Reputation: 123
I want to remove quotes from specific field where there is an integer between quotes.
From this line : "AAA","99"
to this line : "AAA",99
Upvotes: 2
Views: 103
Reputation: 44023
With awk:
awk -F , 'BEGIN { OFS = FS } $2 ~ /^"[0-9]+"$/ { gsub(/"/, "", $2) } 1'
That works as follows:
BEGIN { OFS = FS } # Delimit output the same way as the input
$2 ~ /^"[0-9]+"$/ { # if the second field ($2) consists of only numbers encased
# by double quotes (i.e., matches the regex ^"[0-9]"$)
gsub(/"/, "", $2) # remove the quotes from it
}
1 # print in any case
To make this work for any (and not just a specific field), do the replacement in a loop:
awk -F , 'BEGIN { OFS = FS } { for(i = 1; i <= NF; ++i) if($i ~ /^"[0-9]+"$/) { gsub(/"/, "", $i) } } 1'
The changed bit here is
{ # Do this unconditionally in every line:
for(i = 1; i <= NF; ++i) { # wade through all the fields
if($i ~ /^"[0-9]+"$/) { # then do with them the same thing as before.
gsub(/"/, "", $i)
}
}
}
Upvotes: 1
Reputation: 784958
Using sed
:
s='"AAA","99"'
sed 's/"\([0-9]*\)"/\1/g' <<< "$s"
"AAA",99
Upvotes: 3
Reputation: 88573
Try this with GNU sed:
echo '"AAA","99"' | sed -E 's/"([0-9]+)"$/\1/'
Output:
"AAA",99
Upvotes: 2