|
View previous topic :: View next topic
|
| Author |
Message |
harish5002
New User
Joined: 21 Jul 2010
Posts: 23
Location: chennai
|
|
|
|
Hi,
Can u please guide me how to write the above code. I got the same requirement to find the lines of code for more than 200 programs. So please suggest me any jcl or how to write the code for ISPF panel.
Thanks in advance to all.... |
|
| Back to top |
|
 |
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10888
Location: italy
|
|
|
|
why the fluck reinvent the wheel, aren' t ISPF statistics good enough
( whatever the use for them )
anyway the LOC ( lines of code ) counting is a murky issue ... what do You define as line of code ???
some smart*** might raise its LOC productivity by simply reformatting a program ...
instead of
| Code: |
| call something ( parm1, parm2, ..., parmn ) |
| Code: |
call something ( parm1,
parm2,
...,
parmn)
|
would raise the productivity by 300%
 |
|
| Back to top |
|
|
Bill Woodger
Moderator Emeritus
Joined: 09 Mar 2011
Posts: 7309
Location: Inside the Matrix
|
|
|
|
Please don't tailgate an old topic, but instead start a new one.
What "programs" do you have? Is it a library with ISPF statistics? What actual use does anyone feel that this is?
I can write like this:
| Code: |
| MOVE A TO B. MOVE B TO C. MOVE C TO D. |
or
| Code: |
MOVE
A
TO
B
MOVE
B
TO
C
MOVE
C
TO
D |
Or somewhere "in between". Makes not one jot of difference to what the program does.
Nor does this:
but it can make the code easier to read. If I were to be paid by "lines" I'd consider making my programs much "easier to read" :-) |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10888
Location: italy
|
|
|
|
Hi Bill!
our low esteem for the bean counters made our posts cross over
but I strongly agree that proper reformatting for readability should have the precedence
( after correctness naturally )
but even the dummyiest LOC counter would ignore the completely blank lines  |
|
| Back to top |
|
|
Bill Woodger
Moderator Emeritus
Joined: 09 Mar 2011
Posts: 7309
Location: Inside the Matrix
|
|
|
|
Hi enrico!
I was trying to get across the lack-of-use in "counting" lines, of course.
Blank lines and comments should "go" to approach something sensible, but then also there should be code to deal with what you and I used as examples. So a "micky mouse" counting program, to have worth, has to become the first stage of a compiler. Then you go through all that and the "user" says "Oh, I just wanted a figure for a talk I'm giving to the Board, it doesn't have to be accurate". It can never be "accurate". Sometimes it may be best to code with only a few lines, others it would be best to code with more. You can't just "count" lines and produce any meaningful output, no matter how accurate.
I think we agree on that already...
So if it is a "pointless" task, and already has ISPF Statistics, it is a "good" time to use them :-) |
|
| Back to top |
|
|
don.leahy
Active Member
Joined: 06 Jul 2010
Posts: 765
Location: Whitby, ON, Canada
|
|
|
|
| I like code metrics. They're kind of like sports statistics, fun to argue about but essentially meaningless. |
|
| Back to top |
|
|
Gary McDowell
Active User
Joined: 15 Oct 2012
Posts: 139
Location: USA
|
|
|
|
So for curiosity, assume these 200 PDS Members were added into your system with no ISPF statistics (STATS OFF) and you were asked to provide how many lines of code were in each Member - how would you do it in CLIST/REXX or some JCL Utility?
Of course even turning the STATS ON will not give you the Size (amount of lines) for the member [need to make a change to see Size]. |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10888
Location: italy
|
|
|
|
I beg to disagree
it would be enough to go to option 3.5
| Code: |
Menu RefList Utilities Help
──────────────────────────────────────────────────────────────────────────────
Reset ISPF Statistics Enter required field
Option ===>
R Reset (create/update) ISPF statistics D Delete ISPF statistics
New Userid . . . . . . (If userid is to be changed)
New Version Number . . (If version number is to be changed)
SCLM Setting Enter "/" to select option
3 1. SCLM 2. Non-SCLM 3. As is / Reset Mod Level
/ Reset Sequence Numbers
ISPF Library: / Reset Date/Time
Project . . . / Reset Number of Lines
Group . . . .
Type . . . .
Member . . . (Blank or pattern for member selection
list, "*" for all members)
Other Partitioned Data Set:
Name . . . . . . .
