These were created by someone else in 1992. Yeah, I know there should be ONE
macro that accepts more parms, but it has never been worth the time to do the
enhancement.
Fun fact: just last week I found a bug! The macros considered lines with these
two keys to be duplicate lines:
7475269191210720
7475269191211080
REXX mavens will know why. I've fixed that bug.
Anyway, here's the first one, which excludes all but the first line:
/* rexx - exclude all but first line of any set of duplicate lines,
(checks portion of the line within the current bounds) */
ADDRESS ISREDIT
"MACRO"
ADDRESS ISPEXEC "CONTROL ERRORS RETURN"
progress_interval = 1.0
x = time('E') /* start elapsed time clock */
"NUMBER OFF"
"(NUMLINES) = LINENUM .ZLAST"
linenbr = 0
"(LEFTBND,RIGHTBND) = BOUNDS"
previous_key = nextline()
current_key = nextline()
do while rc = 0
if time('E') > progress_interval then do
call progress_message linenbr,numlines
x = time('R')
end
if previous_key == current_key then do
"LABEL" linenbr "= .A"
"EXCLUDE P'=' .A .A FIRST"
end
previous_key = current_key
current_key = nextline()
end
exit
nextline: procedure,
expose linenbr leftbnd rightbnd rc
linenbr = linenbr + 1
"(THISLINE) = LINE" linenbr
return substr(thisline,leftbnd,rightbnd - leftbnd + 1)
progress_message: procedure
address ISPEXEC
arg line_number,max_lines
complete = (line_number * 100) % max_lines
zerrsm = ''
zerrlm = 'Processing...' complete'% complete'
zerralrm = 'NO'
zerrhm = '*'
zerrtp = 'NOTIFY'
zerrwn = 'NORESP'
"CONTROL DISPLAY LOCK"
"DISPLAY MSG(ISRZ003)"
return
And here's the second, that excludes the entire set. This one doesn't have the
progress message logic, but that's OK; the system is fast enough that you have
to be working on a huge file before you ever see a progress message.
/* REXX - exclude all duplicate lines, (checks portion of the line
within the current bounds) */
ADDRESS ISREDIT
"MACRO"
ADDRESS ISPEXEC "CONTROL ERRORS RETURN"
lineprv = 0
linenbr = 1
"(LEFTBND,RIGHTBND) = BOUNDS"
previous_key = nextlprv()
current_key = nextlnbr()
do while rc = 0
if previous_key == current_key then do
"LABEL" linenbr "= .A"
"EXCLUDE P'=' .A .A FIRST"
"LABEL" lineprv "= .C"
"EXCLUDE P'=' .C .C FIRST"
end
previous_key = current_key
current_key = nextlnbr()
lineprv = lineprv + 1
end
exit
nextlnbr: procedure,
expose linenbr leftbnd rightbnd rc
linenbr = linenbr + 1
"(THISLINE) = LINE" linenbr
return substr(thisline,leftbnd,rightbnd - leftbnd + 1)
nextlprv: procedure,
expose lineprv leftbnd rightbnd
lineprv = lineprv + 1
"(THISLINE) = LINE" lineprv
return substr(thisline,leftbnd,rightbnd - leftbnd + 1)
-----Original Message-----
From: IBM Mainframe Discussion List <IBM-MAIN@LISTSERV.UA.EDU> On Behalf Of
Schmitt, Michael
Sent: Tuesday, January 14, 2025 12:08 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: ISPF edit macro to sort and delete duplicate records
I have macros that exclude duplicate records, comparing a key within the
current bounds. Then you can delete all X.
One macro excludes all but the first of each duplicate set, the other excludes
all of the set. Which one do you want?
-----Original Message-----
From: IBM Mainframe Discussion List <IBM-MAIN@LISTSERV.UA.EDU> On Behalf Of
Peter Ten Eyck
Sent: Tuesday, January 14, 2025 12:05 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: ISPF edit macro to sort and delete duplicate records
Greetings,
Anyone know of an ISPF edit macro to sort and delete duplicate records within
PDS member or dataset?
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
