|
|
| Author |
Message |
PeD
Active User
Joined: 26 Nov 2005
Posts: 297
Location: Belgium
|
|
| Back to top |
|
 |
References
|
 Posted: Wed Apr 30, 2008 3:27 am Post subject: Re: |
 |
|
|
|
Bill O'Boyle
Active User
Joined: 14 Jan 2008
Posts: 296
Location: Orlando, FL, USA
|
|
|
|
Sincere apologies to one and all regarding the GETLGTH Assembler sub-program previously posted. I must have had my head up my A$$ because this previous version WILL NOT WORK and with that, please use the replacement GETLGTH sub-program below.
The Caller's parmlist of four arguments remains the same.
| Code: |
GETLGTH CSECT
USING *,R3 INFORM ASSEMBLER
SAVE (14,12) SAVE REGISTERS
LA R3,0(,R15) R3 IS BASE
LM R5,R8,0(R1) PARM-ADDRESSABILITY
LA R5,0(,R5) CLEAR TOP-BIT
LA R6,0(,R6) SAME
LA R7,0(,R7) SAME
LA R8,0(,R8) SAME
XC 0(4,R8),0(R8) ENSURE X'00'S
LR R1,R6 LOAD END-ADDRESS
SLR R1,R5 REDUCE BY START-ADDRESS
AL R1,0(,R7) ADD LAST-FIELD'S LGTH
LA R15,16 SET 'BAD' RETURN-CODE
CHI R1,0 NEGATIVE?
BL RTN2CLLR YES, RETURN TO CALLER
LA R15,8 RESET RETURN-CODE
BE RTN2CLLR RETURN WHEN ZERO
ST R1,0(,R8) STORE RESULT IN 4TH-PARM
SLR R15,R15 ALL IS WELL
RTN2CLLR EQU *
RETURN (14,12),RC=(15) RESTORE AND RETURN
GETLGTH AMODE 31
GETLGTH RMODE ANY
YREGS REGISTER-EQUATE MACRO
END , END 'GETLGTH'
|
Slowly wiping egg from face.... 
Bill |
|
| Back to top |
|
|
sudhaaa
New User
Joined: 24 Mar 2005
Posts: 47
|
|
|
|
If you have Fileaid, use option 3.8
Thanks,
Sudhaaa |
|
| Back to top |
|
|
kirankumarj
New User
Joined: 18 Dec 2007
Posts: 39
Location: delaware
|
|
|
|
| Option 8 (XREF) will give you the length !! |
|
| Back to top |
|
|
Pankaj Gupta
New User
Joined: 07 May 2008
Posts: 27
Location: Bangalore
|
|
|
|
| Quote: |
| DISPLAY LENGTH OF yourcopybook |
Hello, but I am having a doubt about this syntaxes. The compiler it is not compiling "LENGTH OF". Also before I am mentioned that it is also not liking the EVALUATES syntax. Is the LENTH OF the valid COBOLs? |
|
| Back to top |
|
|
dick scherrer
Global Moderator
Joined: 23 Nov 2006
Posts: 8033
Location: 221 B Baker St
|
|
|
|
Hello,
| Quote: |
| The compiler it is not compiling "LENGTH OF". |
And it should not/will not. . . .
That suggestion was for using fileaid, not compiling a program. |
|
| Back to top |
|
|
Pankaj Gupta
New User
Joined: 07 May 2008
Posts: 27
Location: Bangalore
|
|
|
|
| acevedo wrote: |
what about ?
| Quote: |
| DISPLAY LENGTH OF yourcopybook |
|
Mr dick, I am sure this is meant to be the COBOL |
|
| Back to top |
|
|
dick scherrer
Global Moderator
Joined: 23 Nov 2006
Posts: 8033
Location: 221 B Baker St
|
|
|
|
Hello,
| Quote: |
| Mr dick, I am sure this is meant to be the COBOL |
Why so sure? Your certainty is misplaced if you mean the cobol compiler.
There is no "display length of" in the IBM mainframe cobol language.
The suggestion was for displaying the length using fileaid. |
|
| Back to top |
|
|
Pankaj Gupta
New User
Joined: 07 May 2008
Posts: 27
Location: Bangalore
|
|
|
|
I am really not being so sure. In COBOL we are having the DISPLAY verb and we are also having "LENGTH OF".
Why is then DISPLAY LENGTH OF VARIABLE not being COBOL?
Is DISPLAY also being a command of the FILEAID? If this is being so then I have never been seeing it. |
|
| Back to top |
|
|
PeD
Active User
Joined: 26 Nov 2005
Posts: 297
Location: Belgium
|
|
|
|
I am back to this question and back to my answer.
The handy tool of Mark H. is just perfect for that.
On top of that the poster seems to want to calculate the length outside a program, just online in ISPF.
But maybe noone does the effort to go inside the link and to open the first thread of Mark H.
Ohh these days ...
