Shell script - Read CSV file lines and add prefix for multiple values in a cell

Question

I have a csv file like this.

aaa|c1|bbb|t1
bbb|c1,c2|nnn|t1,t2

The pipe is the delimiter. I want to generate a string with columns 2 and 4. And I need to add a prefix for both column values.

Column 2 = a is the prefix column 4 = b is the prefix

Expected output:

this is final string a.c1=b.t1
this is final string a.c1,a.c2=-b.t1,b,t2

My sample script (in complete):

while read r_line
do
c2 = $(echo $r_line|awk -F'|' '{print $2}')
c4 = $(echo $r_file |awk -F'|' '{print $4}')
out=$("this is final string a.$c2=b.$c4")
done < csv file

Here if the c2 or t2 has comma-separated values I need to apply the prefix for both the values.

Rakesh Sharma · Accepted Answer · 2020-07-07 08:01:13Z

1

You can do it all within bash by reading the csv file into an array and then outputting the 2nd nd 4th fields by first doing parameter substitution within them

while IFS='|' read -ra a;do
  echo This is the final string: \
    "a.${a[1]//,/,a.}=b.${a[3]//,/,b.}"
done < csvfile

answered Jul 7, 2020 at 8:01

Rakesh Sharma

1,1021 gold badge6 silver badges3 bronze badges

Add a comment |

CHEN Yunkai · Accepted Answer · 2020-07-07 07:06:42Z

0

An approach using sed.

prefix1=a.
prefix2=b.
while read r_line; do
    c2=$(echo $r_line | cut -d'|' -f2)
    c4=$(echo $r_line | cut -d'|' -f4)
    s1=$(echo $c2 | sed "s/,/,$prefix1/g" | sed "s/^/$prefix1/g")
    s2=$(echo $c4 | sed "s/,/,$prefix2/g" | sed "s/^/$prefix2/g")
    echo "this is final string $s1=$s2"
done

The script search , and instead with ,$prefix and finally add $prefix at head. And I like to use cut to split strings.

answered Jul 7, 2020 at 7:06

CHEN Yunkai

186 bronze badges

Add a comment |

Prvt_Yadav · Accepted Answer · 2020-07-07 07:20:44Z

0

You can read each line, use cut to get each field and the use sed on them.

while read -r line; do
a="$(cut -d'|' -f2 <<<"$line")"
b="$(cut -d'|' -f4 <<<"$line")"
a="$(echo "$a"|sed -e 's/^/a./' -e 's/,/,a./g')"
b="$(echo "$b"|sed -e 's/^/b./' -e 's/,/,b./g')"
echo "this is final string $a=$b"
done < csv_file

answered Jul 7, 2020 at 7:20

Prvt_Yadav

6,0327 gold badges39 silver badges52 bronze badges

Add a comment |

Archemar · Accepted Answer · 2020-07-07 07:53:13Z

full awk solution

awk file (se.awk )

BEGIN { FS="|" }
{
  a="" ; n=split($2,A,",") ; for (i=1;i<=n;i++) a = a ",a." A[i] ;
  b="" ; n=split($4,B,",") ; for (i=1;i<=n;i++) b = b ",b." B[i] ;
  printf "%s %s=%s\n",prefix,substr(a,2),substr(b,2) ;
}

called as

awk -v prefix="this is final string" -f se.awk c
this is final string a.c1=b.t1
this is final string a.c1,a.c2=b.t1,b.t2

where

-v prefix="this is final string" set var prefix in command line
BEGIN { FS="|" } tell awk to use | as separator
a="" ; n=split($2,A,",") split second field on , and count
for (i=1;i<=n;i++) a = a ",a." A[i] build expected string with a leading unwanted comma
(same goes for b on fourth position )
prefix="this is final string" assemble string using printf, striping comma

if prefix is fixed, it can be insert in awk program

BEGIN { 
  FS="|" ;
  prefix="this is prefix" ; 
}

Stack Exchange Network

Shell script - Read CSV file lines and add prefix for multiple values in a cell

Expected output:

My sample script (in complete):

4 Answers 4

You must log in to answer this question.

Hot Network Questions

Shell script - Read CSV file lines and add prefix for multiple values in a cell

Expected output:

My sample script (in complete):

4 Answers 4

You must log in to answer this question.

Related

Hot Network Questions