#!/bin/bash
# =====================================================
# ERP GENERIC - Health Check Script
# Comprehensive system health check
# =====================================================

set -euo pipefail

# Configuration
BACKEND_URL="${BACKEND_URL:-http://localhost:3000}"
FRONTEND_URL="${FRONTEND_URL:-http://localhost:5173}"
DB_HOST="${POSTGRES_HOST:-postgres}"
DB_PORT="${POSTGRES_PORT:-5432}"
DB_NAME="${POSTGRES_DB:-erp_generic}"
DB_USER="${POSTGRES_USER:-erp_user}"
REDIS_HOST="${REDIS_HOST:-redis}"
REDIS_PORT="${REDIS_PORT:-6379}"
REDIS_PASSWORD="${REDIS_PASSWORD:-}"

# Colors for output
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
NC='\033[0m' # No Color

# Exit code (0 = healthy, 1 = unhealthy)
EXIT_CODE=0

# Helper functions
print_header() {
    echo ""
    echo "=========================================="
    echo "$1"
    echo "=========================================="
}

check_success() {
    echo -e "${GREEN}✓ $1${NC}"
}

check_warning() {
    echo -e "${YELLOW}⚠ $1${NC}"
}

check_failure() {
    echo -e "${RED}✗ $1${NC}"
    EXIT_CODE=1
}

# Main health check
echo "===== ERP Generic Health Check ====="
echo "Started at: $(date)"
echo ""

# 1. Backend API Health Check
print_header "1. Backend API"

if curl -f -s --max-time 5 "${BACKEND_URL}/health" > /dev/null 2>&1; then
    check_success "Backend API is responding"

    # Check detailed health
    HEALTH_RESPONSE=$(curl -s --max-time 5 "${BACKEND_URL}/health")
    echo "Response: $HEALTH_RESPONSE"

    # Parse JSON response (requires jq)
    if command -v jq &> /dev/null; then
        DB_STATUS=$(echo "$HEALTH_RESPONSE" | jq -r '.checks.database.status' 2>/dev/null || echo "unknown")
        REDIS_STATUS=$(echo "$HEALTH_RESPONSE" | jq -r '.checks.redis.status' 2>/dev/null || echo "unknown")

        if [ "$DB_STATUS" = "up" ]; then
            check_success "Database connection: OK"
        else
            check_failure "Database connection: FAILED"
        fi

        if [ "$REDIS_STATUS" = "up" ]; then
            check_success "Redis connection: OK"
        else
            check_failure "Redis connection: FAILED"
        fi
    fi
else
    check_failure "Backend API is not responding"
fi

# 2. Frontend Health Check
print_header "2. Frontend"

if curl -f -s --max-time 5 "${FRONTEND_URL}" > /dev/null 2>&1; then
    check_success "Frontend is accessible"
else
    check_failure "Frontend is not accessible"
fi

# 3. PostgreSQL Database Check
print_header "3. PostgreSQL Database"

PGPASSWORD="${POSTGRES_PASSWORD}"
export PGPASSWORD

if pg_isready -h "$DB_HOST" -p "$DB_PORT" -U "$DB_USER" > /dev/null 2>&1; then
    check_success "PostgreSQL is accepting connections"

    # Check database exists
    if psql -h "$DB_HOST" -p "$DB_PORT" -U "$DB_USER" -lqt | cut -d \| -f 1 | grep -qw "$DB_NAME"; then
        check_success "Database '$DB_NAME' exists"

        # Check connection count
        CONNECTION_COUNT=$(psql -h "$DB_HOST" -p "$DB_PORT" -U "$DB_USER" -d "$DB_NAME" -tAc "SELECT count(*) FROM pg_stat_activity WHERE datname='$DB_NAME';" 2>/dev/null || echo "0")
        MAX_CONNECTIONS=$(psql -h "$DB_HOST" -p "$DB_PORT" -U "$DB_USER" -d "$DB_NAME" -tAc "SHOW max_connections;" 2>/dev/null || echo "100")

        echo "Active connections: $CONNECTION_COUNT / $MAX_CONNECTIONS"

        if [ "$CONNECTION_COUNT" -gt $((MAX_CONNECTIONS * 80 / 100)) ]; then
            check_warning "Connection pool is >80% utilized"
        else
            check_success "Connection pool is healthy"
        fi

        # Check table count
        TABLE_COUNT=$(psql -h "$DB_HOST" -p "$DB_PORT" -U "$DB_USER" -d "$DB_NAME" -tAc "SELECT COUNT(*) FROM information_schema.tables WHERE table_schema NOT IN ('pg_catalog', 'information_schema');" 2>/dev/null || echo "0")
        echo "Tables in database: $TABLE_COUNT"

        if [ "$TABLE_COUNT" -gt 0 ]; then
            check_success "Database has tables"
        else
            check_warning "Database has no tables (not initialized?)"
        fi

