IBM Mainframe Forum

by **zatlas1** » Sun Dec 25, 2011 10:32 am

While investigating an old system I've found a notation that I seem to recall from years past. A quoted meaningful name appears in a commeent before the DFHMDF like this:

Select all

* 'TS25-TRAN-ID' 00000050
TRNID25 DFHMDF POS=(01,001), X00000100
LENGTH=004, X00000200
ATTRB=(PROT,FSET,NORM), X00000300
COLOR=NEUTRAL 00000400
DFHMDF POS=(01,006),ATTRB=(ASKIP,NORM),LENGTH=000 (Stopper) 00000500
* 'TS25-HDG' 00000550
HDG25 DFHMDF POS=(01,008), X00000600
LENGTH=045, X00000700
ATTRB=(PROT,FSET,NORM), X00000800
COLOR=NEUTRAL 00000900
DFHMDF POS=(01,054),ATTRB=(ASKIP,NORM),LENGTH=000 (Stopper) 00001000

I know that today this is not needed as Assembler does accept long field names. My question is more about whether this notation was (quasi) standartised and whthere there were common interpreters that could handle it and generate copybooks with the meaningful names included?
Is so, where could I get such an interpreter?
Thanks
ZA

by **dick scherrer** » Sun Dec 25, 2011 11:23 am

Hello,

I suspect these were just normal comments - possibly required by organizational standards.

If the comments were wanted in the copybook, it would have been been part of the copybook source.

I'm not aware of any common/standard interpreters that used these comments. Lots of organizations wrote their own generators for maps, file definitions, common code, etc, but they were site-specific.

by **zatlas1** » Sun Dec 25, 2011 6:19 pm

Maybe so, but while investigating, I've found that notaion in more then one shop. The shops, indeed have what appear to be their own programs to handle that notaition, but I couldn't find the source code for any of those programs so far.
I may easily write my own interpreter if I need to, but I'd like it to be more generally useful.
Does anybody have any recollection about that notation?
Thanks
ZA

by **enrico-sorichetti** » Sun Dec 25, 2011 7:51 pm

Maybe so

what makes You think that in assembler a line starting with an * would not be a comment ?

The shops, indeed have what appear to be their own programs to handle that notaition, but I couldn't find the source code for any of those programs so far.

if a shop was using comments or generated comments with some <tool> the tool and it's source would not generally be available
outside that organization. searching for themm would be just a waste of time.

I may easily write my own interpreter if I need to,

nobody prevents You from doing it, provided You have the skills, the requirements are clear and Your organization is willing to allocate the <budget/time>

Does anybody have any recollection about that notation?

why should we...
a comment is a comment, so it is just discarded by the <language processor/translator/compiler>
and there is no standard/requirement on the format nor the content of a comment
it' s format/content depend only on the local standards of the organization requiring/using them ( the comments )
see the previous remark about in house developed tools.

by **zatlas1** » Mon Dec 26, 2011 12:29 am

I was not trying to be combative or antagonist, so I will not respond to that comment any farther
ZA

by **enrico-sorichetti** » Mon Dec 26, 2011 1:18 am

nobody was pugnacious or antagonistic ...
I was just replying to each of Your concerns...
and trying to reconfirm that nothing/little can be evinced by comments only looking at them
without knowing the environment where they were enforced/generated...

as far as the code You posted the relation between the names in the comments and the labels in the mdf macros
were not certainly generated automatically

Select all

* 'TS25-TRAN-ID'
TRNID25 .....
* 'TS25-HDG'
HDG25 ....

the logic to go from a long name to a short one is not so intuitive ( if one exists )
drop the first two letters and save the two digits of the first token of the long name
and concatenate them to the last token, but then it will fail for the first name
as an exception if the long name is made up of three tokens drop the vowels of the second one
and concatenate the mangled second token with the third one an append the previous two digits ... :roll:

better consider just a good old handwritten tabular relation

IBM Mainframe Forum

BMS field name in a comment preceing the DFHMDF

BMS field name in a comment preceing the DFHMDF

Re: BMS field name in a comment preceing the DFHMDF

Re: BMS field name in a comment preceing the DFHMDF

Re: BMS field name in a comment preceing the DFHMDF

Re: BMS field name in a comment preceing the DFHMDF

Re: BMS field name in a comment preceing the DFHMDF