CODEWITHSUNDEEP
bashsedawk
krampstudio
krampstudio

Reputation: 3611

bash replace newlines into quotes

I've the following CSV file: 

"test","test2","test
3 and some other

data"
"test4","test5","test6 and some other


data"

I would like to replace all the newlines (windows or unix style) by the character \n, so I would obtain:

"test","test2","test\n3 and some other\n\ndata"
"test4","test5","test6 and some other\n\n\ndata"

I've tried with awk but without success:

cat myFile.csv | awk -F \" -v OFS=\" '{
   for (i=0; i<NF; i++) {
      gsub("\\n", "\\\\n", $i)
   }
   print
}'

Upvotes: 2

Views: 629

Answers (3)

potong
potong

Reputation: 58391

This might work for you (GNU sed):

sed -r ':a;/^("[^"]*",?){3}$/!{$!N;s/\n/\\n/;ta};P;D' file

or at a pinch:

 sed ':a;/"$/!{$!N;s/\n/\\n/;ta};P;D' file

Upvotes: 1

Kent
Kent

Reputation: 195049

does this work for you? two lines with same idea

awk -v RS="\0"  '{gsub(/\n/,"\\n");sub(/\\n$/,"");gsub(/"\\n"/,"\"\n\"");}1' file

or

awk -v RS="\0" -v ORS="" '{gsub(/\n/,"\\n");sub(/\\n$/,"\n");gsub(/"\\n"/,"\"\n\"")}1' file

with your data:

kent$  cat file
"test","test2","test
3 and some other

data"
"test4","test5","test6 and some other


data"

output:

kent$  awk -v RS="\0" -v ORS="" '{gsub(/\n/,"\\n");sub(/\\n$/,"\n");gsub(/"\\n"/,"\"\n\"")}1' file
"test","test2","test\n3 and some other\n\ndata"
"test4","test5","test6 and some other\n\n\ndata"

Upvotes: 1

Chris Seymour
Chris Seymour

Reputation: 85785

Here is one method:

$ awk '!/"$/{sub(/$/,"\\n");printf "%s",$0;next}1' file
"test","test2","test\n3 and some other\n\ndata"
"test4","test5","test6 and some other\n\n\ndata"

Upvotes: 2

Related Questions