        # Check for recent errors in pg_stat_database
        ERROR_COUNT=$(psql -h "$DB_HOST" -p "$DB_PORT" -U "$DB_USER" -d "$DB_NAME" -tAc "SELECT deadlocks FROM pg_stat_database WHERE datname='$DB_NAME';" 2>/dev/null || echo "0")
        if [ "$ERROR_COUNT" -gt 0 ]; then
            check_warning "Database has $ERROR_COUNT deadlocks recorded"
        fi

    else
        check_failure "Database '$DB_NAME' does not exist"
    fi
else
    check_failure "PostgreSQL is not accepting connections"
fi

# 4. Redis Cache Check
print_header "4. Redis Cache"

if [ -n "$REDIS_PASSWORD" ]; then
    REDIS_CLI_CMD="redis-cli -h $REDIS_HOST -p $REDIS_PORT -a $REDIS_PASSWORD"
else
    REDIS_CLI_CMD="redis-cli -h $REDIS_HOST -p $REDIS_PORT"
fi

if $REDIS_CLI_CMD ping > /dev/null 2>&1; then
    check_success "Redis is responding"

    # Check Redis info
    REDIS_VERSION=$($REDIS_CLI_CMD INFO server | grep redis_version | cut -d: -f2 | tr -d '\r')
    echo "Redis version: $REDIS_VERSION"

    CONNECTED_CLIENTS=$($REDIS_CLI_CMD INFO clients | grep connected_clients | cut -d: -f2 | tr -d '\r')
    echo "Connected clients: $CONNECTED_CLIENTS"

    USED_MEMORY=$($REDIS_CLI_CMD INFO memory | grep used_memory_human | cut -d: -f2 | tr -d '\r')
    echo "Used memory: $USED_MEMORY"

    check_success "Redis is healthy"
else
    check_failure "Redis is not responding"
fi

# 5. Disk Space Check
print_header "5. Disk Space"

DISK_USAGE=$(df -h / | awk 'NR==2 {print $5}' | sed 's/%//')

echo "Disk usage: ${DISK_USAGE}%"

if [ "$DISK_USAGE" -lt 70 ]; then
    check_success "Disk space is healthy"
elif [ "$DISK_USAGE" -lt 85 ]; then
    check_warning "Disk space is >70% utilized"
else
    check_failure "Disk space is critically low (>85%)"
fi

# 6. CPU & Memory Check
print_header "6. System Resources"

if command -v free &> /dev/null; then
    MEMORY_USAGE=$(free | awk 'NR==2 {printf "%.0f", $3*100/$2}')
    echo "Memory usage: ${MEMORY_USAGE}%"

    if [ "$MEMORY_USAGE" -lt 80 ]; then
        check_success "Memory usage is healthy"
    else
        check_warning "Memory usage is high (>80%)"
    fi
fi

if command -v top &> /dev/null; then
    CPU_USAGE=$(top -bn1 | grep "Cpu(s)" | awk '{print $2}' | cut -d'%' -f1)
    echo "CPU usage: ${CPU_USAGE}%"

    # Note: This is a simple check, may need adjustment based on system
    if command -v awk &> /dev/null; then
        if awk "BEGIN {exit !($CPU_USAGE < 80)}"; then
            check_success "CPU usage is healthy"
        else
            check_warning "CPU usage is high (>80%)"
        fi
    fi
fi

# 7. Docker Containers Check (if applicable)
print_header "7. Docker Containers"

if command -v docker &> /dev/null; then
    RUNNING_CONTAINERS=$(docker ps --filter "name=erp" --format "{{.Names}}" | wc -l)
    echo "Running ERP containers: $RUNNING_CONTAINERS"

    if [ "$RUNNING_CONTAINERS" -gt 0 ]; then
        check_success "Docker containers are running"

        # List container status
        docker ps --filter "name=erp" --format "table {{.Names}}\t{{.Status}}\t{{.Ports}}"
    else
        check_warning "No ERP containers are running"
    fi
else
    echo "Docker is not installed or not in PATH"
fi

# 8. Log Errors Check (last 1 hour)
print_header "8. Recent Errors in Logs"

if [ -f "/var/log/erp-generic/backend.log" ]; then
    ERROR_COUNT=$(grep -c "ERROR" /var/log/erp-generic/backend.log 2>/dev/null || echo "0")
    echo "Errors in last log file: $ERROR_COUNT"

    if [ "$ERROR_COUNT" -eq 0 ]; then
        check_success "No errors in recent logs"
    elif [ "$ERROR_COUNT" -lt 10 ]; then
        check_warning "Found $ERROR_COUNT errors in logs"
    else
        check_failure "Found $ERROR_COUNT errors in logs (investigate!)"
    fi
else
    echo "Log file not found: /var/log/erp-generic/backend.log"
fi

# Summary
print_header "Health Check Summary"

if [ $EXIT_CODE -eq 0 ]; then
    echo -e "${GREEN}✓ All systems are healthy${NC}"
else
    echo -e "${RED}✗ Some systems are unhealthy (see above)${NC}"
fi

echo ""
echo "Completed at: $(date)"
echo "===== End of Health Check ====="

# Exit with appropriate code (0 = healthy, 1 = unhealthy)
exit $EXIT_CODE