Shell Programming and Scripting

Wondering if someone can help with my task of identifying missing detail lines in transactions.

I have a flat file that contains transaction header, transaction detail, and transaction trailer lines. These lines are identified with a 5-character line identifier "THEAD", "TDETL", and "TTAIL" at the beginning of the line. Similar to this:

Code:

THEAD000000000200000000
TDETL0000000003000000Good
TTAIL0000000004000000

How can I identify which transactions have missing TDETL lines?
In other words, a missing TDETL transaction is one where there is a THEAD and then TTAIL right after on the next line without a TDETL line in between. Here is one such example:

Code:

THEAD000000000200000000
TTAIL0000000003000000

In those conditions, I'd like to print out the THEAD lines to a text file as well as the file name that the missing TDETL transaction was found in as I will be running this script against 1000s of files.

Is SED or GREP the proper tools to use? What would be the best way to go about this?

Code:

nawk '/^THEAD/{h=$0;d="";next} /^TDETL/{d=$0} !d{print FILENAME, h}' myFiles

Works very well. Thank you very much.

Just so I understand: what you're doing is looking for the line to start with "THEAD", once found save the THEAD line in variable h, reset the d variable to blank, and then go to the next line in which case you will then check that it begins with "TDETL", if it does, save the TDETL line in variable d, otherwise print the filename of the file and then variable h (which will be the THEAD line).

Is this right? Please correct me if I'm wrong or if you can explain it in better terms so that I learn.