DSM/Tape Catalog Management Programming Manual

SPI Programming Considerations
DSM/Tape Catalog Management Programming Manual—520481-003
3-5
Constructs Involving Multiple Tokens
Extensible Structured Tokens
An extensible structured token consists of a token map and a value that is an
extensible structure. Extensible structured tokens let programs add a number of
parameters or other items of information to a buffer in a single SPI call. They allow
easy initialization of parameter values, and the subsystem can extend these structures
in future releases to provide new features.
The SPI procedures and definition files provide a mechanism to ensure compatibility of
requesters and servers when these tokens are extended. These structures are typically
used to indicate the attributes of an object being operated on and to return status and
statistics information in responses. The token map describes the structure to SPI so
that SPI can provide compatibility between different versions of the structure. Although
an extensible structured token always implies an extensible structure, a token that has
an internal structure—but is not extensible—is still a simple token.
Each field of a structure has either a value assigned by the application or a null value
assigned by SPI. For more information on extensible structured tokens and the
commands that use them, see Section 4, Common Definitions.
Field Types
Each field in an extensible structured token has a specific type, and some of the field
types are defined by SPI. These field types can have names, and the field types
supported by the DSM/TC subsystem have names of the form:
name
identifies the field. For more information, see Section 4, Common Definitions.
Constructs Involving Multiple Tokens
The DSM/TC subsystem supports multiple occurrences of some of the tokens in its
commands. For example, an ALTER TAPEVOLUME command lets multiple
occurrences of the ZMCS-TKN-TAPEVOLUME token name the files that will change
status in the volume catalog or pool.
Multiple tokens can also be constructed as data lists and error lists. Data lists separate
one or more responses within an SPI buffer. These lists consist of an initial list token, a
response, and an end-list token.
Error lists contain an initial list token, an error token, other (optional) tokens explaining
the error, and an end-list token. Error lists can be nested within other error lists.
ZMCS-TYP-name
Note. The DSM/TC subsystem supports data lists and error lists in responses only.