Volume Serial . . . (If not cataloged)
Data Set Password . . (If password protected) |
|
|
| Back to top |
|
|
Gary McDowell
Active User
Joined: 15 Oct 2012
Posts: 139
Location: USA
|
|
|
|
Well, that's true 3.5 can turn-on the STATS.
But in the spirit of the original poster how would this be done in CLIST/REXX or some JCL Utility? I can't think of a JCL utility to present the PDS Members with the count/Size. |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10888
Location: italy
|
|
|
|
| Quote: |
| But in the spirit of the original poster how would this be done in CLIST/REXX or some JCL Utility? |
that' one of my pet peeves
if somebody asks for a solution using a predetermined tool
it means that such somebody had completed the due diligence
tool comparison, pro's, con' s,
and has checked the availability of the proper skills
so it should not need to ask for anything.
and nobody on a forum will waste time reinventing the wheel and writing code
especially for a unreasonable requirement |
|
| Back to top |
|
|
don.leahy
Active Member
Joined: 06 Jul 2010
Posts: 765
Location: Whitby, ON, Canada
|
|
|
|
One easy way to determine the size of each member of a PDS is to use SUPERC's SRCHFOR to find a string that you are certain will not be found in the PDS. The LNFMTO option (List "not found" member totals only) will create a summary listing the member name plus the number of records searched in each member (which is of course the size). For example:
| Code: |
//SEARCH EXEC PGM=ISRSUPC,
// PARM=(SRCHCMP,
// 'ANYC,LNFMTO')
//NEWDD DD DSN=XXXX.MASTER.SOURCE
// DISP=SHR
//OUTDD DD SYSOUT=(*)
//SYSIN DD *
SRCHFOR X'FF'
/* |
The report looks like this:
| Code: |
MEMBER-SEARCHED LINES-FOUND LINES-PROC
CAL11M 0 62
CAL11T 0 62
CAL11X 0 62
XXXSABEND 0 95
XXXACVOL 0 1209
|
The summary at the end of the report tells you the total number of lines searched. |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10888
Location: italy
|
|
|
|
| pretty smart trick |
|
| Back to top |
|
|
Gary McDowell
Active User
Joined: 15 Oct 2012
Posts: 139
Location: USA
|
|
|
|
| Quote: |
nobody on a forum will waste time reinventing the wheel and writing code
especially for a unreasonable requirement |
Actually our IT Testing area has a similar requirement as the original poster looking for "something" (i.e. clist/REXX, JCL Utility) to provide ID Stats of the last person to make a member update along with Size/count of the member(s). We want it reported (email) on a daily basis. Of course it's easy to do this manual process by sorting PDS by Change date in descending order but I was looking for an automated way to produce the report since we have 50+ PDS datasets to review.
We never found that "something" to get us the PDS STATistics we needed automatically. We currently have a work-around where one of my contract programmers on a daily basis does the manual process mentioned above. In our case, this is not an unreasonable requirement.
I probably should have just started a new post but as I mentioned this post was very similar to our requirement.
Thanks. |
|
| Back to top |
|
|
Bill Woodger
Moderator Emeritus
Joined: 09 Mar 2011
Posts: 7309
Location: Inside the Matrix
|
|
|
|
| IEBPTPCH and then whatever you like. |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10888
Location: italy
|
|
|
|
| Quote: |
| We never found that "something" to get us the PDS STATistics we needed automatically. |
