The examples below are each borken down in two parts. The first part shows a short version of the commands you might use in practice. This is followed by a step by step implementation to take some of the mystery out of the short version.
How to easily convert a name in “FAMILY, GIVEN” form to “GIVEN FAMILY” form?
Traditionally this can be done using a number of Unix commands such as echo and cut. If you break down the task into getting each name separately it is pretty easy using a couple shell variables.
NAME="Doiel, Robert"
GIVEN_FAMILY="$(echo -n "$NAME" | cut -d , -f 1) $(echo -n "$NAME" | cut -d , -f 2)"
echo "$FAMILY $GIVEN"
This uses the subshell syntax and two separate pipe lines. Let’s break it down by parts.
NAME="Doiel, Robert"
echo "Step 1: [$NAME]"
FAMILY_NAME=$(echo -n "$NAME" | cut -d , -f 1)
echo "Step 2: [$FAMILY_NAME]
GIVEN_NAME=$(echo -n "$NAME" | cut -d , -f 2)
echo "Step 3: [$GIVEN_NAME $FAMILY_NAME]"
Each pipeline builds up a name (family and given) and the final echo displays them.
NOTE: The trouble is this doesn’t give you what you want. Notice the leading space. You can fix that but that is just the start of the rabbit whole.
If we think about the family name and given name as elements of an array we can easily reorder them. In this approach we’ll use two commands. The string and jsoncols commands from datatools.
NAME="Doiel, Robert"
string split ", " "$NAME" | jsoncols -i - -d ' ' '.[0]' '.[1]'
First difference you’ll notice is we’re using an Unix pipe to send the output of one command to another. While you can eventually do that in the traditional approach it becomes very complicated very quickly. Using datatools it is easy to move from strings to JSON and back.
Let’s take the datatools approach and output the results of each step rather than using a pipeline.
NAME="Doiel, Robert"
echo "Step 1: [$NAME]"
JSON_ARRAY=$(string split ", " "$NAME")
echo "Step 2: $JSON_ARRAY"
GIVEN_FAMILY=$(jsoncols -d ' ' '.[0]' '.[1]' "$JSON_ARRAY")
echo "Step 3: [$GIVEN_FAMILY]"
NOTE: In this version there is no leading space issue. string command can split on multiple characters and in our case it is splitting on comma space not just on the comma like we get with cut.