how to split csv file, keep header and add file extension - using Terminal

It looks like you already have ".csv" on your file names. Do you see these file extensions when using the "ls" command to see the files in the Terminal? Or are the file extensions only hidden when viewing the files in the Finder? The Finder hides file extensions by default. I know that some macOS GUI apps will have an option to "show" the file extension when saving a file. However, I don't know what these GUI apps do to make the file extension visible in the Finder. The Finder also includes the ability to show all file extensions, but I find this annoying as it shows it for the Applications as well.

In the future please use the "Code Insertion" tool indicated by the "<>" icon when writing a post. The Code Insertion tool will allow better formatting & readability of posted code snippets.

You should enclose your paths in double quotes in case any folders or the filename itself includes any spaces. You should also enclose the third line in double quotes as well to preserve spaces when assigning the output to the variable:

HDR="$(head -1  $FILENAME)"

I guess I misunderstood your request the when I responded previously. You need to modify the "split" command to add the file extension, but I don't know if the bash included with macOS supports the option which is:

 --additional-suffix=".csv"

So the "split" command becomes:

split -l 20000  --additional-suffix=".csv"  "$FILENAME"  xyz 

If macOS doesn't support this option to "split", then you will need to include this command to rename the "xyz" files by inserting this into your script immediately after the "split" line:

for i in xyz*
do
    mv "$i" "$i.csv"
done

I tried both options I suggested in my earlier post on my Linux system. Both the xyz* files and the Part* files had .csv extensions. I cannot see that macOS wouldn't run this script correctly. The only difference between my script and yours is that I used a path to the files as I didn't want to risk accidentally creating or deleting files in an unexpected location. Here is my script for reference which works on Linux:

#!/bin/bash
FILENAME=split.csv
HDR="$(head -1 $FILENAME)"
#split -l 3 --additional-suffix=".csv"  $FILENAME  xyz

split -l 3   $FILENAME  xyz

for i in xyz*
do
    printf "Renaming $i to $i.csv\n"
    mv ./"$i" ./"$i.csv"
done


n=1
for f in xyz*
do
    if [ $n -gt 1 ]; then
        echo "$HDR" > ./Part"$n".csv
    fi
    cat $f >> ./Part"$n".csv
    rm -i "$f"
    ((n++))
done

Here is what happens when I run the script on Linux (I declined to delete the temporary files):

Files in folder before running sript....
split.csv
split-script.sh
****************************************
Running script.....
Renaming xyzaa to xyzaa.csv
Renaming xyzab to xyzab.csv
****************************************
Files in folder after running script....
Part1.csv
Part2.csv
split.csv
split-script.sh
xyzaa.csv
xyzab.csv

Here is the contents of the original "split.csv" file:

This is the header line.....
This is line# 1
This is line# 2
This is line# 3
This is line# 4
This is line# 5

Here is the contents of the "Part2.csv" file:

This is the header line.....
This is line# 3
This is line# 4
This is line# 5

dtarver13 wrote:

I think I found the answer. FINALLY! :D The following splits the file, adds the header into each split and adds the file extension to each file. Whew! I tested it. Regards, ~Dora

That script looks identical to the one in your original post. What is different about it which made it work?