[SOLVED] SED

carolflb · 02-02-2010, 09:01 AM

Hi everybody,

I tried to use the command:

sed -i ''$line_number',/port/s/port/portMARKED/' file

to substitute the word "port" by the word "portMARKED" at the specific line that is saved in "line_number". The substitution happens, but it is copying my line twice... and I have no idea why...

I need to make a single substitution of one word for another at a certain line in my text file. Would somebody know how I could do that?

Thanks!

druuna · 02-02-2010, 09:20 AM

Hi,

sed "${line_number}s/port/portMARKED/" infile

I'm not sure why you added the ,/port/ part, it is not needed.

General form (substitute with address range): sed '[address1[,address2]]s/pattern/replacement/[flags]' infile

flags is not used in your example, but could be g (global replace on whole line).

Hope this helps.

amon · 02-02-2010, 09:27 AM

Some handy sed examples: http://www-h.eng.cam.ac.uk/help/tpl/unix/sed.html

From that link:

Code:

 # substitute (find and replace) "foo" with "bar" on each line
 sed 's/foo/bar/'             # replaces only 1st instance in a line
 sed 's/foo/bar/4'            # replaces only 4th instance in a line
 sed 's/foo/bar/g'            # replaces ALL instances in a line

you have already seen the use of the

Code:

-i

option for inserting text to the file.
Not sure what the other parts were (or if you had got confused with the sed instructions as I often do).

if I take file (tmptxt.txt):

Code:

text1 text text
text2 text text
text3 text3 text3

And run:

Code:

sed -i '2 s/text/textMARKED/' tmptxt.txt

I get:

Code:

text1 text text
textMARKED2 text text
text3 text3 text3

in your script you may need to replace single quotes with double quotes.

carolflb · 02-02-2010, 09:30 AM

Thanks a lot guys!

I saw the comma in another example, but now I understood why it should not be there.