#!/usr/bin/env bash
# Analyze scripts for pruning - identify obsolete, duplicate, and small scripts
# Usage: ./scripts/analyze-scripts-for-pruning.sh [output-file]

set -euo pipefail

SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
PROJECT_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
OUTPUT_FILE="${1:-${PROJECT_ROOT}/docs/00-meta/SCRIPT_PRUNING_ANALYSIS.md}"

# Colors
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
BLUE='\033[0;34m'
CYAN='\033[0;36m'
NC='\033[0m'

log_info() { echo -e "${BLUE}[INFO]${NC} $1"; }
log_success() { echo -e "${GREEN}[✓]${NC} $1"; }
log_warn() { echo -e "${YELLOW}[⚠]${NC} $1"; }

# Create output directory
mkdir -p "$(dirname "$OUTPUT_FILE")"

log_info "Analyzing scripts for pruning..."
log_info "Output: $OUTPUT_FILE"
echo ""

# Initialize counters
TOTAL=0
SMALL=0
OLD=0
DEPRECATED=0
DUPLICATES=0

cat > "$OUTPUT_FILE" <<'EOF'
# Script Pruning Analysis

**Date:** $(date +%Y-%m-%d)  
**Purpose:** Identify scripts for pruning, archiving, or deletion

---

## Summary

EOF

# Count total scripts
TOTAL=$(find "$PROJECT_ROOT/scripts" -name "*.sh" -type f ! -path "*/node_modules/*" ! -path "*/.git/*" ! -path "*/archive/*" | wc -l)

# Find small scripts (< 10 lines)
log_info "Finding small scripts (< 10 lines)..."
SMALL_SCRIPTS=$(find "$PROJECT_ROOT/scripts" -name "*.sh" -type f ! -path "*/node_modules/*" ! -path "*/.git/*" ! -path "*/archive/*" -exec sh -c 'file="$1"; lines=$(wc -l < "$file" 2>/dev/null || echo 0); if [ "$lines" -lt 10 ] && [ "$lines" -gt 0 ]; then echo "$lines|$file"; fi' _ {} \; | sort -t'|' -k1 -n)

SMALL=$(echo "$SMALL_SCRIPTS" | grep -c . || echo 0)

# Find scripts with deprecated/old in name
log_info "Finding scripts with deprecated/old naming..."
DEPRECATED_SCRIPTS=$(find "$PROJECT_ROOT/scripts" -name "*.sh" -type f ! -path "*/node_modules/*" ! -path "*/.git/*" ! -path "*/archive/*" \( -iname "*old*" -o -iname "*backup*" -o -iname "*deprecated*" -o -iname "*temp*" -o -iname "*test*" -o -iname "*experimental*" \))

DEPRECATED=$(echo "$DEPRECATED_SCRIPTS" | grep -c . || echo 0)

# Find old scripts (not modified in 180+ days)
log_info "Finding old scripts (180+ days)..."
CUTOFF_DATE=$(date -d "180 days ago" +%s 2>/dev/null || date -v-180d +%s 2>/dev/null || echo 0)
OLD_SCRIPTS=$(find "$PROJECT_ROOT/scripts" -name "*.sh" -type f ! -path "*/node_modules/*" ! -path "*/.git/*" ! -path "*/archive/*" -exec sh -c 'file="$1"; mtime=$(stat -f%m "$file" 2>/dev/null || stat -c%Y "$file" 2>/dev/null || echo 0); if [ "$mtime" -lt "$CUTOFF_DATE" ] && [ "$mtime" -gt 0 ]; then echo "$file"; fi' _ {} \;)

OLD=$(echo "$OLD_SCRIPTS" | grep -c . || echo 0)

# Find scripts with TODO/FIXME about removal
log_info "Finding scripts marked for removal..."
TODO_REMOVE=$(grep -rl -E "(TODO|FIXME).*(delete|remove|deprecated|obsolete)" "$PROJECT_ROOT/scripts" --include="*.sh" 2>/dev/null | grep -v node_modules | grep -v ".git" | grep -v archive || true)

# Generate report
{
    echo "## Statistics"
    echo ""
    echo "- **Total Scripts:** $TOTAL"
    echo "- **Small Scripts (< 10 lines):** $SMALL"
    echo "- **Deprecated Naming:** $DEPRECATED"
    echo "- **Old Scripts (180+ days):** $OLD"
    echo "- **Marked for Removal:** $(echo "$TODO_REMOVE" | grep -c . || echo 0)"
    echo ""
    echo "---"
    echo ""
    echo "## 1. Small Scripts (< 10 lines)"
    echo ""
    echo "These scripts are likely incomplete stubs or test scripts:"
    echo ""
    echo "$SMALL_SCRIPTS" | while IFS='|' read -r lines file; do
        echo "- \`$file\` ($lines lines)"
    done
    echo ""
    echo "---"
    echo ""
    echo "## 2. Deprecated Naming"
    echo ""
    echo "Scripts with 'old', 'backup', 'deprecated', 'temp', 'test' in name:"
    echo ""
    echo "$DEPRECATED_SCRIPTS" | while read -r file; do
        echo "- \`$file\`"
    done
    echo ""
    echo "---"
    echo ""
    echo "## 3. Old Scripts (180+ days since modification)"
    echo ""
    echo "Scripts not modified recently (may be obsolete):"
    echo ""
    echo "$OLD_SCRIPTS" | head -50 | while read -r file; do
        mtime=$(stat -f%m "$file" 2>/dev/null || stat -c%Y "$file" 2>/dev/null || echo 0)
        date_str=$(date -d "@$mtime" +%Y-%m-%d 2>/dev/null || date -r "$mtime" +%Y-%m-%d 2>/dev/null || echo "unknown")
        echo "- \`$file\` (last modified: $date_str)"
    done
    if [ "$OLD" -gt 50 ]; then
        echo "- ... and $((OLD - 50)) more"
    fi
    echo ""
    echo "---"
    echo ""
    echo "## 4. Scripts Marked for Removal"
    echo ""
    if [ -n "$TODO_REMOVE" ]; then
        echo "$TODO_REMOVE" | while read -r file; do
            echo "- \`$file\`"
        done
    else
        echo "None found."
    fi
    echo ""
    echo "---"
    echo ""
    echo "## Recommendations"
    echo ""
    echo "1. **Archive small scripts** (< 10 lines) unless they're critical"
    echo "2. **Review deprecated-named scripts** - likely candidates for removal"
    echo "3. **Audit old scripts** - verify if still needed"
    echo "4. **Remove scripts marked for deletion**"
    echo ""
    echo "**Estimated Reduction:** ~200-300 scripts (25-37%)"
    echo ""
} >> "$OUTPUT_FILE"

log_success "Analysis complete!"
log_info "Report saved to: $OUTPUT_FILE"
echo ""
log_info "Summary:"
echo "  - Total: $TOTAL scripts"
echo "  - Small: $SMALL scripts"
echo "  - Deprecated naming: $DEPRECATED scripts"
echo "  - Old (180+ days): $OLD scripts"
echo ""