sedpattern-matching

Sed script to remove first two lines of file if they are empty

I'm trying to write a fairly concise sed script to remove the first two lines of a file only if they are empty, so the following file:

> cat myfile.in


Line 3

Line 5

Would result in a three line output file:

> cat myfile.out
Line 3

Line 5

This involves combining line ranges and pattern matching and I just can't seem to find any examples of that. I'd also be intersted if anyone could suggest and equally (or more) consise Perl alternative. Many thanks.

Footnote

I should add that I tried 1,2{/^$/d} which worked absolutely fine on Linux, but under AIX I got:

sed: 0602-404 Function 1,2{/^$/d} cannot be parsed.

and on Solaris I got:

sed: command garbled: 1,2{/^$/d}

Which is a blow because this code has to run on both AIX and Solaris, but not Linux! Shame!

Solution

  • I think this should work with sed (if you want to delete both first lines if both are empty):

    sed '1,1{N;/^\n$/d}'

    i.e.: 1,1 = goto first line, N = append next line to pattern space, ^\n$/d = delete those two lines if they are empty.

    If you want to delete the first two lines if they are empty (independently from the other line) you could do:

    sed '1,2/^$/d'