Requirements

• Python3: to ensure you can execute problems scripts.

GeoServer Cloud

Some steps to reproduce the errors with GeoServer Cloud.

Steps

Problem 1

1. Clone the GeoServer Cloud repository:

   git clone --recurse-submodules https://github.com/geoserver/geoserver-cloud.git
   git checkout v2.26.2.0

2. run the pgconfig file:

   cd ./geoserver-cloud/compose/
   ./pgconfig up -d

3. Now clone this repo:

   git clone https://github.com/romer8/gs-cloud-example.git

4. Run the generate_data.sh file. This will create a table on the postgis database called cities, and it will insert two records. It will use the gscloud_dev_pgconfig-postgis-1 running image:

   cd ./gs-cloud-example/
   ./generate_data.sh

5. Get the IP of the postgis container. We will be using the postgis container that was spin by the pgconfig.sh script :

   docker inspect -f '{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}' gscloud_dev_pgconfig-postgis-1

6. Run the first_problem.py python file. This will create a workspace and a datastore:

   python3 first_problem.py --host <your_ip>  # Replace with your desired IP

7. Access the webui (http://localhost:9090/geoserver/cloud/web/ user:admin, pass:geoserver). Then create a layer based on the created datastore (test_view_datastore1), based on the table cities. You need to create the SQL view manually with the following:

   • sql statement:

         SELECT 
             id,
         name,
             geom 
         FROM 
             cities
         WHERE 
             id = %id%

   • Click on Guess parameters from SQL and complete the SQL view parameters as follows:

     • name: id
     • Default value : 1
     • Validation regular expression : ^[0-9]+$

   • Click on Guess geometry type and srid for the Attributes, and click on Refresh (you will see the list of attributes listed)

8. Once you save the layer, go to layer preview. You should be able to see the error about the special character: "%"

9. if you come back to the layer and edit the sql view and save it, the error will return after you do a new request.

Problem 2

1. Get the IP of the postgis container. We will be using the postgis container that was spin by the pgconfig script :

   docker inspect -f '{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}' gscloud_dev_pgconfig-postgis-1

2. Run the second_problem.py python file. This will create a workspace,a datastore, and a sql view:

   python3 second_problem.py --host <your_ip>  # Replace with your desired IP

3. Once saved, go to layer preview. Choose GeoJSON for the WFS service, You should be able to see that the layer has 0 feature. If you try a WMS service it will show an empty white square.

GeoServer Cloud with DataDir profile

Steps

Problem 1

1. Clone the GeoServer Cloud repository:

   git clone --recurse-submodules https://github.com/geoserver/geoserver-cloud.git
   git checkout v2.26.2.0

2. run the datadir file:

   cd ./geoserver-cloud/compose/
   ./datadir up -d

3. Now clone this repo:

   git clone https://github.com/romer8/gs-cloud-example.git

4. Run the generate_data.sh file. This will create a table on the postgis database called cities, and it will insert two records:

   cd ./gs-cloud-example/
   ./generate_data.sh -c gscloud_dev_datadir-postgis-1

5. Get the IP of the postgis container. We will be using the postgis container that was spin by the datadir.sh script :

   docker inspect -f '{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}' gscloud_dev_datadir-postgis-1

6. Run the first_problem.py python file. This will create a workspace and a datastore:

   python3 first_problem.py --host <your_ip>  # Replace with your desired IP

7. Access the webui (http://localhost:9090/geoserver/cloud/web/ user:admin, pass:geoserver). Then create a layer based on the created datastore (test_view_datastore1), based on the table cities. You need to create the SQL view manually with the following:

   • sql statement:

         SELECT 
             id,
         name,
             geom 
         FROM 
             cities
         WHERE 
             id = %id%

   • Click on Guess parameters from SQL and complete the SQL view parameters as follows:

     • name: id
     • Default value : 1
     • Validation regular expression : ^[0-9]+$

   • Click on Guess geometry type and srid for the Attributes, and click on Refresh (you will see the list of attributes listed)

8. If you are not able to see the layer on layer preview, go to the rest service and login, then come back to the layer preview and you should be able to see it.

Problem 2

1. Get the IP of the postgis container:

   docker inspect -f '{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}' gscloud_dev_datadir-postgis-1

2. Run the second_problem.py python file. This will create a workspace,a datastore, and a sql view:

   python3 second_problem.py --host <your_ip>  # Replace with your desired IP

3. Once saved, go to layer preview. Choose GeoJSON for the WFS service, You should be able to see that the layer has 0 feature. If you try a WMS service it will show an empty white square, or it will give an error, but once you compute the bounds, then you can go back to layer preview and choose WMS service, and it will show an empty white square

Vanilla GeoServer

Some steps to reproduce the errors with Vanilla GeoServer.

Steps

Problem 1

1. Start GeoServer using Docker :

   docker run --name vanila-gs -d -p 80:8080 docker.osgeo.org/geoserver:2.26.2

2. Start a Postgis container"

   docker run --name postgis-vanilla-gs -d -p 5435:5432 -e POSTGRES_PASSWORD=postgres -e POSTGRES_USER=postgis -e POSTGRES_DB=postgis -e POSTGRES_PASSWORD=postgis -e POSTGIS_GDAL_ENABLED_DRIVERS=ENABLE_ALL imresamu/postgis:15-3.4
   GEOSERVER_URL = "http://localhost:9090/geoserver/cloud"

3. Now clone this repo:

   git clone https://github.com/romer8/gs-cloud-example.git

4. Run the generate_data.sh file. This will create a table on the postgis database called cities, and it will insert two records. It will use the gscloud_dev_pgconfig-postgis-1 running image:

   cd ./gs-cloud-example/
   ./generate_data.sh -c postgis-vanilla-gs

5. Get the IP of the postgis container:

   docker inspect -f '{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}' postgis-vanilla-gs

6. Run the first_problem.py python file. This will create a workspace and a datastore:

   python3 first_problem.py --geoserver-url "http://localhost/geoserver" --host <your_ip>  # Replace with your desired IP

7. Access the webui (http://localhost/geoserver user:admin, pass:geoserver). Then create a layer based on the created datastore (test_view_datastore1), based on the table cities. You need to create the SQL view manually with the following:

   • sql statement:

         SELECT 
             id,
         name,
             geom 
         FROM 
             cities
         WHERE 
             id = %id%

   • Click on Guess parameters from SQL and complete the SQL view parameters as follows:

     • name: id
     • Default value : 1
     • Validation regular expression : ^[0-9]+$

   • Click on Guess geometry type and srid for the Attributes, and click on Refresh (you will see the list of attributes listed)

8. Click Compute from data and Compute from native bounds for the Bounding Boxes.

9. Once you save the layer, go to layer preview. You should be able to see the layer succesfully.

Problem 2

1. Get the IP of the postgis container:

   docker inspect -f '{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}' postgis-vanilla-gs

2. Run the second_problem.py python file. This will create a workspace,a datastore, and a sql view:

   python3 second_problem.py --geoserver-url "http://localhost/geoserver" --host <your_ip>  # Replace with your desired IP

3. Once saved, go to layer preview. Choose GeoJSON for the WFS service, You should be able to see that the layer has 0 feature. If you try a WMS service it will show an empty white square, or it will give an error, but once you compute the bounds, then you can go back to layer preview and choose WMS service, and it will show an empty white square