i have a pipe delimited file that looks like this:

123| ste hen| 456| out put 
143 | ste hen| 456| out put

i want to delete the spaces--- something like this

123,ste hen,456,out put
143,ste hen,456,out put

this is my code but dont works..

 awk '{gsub(/^  |  $/,"|"); gsub(/ *, */,"|")}1'  test2.txt

CodePudding user response：

awk '{gsub(/ *| */, "|");print}' test2.txt

CodePudding user response：

My very first and fast approach is the next one:

cat testFile.txt | sed 's/|/,/g' | sed 's/\s*,\s*/,/g''

The past command display the following result:

123,ste hen,456,out put 
143,ste hen,456,out put

from the following input:

123| ste hen| 456| out put 
143 | ste hen| 456| out put

CodePudding user response：

I would use GNU AWK following way, let file.txt content be

123| ste hen| 456| out put 
143 | ste hen| 456| out put

then

awk '{gsub(/ *\| */,",");print}' file.txt

output

123,ste hen,456,out put 
143,ste hen,456,out put

Changes done as compared to {gsub(/^ | $/,"|"); gsub(/ *, */,"|")}1

• eliminated ^ i.e. begin
• eliminated $ i.e. end
• escaped | as you need literal | not alternative
• looking for zero or more (*) spaces around each | rather than at least 1 at each side
• replaced two gsubs with single one
• trigger printing by print rather 1 (matter of style, consult style guide of your organization to use solution appropriate in your case)

(tested in gawk 4.2.1)

CodePudding user response：

A sed solution:

sed 's/^ *//; s/ *$//; s/ *| */,/g' file