Cheers
Pierre |
|
| Back to top |
|
|
dick scherrer
Global Moderator
Joined: 23 Nov 2006
Posts: 8033
Location: 221 B Baker St
|
|
|
|
Hello,
| Quote: |
In COBOL we are having the DISPLAY verb and we are also having "LENGTH OF".
Why is then DISPLAY LENGTH OF VARIABLE not being COBOL?
|
You cannot "mix and match" as you seem to want to do. It is not part of cobol because the people responsible for the development of the language have seen no reason to include this.
You must use the syntax as intended (which is in the manual which you are not using). At the top of the web page is a link to "IBM Manuals" - the cobol manuals are the first group. If you look in the manuals you will not find "display length of" as an opton. You will also not find (and cannot compile) "MULTIPLY FILE-SECTION BY SPECIAL-NAMES GIVING WORKING-STORAGE" even though all of these too are valid parts of the language.
If your goal is to get the length and you have not looked at Pierre's link, you should do so. You might also get help from someone who speaks assembler and look at Bill's suggestion.
If you want the compiler to be changed to perform as you believe it should, you will need to manage some disappointment  |
|
| Back to top |
|
|
Srinivas George
New User
Joined: 31 May 2008
Posts: 4
Location: bangalore
|
|
|
|
Grettings Pankaj Gupta,
While "ADD WORKING-STORAGE TO FILE-SECTION" is certainly yes most amusing, you are not so wrong in trying to use
| Code: |
| DISPLAY LENGTH OF VARIABLE |
it is yes but one step away. We must certainly assign LENGTH OF to a variable and then yes we can display the variable. I elaborate now in the code for you:
| Code: |
SET VARIABLE TO LENGTH OF TOP-VARIABLE-OF-COPYBOOK
DISPLAY VARIABLE |
This you will find will certainly compile. Unless you are yes using the long teethed OSVS COBOL.
I am certainly intrigued by the use of an assembler routine to determine the length of a copybook if we are yes using the OSVS COBOL. In this situation we can certainly say this:
| Code: |
EXAMINE TOP-LEVEL-OF-COPYBOOK
TALLYING VARIABLE FOR CHARACTERS |
NO, We cannot. . .
After this code has executed then yes the variable VARIABLE will certainly contain the length of the copybook. An assembler subroutine seems certainy over the lid. |
|
| Back to top |
|
|
dick scherrer
Global Moderator
Joined: 23 Nov 2006
Posts: 8033
Location: 221 B Baker St
|
|
|
|
Hello Srinivas George,
Before posting did you do testing - on an IBM mainframe?
I didn't trust my memory, so i looked at the COBOL manuals available via the "IBM Manuals" link at the top of the page and as i recalled, EXAMINE is no longer supported. To do the same thing, INSPECT is now used. If you code EXAMINE and use Enterprise (the most current) COBOL, EXAMINE causes this:
| Code: |
| 955 IGYPS2072-S "EXAMINE" WAS INVALID. SKIPPED TO THE NEXT VERB |
If one uses INSPECT (with Enterprise COBOL) and tries to TALLY the length of the copybook, the result is not (always?) the correct length.
If you are going to post "solutions", you need to test them and the test must be done in an environment covered by these forums (i.e. only IBM mainframe - usng the IBM compilers). |
|
| Back to top |
|
|
Srinivas George
New User
Joined: 31 May 2008
Posts: 4
Location: bangalore
|
|
|
|
Mr Dick, my humble apologies. By a slip of the keboard I have certainly typed "EXAMINE" when I meant to type "INSPECT".
On OSVS CObol, as my post was certainly concerning, this WILL give the correct length. Note that yes you should certainly be setting the counter to the zero first:
| Code: |
MOVE ZERO TO VARIABLE
INSPECT TOP-LEVEL-OF-COPYBOOK
TALLYING VARIABLE FOR CHARACTERS |
After this code the field "VARIABLE" WILL contain yes the correct length of everything that is certainly below the top level. This code I am using yes every day. It is certainly correct and certainly works.
I apologise certainly once more for typing the wrong verb. |
|
| Back to top |
|
|
dick scherrer
Global Moderator
Joined: 23 Nov 2006
Posts: 8033
Location: 221 B Baker St
|
|
|
|
Hello,
| Quote: |
I apologise certainly once more for typing the wrong verb.
|
Not a problem - we all make that mistake (if we've coded long enough to remember EXAMINE)
As far as the INSPECT, when i ran a couple of incomplete tests, the value of length worked for a very simple copybook, but i did not get what i expected for a complicated copybook (it is over 14k in length and contains many redefinitions and several repeating groups). Time permitting, i'll try a bit more today.
I mentioned using an IBM mainframe because we have received posts for unix and windows that have compiler features (extensions) the ibm cobol compiler does not support. Sorry for the misunderstanding |
|
| Back to top |
|
|
|
|
