On Sun, 12 Jan 2020 15:19:14 -0600, Lionel B Dyck wrote:
>Thanks for the idea - what I'm trying to avoid is copying (cp) the z/OS
>dataset to OMVS unless it has changed by comparing it's hash to the hash of
>the OMVS copy. Guess either way a copy will have to be done so I might as
>well just copy the z/OS to OMVS regardless.
>
Yup. It might perform better to use "cmp" rather than "cksum". Either
way you need to use both files:
https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.3.0/com.ibm.zos.v2r3.bpxa500/cmp.htm
Rename if different; remove if same.
------------------
On Sun, 12 Jan 2020 15:25:14 -0600, Al Ferguson wrote:
>
>Another option would be to use a REXX that calls the CSF API’s to do this. I
>think this is the coding you would need:
>
>RC=COPIES('FF'x,4); RsC=COPIES('FF'x,4)
>
>ExitDataLnth=COPIES('00'x,4); ExitData=''
>
>RuleArrayCnt=RIGHT(D2C(1),4,'00'x); RuleArray=LEFT('SHA-256',8)
>TextLnth=RIGHT(D2C(LENGTH(Key)),4,'00'x); Text=Key
>CVectorLnth=RIGHT(D2C(128),4,'00'x); CVector=COPIES('00'x,C2D(CVectorLnth))
>HashLnth=RIGHT(D2C(32),4,'00'x); Hash=COPIES('00'x,C2D(HashLnth))
>
>
>ADDRESS "LINKPGM" "CSNBOWH RC RsC ExitDataLnth ExitData", ...
> ...
>CALL CSNBOWH ( ICSF One Way HASH function
>
LINKPGM? CALL? Which? (I suspect LINKPGM if Rexx; CALL if Assembler.)
> ...
> text_length, <=== Integer Length of Text to be HASHed (multiple
> of 128 will work for all 3 SHA's)
> text, <=== String To be HASHed (Pad with ' ' to match
> text_length)
> ...
How is it supposed to tell padding blanks from actual blanks in the data?
Is there a maximum text_length?
Beware line boundary ambiguity.
And you need a step to read the file into the text string.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
