This a question about getting Macs and Windows PCs to play nicely with each other.
SCENARIO: I use a Mac. I have a bunch of data files that need some slicing and dicing to use with a visualization toolkit. The visualization software is for Windows only, which is no problem since I have a virtual Windows machine. awk was my tool of choice for this problem, and it generated the input files beautifully. Unfortunately, when I generate the input files for the visualization tool and send them over to the PC, there are no new lines; i.e., on the Mac I open (say) output001.txt:
header.info
# file automatically generated by awk
BEGIN File
...
<columns of data>
...
END File
Looks great, but then I open in on the PC and have:
header.info# file automatically generated by awkBEGIN File...<columns of data>...END File
I understand that Windows and Mac OS treat end of line characters differently. I've read that \n on a Mac stands for newline AND character return whereas Windows treats them separately, necessitating \n\r. But here, the problem is somewhat different. I never used a \n character generating the text file -- awk did all the newlines.
Is there a character that I can have awk put at the end of each line it writes so that Windows will understand where the newlines go?
Sure, make the "Output Record Separator" a Carriage Return plus Linefeed. So instead of
seq 1 10 | awk '1' | cat -vet
1$
2$
3$
4$
5$
6$
7$
8$
9$
10$
use
seq 1 10 | awk '1' ORS='\r\n' | cat -vet
1^M$
2^M$
3^M$
4^M$
5^M$
6^M$
7^M$
8^M$
9^M$
10^M$
Notes:
seq 1 10 just generates the numbers from 1 to 10.
1 amounts to true in awk which means it should just do its default thing - which means to just print the current line.
I just use cat -vet to show the Carriage Returns, i.e ^M.