Without seeing your code I'll guess you don't have your single and double quotes correct.
-----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Micheal Butz Sent: Tuesday, April 01, 2014 2:31 PM To: [email protected] Subject: Re: Rexx substr not retiring string I just did that I assigned the final value to outvar.0 And get a error message positional paramter not valid Sent from my iPhone > On Apr 1, 2014, at 4:24 PM, "Dale R. Smith" <[email protected]> > wrote: > >> On Tue, 1 Apr 2014 14:40:40 -0400, Micheal Butz <[email protected]> >> wrote: >> >> That was it thanks >> >> Sent from my iPhone >> >>> On Apr 1, 2014, at 7:26 AM, Andrew Armstrong <[email protected]> >>> wrote: >>> >>> Missing a . after the stem name. Try: >>> >>> execio * DISKW outds (finis stem outvar. > > Just a word of caution, you should never use "*" for the number of records to > write with DISKW when using the STEM option. > By REXX definition/usage a stem of the form "stemname." should have a count > of the entries in the stem in variable "stemname.0". So in your example, > outvar.0 should contain the number of entries in the outvar. stem. If you > are creating outvar. yourself, then you must keep count of the number of > entries you are creating and set outvar.0 to that number when you are done. > Your EXECIO command should then look like this: > > 'EXECIO' outvar.0 'DISKW OUTDS (FINIS STEM OUTVAR.' > > So why not use "*"? It will work most of the time and some of you are > probably using it now with no problems. From the "TSO REXX Reference" on > EXECIO: > > STEM var-name > the stem of the list of variables from which information is to be > written. To write information from compound variables, which allow for > indexing, the var-name should end with a period, MYVAR., for example. > When three lines are written to the data set, they are taken from > MYVAR.1, MYVAR.2, MYVAR.3. When * is specified as the number of lines > to write, the EXECIO command stops writing information to the data set > when it finds a null line or an uninitialized compound variable. In > this case, if the list contained 10 compound variables, the EXECIO > command stops at MYVAR.11. > > The 0th variable has no effect on controlling the number of lines > written from variables. > . . . . > > Let's say you have a REXX program that reads several different files/members > and writes those files/members to different outputs. You read the > file/member into a stem using DISKR and write the same stem out using DISKW > and "*". > > 'EXECIO * DISKR FILEIN (FINIS STEM REC.' /* Lets say 20 records were > read, rec.0 = 20 */ > 'EXECIO * DISKW FILEOUT (FINIS STEM REC.' /* Writes 20 records out > because rec.21 is unassigned rec.21 = 'REC.21' */ > (allocate new files to FILEIN and FILEOUT) > 'EXECIO * DISKR FILEIN (FINIS STEM REC.' /* Lets say 10 records were > read, rec.0 = 10 */ > 'EXECIO * DISKW FILEOUT (FINIS STEM REC.' /* Writes 20 records out! > rec.1-rec.10 from current file, rec.11-rec.20 from previous file! */ > 'EXECIO' rec.0 'DISKW FILEOUT (FINIS STEM REC.' /* Writes 10 records > out */ > > Of course, you could do a REXX "Drop rec." command between the reads and that > would allow "*" to work, but why bother? > > -- > Dale R. Smith > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, send > email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ______________________________________________________________________ CONFIDENTIALITY NOTICE: This email from the State of California is for the sole use of the intended recipient and may contain confidential and privileged information. Any unauthorized review or use, including disclosure or distribution, is prohibited. If you are not the intended recipient, please contact the sender and destroy all copies of this email. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
