A question on using sed or awk

UNIX for Advanced & Expert Users
1

Hi I have a pattern like this. repeating many lines
CHANGE #13 TYP:22 CLS: 21 AFN:12 DBA:0x0040a15f OBJ:41142 SCN:0x0000.00036b3e SEQ:1 OP:11.2
CHANGE #15 TYP:32 CLS: 32 AFN:212 DBA:0x0040a15f OBJ:41143 SCN:0x0000.00046b3e SEQ:1 OP:13.3
.
And i am trying to do the following:
a) I need to get an output like this
OBJ:41142 SCN:0x0000.00036b3e SEQ:1 OP:11.2
OBJ:41143 SCN:0x0000.00046b3e SEQ:1 OP:13.3
So basically strip all content from Change # till the start of OBJ.
.
b)Secondly once i get (a) done i want some kind of a calculation like
OBJ:41142 SCN:224062 SEQ:1 OP:11.2
OBJ:41143 SCN:289598 SEQ:1 OP:13.3
where in the scn values are derived based on some calculation like
the decimal equivalent of first *4GB+the decimal equivalent of second.
.
Are these possible using sed or awk?
Thanks,
Hare.

2

For the qn 1,

sed 's/\(.*\) OBJ:\(.*\)/OBJ:\2/' filename

For qn2, could you explain it in detail what exactly u need?

3

Hi,
Thanks. That works of sed 's/\(.\) OBJ:\(.\)/OBJ:\2/'
.
Assuming the output now i get is this after executing above line
OBJ:41142 SCN:0x0000.00036b3e SEQ: 1 OP:11.2
OBJ:......
.
I want a final output as
OBJ:41142 SCN:0x0000.00036b3e SEQ: 1 OP:11.2
where
the decimal equivalent of first part say 0x0000 is 0 and of
second part namely 00036b3e is 224062
so 0+224062=224062
So the output should become
OBJ:41142 SCN:224062 SEQ: 1 OP:11.2
.
Thanks,
Hare