Count how many tokens your codebase is

count-typescript-stats.sh

#!/bin/bash

# Count TypeScript files statistics in src directory

echo "=== TypeScript Files Statistics ==="
echo

# Count files
file_count=$(find src -name "*.ts" -o -name "*.tsx" | wc -l)
echo "Total files (.ts and .tsx): $file_count"

# Count words
word_count=$(find src -name "*.ts" -o -name "*.tsx" | xargs wc -w | tail -1 | awk '{print $1}')
echo "Total words: $word_count"

# Estimate tokens (rough approximation: 1 word ≈ 1.3 tokens for code)
tokens_low=$(echo "$word_count * 1.3" | bc | cut -d. -f1)
tokens_high=$(echo "$word_count * 1.5" | bc | cut -d. -f1)
echo "Estimated tokens: $tokens_low - $tokens_high"

echo
echo "=== Breakdown by directory ==="
echo

# Show breakdown by major directories
for dir in src/lib src/model src/routes src/services src/views src/workflows; do
    if [ -d "$dir" ]; then
        dir_files=$(find "$dir" -name "*.ts" -o -name "*.tsx" | wc -l)
        dir_words=$(find "$dir" -name "*.ts" -o -name "*.tsx" | xargs wc -w 2>/dev/null | tail -1 | awk '{print $1}')
        dir_tokens_low=$(echo "$dir_words * 1.3" | bc | cut -d. -f1)
        dir_tokens_high=$(echo "$dir_words * 1.5" | bc | cut -d. -f1)
        printf "%-20s %3d files, %6d words, %6d - %6d tokens\n" "$dir:" "$dir_files" "$dir_words" "$dir_tokens_low" "$dir_tokens_high"
    fi
done

echo
echo "=== Top 10 largest files by word count ==="
echo

# Show top 10 largest files
find src -name "*.ts" -o -name "*.tsx" | xargs wc -w | sort -nr | head -11 | tail -10