Simple script outputting who likely has the most knowledge about an area of your code base

who-knows-about.sh

#!/bin/bash

##
# Outputs the top 10 committers on the provided area, for the period provided,
# based on the number of commits, in the form of 4 columns:
# Lines added, Lines deleted, Number of commits, Author
# (This allows for easily re-sorting the output by piping it into `sort -nr -k COLUMN_NUMBER`)
#
# Tip: replace tabs with commas everywhere below to output valid CSV content.
##

##
# Sample output:
#   $ ./who-knows-about.sh some/area/of/your/repo 2022-09-01
#   Lines added   Lines deleted   Commits    Author
#   5635          1813            91         "Jane Doe <jane.doe@acme.com>"
#   1283          874             79         "Someone Else <someone@example.com>"
#   ...
##

readonly PATH_WITHIN_REPO=$1
readonly AFTER_DATE=$2

if [[ -z "$PATH_WITHIN_REPO" || -z "$AFTER_DATE" ]]; then
  >&2 echo -e "Usage $0 PATH_WITHIN_REPO AFTER_DATE\n\twith AFTER_DATE being a date following the format YYYY-MM-DD"
  exit 1
fi

readonly SHORTLOG="$(git shortlog --no-merges -se --after="$AFTER_DATE" -- "$PATH_WITHIN_REPO" | sort -nr | head -n 10 | awk '{printf $1 "\t\t\""; for (i=2; i<NF; i++) printf $i " "; print $NF  "\"" }')"

echo -e "Lines added\tLines deleted\tCommits\t\tAuthor"                                                                                                                                                                                              
IFS=$'\n'
for shortlog_entry in $SHORTLOG; do
  AUTHOR="$(echo "$shortlog_entry" | sed -E 's/.*<(.*?)>.*/\1/')"
  git log --no-merges --numstat --author="$AUTHOR" --after="$AFTER_DATE" -- "$PATH_WITHIN_REPO" | echo -e "$(awk 'NF==3 {plus+=$1; minus+=$2} END {printf("%d\t\t%d\n", plus, minus)}')\t\t$shortlog_entry"
done
unset IFS