Edit Compare between zOS and Unix Files
Karlheinz Wittemann
Paul Gilmartin
> Hallo,
> what i did is: Write some records to an zos file (PS or PO-Member) and copy this to an unix file. When I do a comp in edit, I get a result, that all blank-lines are different between zos and unix.
What utility did you use to copy PS/PO to UNIX?
Were you unable to write directly to the UNIX file, bypassing the copy?
> I tried a lot of different things, at least to interpret the screen buffer but this is horrible and without sense if lrecl is several 100 byte long.
> Any idea?
Long ago, the Data Management manual stated that the minimum allowed
value for the count in an RDW was 5. ISPF accommodated that by padding
any empty line with a single blank. Later, Using Data Sets relaxed
that minimum to 4. ISPF long, perhaps still, retained the behavior
to comply with a rule no longer in force.
Programmers familiar with desktop editors such as vi and Notepad++
expect editors not to treat blanks specially and perceive any
meddling with blanks by an editor as data corruption, causing
false positives in diff.
--
gil
Marvin Knight
When you Edit the Unix file make sure the “Preserve VB record length” field is selected on the Edit Entry panel.
EDIT Entry Panel
Object Name:
/u/knightm/copy1
Initial Macro . .
Profile Name . . . (Blank defaults to Type)
Format Name . . .
Panel Name . . . . (Leave blank for default)
Record Length . .
Line Command Table
Options Data Encoding
/ Confirm Cancel/Move/Replace 1. ASCII
EDIT Mixed Mode 2. UTF-8
/ Preserve VB record length
Warn on First Data Change
Press ENTER to continue. Press CANCEL to cancel action.
Marvin Knight
From: ispf-...@nd.edu <ispf-...@nd.edu>
On Behalf Of Karlheinz Wittemann
Sent: Tuesday, April 2, 2024 5:28 AM
To: ispf-...@nd.edu
Subject: [EXTERNAL] [ISPF-L] Edit Compare between zOS and Unix Files
Hallo, what i did is: Write some records to an zos file (PS or PO-Member) and copy this to an unix file. When I do a comp in edit, I get a result, that all blank-lines are different between zos and unix. I tried a lot of different things, at
ZjQcmQRYFpfptBannerStart
|
ZjQcmQRYFpfptBannerEnd
--
You received this message because you are subscribed to the Google Groups "ISPF discussion list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
ispf-l-list...@nd.edu.
To view this discussion on the web visit https://groups.google.com/a/nd.edu/d/msgid/ispf-l-list/1569001413.249217.1712053704671%40email.df.eu.
Paul Gilmartin
> When you Edit the Unix file make sure the “Preserve VB record length” field is selected on the Edit Entry panel.
> EDIT Entry Panel
I believe that option, faithful to its name, will add or remove
a corresponding number of trailing blanks when characters are
deleted or inserted earlier in the record in order to
“Preserve VB record length”. This is not the behavior that
programmers have come to expect with vi and desktop editors,
which leave blanks unchanged when changes are made elsewhere in
a record.
Is there any option to leave blanks unchanged when changes
are made elsewhere in a record?
> *From: Karlheinz Wittemann
> *Sent:* Tuesday, April 2, 2024 5:28 AM
> Hallo, what i did is: Write some records to an zos file (PS or PO-Member) and copy this to an unix file. When I do a comp in edit, I get a result, that all blank-lines are different between zos and unix. I tried a lot of different things, at least to interpret the screen buffer but this is horrible and without sense if lrecl is several 100 byte long.
gil
Marvin Knight
Marvin
You received this message because you are subscribed to the Google Groups "ISPF discussion list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ispf-l-list...@nd.edu.
Paul Gilmartin
> ISPF Edit removes trailing spaces from Variable records when it writes the file. For a UNIX file there is no requirement for at least 1 blank character. So ISPF removes all blanks. For non-unix files ISPF keeps at least 1 space.
Where is there a requirement "for at least 1 blank"? For example, in:
<https://www.ibm.com/docs/en/zos-basic-skills?topic=set-data-record-formats>,
I read:
V (Variable)
This format has one logical record as one physical block.
A variable-length logical record consists of a record
descriptor word (RDW) followed by the data. The record
descriptor word is a 4-byte field describing the record.
The first 2 bytes contain the length of the logical record
(including the 4-byte RDW). The length can be from 4 to
32,760 bytes.
If the length of the logical record is 4 bytes, it consists
solely of the RDW and no blank should be required.
