git-contributors

git-contributors.sh

#!/usr/bin/env bash
#
# git-contributors [options] <path>
#
# Show the main contributors for a given path, ranked by commits, with
# lines added/removed, the date range of their activity, and a per-author
# sparkline of commit activity over time (one cell per month).
#
# Defaults to the last 12 months. Pass --since "" to include all history.

set -euo pipefail

usage() {
    cat <<'EOF'
git-contributors — who are the active contributors for a path?

Usage:
  git-contributors [options] <path>

Options:
  -s, --since <when>   Window to consider (default: "12 months ago").
                       Use --since "" for all history.
  -n, --top <N>        Show only the top N contributors (default: all).
  -e, --email          Group by email instead of name.
  -h, --help           Show this help.

Examples:
  git-contributors .                              # this directory, last 12 months
  git-contributors internal/activities            # a subdirectory
  git-contributors --top 10 .                      # top 10 contributors
  git-contributors --since "3 months ago" .        # narrower window
  git-contributors --since "" services/pipeline    # all history
  git-contributors --email .                        # group by email

The ACTIVITY column is a sparkline of commits per month (oldest -> newest)
across the selected window, so you can see who is ramping up or trailing off.
EOF
}

# No arguments at all -> show help.
if [[ $# -eq 0 ]]; then
    usage
    exit 0
fi

since="12 months ago"
top=0
group_by="name"
path=""

while [[ $# -gt 0 ]]; do
    case "$1" in
        -s|--since) since="$2"; shift 2 ;;
        -n|--top)   top="$2"; shift 2 ;;
        -e|--email) group_by="email"; shift ;;
        -h|--help)  usage; exit 0 ;;
        -*)         echo "unknown option: $1" >&2; usage >&2; exit 1 ;;
        *)          path="$1"; shift ;;
    esac
done

if [[ -z "$path" ]]; then
    echo "error: no path given" >&2
    echo >&2
    usage >&2
    exit 1
fi

if ! git rev-parse --is-inside-work-tree >/dev/null 2>&1; then
    echo "error: not inside a git repository" >&2
    exit 1
fi

since_args=()
[[ -n "$since" ]] && since_args=(--since="$since")

if [[ "$group_by" == "email" ]]; then
    fmt="%ae"
else
    fmt="%an"
fi

total_commits=$(git log "${since_args[@]}" --no-merges --oneline -- "$path" | wc -l | tr -d ' ')

echo "Contributors for: $path"
if [[ -n "$since" ]]; then
    echo "Window: since $since"
else
    echo "Window: all history"
fi
echo "Total commits: $total_commits"
echo

if [[ "$total_commits" == "0" ]]; then
    echo "No commits found."
    exit 0
fi

# Emit per-commit author + month, plus numstat lines. awk aggregates totals
# and a per-author per-month commit count, then renders a sparkline spanning
# the full month range present in the data.
git log "${since_args[@]}" --no-merges \
    --pretty=format:"C%x09${fmt}%x09%ad" --date=format:'%Y-%m' --numstat -- "$path" \
| awk -F'\t' '
    function spark(count,   levels, i) {
        if (count <= 0) return " "
        # 8 levels of block characters
        split("▁▂▃▄▅▆▇█", levels, "")
        i = count
        if (i > 8) i = 8
        return levels[i]
    }
    /^C\t/ {
        author = $2
        month  = $3
        commits[author]++
        cell[author SUBSEP month]++
        months[month] = 1
        if (!(author in first) || month < first[author]) first[author] = month
        if (!(author in last)  || month > last[author])  last[author]  = month
        next
    }
    NF >= 3 && $1 ~ /^[0-9]+$/ {
        added[author]   += $1
        removed[author] += $2
    }
    END {
        # Ordered list of all months seen.
        n = 0
        for (m in months) order[++n] = m
        # insertion sort (small n)
        for (i = 2; i <= n; i++) {
            key = order[i]; j = i - 1
            while (j >= 1 && order[j] > key) { order[j+1] = order[j]; j-- }
            order[j+1] = key
        }

        printf "%-26s %7s %8s %8s   %-*s  %s\n", "AUTHOR", "COMMITS", "+LINES", "-LINES", (n>9?n:9), "ACTIVITY", "RANGE"

        for (a in commits) {
            line = ""
            for (i = 1; i <= n; i++) line = line spark(cell[a SUBSEP order[i]])
            printf "%d\t%-26s %7d %8d %8d   %-*s  %s -> %s\n", \
                commits[a], a, commits[a], added[a], removed[a], (n>9?n:9), line, first[a], last[a]
        }
    }
' \
| { read -r header; echo "$header"; sort -rn -k1,1 | cut -f2-; } \
| { [[ "$top" -gt 0 ]] && { read -r header; echo "$header"; head -n "$top"; } || cat; }

if [[ -n "$since" ]]; then
    echo
    echo "(ACTIVITY: one cell per month, oldest -> newest; taller = more commits)"
fi