[Geoserver-users] Geoserver 1.3.0 Pre Release

Hi all,

We are happy to announce the pre-release of GeoServer 1.3.0. This will be our final 1.3.0 release candidate. Download is available from:

http://docs.codehaus.org/display/GEOS/Latest

Please download and try it out. Helping us test and report bugs will help us flush out any final issues for 1.3.0.

Special thanks to the following for bug fixes and contributions this release:

Holger Lang: Patch allowing GetCapabilties to be based on service level

Akira Sugawara: Patch allowing support for Japanese and other multi-byte encoded languages.

And thanks to everyone else who helped us test and file bug reports.

- The GeoServer Team

Hi all,

It seems that postgresql datastore (for postgresql 7.2.2, postgis 1.0) is
broken with PR1 (may be RC7 too, not in RC6 and before).

I am having the following error :

org.geotools.data.DataSourceException: Error querying relation:serie_rouge.
Possible cause:ERROR: parser: parse error at or near "." at
org.geotools.data.postgis.PostgisDataStore.buildSchema(PostgisDataStore.java:591)

Looking at PostgisDataStore.java, the query launched against postgresql is :

st.execute("SELECT * FROM \"" + config.getDatabaseSchemaName() +
"\".\""+typeName + "\" LIMIT 0;");

In the case of postgres 7.2, we don't have a dictionnary except through pg_*
tables.

Am I right or did I missed something ?

didier

Hi Richard,

This is an issue with schema support in earlier versions of postgres. If I am not mistaken schema support was still in an experimental state pre 7.3, which means there are probably issues with the jdbc driver playing nice with schemas in 7.2. Upgrading your postgres database to a post 7.3 should fix this problem.

However, this upgrade may not be possible for some users. We could patch the data store to check which version of postgres is being used and form queries appropriatly, however that could be messy.

What do other people think?

-Justin

Richard didier wrote:

Hi all,

It seems that postgresql datastore (for postgresql 7.2.2, postgis 1.0) is broken with PR1 (may be RC7 too, not in RC6 and before).

I am having the following error :

org.geotools.data.DataSourceException: Error querying relation:serie_rouge. Possible cause:ERROR: parser: parse error at or near "." at org.geotools.data.postgis.PostgisDataStore.buildSchema(PostgisDataStore.java:591)

Looking at PostgisDataStore.java, the query launched against postgresql is :

st.execute("SELECT * FROM \"" + config.getDatabaseSchemaName() + "\".\""+typeName + "\" LIMIT 0;");

In the case of postgres 7.2, we don't have a dictionnary except through pg_* tables.

Am I right or did I missed something ?

didier

--
Justin Deoliveira
The Open Planning Project
http://topp.openplans.org

I think you should deprecate support for earlier versions of PgSQL. 7.3 is starting to look pretty old.
P

On Jan 19, 2006, at 2:42 PM, Justin Deoliveira wrote:

Hi Richard,

This is an issue with schema support in earlier versions of postgres. If I am not mistaken schema support was still in an experimental state pre 7.3, which means there are probably issues with the jdbc driver playing nice with schemas in 7.2. Upgrading your postgres database to a post 7.3 should fix this problem.

However, this upgrade may not be possible for some users. We could patch the data store to check which version of postgres is being used and form queries appropriatly, however that could be messy.

What do other people think?

-Justin

Richard didier wrote:

Hi all,
It seems that postgresql datastore (for postgresql 7.2.2, postgis 1.0) is broken with PR1 (may be RC7 too, not in RC6 and before).
I am having the following error :
org.geotools.data.DataSourceException: Error querying relation:serie_rouge. Possible cause:ERROR: parser: parse error at or near "." at org.geotools.data.postgis.PostgisDataStore.buildSchema(PostgisDataStore.java:591)
Looking at PostgisDataStore.java, the query launched against postgresql is :
st.execute("SELECT * FROM \"" + config.getDatabaseSchemaName() + "\".\""+typeName + "\" LIMIT 0;");
In the case of postgres 7.2, we don't have a dictionnary except through pg_* tables.
Am I right or did I missed something ?
didier

--
Justin Deoliveira
The Open Planning Project
http://topp.openplans.org

-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
geoserver-devel List Signup and Options

Hi all,

Is it possible (with geoserver) to group differents rasters in order just to simplify the url. Instead of having to say i want this one, this one, this one…, i’d like to say “I want this group of the layers in this scale ?” (some scale have more than 100 files so it a huge url) ?

