sed-replacing and using a pattern with multi-lines

Question

I've a CSV file with multiple rows. I wanted to copy the same data multiple times and each time replacing a specific character with another character from a different file.

Example:

Below are two files, a csv and a ref file (containing A to Z) characters. I wanted to copy the csv content 26 times (A to Z) and each time replacing the variable VAR with A,B,C,....Z.

cat csv_file

some text with VAR,data 2,data 3,data 4,some text and VAR
,,,data 4,some text
,,,data 4,some text

cat ref

A
B
C
...so on upto 
Z

Result:

some text with A,data 2,data 3,data 4,some text and A
,,,data 4,some text
,,,data 4,some text
some text with B,data 2,data 3,data 4,some text and B
,,,data 4,some text
,,,data 4,some text
..
... so on

what I'm doing is

DATA="some text with & following text,data 2,data 3,data4,some text and_&\n,,,data4,some text\n,,,data 4,some text"

sed -e "s/^.*$/$DATA/g" ref

Is there any other way of achieving the same result but with file name instead of variable (DATA)?

Answer 1

Below command every time it replaces VAR with {A..Z}

Its working fine

for i in `cat /var/tmp/ref`; do  sed "s/VAR/$i/g" csv_file ;done

Sample output

some text with A,data 2,data 3,data 4,some text and A
,,,data 4,some text
,,,data 4,some text
some text with B,data 2,data 3,data 4,some text and B
,,,data 4,some text
,,,data 4,some text