Sorting comma delimited file?

[perumalnk]

Hi,

I have a comma delimited file having 35 fields with each field value is separated by a comma.Depending upon the value of the field the position occupied by a particular field differs.For example,

Rec # 1 - 3,10.1.2.45,sxcd,34566........
Rec # 2 - 3,10.22.33.456,sxcd,34566..........

Here i need sort the second field and fourth field.Is there any way to accomplish this?

Thanks,
Perumal.N.K.

[Frank Yaeger]

You can use DFSORT's PARSE function to do this. However, in order to sort correctly, you need to "normalize" the timestamp in your second field. For example, if you had:

4.12.20.452
11.3.18.32
4.12.5.3

it wouldn't sort correctly unless you treated it as:

04.12.20.452
11.03.18.320
04.12.05.300

The DFSORT job below will do what you want. It will extract the fields you want to sort on and normalize them for sorting. I used a better set of test data in SORTIN. I assumed your input file has RECFM=FB and LRECL=80, but you can change the job appropriately for other attributes.

[perumalnk]

Hi Frank,

Thanks a lot for your reply and the detailed example !

I tried the solution in my application Mainframe machine.It looks like the function PARSE is not supported in my Mainframe machine.But this example gave me idea about using the DFSORT.Once again thanks a lot !

Is there anyother way driven by the number of occurance of commas in the record.Like to sort 12th field, by checking the occurance of comma for 11 times.

Thanks,
Perumal.N.K.

[Frank Yaeger]

[perumalnk]

Hi Frank,

Thanks a lot !
I will discuss about this with my Tech support team.

Thanks,
Perumal.N.K.