Jérémy

No, not yet. I am hoping that this will indeed be possible with the implementation of the ComplexFeature Store and further development of the WCS branch.

Alex

On 1/20/06, jeremy.deroyer@…412… <jeremy.deroyer@…412… > wrote:

Hi all,

Is it possible (with geoserver) to group differents rasters in order just to simplify the url. Instead of having to say i want this one, this one, this one…, i’d like to say “I want this group of the layers in this scale ?” (some scale have more than 100 files so it a huge url) ?

Jérémy

Hrm, I'm not sure that getting all up in to complex and wcs would be the
way to do it. I feel like there could be an easier way to do it, that
doesn't get all in to datastore definitions, but allows a more light
weight defining of a 'virtual layer', that is just made up of a group
of others. David Blasby had a few thoughts in that direction. It's
definitely a capability I'd like to see, it would be useful indeed.

best regards,

Chris

Quoting Alexander Petkov <greenkov@anonymised.com>:

No, not yet. I am hoping that this will indeed be possible with
the
implementation of the ComplexFeature Store and further development of
the
WCS branch.

Alex

On 1/20/06, jeremy.deroyer@anonymised.com <jeremy.deroyer@anonymised.com>
wrote:
>
>
> Hi all,
>
> Is it possible (with geoserver) to group differents rasters in
order just
> to simplify the url. Instead of having to say i want this one, this
one,
> this one..., i'd like to say "I want this group of the layers in
this scale
> ?" (some scale have more than 100 files so it a huge url) ?
>
> Jérémy

***
Chris Holmes
The Open Planning Project
thoughts at: http://cholmes.wordpress.com

----------------------------------------------------------
This mail sent through IMP: https://webmail.limegroup.com/

Hi,

As my url is too long (about 100 rasters), I’ve tried to use geoserver with a post method. Below is my html source and the message I received.

Any idea how to make geoserver works with so many files ?

May be are worldimage not the best image format for such an application… what kind of format should I ask for ?

Thank’s,

Jérémy

HTML :

winput type="text" name="request" value="GetMap">

Response from geoserver :

<?xml version="1.0" encoding="UTF-8" standalone="no" ?> line 1 column 1 -- Content is not allowed in prolog.

Chris Holmes cholmes@anonymised.com

20/01/2006 20:30

A
Alexander Petkov <greenkov@anonymised.com…>

cc
jeremy.deroyer@anonymised.comjeremy.deroyer@anonymised.com, “geoserver-users@anonymised.comet” geoserver-users@lists.sourceforge.net

Objet
Re: [Geoserver-users] Rasters & scales

Hrm, I'm not sure that getting all up in to complex and wcs would be the
way to do it. I feel like there could be an easier way to do it, that
doesn't get all in to datastore definitions, but allows a more light
weight defining of a 'virtual layer', that is just made up of a group
of others. David Blasby had a few thoughts in that direction. It's
definitely a capability I'd like to see, it would be useful indeed.

best regards,

Chris

Quoting Alexander Petkov <greenkov@anonymised.com>:

> No, not yet. I am hoping that this will indeed be possible with
> the
> implementation of the ComplexFeature Store and further development of
> the
> WCS branch.
>
> Alex
>
> On 1/20/06, jeremy.deroyer@anonymised.com <jeremy.deroyer@anonymised.com>
> wrote:
> >
> >
> > Hi all,
> >
> > Is it possible (with geoserver) to group differents rasters in
> order just
> > to simplify the url. Instead of having to say i want this one, this
> one,
> > this one..., i'd like to say "I want this group of the layers in
> this scale
> > ?" (some scale have more than 100 files so it a huge url) ?
> >
> > Jérémy
>

***
Chris Holmes
The Open Planning Project
thoughts at: http://cholmes.wordpress.com

----------------------------------------------------------
This mail sent through IMP: https://webmail.limegroup.com/

Just use the xml example on www.geoserver.org.

