Current location:

Setting up and managing a metadata repository

Metadata is a valuable resource and may not be created or maintained for all data. Indeed, under the INSPIRE directive, only certain datasets in use at an authority may have metadata records. Data custodians should consider which datasets require metadata and design a data strategy accordingly.

Cadcorp currently only supports the linking of metadata records to database tables. It is therefore a requirement to store the geographic data you require to link to metadata records in a database. The Metadata Manager can retrieve and link to data held in tables stored in Oracle, PostGIS or SQL Server 2008.

The general recommended workflow should be to harvest metadata records from a dedicated database and then to manually add additional information required to complete the full metadata record.

It is recommended that a dedicated spatial database is created for geographic datasets that require metadata records. This makes the maintenance of metadata and the geographic data easier.
Ensure that the geographic data is ready to be published and complete.
Ensure that the Metadata Manager is installed.
The first time that you try and use the Metadata Manager, you will need to set the following information:
- GeognoSIS URL
- GeognoSIS Source Folder
The GeognoSIS URL is the address of the GeognoSIS server and is defined like this:
http://[machine name].[domain]:[port number]/OgcService/CSW.impl

e.g. http:// CSUPP959.cadcorp.net:4326/OgcService/CSW.impl.

The URL is used to identify the server. The :4326 may change and is the port number used for the instance created in GeognoSIS. This can be found by looking at the Properties of the GeognoSIS instance:
The GeognoSIS Source Folder is defined on the GeognoSIS server and is where Gemini files for import can be copied. This is required to ensure that GeognoSIS has direct access to the required files and that the reading of the file for import is not impeded by network or user permissions. This should be defined using a UNC path name.
You will be directed to the Administration options to set this information if you haven’t already done so.
Once defined, use the Harvest command to create your initial metadata repository:
- Go to Applications - Harvest You will see the following warning message:
  
  This serves as a reminder that a harvest operation is powerful and will overwrite any existing metadata records that have been either harvested or created from tables within the database you wish to harvest.
  
  A harvest is used to initially create a metadata record for each table in a defined database.
  
  It can also be used to add metadata records to an existing metadata repository from a new database source.
  
  So you can harvest the data from PostGIS database P1 to create metadata records for each of the tables in database P1. If they then want to harvest the data from a different PostGIS, Oracle or SQL Server 2008 database (P2, O1 or S1), then they can perform another harvest from each of these databases and new metadata records will be created alongside those previously created from database P1.
  
  However, if you choose to harvest PostGIS database P1 again, then all metadata records created by the previous harvest of P1, will be overwritten. All other records that are linked back to database P1 will also be overwritten. Only metadata records linked back to other databases will be unaffected by a subsequent harvest of P1.
  
  It is also possible to harvest OGC Web Mapping Servers (WMS) and OGC Web Feature Servers (WFS) using the GeognoSIS Manager. Any WMS or WFS metadata records harvested by GeognoSIS Manager will be added to the metadata repository.
- Choose the correct database type (Oracle, PostGIS or Microsoft SQL Server) and enter the database connection details for the database you want to harvest. The Metadata Manager will only harvest tables which are owned by the user entered in this dialog. You may need to use an administrative user and password to ensure that you harvest all of the tables you require.
  
  You will be notified when the harvest is complete:
At this point you should check the metadata records that have been created. Go to the Administration command and press the Records option.

Press the View All Records button to display all of the records in the metadata repository.

From this dialog you can also Export, Edit, View and Delete one or multiple Metadata records.
You can now start to edit the metadata records. You can select a metadata record in the main table view, then use the View and Edit buttons to examine and update the record.
The Save Changes button can be used at any time during the edit process. You may move between panels and use the Save Changes to update all changes made during your edit session. You do not need to Save Changes on every panel, although this is often good practice.
You can access the same interactive view of the metadata through the Map Control Bar.

Use the right mouse button on an overlay created from a database table to display the context sensitive options to View or Edit the metadata record.

The metadata fields are described, where possible, in the interface, but to ensure that your metadata is as accurate as possible, it is recommended that the Gemini 2.1 specification is used for reference. This is available from:

http://location.defra.gov.uk/wp-content/uploads/2010/09/GEMINI2_1_published.pdf

Setting up and managing a metadata repository

Related topics