[SOLVED] Using sed to remove string between two other stings in specific column

Issue

I want sed to look at only the first column of a tab-delimited file and remove everything between some string and the first tab for all rows. Sorry if this is repetitive, have searched other entries and cannot seem to find anything quite right.

For example, input:

blah-a_blah-b.13_blah-x_blah-y        0       0       0       0       0       17.983559

desired output:

blah-a_blah-b.13        0       0       0       0       0       17.983559

Tried unsuccessfully various iterations of below:

sed -i 's/\(\.[0-9]\).*\([^\t]\)//1' file

Advice?

Solution

This should work in gnu-sed:

sed -E 's/^([^\t]*\.[0-9]+)[^\t]+/\1/' file

blah-a_blah-b.13    0   0   0   0   0   17.983559

For non-gnu sed use:

sed -E 's/^([^[:blank:]]*\.[0-9]+)[^[:blank:]]+/\1/' file

Answered By - anubhava

Answer Checked By - Clifford M. (WPSolving Volunteer)