Hi All,

I have a huge unix flat file delimted by @ at the end of the line. I need to find out if there is any line/s containing @ in between the line so that I can remove that and put the file for processing.

Thanks in advance for your help.

Hi
I'm not quite clear on your post.
Are all lines terminated with an @?
What do you mean 'in between the line'?
Can you give (an abbreviated) example of what you are trying to do?
Peter

Hi peter,

The few lines from example file look like;

0000000014471;S;00;05; ;-0038044;000000000;000000000;01112007; @
0000000015630;S;00;12;CSH;-0007017;000000000;000000000;01112007; @
0000000075945;S;00;25;ETO;00038227;000038227;000000000;01112007; @

All lines are are terminated with delimiter '@'. Sometimes within the line we get @ thereby creating a mismatch between the records in the file and records in the trailer (File count > Trailer Count). For example;

5000006267 ;U;DR &; ;HARPER @ ;2 MARITIME ;24062002;+000023772; ;000000;00000000;23102007;+000000000;2306
2002;+000023772;D;11; ; ; ;0000; ;COT;06111998;00@

here after HARPER there is a '@' which creates count mismatch.

We usually use a macro to identify the extra delimiter within the line and delete it manually and then run a validation script to see if it passes the record count validation .

I need to find out two scenarios;

1. Lines containing '@' within
2. Lines missing '@' at the end of line

Regards,
Bhausab

there may be a lot of way , Just an idea .. Remove all @ by using sed , and the insert @ at every endof line ...

PLease let me know if you need the sed commands

1. Lines containing '@' within
This is easy if you use egrep:

$ cat file.txt | egrep '@.+'

2. Lines missing '@' at the end of line
Use -v to get those lines that don't match:

$ cat file.txt | grep -v '@$'


Peter