When you access
ADABAS data through the
SAS/ACCESS interface, the interface view engine
maps the
ADABAS data into
SAS observations. You need to be aware of how the interface view
engine maps multiple-value fields and periodic groups. That is,
multiple-value field occurrences are mapped to multiple SAS variables,
and periodic group occurrences are mapped to multiple SAS observations.For example, suppose
an
ADABAS file has the data
fields and values shown in the following output. LASTNAME is an elementary
field, JOBTITLE is a multiple-value field, and EDUCATION is a periodic
group consisting of the data fields COLLEGE, DEGREE, and YEAR.
ADABAS Data
______________________________________________________________
| LASTNAME | JOBTITLE | EDUCATION |
|___________|___________________|_____________________________|
| Reid | Systems Analyst | Purdue | BA | 1973 |
|___________|-------------------|-----------------------------|
| DBA | Harvard | MBA | 1975 |
|___________________|_____________________________|
The interface view engine
would map the
ADABAS data
into two SAS observations, as shown in the following output.
ADABAS Data Mapped
into SAS Observations
LASTNAME JOBTITL1 JOBTITL2 COLLEGE DEGREE YEAR
Reid Systems Analyst DBA Purdue BA 1973
Reid Systems Analyst DBA Harvard MBA 1975
If you were browsing
the
ADABAS data, such as
with the FSVIEW procedure, the results would be similar to
ADABAS Data Mapped into SAS Observations , with LASTNAME, JOBTITL1, and JOBTITL2 repeated for each
set of COLLEGE, DEGREE, and YEAR values. Actually though, the value
Reid is stored in the
ADABAS file only once. For retrievals, the results are straightforward.
When updating, however, you need to keep in mind how the interface
view engine maps multiple-value fields and periodic groups.
Suppose you want to
change the spelling of a last name using the FSVIEW procedure. To
change Reid to Reed, all you need to do is type REED over one of
the REID values, and, with a single update operation, the last names
are all changed. On the other hand, suppose you want to delete an
observation for Reid using the FSEDIT procedure. Each observation
for his job titles and education data would display his last name.
If you deleted an observation. For example, the one for Purdue, the
deletion would not affect the last name or the job title data, but
the Purdue observation would be gone. For more information and an
example of deleting an observation from
ADABAS data, see
Browsing and Updating ADABAS Data.