#!/bin/bash

# Content Management Shared Library
# This library provides common functions used across content scripts

set -e

# Configuration defaults
CONTENT_DIR="${SITE_CONTENT_PATH:-site/content}"
CONTENT_KINDS_FILE="${CONTENT_DIR}/content-kinds.toml"
DEFAULT_LANGUAGES=("en" "es")

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

# =============================================================================
# LOGGING FUNCTIONS
# =============================================================================

log_error() {
    echo -e "${RED}❌ ERROR: $1${NC}" >&2
}

log_warning() {
    echo -e "${YELLOW}⚠️  WARNING: $1${NC}"
}

log_success() {
    echo -e "${GREEN}✅ $1${NC}"
}

log_info() {
    echo -e "${BLUE}ℹ️  $1${NC}"
}

log_step() {
    echo -e "${CYAN}🔧 $1${NC}"
}

# =============================================================================
# CONTENT-KINDS.TOML PARSING
# =============================================================================

# Get enabled content types from content-kinds.toml
get_enabled_content_types() {
    if [ ! -f "$CONTENT_KINDS_FILE" ]; then
        log_warning "content-kinds.toml not found at $CONTENT_KINDS_FILE, using defaults"
        echo "blog recipes"
        return
    fi
    
    # Robust TOML parsing - process line by line and track state
    local in_content_kinds_section=false
    local current_name=""
    local current_directory=""
    local current_enabled=""
    local enabled_types=()
    
    while IFS= read -r line; do
        # Skip comments and empty lines
        [[ "$line" =~ ^[[:space:]]*# ]] && continue
        [[ -z "${line// }" ]] && continue
        
        # Check for new content_kinds section
        if [[ "$line" =~ ^\[\[content_kinds\]\] ]]; then
            # Process previous section if we have one
            if [[ "$in_content_kinds_section" == true && "$current_enabled" == "true" && -n "$current_directory" ]]; then
                if [ -d "${CONTENT_DIR}/${current_directory}" ]; then
                    enabled_types+=("$current_directory")
                    log_info "Content type '$current_name' -> directory '${current_directory}' enabled" >&2
                else
                    log_warning "Content type '$current_name' enabled but directory '${CONTENT_DIR}/${current_directory}' missing - SKIPPING" >&2
                fi
            fi
            
            # Reset for new section
            in_content_kinds_section=true
            current_name=""
            current_directory=""
            current_enabled=""
            continue
        fi
        
        # If we're in a content_kinds section, parse the fields
        if [[ "$in_content_kinds_section" == true ]]; then
            # Check for another section starting (end current one)
            if [[ "$line" =~ ^\[\[ ]]; then
                # Process current section before moving to next
                if [[ "$current_enabled" == "true" && -n "$current_directory" ]]; then
                    if [ -d "${CONTENT_DIR}/${current_directory}" ]; then
                        enabled_types+=("$current_directory")
                        log_info "Content type '$current_name' -> directory '${current_directory}' enabled" >&2
                    else
                        log_warning "Content type '$current_name' enabled but directory '${CONTENT_DIR}/${current_directory}' missing - SKIPPING" >&2
                    fi
                fi
                in_content_kinds_section=false
                continue
            fi
            
            # Extract fields
            if [[ "$line" =~ ^name[[:space:]]*=[[:space:]]*\"([^\"]+)\" ]]; then
                current_name="${BASH_REMATCH[1]}"
            elif [[ "$line" =~ ^directory[[:space:]]*=[[:space:]]*\"([^\"]+)\" ]]; then
                current_directory="${BASH_REMATCH[1]}"
            elif [[ "$line" =~ ^enabled[[:space:]]*=[[:space:]]*true ]]; then
                current_enabled="true"
            fi
        fi
    done < "$CONTENT_KINDS_FILE"
    
