May 18, 2009

Converting Data Sets from WYLBUR’s Edit Format to Another Format

CIT will drop support for NIH WYLBUR on December 31, 2009 (see the Titan News article WYLBUR Will be Retiring). One issue that will affect all WYLBUR users is the format of their data that was saved in WYLBUR. Edit format is the default format used by WYLBUR to store data sets because it uses disk space more efficiently than other formats -- but it is a format that cannot be used by other software. To use your data sets outside of WYLBUR, you will need to convert them to a format other than WYLBUR edit format.

You can convert your edit format data sets to another format using WYLBUR, ISPF, TSO, or batch processing.

Using WYLBUR

The simplest way to change the format is to use WYLBUR itself. First “use” the data set in question, then use the “resave” command to change the format of the data set. The formats available are Card, Print, TSO, Numbered, and Variable. They have these characteristics:

Name	Line numbers	LRECL	RECFM
Card	No	80	FB
Print*	No	133	FB
TSO	Yes	80 (72 plus 8 for line number at end)	FB
Numbered	Yes	80 (72 plus 8 for line number at end)	FB
Variable*	No	1004	VB

*Line numbers can be specified for Print and Variable. However, since ISPF displays line numbers if you open a data set for editing, line numbers are not as important as they are in WYLBUR. For more information on specifying line numbers, see the WYLBUR General Editing manual.

The following example resaves the data set stored in the active file TEST1 in variable format:

resave act test1 variable

Using ISPF or TSO

Another way to change the format is to use a TSO REXX EXEC, either natively through TSO or from ISPF. With this method the reformatted data set is saved separately under another name and you must be sure to go back and delete the original data set. The REXX EXECs shown below will convert data sets. Choose an EXEC that produces a data set with a LRECL large enough to hold the longest line in the output data set.

REXX EXEC	LRECL	RECFM
WYLFMEV*	1004	VB
WYLFMEC	80	FB
WYLFMEP	133	FBA (handles ASA print-control characters)
WYLFME	132	FB
WYLFME1	256	FB

*WYLFMEV is the recommended method since it will handle any edit format data set. Because the variable length record format saves disk space, it will create the most cost-effective data set.

You can use these EXECs in ISPF or on the TSO command line.

Using ISPF

In ISPF, select option 3.4, and type the EXEC name on the line that lists the data set.

In the example below, the EXEC name is shown in red.



ISPF will create the new data set and give it a name of the format originalname. followed by a character. The system will return a message that tells you the name of the new data set. If you want to control the name of the new data set, specify this by adding the new data set name after the current one. Once you have verified that the new data set is the way you want it, remember to delete the original data set.

In the example below, the EXEC name is shown in red and the user specifies a name for the new data set.



Using the TSO Command Line

The syntax is: TSO execname input output

where

execname is the name of one of the REXX EXECs listed in the table
input is the name of an edit format data set
ouput is the name of the new data set

For example: TSO WYLFMEV userid.test userid.newtest

Don’t forget to delete the original data set.

Using Batch Processing

The EDSUTIL package converts data sets from edit format to non-edit format. You can specify the LRECL and RECFM. The LRECL value specified must be long enough to hold the longest line in the data set.

EDSUTIL will continue to be supported after WYLBUR retires. For information on using EDSUTIL, please see the online versions of the Titan Batch Processing manual or the Edit Format Utility Package manual.

For Assistance

For additional assistance contact the NIH Help Desk at http://ithelpdesk.nih.gov/support or 301-496-4357, 301-496-8294 (TTY), or toll free at 866-319-4357.

Back to top of page