SAS Open
Metadata Architecture enables an administrator to define metadata
objects that are common to one or more SAS client applications. For
example, you can describe data sources and set security that supplements
protections from the host environment and other systems.
In most cases, an administrator
maintains the metadata by using products like SAS Management Console,
SAS Data Integration Studio, or SAS Enterprise Guide. However, an
administrator can also maintain metadata by running a SAS program
in batch or from the SAS windowing environment. The code that can
be submitted in a SAS session uses the SAS metadata language elements.
Many of the
metadata language elements enable you to maintain metadata that defines
a data source. A convention in the SAS Open Metadata Architecture
is to refer to data in terms of SAS libraries, tables, rows, and columns.
-
A data source is defined in metadata
as a table.
-
SAS tables are organized by being
stored in a library.
-
In SAS documentation, a row in
a table is often called an observation, and a column is called a variable.
A SAS Metadata Server
manages access to metadata in SAS metadata repositories. Some of the
metadata language elements can be used to monitor and maintain the
SAS Metadata Server.
This book is a reference
to the metadata language elements. For information about metadata
and SAS Metadata Server administration tasks, see the
SAS Intelligence Platform: System Administration Guide.
The SAS metadata language
elements described in this book include:
Use the system options
to set defaults for metadata access. They are organized into three
groups: connection to the SAS Metadata Server, client encryption,
and resources.
Metadata LIBNAME statement
As with other SAS engines,
an administrator can assign a libref to serve as a shorthand for users.
With the metadata engine, the underlying LIBNAME information is stored
in metadata objects. The metadata engine helps implement security
across an enterprise.
Data set options for the metadata engine
You can apply these
data set options to one table, rather than to an entire library.
You can use the following
procedures to perform many common maintenance tasks on metadata and
the SAS Metadata Server.
-
PROC METALIB automates the creation
and update of table metadata for a specified SAS library. (The SAS
library must be defined in a SAS Metadata Repository using SAS Management
Console or SAS Data Integration Studio, first.)
-
PROC METADATA enables clients to
submit XML-formatted SAS Open Metadata Interface method calls that
read and write metadata objects of all SAS Metadata Model metadata
types from within SAS. It also enables you to issue status requests
that query the SAS Metadata Server’s configuration, the server’s
backup configuration and history, and the server’s availability.
PROC METADATA returns XML output that mirrors the input, except the
requested values are filled in. To process the output with SAS, you
can define an XML map that can be read with the XML LIBNAME engine.
-
PROC METAOPERATE pauses, resumes,
refreshes, backs up, recovers, and stops the SAS Metadata Server.
The DATA step functions
cover the same metadata functionality as PROC METADATA, and return
data to the DATA step, which can then be arranged in a SAS data set.
Because the DATA step functions execute within a DATA step, you can
use the output from one function as the input to another function.
The SAS
commands METABROWSE, METACON, and METAFIND are documented in the online
Help that is available from the SAS windowing environment.