Need to convert the below awk command into sed [du

Question

This question already has an answer here:

• Find and remove in unix 4 answers

I need to convert the below awk command into sed,
awk -F ',' '$2 ~ /^ *[0-9]*(\.[0-9]+)?" *$/{sub(/"/, "", $2); print $2}'

Below is my input file:

sample.txt:

3",3"
6-position,6-position
7' 4" to 10' 3-1/2",7' 4" to 10' 3-1/2"
4.8",4.8"
Adjustable from 99" to 111" - max 148,Adjustable from 99" to 111" - max 148

And i need the output as below,

output.txt:

3",3
6-position,
7' 4" to 10' 3-1/2",
4.8",4.8
Adjustable from 99" to 111" - max 148,
So basically I need to print the numeric value for the " symbol, other non-numeric text needs to be replaced by a empty line, the problem with awk command is the 2,3 and 5 line is getting removed.

Answer 1

as I commented, the awk one-liner doesn't make sense for your input file. However you could try this sed line:

sed -r '/^[0-9.]+"$/{s/"$//;n;};s/.*//' file

this works for your input file.

kent$  cat file
3"
6-position
7' 4" to 10' 3-1/2"
4.8"
Adjustable from 99" to 111" - max 148    
kent$  sed -r '/^[0-9.]+"$/{s/"$//;n;};s/.*//' file
3


4.8
(here, the last line is empty too.)

EDIT

remove -r option, and adjust the sed to work with new input/output:

sed  's/^\([^,]*,\)\([0-9.]*\)".*$/\1\2/;t;s/^\([^,]*,\).*/\1/' file

This line worked with your new input example:

kent$  cat file
3",3"
6-position,6-position
7' 4" to 10' 3-1/2",7' 4" to 10' 3-1/2"
4.8",4.8"
Adjustable from 99" to 111" - max 148,Adjustable from 99" to 111" - max 148

kent$  sed  's/^\([^,]*,\)\([0-9.]*\)".*$/\1\2/;t;s/^\([^,]*,\).*/\1/' file
3",3
6-position,
7' 4" to 10' 3-1/2",
4.8",4.8
Adjustable from 99" to 111" - max 148,

Answer 2

This works for me:

sed -e 's/^\([0-9]\+\(\.[0-9]\+\)*\)"$/\1/g' \
-e '/^\([0-9]\+\(\.[0-9]\+\)*\)$/! s/^.*$//g' sample.txt

Answer 3

You can do this with both, awk and sed. Since I have no Linux (shell) on my phone I'll try to help out with a sed solution:

sed -e 's/^([0-9.]+)"$|.*/\1/g' < file

Should do the job