    # Don't forget to process the last section
    if [[ "$in_content_kinds_section" == true && "$current_enabled" == "true" && -n "$current_directory" ]]; then
        if [ -d "${CONTENT_DIR}/${current_directory}" ]; then
            enabled_types+=("$current_directory")
            log_info "Content type '$current_name' -> directory '${current_directory}' enabled" >&2
        else
            log_warning "Content type '$current_name' enabled but directory '${CONTENT_DIR}/${current_directory}' missing - SKIPPING" >&2
        fi
    fi
    
    if [ ${#enabled_types[@]} -eq 0 ]; then
        log_warning "No enabled content types found, discovering from directory structure" >&2
        # Fallback: discover from directory structure
        local discovered_types=()
        for dir in "${CONTENT_DIR}"/*; do
            if [ -d "$dir" ]; then
                local dirname=$(basename "$dir")
                # Skip non-content directories
                if [[ "$dirname" != "locales" && "$dirname" != "themes" && "$dirname" != "menus" && "$dirname" != "footer" && "$dirname" != "tmp" && "$dirname" != "routes" && "$dirname" != "pages" ]]; then
                    discovered_types+=("$dirname")
                fi
            fi
        done
        if [ ${#discovered_types[@]} -eq 0 ]; then
            echo "blog recipes"
        else
            echo "${discovered_types[@]}"
        fi
    else
        echo "${enabled_types[@]}"
    fi
}

# Get supported languages from directory structure
get_supported_languages() {
    local content_types=($(get_enabled_content_types))
    local languages=()
    
    for content_type in "${content_types[@]}"; do
        local content_dir="${CONTENT_DIR}/${content_type}"
        if [ -d "$content_dir" ]; then
            for lang_dir in "$content_dir"/*/; do
                if [ -d "$lang_dir" ]; then
                    local lang=$(basename "$lang_dir")
                    if [[ ! " ${languages[@]} " =~ " ${lang} " ]]; then
                        languages+=("$lang")
                    fi
                fi
            done
        fi
    done
    
    if [ ${#languages[@]} -eq 0 ]; then
        echo "${DEFAULT_LANGUAGES[@]}"
    else
        echo "${languages[@]}"
    fi
}

# =============================================================================
# FRONTMATTER PROCESSING
# =============================================================================

# Extract frontmatter from markdown file
extract_frontmatter() {
    local file="$1"
    
    if [ ! -f "$file" ]; then
        return 1
    fi
    
    # Check if file starts with frontmatter
    if ! head -1 "$file" | grep -q "^---$"; then
        return 1
    fi
    
    # Extract YAML frontmatter between --- delimiters
    awk '/^---$/{flag++; next} flag==1 && /^---$/{flag++; exit} flag==1{print}' "$file"
}

# Extract specific field from frontmatter
extract_field() {
    local frontmatter="$1"
    local field="$2"
    local default_value="$3"
    
    # Handle different field types
    case "$field" in
        "tags"|"translations")
            # Extract array values and format as JSON
            local value=$(echo "$frontmatter" | grep -E "^${field}:" | sed 's/.*: *\[//' | sed 's/\]//' | tr -d '"' | sed 's/,/, /g')
            if [ -n "$value" ]; then
                # Convert to proper JSON array format - split by comma and quote each item
                echo "[$value]" | sed 's/\([^,]*\),\s*/"\1",/g' | sed 's/\([^,]*\)$/"\1"/' | sed 's/^\[,/[/' | sed 's/,"]/"]/'
            else
                echo "[]"
            fi
            ;;
        "featured"|"published")
            # Boolean fields
            local value=$(echo "$frontmatter" | grep -E "^${field}:" | sed 's/.*: *//' | tr '[:upper:]' '[:lower:]')
            case "$value" in
                true|yes|1) echo "true" ;;
                false|no|0) echo "false" ;;
                *) echo "${default_value:-false}" ;;
            esac
            ;;
        *)
            # String fields
            local value=$(echo "$frontmatter" | grep -E "^${field}:" | sed 's/.*: *//' | sed 's/^"*//' | sed 's/"*$//')
            if [ -n "$value" ]; then
                echo "$value"
            else
                echo "${default_value:-}"
            fi
            ;;
    esac
}

# Extract ID from frontmatter or use filename as fallback
extract_id() {
    local file="$1"
    local frontmatter=$(extract_frontmatter "$file")
    
    if [ -n "$frontmatter" ]; then
        local id=$(extract_field "$frontmatter" "id")
        if [ -n "$id" ]; then
            echo "$id"
            return
        fi
    fi
    
    # Fallback to filename without extension
    basename "$file" .md
}

# =============================================================================
# FILE AND DIRECTORY UTILITIES
# =============================================================================

