Hi,
I've just upgraded to Geoserver 1.5.2 and wanted to try out the GeoRSS support. So I put this in firefox: http://localhost:8080/geoserver/wms/reflect?layers=topp:states&format=rss and got
<ServiceExceptionReport version="1.1.1">
−
<ServiceException code="InvalidFormat">
There is no support for creating maps in rss format
</ServiceException>
</ServiceExceptionReport>
The same thing happens when I tried using a full wms request with format=application/rss+xml (since that's what's listed in the capabilities doc), does anyone else have this problem?
On a different topic, I used the new wms-gs.js with some of the same feature types I had from 1.5.1 and it seems like the output is shifted in different directions at different zoom levels. I don't know if it's just my particular data, but it seems to me that the old one was more accurate (as far as I can tell on the map, since a lot of my point data are on roads)? Using the old wms-gs.js with 1.5.2 makes it worse though...is there some tweaks I can play around with and see how to make it better?
Thanks!
Doris
dlamoris@anonymised.com wrote:
Hi,
I've just upgraded to Geoserver 1.5.2 and wanted to try out the GeoRSS support. So I put this in firefox: http://localhost:8080/geoserver/wms/reflect?layers=topp:states&format=rss and got
<ServiceExceptionReport version="1.1.1">
−
<ServiceException code="InvalidFormat">
There is no support for creating maps in rss format
</ServiceException>
</ServiceExceptionReport>
The same thing happens when I tried using a full wms request with format=application/rss+xml (since that's what's listed in the capabilities doc), does anyone else have this problem?
Sigh... It looks like accessing the format via just "rss" was removed. Due to a cite test issue it was removed from the capabilities document as an available format... but i believe the intent was to still allow the short hand. Andrea: Can you confirm?
I can also confirm that "application/rss+xml" does not work? It appears the "+" has to be url encoded. So if you use:
"format=application/rss%2Bxml"
It should work. Which is a bit confusing because other formats with "+" sign in them work without url encoding. for instance:
application/vnd.google-earth.kml+XML
I will look into this further.
On a different topic, I used the new wms-gs.js with some of the same feature types I had from 1.5.1 and it seems like the output is shifted in different directions at different zoom levels. I don't know if it's just my particular data, but it seems to me that the old one was more accurate (as far as I can tell on the map, since a lot of my point data are on roads)? Using the old wms-gs.js with 1.5.2 makes it worse though...is there some tweaks I can play around with and see how to make it better?
No.. the file was recently changed to use a different projection (the "google projection"). See this recent thread on the developers list for the details:
http://www.nabble.com/Google-projection,-first-cut-in-t4213712.html
Thanks!
Doris
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users
!DSPAM:4007,46b791fe15971336712104!
--
Justin Deoliveira
The Open Planning Project
http://topp.openplans.org
Hi,
I just had a similar experience using geoserver-1.5.2-war.zip
http://localhost/geoserver152/wms/reflect?layers=states&format=application/r
ss%2Bxml does work.
http://localhost/geoserver152/wms/reflect?layers=states&format=rss or georss
does not:
<?xml version="1.0" encoding="UTF-8" standalone="no" ?>
<!DOCTYPE ServiceExceptionReport (View Source for full doctype...)>
- <ServiceExceptionReport version="1.1.1">
<ServiceException code="InvalidFormat">There is no support for creating
maps in rss format</ServiceException>
</ServiceExceptionReport>
Thanks
Randy
-----Original Message-----
From: geoserver-users-bounces@lists.sourceforge.net
[mailto:geoserver-users-bounces@lists.sourceforge.net] On Behalf Of
dlamoris@anonymised.com
Sent: Monday, August 06, 2007 3:26 PM
To: geoserver-users@lists.sourceforge.net
Subject: [Geoserver-users] GeoRSS output and google maps
Hi,
I've just upgraded to Geoserver 1.5.2 and wanted to try out the GeoRSS
support. So I put this in firefox:
http://localhost:8080/geoserver/wms/reflect?layers=topp:states&format=rss
and
got
<ServiceExceptionReport version="1.1.1">
−
<ServiceException code="InvalidFormat">
There is no support for creating maps in rss format
</ServiceException>
</ServiceExceptionReport>
The same thing happens when I tried using a full wms request with
format=application/rss+xml (since that's what's listed in the
capabilities doc), does anyone else have this problem?
On a different topic, I used the new wms-gs.js with some of the same
feature types I had from 1.5.1 and it seems like the output is shifted
in different directions at different zoom levels. I don't know if it's
just my particular data, but it seems to me that the old one was more
accurate (as far as I can tell on the map, since a lot of my point
data are on roads)? Using the old wms-gs.js with 1.5.2 makes it worse
though...is there some tweaks I can play around with and see how to
make it better?
Thanks!
