Bash: delete characters after colon in specific columns

[Muriel_Gros-Baltha]

Hello !

I have this tab-delimited file :


KE339719.1 13 0/0:2,0:2:33:0,33,428 0/0:1,0:1:9:0,9,148
KE332739.1 14 md 0/0:3,1:4:15:0,15,276
KE336330.1 14 0/1:5,12:17:99:157,0,382 0/1:2,7:9:99:100,0,147

In the third and fourth column, I want to delete everything that is after ":", to obtain :


KE339719.1 13 0/0 0/0
KE332739.1 14 md 0/0
KE336330.1 14 0/1 0/1

I wanted to combine awk and sed but don't manage to find a solution :(

Any help ?

Thanks !

Muriel

[Bucky Ball]

Thread moved to Programming Talk.

[sisco311]

This is how you do it in Bash:


while read -r f1 f2 f3 f4
do
printf '%s\t' "$f1" "$f2" "${f3%%:*}" "${f4%%:*}"
printf '\n'
done < file

See: BashFAQ 001 (link in my sig) and
http://mywiki.wooledge.org/BashGuide/Parameters#Parameter_Expansion

[Vaphell]

sed -r 's/:[^ \t]+//g'
though it won't necessarily clean up the formatting


while read -r c1 c2 c3 c4; do printf "%s\t%s\t%s\t%s\n" "$c1" "$c2" "${c3%%:*}" "${c4%%:*}"; done < file



$ sed -r 's/:[^ \t]+//g' tab_delim.txt
KE339719.1 13 0/0 0/0
KE332739.1 14 md 0/0
KE336330.1 14 0/1 0/1
$ while read -r c1 c2 c3 c4; do printf "%s\t%s\t%s\t%s\n" "$c1" "$c2" "${c3%%:*}" "${c4%%:*}"; done < tab_delim.txt
KE339719.1 13 0/0 0/0
KE332739.1 14 md 0/0
KE336330.1 14 0/1 0/1

[Muriel_Gros-Baltha]

Hello !

Thank you both for your answers.

I tried the sed option and don't have formatting problems so it's perfect !

Mu