--
gil
Marvin Knight
With UNIX it will actually remove the 1 blank.
"Preserve VB record length" will keep the existing RDW unless you make the length longer in Edit by adding non blank data that goes past the existing RDW length.
Marvin
-----Original Message-----
From: 'Paul Gilmartin' via ISPF discussion list <ispf-...@nd.edu>
Sent: Wednesday, April 3, 2024 5:24 PM
To: ispf-...@nd.edu
Subject: [EXTERNAL] Re: [ISPF-L] Edit Compare between zOS and Unix Files
You received this message because you are subscribed to the Google Groups "ISPF discussion list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ispf-l-list...@nd.edu.
Paul Gilmartin
> Edit has always kept at least 1 blank for a non unix file.
Why?
--
gil
Marvin Knight
I don’t know if it is still a DFSMS restriction, but the JCL manual still shows LRECL has a minimum length of 5
https://www.ibm.com/docs/en/zos/3.1.0?topic=lp-subparameter-definition-3
Note: When RECFM is F or U, the length must not exceed DCB BLKSIZE. For RECFM=D or V, the length must be a minimum of 5 and a maximum of BLKSIZE minus 4 to account for the 4 byte record descriptor word (RDW) preceding the data in every record. For RECFM=VS, the length can exceed BLKSIZE. For unblocked records when DCB RKP=0, the length is for only the data portion of the record. LRECL=0 is valid only for RECFM=U.
Marvin
-----Original Message-----
From: 'Paul Gilmartin' via ISPF discussion list <ispf-...@nd.edu>
Sent: Thursday, April 4, 2024 10:04 PM
To: ispf-...@nd.edu
Subject: [EXTERNAL] Re: [ISPF-L] Edit Compare between zOS and Unix Files
You received this message because you are subscribed to the Google Groups "ISPF discussion list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ispf-l-list...@nd.edu.
Paul Gilmartin
> Gil,
> I don’t know if it is still a DFSMS restriction, but the JCL manual still shows LRECL has a minimum length of 5
That's irrelevant. It has always been possible for individual records
with RECFM=V[B] to have RDW values less than LRECL, down to the
DFSMS minimum documented in Using Data Sets as 4. Edit is enforcing a
"rule" that doesn't exist.
This thread started when a user edited Unix and PO/PS files expecting
similar results but discovered inconsistencies in the treatment of
blanks. ISPF should remedy this inconsistency by treating blanks rhe
same in PS/PO data sets rather than enforcing an imaginary rule.
How many users have found it beneficial that with PRESERVE checked,
when characters are inserted or deleted earlier in the line, Edit
adjusts trailing blanks to "preserve" the value of the RDW?
<https://www.ibm.com/docs/en/zos/3.1.0?topic=records-record-descriptor-word-rdw>
... The first 2 bytes contain the length (LL) of the logical record
(including the 4-byte RDW). The length can be from 4 to 32,760.
(32,760? 32,756 because of the BDW? BDW-4?)
--
Thanks,
gil
Roger Bolan
--
You received this message because you are subscribed to the Google Groups "ISPF discussion list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ispf-l-list...@nd.edu.
To view this discussion on the web visit https://groups.google.com/a/nd.edu/d/msgid/ispf-l-list/6dd08049-b6c6-4d8f-abbe-1277299ca763%40AIM.com.
Paul Gilmartin
> The only time I really find it necessary to preserve trailing blanks is if you're editing a file with architected structured fields like AFP records. That's because some fields have defined lengths and the AFP structured field itself has an encoded length.
>
"AFP" = "Advanced Function Printing"?
How does AFP handle variable width fonts, such as the popular Utf-8?
--
gil
Roger Bolan
--
You received this message because you are subscribed to the Google Groups "ISPF discussion list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ispf-l-list...@nd.edu.
To view this discussion on the web visit https://groups.google.com/a/nd.edu/d/msgid/ispf-l-list/4fce115c-f295-4c21-9d04-1c0aed4e0be6%40AIM.com.
robhb...@gmail.com
This task has come up often enough that I should have automated it before: I want to search for one or more strings in each of the libraries from which my client submits production jobs. Each time I need to do this I go to ISPF 3.14, fill in a character string and one of the production JCL libraries, and fire it off. Then another production library, and another, and so on. Long past time I write a REXX to do this, because it’s going to come up again.
I’m sure I’ve done something like it before; there’s an ISPF service for it, isn’t there? Or maybe something else, because I’m looking in the ISPF dialogue services guide and I don’t see it. Where can I find it, please?