Doris
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users
I can also confirm that "application/rss+xml" does not work? It appears the "+" has to be url encoded. So if you use:
"format=application/rss%2Bxml"
It should work. Which is a bit confusing because other formats with "+" sign in them work without url encoding. for instance:
application/vnd.google-earth.kml+XML
Looks like KML does a special check:
format.startsWith("application/vnd.google-earth.kml").
Opening up a bug for this. Should be simple to fix. Should we get it into 1.5.2a?
http://jira.codehaus.org/browse/GEOS-1271
This also begs the larger question if we should put in a "global" check so this does not happen to other formats which use a "+" sign.
-Justin
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users
--
Justin Deoliveira
The Open Planning Project
http://topp.openplans.org
Hi,
Sorry for nagging...but I'm a bit confused how using the old wms-gs.js with 1.5.2 would change the output compared to 1.5.1? Since the script only returns a wms request url, shouldn't both be rendered the same way, unless it's handled differently in 1.5.2?
Thanks,
Doris
On a different topic, I used the new wms-gs.js with some of the same feature types I had from 1.5.1 and it seems like the output is shifted in different directions at different zoom levels. I don't know if it's just my particular data, but it seems to me that the old one was more accurate (as far as I can tell on the map, since a lot of my point data are on roads)? Using the old wms-gs.js with 1.5.2 makes it worse though...is there some tweaks I can play around with and see how to make it better?
No.. the file was recently changed to use a different projection (the "google projection"). See this recent thread on the developers list for the details:
http://www.nabble.com/Google-projection,-first-cut-in-t4213712.html
Thanks!
Doris
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users
!DSPAM:4007,46b791fe15971336712104!
Justin Deoliveira ha scritto:
dlamoris@anonymised.com wrote:
...
Sigh... It looks like accessing the format via just "rss" was removed. Due to a cite test issue it was removed from the capabilities document as an available format... but i believe the intent was to still allow the short hand. Andrea: Can you confirm?
You are all right about the intentions, but given your test it seems
I failed to turn them into practice 
Justin, I see from the svn mailing list you already fixed this one, right? What about the "release"?
Cheers
Andrea
dlamoris@anonymised.com ha scritto:
Hi,
...
On a different topic, I used the new wms-gs.js with some of the same feature types I had from 1.5.1 and it seems like the output is shifted in different directions at different zoom levels. I don't know if it's just my particular data, but it seems to me that the old one was more accurate (as far as I can tell on the map, since a lot of my point data are on roads)?
Hum.... the old one was adding (depending on the zoom) shifts of up to tens of kilometers on the GeoServer sample data. The new one has been
tested with the same data and the output looks the same as putting
data right into Google Earth thru KML. Oh, the new one requires the
100003 projection code in $GEOSERVER_DATA_DIR/user_projections/epsg.properties, you have it, right?
There has been another change that may explain the issue. Go to the
feature type configuration panel, and see if you have a drop down
list allowing you to choose how to handle differences between then
native CRS and the declared one (you won't find it if your data
does not have a native CRS, such as a shapefile with a .prj file
associated). Anyways, if you have that dropdown, try the different options and tell me if it makes any difference.
Cheers
Andrea
You are all right about the intentions, but given your test it seems
I failed to turn them into practice
Justin, I see from the svn mailing list you already fixed this one, right? What about the "release"?
Yeah. the fix is committed for georss. From what I understand the release is already out so we are just going to live with this one?
Cheers
Andrea
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users
!DSPAM:4007,46b8110d88532090977483!
--
Justin Deoliveira
The Open Planning Project
http://topp.openplans.org
Justin Deoliveira ha scritto:
You are all right about the intentions, but given your test it seems
I failed to turn them into practice
Justin, I see from the svn mailing list you already fixed this one, right? What about the "release"?
Yeah. the fix is committed for georss. From what I understand the release is already out so we are just going to live with this one?
I guess so
Bah, next time it's better if we do an RC even for a minor release....
Cheers
Andrea
I do have the 100003 projection, and I tried the different dropdown options, but there's no difference...
If it helps, I've attached some screenshots of what it looks like. new means it's using 1.5.2 with 100003 and old means 1.5.1 with the old wms-gs.js. The numbers attached are the zoom levels. You can see that for the new one, it suddenly shifts in the opposite direction going from zoom level 2 to 1. (The actual data should be mostly on the roads, at least on the left, so the old ones are actually more accurate...) The same shifts happen for all my data that were fine using the old setup. I wouldn't mind reverting back, but I really like the new kml improvements (it works fine in google earth). It's not urgent, but is there a way I can get the old script to work with 1.5.2?
Thanks!
Doris
Andrea Aime wrote:
dlamoris@anonymised.com ha scritto:
Hi,
...
On a different topic, I used the new wms-gs.js with some of the same feature types I had from 1.5.1 and it seems like the output is shifted in different directions at different zoom levels. I don't know if it's just my particular data, but it seems to me that the old one was more accurate (as far as I can tell on the map, since a lot of my point data are on roads)?
