Exclude redundant records - Generic

rkarthik22 · New User Joined: 18 Apr 2009 Posts: 47 Location: India

Hi Experts,

My input file is FB. LRECL is 95

Pos 1 to 8 is Date - Length 8 bytes

Pos 11 to 54 is file name - Length 45 bytes

Below is the input file having correct records and Redundant (unwanted records which truncated) - Showing only file name field

gylbharat · Posted: Fri Aug 05, 2011 12:41 pm

why 'CA7TRI' and 'CA7TRIG.SYST' are redundant?

enrico-sorichetti · Posted: Fri Aug 05, 2011 12:54 pm

because the TS knows

but a SS looking for a ". " might get rid of some of them

rkarthik22 · New User Joined: 18 Apr 2009 Posts: 47 Location: India

gylbharat

Actually those are reduntant records.

As i already mentioned

'CA7TRI' , 'CA7TRIG.SYST' etc. are all truncated records populated tin the file.

enrico,

Great Thanks for your suggestion.

I took some sample records in my input file.

I try to remove the first record in the below file.

enrico-sorichetti · Posted: Fri Aug 05, 2011 2:17 pm

in this case You forgot the blank after the dot
should be '. ' not '.'

but for the other cases it looks a bit more difficult

we have noooo way of telling that

gcicchet · Posted: Fri Aug 05, 2011 2:54 pm

Hi,

rkarthik22 · New User Joined: 18 Apr 2009 Posts: 47 Location: India

enrico,

Great Thanks.

Yes, We can remove those records with '. ' (blank after .)

Sorry, i forgot to include that blank.

There are some records having " in it. We can also remove this using 'SS'

But for other type of records really i dont have any clue.

As you suggested, I will try to write some REXX code.

But if anyone has somemore idea regarding this, Please kindly
share your ideas.

It would be very helpful for me.

Thanks,

Gerry,

Yes, you are correct. It is 44 bytes, not 45.

Thanks for pointing out my mistake.

kratos86 · Posted: Fri Aug 05, 2011 4:24 pm

Try to come up with proper selection criteria to select records as redundant, otherwise no programming language can help you apart from reviewing and removing it manually.

enrico-sorichetti · Posted: Fri Aug 05, 2011 4:29 pm

the include/exclude criteria are clear...
the TS wants to keep only EXISTING datasets
that' the reason why no utility will be able to do it
the only thing that a general purpose utility can do is pattern matching
but that, as already said, does not satisfy the requirement

the only thing that needs revisiting is the relative gdg stuff !
it is not clear the <goodness> criteria why (+1) is good and (0) is bad
apart that it can be done quite easily

kratos86 · Posted: Fri Aug 05, 2011 4:32 pm

Bill Woodger · Posted: Fri Aug 05, 2011 4:38 pm

Everything says rexx.

Maybe you don't even need to check for the existence of the dataset (your choice), just that the format is valid.

You could include code to exclude the valid-looking datasets that you do not want (whether or not they exist).

rkarthik22 · New User Joined: 18 Apr 2009 Posts: 47 Location: India

Hi experts,

Thanks for each and everyone, Spending your valuable time.

Regarding my doubt,

We found some issues in the Sortcard that creates
my final Output file (which is having this truncated or Redundant records)

We put a fix by including all the Possible conditions that not to
truncate the dataset names.

Now we are able to get the final file with all correct file names populated.

Again thanks for everyone.

Moderator,

kindly, Can you please lock this thread.

gylbharat · Posted: Tue Aug 23, 2011 11:15 am

Can you post the solution...

rkarthik22 · New User Joined: 18 Apr 2009 Posts: 47 Location: India

enrico-sorichetti · Posted: Tue Aug 23, 2011 4:53 pm

rkarthik22 · New User Joined: 18 Apr 2009 Posts: 47 Location: India

Mr. Enrico,

I respect your words....

No one scolded me like this ever.....

I am with this forum for nearly two and half a year...

Sincerely, i know the rules of posting a question seeking for help.

How can i waste other peoples valuable time....( i will never do that )

I asked for a help, Got some suggestions (Rexx), but during working on it we found simple approach (instead of eliminating / filtering the redundant records we can stop populating this redundant records)

what i said to gylbharat is , we just used exclude and include conditions in the SAR unload file (input file) to keep in mind that, we should not get any truncated data.....just simple include/exclude/split/build keywords ( this is what i said about internal process ).

More....

Earlier we received this final file from other system with lot of truncated datas, so we thought we can go for some REXX to eliminate this redundant records.

But after we analyzed the original input file, we ourselves took some time to incorporate this.

I am explaining everything in detail in order to prove , nothing is wrong from my side.

If anyone feels, anything is wrong on my side, Please you can give warning.

I will take care about this in my future postings, that this should not happen.

sincere apologies to each and everyone who replied to this thread.

kratos86 · Posted: Tue Aug 30, 2011 2:16 pm

Hi Karthik,

You gave us the edited input file and why don't you show us the card which can be used to achieve the edited output you have posted.

We are not asking you the confidential information which could impact your business.

Note:Everyone is here to share their valuable learnings. So what you learnt with this scenario might help others.