Sorry,

Jérémy

jeremy.deroyer@anonymised.com
Envoyé par : geoserver-users-admin@lists.sourceforge.net

28/01/2006 15:39

A
Chris Holmes cholmes@anonymised.com

cc
geoserver-users@anonymised.comrge.netgeoserver-users@lists.sourceforge.net, Alexander Petkov greenkov@anonymised.com

Objet
Re: [Geoserver-users] Rasters & scales

Hi,

As my url is too long (about 100 rasters), I’ve tried to use geoserver with a post method. Below is my html source and the message I received.

Any idea how to make geoserver works with so many files ?

May be are worldimage not the best image format for such an application… what kind of format should I ask for ?

Thank’s,

Jérémy

HTML :

winput type="text" name="request" value="GetMap">

Response from geoserver :

<?xml version="1.0" encoding="UTF-8" standalone="no" ?> line 1 column 1 -- Content is not allowed in prolog.

Chris Holmes cholmes@anonymised.com

20/01/2006 20:30

A
Alexander Petkov greenkov@anonymised.com

cc
jeremy.deroyer@anonymised.comjeremy.deroyer@anonymised.com, “geoserver-users@anonymised.comet” geoserver-users@lists.sourceforge.net

Objet
Re: [Geoserver-users] Rasters & scales

Hrm, I'm not sure that getting all up in to complex and wcs would be the
way to do it. I feel like there could be an easier way to do it, that
doesn't get all in to datastore definitions, but allows a more light
weight defining of a 'virtual layer', that is just made up of a group
of others. David Blasby had a few thoughts in that direction. It's
definitely a capability I'd like to see, it would be useful indeed.

best regards,

Chris

Quoting Alexander Petkov <greenkov@anonymised.com>:

> No, not yet. I am hoping that this will indeed be possible with
> the
> implementation of the ComplexFeature Store and further development of
> the
> WCS branch.
>
> Alex
>
> On 1/20/06, jeremy.deroyer@anonymised.com <jeremy.deroyer@anonymised.com>
> wrote:
> >
> >
> > Hi all,
> >
> > Is it possible (with geoserver) to group differents rasters in
> order just
> > to simplify the url. Instead of having to say i want this one, this
> one,
> > this one..., i'd like to say "I want this group of the layers in
> this scale
> > ?" (some scale have more than 100 files so it a huge url) ?
> >
> > Jérémy
>

***
Chris Holmes
The Open Planning Project
thoughts at: http://cholmes.wordpress.com

----------------------------------------------------------
This mail sent through IMP: https://webmail.limegroup.com/

Richard didier wrote:

Le Jeudi 19 Janvier 2006 23:42, Justin Deoliveira a écrit :

Hi Richard,

This is an issue with schema support in earlier versions of postgres. If
I am not mistaken schema support was still in an experimental state pre
7.3, which means there are probably issues with the jdbc driver playing
nice with schemas in 7.2. Upgrading your postgres database to a post 7.3
should fix this problem.

However, this upgrade may not be possible for some users. We could patch
the data store to check which version of postgres is being used and form
queries appropriatly, however that could be messy.

If not, the postgresql datastore should not be loaded at all (failure comes only when accessing featureTypes, not when reading datastore config). In that case, geoserver documentation should say : "only post 7.3 postgreSQL datastores are supported".
On the other hand, it was working up to RC6.

What do other people think?

I have migrated my databases once from 7.0 to 7.2.2 and it was not so easy.
But, migrating databases is one day mandatory ...

I agree, upgrading database versions is no easy task which is why i am open to having the datastore be smarter about queries are constructed.

Suggestion: if the first test failed, would it be possible to check with the "old syntax way" for the featureType to exist with pg_* tables in PostgisDataStore.java (JDBCDataStore.java might also be changed) ?

I am not too sure what you mean about the pg_* system system tables. As written the datastore doesn't really touch them, all that is abstracted away through the jdbc driver.

I beleive that all is required is to change instances of

"SELECT ... FROM " + config.getDataBaseSchemaName() + "." + typeName + "..."

to

"SELECT ... FROM " + typeName + "..."

-Justin

didier

--
Justin Deoliveira
The Open Planning Project
http://topp.openplans.org