Hum.... the old one was adding (depending on the zoom) shifts of up to tens of kilometers on the GeoServer sample data. The new one has been
tested with the same data and the output looks the same as putting
data right into Google Earth thru KML. Oh, the new one requires the
100003 projection code in $GEOSERVER_DATA_DIR/user_projections/epsg.properties, you have it, right?
There has been another change that may explain the issue. Go to the
feature type configuration panel, and see if you have a drop down
list allowing you to choose how to handle differences between then
native CRS and the declared one (you won't find it if your data
does not have a native CRS, such as a shapefile with a .prj file
associated). Anyways, if you have that dropdown, try the different options and tell me if it makes any difference.
Cheers
Andrea
Doris Lam ha scritto:
I do have the 100003 projection, and I tried the different dropdown options, but there's no difference...
If it helps, I've attached some screenshots of what it looks like. new means it's using 1.5.2 with 100003 and old means 1.5.1 with the old wms-gs.js. The numbers attached are the zoom levels. You can see that for the new one, it suddenly shifts in the opposite direction going from zoom level 2 to 1.
Hem, that's what was happening to me with the old script, stuff jumped
on and off a river I used as a visual reference as I zoomed in. With
the new one data are stable instead.... something is really going on,
I just can't figure out what is it.
(The actual data should be mostly on the roads, at least on the left, so the old ones are actually more accurate...) The same shifts happen for all my data that were fine using the old setup. I wouldn't mind reverting back, but I really like the new kml improvements (it works fine in google earth). It's not urgent, but is there a way I can get the old script to work with 1.5.2?
I am not aware of anything we did in GeoServer that may have made the old script fail. Can you send me, by private mail, the dataset and the
html file you're using so that I can reproduce the issue and investigate?
Cheers
Andrea
Andrea Aime wrote:
Justin Deoliveira ha scritto:
You are all right about the intentions, but given your test it seems
I failed to turn them into practice
Justin, I see from the svn mailing list you already fixed this one, right? What about the "release"?
Yeah. the fix is committed for georss. From what I understand the release is already out so we are just going to live with this one?
I guess so
Bah, next time it's better if we do an RC even for a minor release....
Yeah, we could call it 1.5.3, and make that the one with the big announcement, which right now I'm not 100% sure we'll get out tomorrow.
Chris
Cheers
Andrea
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users
!DSPAM:4005,46b817d292372090977483!
Doris Lam ha scritto:
I do have the 100003 projection, and I tried the different dropdown options, but there's no difference...
Doris,
I tracked the problem down to the meta-tiler. In the new gs-wms.js
file set tiled=false and points should get to their usual location.
Can you confirm this?
Cheers
Andrea
Doris Lam ha scritto:
I do have the 100003 projection, and I tried the different dropdown options, but there's no difference...
Well, the issue was in fact in the wms-js file.
Now, can I ask you to try out the version I've attached to this
mail? It's the new wms-js with a few fixes, and I also expanded
it so that you can specify filters in various ways.
For consistency with the other params, CQL is specificed by the myCQL
field instead of the 'cql' one you used.
I tried this wms-js with 4 different datasets and it seems to
work fine, can you confirm?
Cheers
Andrea
Andrea Aime ha scritto:
Doris Lam ha scritto:
I do have the 100003 projection, and I tried the different dropdown options, but there's no difference...
Well, the issue was in fact in the wms-js file.
Now, can I ask you to try out the version I've attached to this
mail?
Hem.... attachment included this time...
Cheers
Andrea
wms-gs.js (5.13 KB)
Sorry I just got to work. Yes, it works fine now, thank you very much!
One thing I noticed is that I had to comment out this line
if (typeof(window['this.myStyles'])=="undefined") this.myStyles="";
in order for it to use myStyles.
What's the difference between CQL and FILTERS?
Doris
Quoting Andrea Aime <aaime@anonymised.com>:
Andrea Aime ha scritto:
Doris Lam ha scritto:
I do have the 100003 projection, and I tried the different dropdown options, but there's no difference...
Well, the issue was in fact in the wms-js file.
Now, can I ask you to try out the version I've attached to this
mail?
Hem.... attachment included this time...
Cheers
Andrea
dlamoris@anonymised.com ha scritto:
Sorry I just got to work. Yes, it works fine now, thank you very much!
One thing I noticed is that I had to comment out this line
if (typeof(window['this.myStyles'])=="undefined") this.myStyles="";
in order for it to use myStyles.
Ah, cool, I'll have a look at this.
What's the difference between CQL and FILTERS?
Well, CQL is a filter expressed in CQL syntax, while
FILTER is a filter expressed in OGC syntax, that is,
as XML. You can find some sample of the two in the
sample requests.
Usually CQL filters are much easier to setup and read,
but they have a limitation: at the moment there is
no way use a feature id for filtering in CQL.
OGC filters do not have such a limitation.
Cheers
andrea