Looking to run an AWK script that is too slow in Perl?

[gimiluv]

Hello, I have a perl script that seems to be running too slow and I was told than awk could possible speed up 2M+ lines of parsing. The output I'm going through looks as follows:
07:07:05.048 I: QFEED: S=BOL P=44.25 B=23.5 A=15.43 SZ=4200x1500 L=0 D=0(42) CD=0 N=1

What I need to do is find out if the value of A is less than the value of B, if So, print the line. But it's doing it on about 2-4M lines of code. I know k-shell is only capable of doing integers, so I thought I would try awk. Any ideas?

 

[ranjit]

[tt]

Perhaps this...

$ cat script
/[A-Z]/ {
b=sprintf("%s", substr($6,index($6,"=")+1,5))
a=sprintf("%s", substr($7,index($7,"=")+1,5))

if(a<b) {print $0}
}

$ cat data
07:07:05.048 I: QFEED: S=BOL P=44.25 B=23.5 A=15.43 SZ=4200x1500 L=0 D=0(42) CD=0 N=1
07:07:05.048 I: QFEED: S=BOL P=44.25 B=23.5 A=55.43 SZ=4200x1500 L=0 D=0(42) CD=0 N=1
07:07:05.048 I: QFEED: S=BOL P=44.25 B=23.5 A=85.43 SZ=4200x1500 L=0 D=0(42) CD=0 N=1


$ awk -f script data
07:07:05.048 I: QFEED: S=BOL P=44.25 B=23.5 A=15.43 SZ=4200x1500 L=0 D=0(42) CD=0 N=1


[/tt]