sed command to delete points in a certain colum

Shell Programming and Scripting
1

Dear Unix users,

I have a (I think simple) question;

I have a file (out.dat) like below,
the file contains some line which include 'LOS' string.
.
LOS 46.5360 91.0220 200708.2515.4900. 5400 64 1100010
.
.
I would like to delete the points in 4th column.

When I used the below sed command:

sed -e '/LOS/s/\.//g' out.dat

it gives
.
LOS 465360 910220 20070825154900 5400 64 1100010
.

it deletes all the points in the line.

How can I specify that I only want to delete the points at 4th column (i.e 200708.2515.4900.)

Thank you very much,
Murat

2

Try this :-

awk '/LOS/ {gsub(/\./,"",$4);print}' <out.dat

On solaris i had to use /usr/xpg4/bin/awk

3

Hi lavascript,

Thank you very much for your reply.

your command worked, however my out.dat file contains
other lines besides the ones begin LOS. So your comment
change only lines begin with LOS and print only these lines.

...
565 233 33
LOS 46.5360 91.0220 200708.2515.4900. 5400 64 1100010
333 20
123 20 330 20
233
...

How can I print other lines as well?.
Thank you very much

4 
awk '/LOS/ {gsub(/\./,"",$4);print}'

Should be:

awk '/LOS/ {gsub(/\./,"",$4)}{print}'
5 
nawk '{
        gsub(/\./,"",$4)
        print
        }' filename