Shell Programming and Scripting

I have a file which has entries like as below,

10:29:07.32 GOOD 122
10:29:07.35 GOOD 124
10:29:09.97 BAD 122
10:29:09.97 BAD 124
10:29:18.66 GOOD 128
10:29:19.12 BAD 128
10:30:23.20 GOOD 148
10:30:23.23 GOOD 156
10:30:25.55 BAD 148
10:30:25.55 BAD 156
10:30:35.53 GOOD 167
....
.......
..........


I need the following info from this, can anyone suggest a shellscript for this?

1) No of GOOD :

2) NO of BAD:

3) Time Diff(refer 1st column) of (GOOD-BAD) which has the same field value in the third column? (seperate one)

4) Total Time Diff between GOOD-BAD

Thanks!

How to compare a field value of a same column?

One part of your query:

Code:

$ awk '{count[$2]++}END{for(j in count) print j,"("count[j]")"}' gobinath.txt
BAD (5)
GOOD (6)

//Jadu

or

Code:

$ awk '$2 ~ /BAD/{++c} END {print c}' gobinath.txt

Code:

$ awk '$2 ~ /GOOD/{++c} END {print c}' gobinath.txt

and regarding your query:
How to compare a field value of a same column?

could you please put an example what you want to compare, would be very helpful to answer.

//Jadu

Thanks for your immediate response....

I need the output like this,

Time diff of 122 : (10:29:09.97-10:29:07.32) = 00:00:02.65
Time diff of 124 : (10:29:09.97-10:29:07.35) = 00:00:02.65
Time diff of 128: (10:29:19.12-10:29:18.66) = 00:00:00:46
Time diff of 148:
Time diff of 156:
Time diff of 167:
================================================
Consolidated time diff= 00:00:05:76

Thanks!

file name is a

Code:

awk '{
if($2=="GOOD")
	good++
else
	bad++

if (a[$3]=="")
	a[$3]=$1
else
	a[$3]=sprintf("%s-%s",$1,a[$3])
}
END{
print "No of GOOD:"good
print "No of BAD:" bad
for (i in a)
{
	print "Time diff of "i" "a[i]
}
}' a > a.t

sed '1,2d' a.t | cut -d' ' -f 5 |sed 's/\./:/g;s/-/:/' | awk 'BEGIN{FS=":"}
{
	d=$4-$8
	if (d<0)
	{
		d=100+d
		c=$3-$7-1
	}
	else
		c=$3-$7
	if (c<0)
	{
		c=60+c
		b=$2-$6-1
	}
	else
		b=$2-$6
	if (b<0)
	{
		b=60+b
		a=$1-$5-1
	}
	else
		a=$1-$5
	print a":"b":"c":"d
}'> a.tt

nawk '{
if(NR==FNR)
	a[NR]=$0
else
{
	if (NF<5)
		print
	else
	{
		$5=a[FNR-2]
		print
	}
}
}' a.tt a.t
rm a.t a.tt

Thanks to all for your solution...it is working

I have an almost similar problem, but in this case the fields 122, etc can re-occur more than once.

10:29:07.32 GOOD 122
10:29:07.35 GOOD 124
10:29:09.97 GOOD 122
10:29:09.97 BAD 124
10:29:18.66 GOOD 128
10:29:19.12 BAD 128
10:30:23.20 GOOD 148
10:30:23.23 GOOD 156
10:30:25.55 GOOD 148
10:30:25.55 BAD 156
10:30:35.53 GOOD 167
10:30:25.55 BAD 122
10:30:35.53 BAD 122
10:30:25.55 BAD 148
10:30:25.55 BAD 148

How can I consolidate the total time for a record (ie 122), less any total time that is flagged as Bad?