Creating File names inside script

[lenjturnbull]

I am splitting information inside a script and want to establish a file name from data read. Right now I have a script that separates and fills a file but the file name is one I create. In the file I am reading the 2nd text line(record) has the 9th field that I want to id the file.
I am beating my brains trying to figure this one out.

here is a sample of the script I am using.
It takes a file and creates single files for each ISA/IEA encountered.


/^ISA/ && OUTFILE == "" {
OUTFILE = "ACK997X"sprintf("%03d",outnum++)".999"
}
OUTFILE != ""{
print > OUTFILE
}
/^IEA/ && OUTFILE != ""{
close(OUTFILE)
OUTFILE=""

In the 2nd text line, it always starts with a "GS" and the 9th field always has a value
Example
ISA*00* *00* *01*003897733 *02*HBGI *000210*072
8*U*00300*000021127*0*P*^
GS*FA*COMPAQSTS*HBGI*000210*0728*21127*X*003040
ST*997*0001
AK1*QM*14975
AK9*A*1*1*1
SE*4*0001
GE*1*21127
IEA*1*000021127

In the above I want to create a file id "ACK997X"+"3040".suffix

Thanks for any assistance.


Len

 

[Chapter11]

Erm, what scripting language is that?

 

[lenjturnbull]

I am using this in a RS6000 Aix environment.
The application id is splitxxx.awk

One other thing I have a question on
since I pass the file into this script.
Is the file name available to use when the line
print > outfile is in play ???

I wonder if I need to set a filename var in the Begin ??

Thanks,

Len

 

[vgersh99]

hi,
as far as I can recall, I've posted this solution for the previous thread.

The output file name ["OUTFILE"] is derived. I odn't quite understand what needs to be achieved though....

To answer your second question - the name fo the currently processed file is available within awk by the reference to the internal variable FILENAME. The variable is assigned appropriate filename value implicitly. vlad
+---------------------------+
|#include<disclaimer.h> |
+---------------------------+

 

[lenjturnbull]

Vgersh99, yes you answered last week.

I got the file to split with the script you sent.

So when you say the current processing file is refered by the internal variable FILENAME. Can I pass it into the awk -f line to have it be the file name that is generated by the awk script ????
Am I clear on that???

File with data is sent to the awk script .
It parses data into individual files.
Before it was file names =file +xxx(incrementing)with txt suffix.
I want to be able to name the files with the processing file name + xxx(incrementing) suffix.


thanks,


Len

 

[vgersh99]

/^ISA/ && OUTFILE == &quot;&quot; {
OUTFILE = FILENAME sprintf(&quot;%03d&quot;,outnum++) &quot;.999&quot; vlad
+---------------------------+
|#include<disclaimer.h> |
+---------------------------+