: Re: [Geoserver-users] WMS broken after GeoServer
Hi Andrea,
hi Jody,
thanks for that! The "Skip the counting of the numberMatched attribute"
did the trick :-) That's what I was looking for. Works great with both
GML and JSON format.
My client does not require any "numberXxx" attributes. Nevertheless,
with my version 2.22.3, numberMatched i
Hi Jody,
actually... the WFS 2.0 specification leaves out a possibility to return
"unknown" in the count numberMatched field, from the schemas:
To use it, configure the layer to skip number matched (publishing/wfs):
[image: image.png]
I believe this setting is not used often
13 matches
Mail list logo