# Get all markdown files for a content type and language
get_markdown_files() {
    local content_type="$1"
    local language="$2"
    local content_dir="${CONTENT_DIR}/${content_type}/${language}"
    
    if [ -d "$content_dir" ]; then
        find "$content_dir" -name "*.md" -type f | sort
    fi
}

# Check if a content directory exists
content_dir_exists() {
    local content_type="$1"
    local language="$2"
    [ -d "${CONTENT_DIR}/${content_type}/${language}" ]
}

# Create directory if it doesn't exist
ensure_dir() {
    local dir="$1"
    if [ ! -d "$dir" ]; then
        mkdir -p "$dir"
        log_info "Created directory: $dir"
    fi
}

# =============================================================================
# VALIDATION UTILITIES
# =============================================================================

# Validate JSON file
validate_json() {
    local file="$1"
    
    if command -v jq >/dev/null 2>&1; then
        if jq empty "$file" 2>/dev/null; then
            return 0
        else
            return 1
        fi
    else
        log_warning "jq not available, skipping JSON validation"
        return 0
    fi
}

# Check if file has valid frontmatter
has_valid_frontmatter() {
    local file="$1"
    [ -f "$file" ] && extract_frontmatter "$file" >/dev/null 2>&1
}

# =============================================================================
# CONTENT PROCESSING UTILITIES
# =============================================================================

# Process a single markdown file and extract metadata
process_markdown_file() {
    local file="$1"
    local content_type="$2"
    local language="$3"
    
    if ! has_valid_frontmatter "$file"; then
        log_warning "Skipping file without valid frontmatter: $file"
        return 1
    fi
    
    local filename=$(basename "$file" .md)
    local frontmatter=$(extract_frontmatter "$file")
    
    # Extract all standard fields
    local id=$(extract_field "$frontmatter" "id" "$filename")
    local title=$(extract_field "$frontmatter" "title" "Untitled")
    local subtitle=$(extract_field "$frontmatter" "subtitle")
    local excerpt=$(extract_field "$frontmatter" "excerpt")
    local author=$(extract_field "$frontmatter" "author")
    local date=$(extract_field "$frontmatter" "date")
    local read_time=$(extract_field "$frontmatter" "read_time")
    local category=$(extract_field "$frontmatter" "category")
    local difficulty=$(extract_field "$frontmatter" "difficulty")
    local prep_time=$(extract_field "$frontmatter" "prep_time")
    local tags=$(extract_field "$frontmatter" "tags")
    local featured=$(extract_field "$frontmatter" "featured" "false")
    local published=$(extract_field "$frontmatter" "published" "true")
    local translations=$(extract_field "$frontmatter" "translations")
    
    # Output as JSON object
    cat << EOF
{
  "id": "$id",
  "title": "$title",
  "subtitle": ${subtitle:+\"$subtitle\"},
  "excerpt": ${excerpt:+\"$excerpt\"},
  "author": ${author:+\"$author\"},
  "date": ${date:+\"$date\"},
  "read_time": ${read_time:+\"$read_time\"},
  "category": ${category:+\"$category\"},
  "difficulty": ${difficulty:+\"$difficulty\"},
  "prep_time": ${prep_time:+\"$prep_time\"},
  "tags": $tags,
  "featured": $featured,
  "published": $published,
  "language": "$language",
  "source_file": "${CONTENT_DIR}/$content_type/$language/$filename.md",
  "html_file": "/$content_type/$language/$filename.html",
  "translations": $translations
}
EOF
}

# =============================================================================
# INITIALIZATION
# =============================================================================

# Initialize library (call this from scripts that source this library)
init_content_lib() {
    local script_name="$1"
    log_info "Initializing content library for: $script_name"
    log_info "Content directory: $CONTENT_DIR"
    
    if [ ! -d "$CONTENT_DIR" ]; then
        log_error "Content directory not found: $CONTENT_DIR"
        exit 1
    fi
    
    local content_types=($(get_enabled_content_types))
    local languages=($(get_supported_languages))
    
    log_info "Enabled content types: ${content_types[*]}"
    log_info "Supported languages: ${languages[*]}"
}

# Export functions that may be used by other scripts
export -f log_error log_warning log_success log_info log_step
export -f get_enabled_content_types get_supported_languages
export -f extract_frontmatter extract_field extract_id
export -f get_markdown_files content_dir_exists ensure_dir
export -f validate_json has_valid_frontmatter process_markdown_file
export -f init_content_lib