Replace first few lines with first few lines from other file

Question

I am working on Linux. I have 2 files - file1.dat and file2.dat.

cat file1.dat
1
2
3
4
5
6
7
8
9
10

and for file2:

cat file         


        

Answer 1

Following awk may also help you in same, tested codes in GNU awk.

Solution 1st:

awk 'FNR==NR && FNR<4{print;next} FNR>4 && FNR!=NR' file2.dat file1.dat

Solution 2nd:

awk 'FNR==NR && FNR==4{nextfile} FNR==NR{print;next} FNR>4 && FNR!=NR' file2.dat file1.dat
OR
awk 'FNR==NR{if(FNR==4){nextfile};print;next}  FNR>4 && FNR!=NR' file2.dat file1.dat

Solution 3rd: Using awk and head and tail command's combinations here.

awk 'FNR==1{system("head -n3 file2.dat");next} 1' <(tail -n +4 file1.dat)

Answer 2

awk is your friend

Script

# awk 'NR==FNR && FNR<=3 || NR>FNR && FNR>4' file2 file1

Output

1a
2a
3a
5
6
7
8
9
10

Tips

• NR - Total number of records processed
• FNR - Total number of records processed but resets when reading a new file.
• When a condition evaluates to true and no extra commands are given,awk just prints.

  ---

All good :-)

Answer 3

Assuming GNU sed

$ sed '3q' f2 | sed -e '3r /dev/stdin' -e '1,4d' f1
1a
2a
3a
5
6
7
8
9
10

• sed '3q' f2 gives the first three lines from second file
• -e '3r /dev/stdin' use stdin data
• -e '1,4d' delete required lines
• order is important - first r then d

For small number of lines, you can also use

sed -e '3R f2' -e '3R f2' -e '3R f2' -e '1,4d' f1

R command reads one line at a time

With GNU coreutils, this would probably be better for all/most scenarios

head -n3 f2; tail -n +5 f1