On Sun, Mar 14, 2004 at 04:30:40PM -0600, Steven N. Fettig wrote: > Sorry for posting an off-topic question to the list, but this is > somethin that has been driving me nuts for weeks now and I can't figure > it out. I want to pass a text file through sed that replaces all > whitespaces with a carriage return. I.e., if I have the file > my_test_text_document.txt that is a few paragraphs of writing, I want to > take the following input: > > I have just written five paragraphs of absolute jibberish and wish that > I could get sed to work the way that I want. Oh how this question has > plagued me! > > And have sed output: > I > have > just > written > five > paragraphs > of > absolute > jibberish > and > ... you get the point. > > I can't figure out what the newline character is... I've tried \n \r &\, > etc. with no avail. I run the following: > > sed 's/[ ]/\n/g' my_test_text_document.txt > > and the output never has a newline added regardless of what I have > substituted \n with. I have also used " instead of ' and that hasn't > helped... > Sorry for the question, but I'd really appreciate the help!
sed(1) can do it, but it's cleaner and simpler to use tr(1): % cat foo I have just written five paragraphs of absolute jibberish and wish that I could get sed to work the way that I want. Oh how this question has plagued me! % tr -s ' ' '\n' < foo I have just written five paragraphs of absolute jibberish and wish that I could get sed to work the way that I want. Oh how this question has plagued me! Cheers, Matthew -- Dr Matthew J Seaman MA, D.Phil. 26 The Paddocks Savill Way PGP: http://www.infracaninophile.co.uk/pgpkey Marlow Tel: +44 1628 476614 Bucks., SL7 1TH UK
pgp00000.pgp
Description: PGP signature