Jody,
Yeah of course you're right I was a bit confused. It is quite
straight-forward to solve.
Just need to overwrite getInfo in WFSContentFeatureSource, and need to
write some extra stuff in the strategy to take that info out of the
getcapabilities.
Will do.
Regards
Niels
On 30/05/14 13:57, Jody Garnett wrote:
In this case WFS Client has much better information then what the
generic ContentDataStore has access to, this data structure was
originally made so that WFS Server could communicate the title and
description information back. Although JDBC implementations can check
table metadata for title / description.
So yeah fill in what you can (and WFS should be able to fill in
everything). As you have seen uDig uses this data structure to figure
out what content is made available, and it should populate the
defaults in GeoServer as well.
--
Jody
Jody Garnett
On Fri, May 30, 2014 at 6:28 PM, Niels Charlier <ni...@scitus.be
<mailto:ni...@scitus.be>> wrote:
Hello Jody,
Regarding the info: the getInfo that passes on this information is
implemented in ContentDataStore in gt-data, a class written by you
two (Jody and Justin).
It seems to me you want me to pass on information from the
GeoServer catalog, but how is that information even available from
within a geotools module? That seems to me the reason why CRS and
bounds can be provided but not the other stuff. I don't know if
there is a straight-forward way to do this.
Either way, it seems to me this might be out of scope here because
this regards a more general issue (ContentDataStore) rather than a
wfs-ng specific issue. I'll leave it up to you two to decide...
Kind Regards
Niels
On 29/05/14 03:25, Jody Garnett wrote:
Aside: In testing this patch for uDig I noticed that wfs-ng
(before your pull request) is not filling in the Info data
structure for topp:states
title is null, but should be "USA Population"
description is null, but should be "This is some census data on
the states."
bounds is correct.
crs is correct.
resourceInfo.getSchema() fails with a NPE inside GeoTools code:
java.lang.NullPointerException
at java.net.URI$Parser.parse(URI.java:3023)
at java.net.URI.<init>(URI.java:595)
at
org.geotools.data.store.ContentFeatureSource$1.getSchema(ContentFeatureSource.java:267)
Checking DescribeFeatureType header this result should be
"http://www.openplans.org/topp";
With the fallback code the schema is set to
"http://localhost:8080/geoserver/ows?service=wfs&version=1.0.0&request=GetCapabilities";
keywords came back empty, but should be
"census","united","boundaries","state","states"
Jody Garnett
On Thu, May 29, 2014 at 9:44 AM, Jody Garnett
<jody.garn...@gmail.com <mailto:jody.garn...@gmail.com>> wrote:
Just a remark first: I did not write any of the code we are
writing about, I am only reporting what I find. I don't feel
one way or the other about it, so if you reckon it needs to
be changed it is fine for me, as long as you specify exactly
what you want to change about it. I'm just the messenger,
basically.
Thanks Niels, I expect we are both being too polite. I will
try and be succinct, feel free to push back if I am creating
more work for you.
1) access to version negotiation
uDig was making use of a method that allowed passing in a URL
(of WFS Service without VERSION) and it would return the
result of version negotiation between client and server (i.e.
a URL with VERSION=X.X.X). This is required so we get a
stable ID for recording the datastore (and thus can notices
if two URLs end up talking to the same WFS using the same
protocol).
2) Default output format WFS=1.0.0 = GML2, WFS 1.1 = GML3,
WFS=2.0 = GML?
3) namespace
Namespace should be not only optional, but marked as
LEVEL="advanced" (or even LEVEL="program" if it something
that geoserver fills in automatically to match the workspace
namespace)
4) udig did a null check, just in case.
Agreed.
Thanks Niels, sorry for the epic discussion thread.
Jody
------------------------------------------------------------------------------
Time is money. Stop wasting it! Get your web API in 5 minutes.
www.restlet.com/download
http://p.sf.net/sfu/restlet
_______________________________________________
GeoTools-Devel mailing list
GeoTools-Devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geotools-devel
