Copy specific Column from multiple csv file and write it to new csv file. Shell Scripting

Question

I have multiple CSV files from these I want to extract 3rd column and write it new file, in the same manner, the file is sequenced in a folder. Append should be in such a manner that all the columns come side by side and not one below another. I used multiple commands

paste -d "," *csv >> Main.csv

Resulted in improper sequences of the data appending. Files sequencing was wrong. (wasn't able to understand how to append-only specific columns, entire file was appended but were side by side)

cut -d ',' -f1-2 *.csv t.csv

Resulted in appending data one below another instead of side by side. (and appended two columns which I can change to append one column).

cut -d ',' -f1-2 File1.csv | paste File2.csv - > Main.csv

This method worked for 2 files but again how to extend it to reading multiple files.

Eg:-
File_1.csv
a,32,37
b,26,34 
c,56,65

File_2.csv
a,92,60
b,48,63 
c,52,23

File_3.csv
a,2,37
b,45,53 
c,56,63
.
.
.
.


Output_File.csv
a,37,60,37 . . . .
b,34,63,53 . . . .
c,65,23,63 . . . .

Answer 1

The following assumes that

1. The CSV data is "simple", meaning its fields do not contain embedded commas or newlines.
2. All files have the same number of rows.
3. The current directory contains at least one CSV file.

Let us start by getting a list of the files:

rm -f out.csv
filelist=( *.csv )

This creates an array called filelist containing the names of all files in the current directory that match the pattern *.csv. I started by removing out.csv as we will use that name for the result file and we don't want to include that in the processing if it exists.

Then we extract the first column from the first of these files. We save the column to a new file called out.csv.

cut -d , -f 1 -- "${filelist[0]}" >out.csv

We then loop over the files, pulling out the 3rd column of each file and adding it to the existing out.csv file by means of cut, paste, and an intermediate file called out.tmp.

for file in "${filelist[@]}"; do
    cut -d , -f 3 -- "$file" | paste -d , out.csv - >out.tmp &&
    mv out.tmp out.csv
done

The whole thing, but rewritten for /bin/sh (which has no arrays):

rm -f out.csv
set -- *.csv

cut -d , -f 1 -- "$1" >out.csv

for file do
    cut -d , -f 3 -- "$file" | paste -d , out.csv - >out.tmp &&
    mv out.tmp out.csv
done

Answer 2

Code i am finding it easy to use is .

  paste -d"," *csv>>OG.csv
  cut --complement -d',' -f3,4,5,7,8,10,.. OG.csv>>Data.csv

if there are any alternate ways please let time know.