ISPF updates the statistics AUTOMATICALLY
to do automatically something according to Your whims is asking too much
ISPF provides the LM.... and LMM.... services to extract the statistics
pretty easy to do it
here is a posts discussing some of it
www.ibmmainframes.com/viewtopic.php?t=52204&highlight=lmmlist
and
did not take more than 10 minutes to test a silly snippet
| Code: |
EDIT ENRICO.ISPF.EXEC(ZSTATS) - 01.01 Columns 00001 00072
Command ===> Scroll ===> CSR
****** ***************************** Top of Data ******************************
000001 /*REXX - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - */
000002 /* */
000003 /* - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - */
000004 Trace "O"
000005
000006 Parse Source _sys _how _cmd .
000007
000008 If Sysvar(SYSISPF) \= "ACTIVE" Then Do
000009 Say left(_cmd,8)"- Ispf is not active. Command not executed"
000010 Exit 4
000011 End
000012
000013 call $ispex "CONTROL ERRORS RETURN"
000014
000015 if $isred("MACRO (ZPARMS) NOPROCESS ") = 0 then do
000016 _parms = strip(translate(zparms))
000017 end
000018 else do
000019 zerrsm = "Invocation ERROR"
000020 zerrlm = left(_cmd,8)"- Must be invoked as an edit macro"
000021 call $ispex "SETMSG MSG(ISRZ002) "
000022 Exit 1
000023 end
000024
000025 call $isred "(DATASET) = DATASET"
000026 call $isred "(CURRNT) = MEMBER"
000027 call $isred "(DATAID) = DATAID"
000028
000029 lm0_rc = $ispex("LMOPEN DATAID("dataid") OPTION(INPUT) ")
000030
000031 member = ""
000032
000033 lmmopts = "OPTION(LIST) STATS(YES)"
000034 lmmlist = "LMMLIST DATAID("dataid") MEMBER(MEMBER) " lmmopts
000035 do while ($ispex(lmmlist) = 0 )
000036 say left(member,8) left(zluser,8)
000037 end
000038
000039 lm0_rc = $ispex("LMMLIST DATAID("dataid") OPTION(FREE) " )
000040 lm0_rc = $ispex("LMCLOSE DATAID("dataid") " )
000041
000042 Exit
000043
000044 /* */ */
000045 $ispex:
000046 isp_0tr = trace("O")
000047 Address ISPEXEC arg(1)
000048 isp_0rc = rc
000049 trace value(isp_0tr)
000050 return isp_0rc
000051
000052 /* */
000053 $isred:
000054 isr_0tr = trace("O")
000055 Address ISREDIT arg(1)
000056 isr_0rc = rc
000057 trace value(isr_0tr)
000058 return isr_0rc
000059
000060 /* */
000061 $tsoex:
000062 tso_0tr = trace("O")
000063 Address TSO arg(1)
000064 tso_0rc = rc
000065 trace value(tso_0tr)
000066 return tso_0rc
000067
****** **************************** Bottom of Data **************************** |
just as a POC I displayed only the ZLUSER
just a couple more statements to check if the stats are ON for a member
and display all the relevant stats fields |
|
| Back to top |
|
|
Gary McDowell
Active User
Joined: 15 Oct 2012
Posts: 139
Location: USA
|
|
|
|
| Bill Woodger wrote: |
| IEBPTPCH and then whatever you like. |
Thanks, but IEBPTPCH does not give STATistics. It reprots the PDS Member data. |
|
| Back to top |
|
|
Gary McDowell
Active User
Joined: 15 Oct 2012
Posts: 139
Location: USA
|
|
|
|
| Quote: |
| ISPF updates the statistics AUTOMATICALLY |
By AUTOMATICALLY, as my post noted, I was referring to an automated email report we would like produced daily - not MANUALLY reviewing 50+ PDS datasets for the STATistics as we do today.
| Quote: |
| to do automatically something according to Your whims is asking too much |
Too funny! I'm in the companies IT Automated Testing group which is why I was looking for "something" for our requirement.
Hey, thanks for the ZSTATS REXX. I'll give a try hopefully soon. |
|
| Back to top |
|
|
Bill Woodger
Moderator Emeritus
Joined: 09 Mar 2011
Posts: 7309
Location: Inside the Matrix
|
|
|
|
| Ah. Didn't notice you were an "interloper" with a different question, I just thought you were the Devil's Advocaat :-) |
|
| Back to top |
|
|
Pedro
Global Moderator
Joined: 01 Sep 2006
Posts: 2594
Location: Silicon Valley
|
|
|
|
| Quote: |
| to provide ID Stats of the last person to make a member |
If this is an accountability issue... you should be aware that the userid and time of last update are not secure fields. Anybody with update authority can set bogus information in those fields.
You can use the LMMSTATS service to set ISPF statistics. Or just use ISPF option 3.5. |
|
| Back to top |
|
|
Gary McDowell
Active User
Joined: 15 Oct 2012
Posts: 139
Location: USA
|
|
|
|
| Yea, I am aware date/time of last update and userID are not secure fields. Of course setting bogus information for these datasets in our company is also grounds for dismissal. |
|
| Back to top |
|
|
|
|
