processing the file based on the string in the file.

[dodda]

Hi all ,

Iam bit newbie to the shell scripting. And my requirement is as folows

Iam having txt file and it contains data as :::

HEADER_FILE;SPD;TCE;20080417 11:20

118982_101;20RTG STD;GRE;118982_101;ATK;1;1;01012008 00:00;01012100 00:00;1;1;STD;0;
111159_101;20RTG STD;GRE;111159_101;ATK;1;1;01012008 00:00;01012100 00:00;1;1;STD;0;
036964_101;20RTG STD;GRE;036964_101;ATK;1;1;01012008 00:00;01012100 00:00;1;1;STD;0;
015402_101;20RTG STD;GRE;015402_101;ATK;1;1;01012008 00:00;01012100 00:00;1;1;STD;0;

FOOTER_FILE;4

I want to search for a string FOOTER_FILE in the last line of the line. If the string exists then i need to process the file. Else the file need not to be processed.


Can you share the piece of code for the above requirement.

Your help is greatly appreciated.

Thanks in advance.

[ArndW]

The simplest way would be to check for the footer from a Job Sequence and then decide whether or not to call up the processing job.

[chulett]

Is this in an attempt to ensure a file, which is being transferred to your system via ftp, is complete?

Me, I'd grep for the footer record and count the result:

Code: Select all

grep FOOTER_FILE <filename> | wc -l

0 = not there, 1 = there. Easy to make a decision from there.

[ArndW]

Adding to Craig's post to make it bulletproof:

Code: Select all

tail -1 <filename> | grep FOOTER_FILE | wc -l

if that string could show up elsewhere in the file

[chulett]

Of course.

[dodda]

I wrote the following piece of code :::

#!/bin/ksh
export count
count =` grep -i "FOOTER_FILE" /home/user/dsstage/test/Input.txt|wc -l`
echo $count

But it is throwing an error.
count not found.

Can anyone correct the error.

[iDomz]

Remove export
If you have not declared count before, no point in exporting it,right?

[dodda]

Eventhough i removed the export before it is throwing the same error

footer.ksh[3]: count: not found


Any help....

[iDomz]

Try this on command line

count=` grep -i "FOOTER_FILE" /home/user/dsstage/test/Input.txt|wc -l`

No spaces on either side of =

[dodda]

i tried in the command prompt as


>= `grep -i "FOOTER_FILE" /home/user/dsstage/test/Input.txt|wc -l`

It throws an error. ksh not found.

but i tried as
> grep -i "FOOTER_FILE" /home/user/dsstage/test/Input.txt|wc -l

It results 1. without blackquote it's is working fine.

[iDomz]

There should be no spaces between the VARIABLE (count) , the assignment operator ( = ) and the operand (`grep -i "FOOTER_FILE" /home/user/dsstage/test/Input.txt|wc -l`). I am not sure if it is copy-paste or phpBB, but I can see spaces in both posts.

Copy this line and execute

count=`grep -i "FOOTER_FILE" /home/user/dsstage/test/Input.txt|wc -l`

then execute

echo $count

Ping back if it still throws error

[dodda]

Thanks for your immediate reply.Thanks a lot.

It's working fine now. This solves my problem.

[chulett]

Then please mark the topic as Resolved.

ps. You don't need the quotes around FOOTER_FILE. Don't hurt but they aren't needed. And the "-i" means "ignore case", something I wouldn't do in this case.