I'm trying to extract a range of bytes from a file. File contains continuous 16 bit sample data. I would think cut -b should work but I am getting error in the data.
Extracting 20 bytes (10 samples)
cut -b188231561-188231580 file.dat > out.dat
I expect it to create a 20 byte file with 10 samples (last sample should be the -79). However it creates a 5749bytes file, with the following contents (displayed using od -s)
0000000 -69 -87 -75 -68 -83 -94 -68 -67
0000020 -82 -79 2570 2570 2570 2570 2570 2570
0000040 2570 2570 2570 2570 2570 2570 2570 2570
*
0013140 -65 -67 -69 -69 -71 -66 -72 -68
0013160 -69 -80 10
0013165
As you can see, there is a whole bunch of repeated 2570 values where cut was supposed to stop.
What am I doing wrong here? Also the Wikipedia article on cut says -b is limited to 1023 byte lines. Although the man pages for cut don't seem to mention this limitation.
Is there a better bash command to extract bytes N-M from a binary file? I already wrote a Perl script to do it. I'm just curious.
cut -b is used to get certain bytes from each line, it can't be used to get bytes from the file as a whole.
You can use head/tail instead:
N=120
M=143
tail -c +$N file | head -c $((M-N))