A tool for unpacking a packed file

vickey_tcs · New User Joined: 14 Mar 2009 Posts: 5 Location: At Desk

Hi,

I m Planning to write a tool to unpack a packed dataset.My input will be a dataset where there can be few packed data in the file. I want my new tool to unpack the packed data and write all the fields into a new dataset (fully unpacked data).Can any one help me out on this??

superk · Posted: Sun Apr 26, 2009 7:22 pm

I must admit that I don't know much about this concept of "packed" datasets, but I thought that ISPF services can handle all of the pack/unpack processes?

dbzTHEdinosauer · Posted: Sun Apr 26, 2009 7:30 pm

vickey_tcs,
are you talking about a file that contains packed-decimal fields?

mtaylor · Posted: Sun Apr 26, 2009 7:36 pm

You're unpacking tool must also have as input the locations of all packed fields in the file because there is no way to differentiate from the actual file where packed fields begin/end.

This requirement can complicate the implementation a lot. One approach is to give as input the copybook corresponding to the file, but then your tool must parse the copybook. This isn't impossible or especially difficult but maybe more work than you had anticipated.

Marso · Posted: Sun Apr 26, 2009 8:05 pm

Or are you talking about a dataset with packed members ????

There are a few different things that can be packed. Please be more precise.

vickey_tcs · New User Joined: 14 Mar 2009 Posts: 5 Location: At Desk

HI,

Yes the tool will first get the file(dataset), then the layout(copybook) of the file, the tool should map the layout and the file so as to unpack the packed file and create a new file by calculating the new LRECL of the file and then load all the fields into it..is this possible??? Please suggest

enrico-sorichetti · Posted: Sun Apr 26, 2009 11:06 pm

Yes ?? to what question are You replying...

1) ISPF packed data
2) some form of data compression
3) packed decimal numeric fields, binary numeric fields

the question might be clear to You, You did little to make it clear to everybody else

enrico-sorichetti · Posted: Sun Apr 26, 2009 11:46 pm

it might be easier for You
( if any of them is available )
to rely on the formatting capabilities of FILEAID or FILEMANAGER

I can tell for FILEMANAGER

see this topic for more info on the subject... the issue is almost the same
ibmmainframes.com/viewtopic.php?t=31446&highlight=rexx+copybook+processing

Robert Sample · Posted: Mon Apr 27, 2009 12:06 am

vickey_tcs: be aware there is an option under ISPF to pack files. This is a form of compression and has absolutely nothing to do with packed decimal fields. Packed files must be unpacked before they can be used as input to a compiler, text editor, or any other application.

Your posts do not make it clear if you have these packed files that you want to unpack, or if you have individual packed decimal fields you want to unpack, or if you are talking about some form of compression that was used to "pack" the data.

Further, there has been discussion on this forum about these topics in the past -- have you used the search function to review them before posting?

Craq Giegerich · Posted: Mon Apr 27, 2009 12:33 am

If you are talking about unpacking packed decimal fields, WHY? Packed decimal is smaller and more efficient for use. Is this because you are a new programmer and haven't learned how to decipher PD?

Bill Dennis · Posted: Mon Apr 27, 2009 7:46 pm

Since we're all speculating, my guess is unpacking PD fields to prepare for FTP. I hope vicky_tcs returns soon!

mtaylor · Posted: Mon Apr 27, 2009 8:24 pm

I agree with Bill, and I thought it was pretty obvious that's what vickey_tcs was referring to. I used to get the request to unpack packed decimal data all the time. Having a tool to do this would be very helpful. But then why not add other formatting options:

expand packed decimal and comma separate fields?

expand packed decimal and convert to xml output?

At some point you'll end up rewriting Easytrieve.