Replace all surnames following 'Mr' prefix using bash script

I have a csv file where each line contains a person’s name like this.

| 11th October  | 06 6606 1161 | Mr Bean is 43 years old. Mr Bean likes apples. |

My goal is to write a bash script to replace the surnames in the file with a new surname.

So for example with a script replace_surname.sh –

#!/bin/sh

if [ ! -z $1 ]
then
    #replace surname following 'Mr' with $1
fi

I’d expect, given the above csv and the command replace_surname.sh Smith

| 11th October  | 06 6606 1161 | Mr Smith is 43 years old. Mr Smith likes apples. |

What I’ve tried –

sed -i 's/^Mr .*$/Mr PacMan/' test.csv

And a few other from here
Which doesn’t affect the file at all.

I’m a bit stuck and my csv file has hundreds of records so I wouldn’t like to do it all by hand! Any help is very much appreciated 🙂

>Solution :

This should do it:

#!/bin/sh

if [ -n "$1" ]; then
  new_surname="$1"
  sed -i "s/Mr\s\+\(\w\+\)/Mr ${new_surname}/g" /tmp/mr
fi

that is if the old surname is 1 word

Regex can be further improved, but you should get the idea