Extract specific data content from a long list of data

Shell Programming and Scripting
1

My input:

Data name: ABC001
Data length: 1000
Detail info
Data Direction Start_time End_time Length
1 forward 10 100 90       
1 forward 15 200 185
2 reverse 50 500 450

Data name: XFG110
Data length: 100
Detail info
Data Direction Start_time End_time Length
1 forward 50 100 50       
2 reverse 50 90 40

Data name: ABC010
Data length: 200
Detail info
Data Direction Start_time End_time Length
1 forward 150 190 40       
.
.
.

My output:

ABC001 10 100
ABC001 15 200
ABC001 50 500

XFG110 50 100
XFG110 50 90

ABC010 150 190
.
.
.

Based on individual Data name, I would like extract its Start_time and End_time detail and put it at the column 2 and 3, respectively.
Thanks for any suggestion or hit to archive this goal.

2

If i'n reading your requiremenst right....one way:

#  awk '/Data name/{print"";t=$3}/forward/||/reverse/{print t,$3,$4}' infile

ABC001 10 100
ABC001 15 200
ABC001 50 500

XFG110 50 100
XFG110 50 90

ABC010 150 190

HTH

3

Try:

awk '{ if($0 ~ /^Data name:/) str=$3; if($0 ~ /^[0-9]/) print str" "$3" "$4; }' file
4 
while read -r line
do
    case "$line" in
        "Data name"* )
            dname=${line#*: }
            ;;
        *forward*|*reverse* )
            set -- $line
            echo $dname $3 $4
    esac
done <"file"
5

Thanks a lot, Tytalus.
Your awk code worked perfectly and very fast for my case :slight_smile:
I just check it along all my long list of data.
I found out something like this happen if my input data is shown like this:
My input

Data name: LCM50
Data length: 100
Detail info
Data Direction Start_time End_time Length
1 reverse 10 20 10       
2 reverse 50 90 40

Data name: BCD011
Data length: 1000
Detail info
Data Direction Start_time End_time Length

Data name: FGG010
Data length: 500
Detail info
Data Direction Start_time End_time Length
1 reverse 10 190 180

My desired output:

LCM50 10 20
LCM50 50 90

FGG010 10 190

Problem facing by the awk code given this output:

LCM50 10 20
LCM50 50 90


FGG010 10 190

Do you know how to archive the desired output by ignoring those data that got Data name, but don't have Start_time and End_time?
Thanks, Tytalus.

---------- Post updated at 05:18 AM ---------- Previous update was at 05:15 AM ----------

Hi, dennis.jacob.
I just try your code. It is worked nice as well.
Just it can't add a newline between two different Data name :frowning:
Thanks for your any advice and suggestion.

---------- Post updated at 05:22 AM ---------- Previous update was at 05:18 AM ----------

Thanks a lot too, ghostdog74.
Your code can work as well.
Just it also unable to add a new line between two different Data name detail :frowning:
From your opinion, is it ok to archive my target by running your code first , then take the output file and let it add new lane in between two different data name after then?
Thanks for advice.

6

if you want to add a newline, just do printf with \n. i will leave it to you to do that.

7

Thanks a lot, ghostdog74.
I get it already :slight_smile:

8 
$ awk 'BEGIN{RS="";FS="\n"}
/Data name/ { split($1,a," ")
     for (i=2;i<=NF;i++) 
         {if (($i~/forward/)||($i~/reverse/)) 
              { split($i,b," "); print a[3],b[3],b[4]}}
     {printf "\n"}}'  urfile

ABC001 10 100
ABC001 15 200
ABC001 50 500

XFG110 50 100
XFG110 50 90

ABC010 150 190
9

Hi rdcwayx, your code is worked fine too. Just it lack of put a ";" before the for.
I already edited it for you :slight_smile:
Thanks a lot.

$ awk 'BEGIN{RS="";FS="\n"}
/Data name/ { split($1,a," ") ;
     for (i=2;i<=NF;i++) 
         {if (($i~/forward/)||($i~/reverse/)) 
              { split($i,b," "); print a[3],b[3],b[4]}}
     {printf "\n"}}'  urfile

ABC001 10 100
ABC001 15 200
ABC001 50 500

XFG110 50 100
XFG110 50 90

ABC010 150 190
10

Interesting, I needn't the ; in my system.

Anyway, that's good to hear the problem is fixed.

11 
my $flag=0;
my $name;
while(<DATA>){
	if(/^Data name:\s*(.*)/){
		$name = $1;
	}
	elsif(/Start_time End_time/){
		$flag=1;
	}
	elsif(/^\s*$/){
		print "\n";
		$flag=0;		
	}
	else{
		if($flag eq 1){
			my @tmp = split;
			print $name," ",$tmp[2]," ",$tmp[3],"\n";
		}
	}
}
__DATA__
Data name: ABC001
Data length: 1000
Detail info
Data Direction Start_time End_time Length
1 forward 10 100 90       
1 forward 15 200 185
2 reverse 50 500 450

Data name: XFG110
Data length: 100
Detail info
Data Direction Start_time End_time Length
1 forward 50 100 50       
2 reverse 50 90 40

Data name: ABC010
Data length: 200
Detail info
Data Direction Start_time End_time Length
1 forward 150 190 40
12

thanks summer, I trying your suggestion now :slight_smile:
Thanks first ^^