More Working with CSV Files from the Command Line

Not something I use very often, it seems a bit "involved" so it doesn't always occur to me at first, but it is useful.

According to wikipedia ksh, zsh, and rc also support process substitution.

Here's a shell script that contains the code from this video and the previous one:

#!/bin/bash

# Extract emails:
# Could also do: grep -i -o '[^,]*@[^,]*' stuff.csv
cut -f 3 -d , stuff.csv
echo

# Use emails to extract lines:
for i in $(cut -f 3 -d , stuff.csv | sort --ignore-case | uniq --ignore-case)
do
        grep -i -F "$i" stuff.csv | head -n 1
done
echo

# Make sure posix mode is OFF so process substitution works "<(...)" syntax.
set +o posix

# Use emails to de-dupe a file and then extract lines:
for i in $(comm -23 \
                <(cut -f 3 -d , stuff.csv | sort --ignore-case | uniq --ignore-case) \
                <(cut -f 3 -d , nostuff.csv | sort --ignore-case | uniq --ignore-case))
do
        grep -i -F "$i" stuff.csv | head -n 1
done

Here's stuff.csv:

1,a,joe@example.com,e
2,b,jim@example.com,f
3,c,sally@example.com,g

And nostuff.csv:

3,c,sally@example.com,g