Hello,
Can you please let me know how to find all lines that don't begin with pattern1, pattern2, pattern3 and pattern4?
Here is my awk script that gives an error.

awk 'BEGIN { NAME="$FILE"
GSNO=0 }
/^ISA/ { FIRST=$0;
LAST="IEA*1*" + "substr($0,90,9}" }
/^GS/ { GSNO=++;
NAME="$FILE"$GSNO;
print "$FIRST" >> $NAME;
print $0 >> $NAME}
[/!^ISA/[,/!^GS/[,/!^GE/[,/!^IEA/]]]]{
print $0 >> $NAME }
/^GE/ { print $0 >> $NAME;
print $LAST >> $NAME }
/^IEA/ {} ' $FILE

The red part should mean 'if it does not begins with ISA, GS, GE, IEA. I don't know the syntax.

Thanks.

I'm having trouble reading your code, please use code tags.

Can you write a sentence that says what you are trying to do with the whole code block? Not just the part of pattern matching?

OK, this is what I want.
There is a file like this:

ISAtext1
GStext2
text3
text4
...
textn
GE
IEA

I want to do this:

if line begins ISA - do command1.
if line begins GS -do command2.
all other lines (means all except ISA, GS, GE, IEA) - do command3. - how to write this in awk???
if line begins GE - do commad4
if line begins IEA -do command5.

Hope I was more clear.

Thanks

I have fixed some of the syntax errors...

awk 'BEGIN { NAME="$FILE" <-- use -v to pass command line parameters
LAST="IEA*1*" + "substr($0,90,9}" } <-- use space for concatenation and don't surround functions by double quotes
/^GS/ { GSNO=++; <-- use either GSNO++ or GSNO+=1
NAME="$FILE"$GSNO; <-- there is no dollar sign when referring to variables
print "$FIRST" >> $NAME; <-- when writing to a new file just use > not >> (unless appending to an existing file)

Thank you very much Igor.
It works!