Home/ Questions/Q 6964893
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-05-27T15:58:57+00:00

This maybe extension from the question: Incorporate variables into bash code line I just

  • 0

This maybe extension from the question:
Incorporate variables into bash code line

I just realize in my text, lines actually come in variable format.

2   118610455   P2_PM_2_5034    T   <DUP:TANDEM>    40  .   END=118610566;SVLEN=110;SVTYPE=TDUP;CIPOS=-100,55;CIEND=-56,100;IMPRECISE;DBVARID=esv7540;VALIDATED;VALMETHOD=CGH;SVMETHOD=RP 

1   859214  P2_M_061510_1_73    C   <DEL>   .   .   CIEND=-130,50;CIPOS=-57,93;END=860180;IMPRECISE;SVLEN=-966;SVTYPE=DEL;VALIDATED;DBVARID=esv10036;VALMETHOD=CGH;SVMETHOD=RD,RP

What I need is 

2 118610455 118610566
1 859214 860180

Just as shown in above, this "END=#" may come in different positions at 8th column. So basically I need to find “END=..” part from 8th column first, then grep the number.
So this is actually about how to grep specific pattern from string ( in this case, the pattern is “END=”)

But how can I do that?
thx

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    Grep:

    You can use the -o option of grep for your search:

    Test:

    [jaypal:~/Temp] grep -o "END=[0-9]\+;" file | tr -ds 'END=|;' ''
118610566
860180

    But if you are looking for a complete solution then how about using awk (sorry I know this wasn’t your requirement. But here are two solutions:

    Awk:

    If the first and second parameters you want do not vary in position, then we can split each values in specific fields and then loop over each of them. As soon as we reach a field that is END we print the $1 and $4 and then print the column next to END.

    awk -v FS="[ ;=]" '{for(i=1;i<=NF;i++) if ($i=="END") print $1,$4,$(i+1)}' file

    Test:

    [jaypal:~/Temp] cat file
2   118610455   P2_PM_2_5034    T   <DUP:TANDEM>    40  .   END=118610566;SVLEN=110;SVTYPE=TDUP;CIPOS=-100,55;CIEND=-56,100;IMPRECISE;DBVARID=esv7540;VALIDATED;VALMETHOD=CGH;SVMETHOD=RP 
1   859214  P2_M_061510_1_73    C   <DEL>   .   .   CIEND=-130,50;CIPOS=-57,93;END=860180;IMPRECISE;SVLEN=-966;SVTYPE=DEL;VALIDATED;DBVARID=esv10036;VALMETHOD=CGH;SVMETHOD=RD,RP

[jaypal:~/Temp] awk -v FS="[ ;=]" '{for(i=1;i<=NF;i++) if ($i=="END") print $1,$4,$(i+1)}' file
2 118610455 118610566
1 859214 860180

    GNU AWK:

    If you have gawk then it has a built-in function called gensub. That supports back references. So you can also do the following –

    gawk '{print $1,$2,gensub(/.*\<END\>=(.[^;]*);.*/,"\\1",$0)}' file

    Test:

    [jaypal:~/Temp] gawk '{print $1,$2,gensub(/.*\<END\>=(.[^;]*);.*/,"\\1",$0)}' file
2 118610455 118610566
1 859214 860180
    • 0