website-htmx-rustelo-code/scripts/others/populate-content-indices-simple.sh

#!/bin/bash

# Simple script to populate content index.json files from existing markdown files

set -e

# Colors
GREEN='\033[0;32m'
BLUE='\033[0;34m'
NC='\033[0m'

log_info() { echo -e "${BLUE}ℹ️  $1${NC}"; }
log_success() { echo -e "${GREEN}✅ $1${NC}"; }

# Function to extract value from YAML frontmatter
get_yaml_value() {
    local file="$1"
    local key="$2"
    
    # Extract frontmatter and get the value
    awk '/^---$/{if(p==1){exit} p++; next} p==1 && /^'$key':/ {gsub(/^'$key': */, ""); gsub(/^"/, ""); gsub(/"$/, ""); print}' "$file"
}

# Function to extract tags array from YAML
get_yaml_tags() {
    local file="$1"
    
    # Extract tags array - handle both inline and multiline formats
    awk '/^---$/{if(p==1){exit} p++; next} 
         p==1 && /^tags:/ {
           if(/\[.*\]/) {
             # Inline array format: tags: ["tag1", "tag2"]
             gsub(/^tags: *\[/, ""); gsub(/\].*$/, ""); gsub(/"/, ""); gsub(/, */, "\",\""); 
             if(length($0) > 0) print "\"" $0 "\""
           } else {
             # Multiline format - skip for now, would need more complex parsing
             next
           }
         }' "$file"
}

# Function to process a single markdown file
process_markdown_file() {
    local md_file="$1" 
    local content_type="$2"
    local language="$3"
    
    # Extract values
    local id=$(get_yaml_value "$md_file" "id")
    local title=$(get_yaml_value "$md_file" "title")
    local slug=$(get_yaml_value "$md_file" "slug")
    local excerpt=$(get_yaml_value "$md_file" "excerpt")
    local date=$(get_yaml_value "$md_file" "date")
    local read_time=$(get_yaml_value "$md_file" "read_time")
    local featured=$(get_yaml_value "$md_file" "featured")
    local category=$(get_yaml_value "$md_file" "category")
    local tags=$(get_yaml_tags "$md_file")
    
    # Defaults
    id=${id:-$(basename "$md_file" .md)}
    slug=${slug:-$id}
    title=${title:-"Untitled"}
    excerpt=${excerpt:-""}
    date=${date:-$(date +"%Y-%m-%d")}
    read_time=${read_time:-"5 min read"}
    featured=${featured:-false}
    
    # Infer category from directory if not set
    if [ -z "$category" ]; then
        local dir_name=$(basename "$(dirname "$md_file")")
        if [ "$dir_name" != "$language" ]; then
            category="$dir_name"
        fi
    fi
    
    # Build categories array
    local categories=""
    if [ -n "$category" ]; then
        categories="\"$category\""
    fi
    
    # Build tags array
    if [ -z "$tags" ]; then
        tags=""
    fi
    
    # Build source_file path (relative to content root, includes category directory)
    # Use basename and dirname to manually construct the relative path since macOS realpath doesn't have --relative-to
    local file_dir=$(dirname "$md_file")
    local content_relative_dir=${file_dir#content/}
    local filename=$(basename "$md_file")
    local relative_path="$content_relative_dir/$filename"
    
    # Create JSON using jq for safety
    jq -n \
        --arg id "$id" \
        --arg title "$title" \
        --arg slug "$slug" \
        --arg language "$language" \
        --arg content_type "$content_type" \
        --argjson categories "[$categories]" \
        --argjson tags "[$tags]" \
        --argjson featured "$featured" \
        --arg excerpt "$excerpt" \
        --arg content "" \
        --arg created_at "$date" \
        --arg read_time "$read_time" \
        --arg source_file "$relative_path" \
        '{
            id: $id,
            title: $title,
            slug: $slug,
            language: $language,
            content_type: $content_type,
            categories: $categories,
            tags: $tags,
            emoji: null,
            featured: $featured,
            excerpt: $excerpt,
            content: $content,
            created_at: $created_at,
            updated_at: null,
            read_time: $read_time,
            source_file: $source_file
        }'
}

# Function to process content type and language
process_content_lang() {
    local content_type="$1"
    local language="$2"
    local lang_dir="content/${content_type}/${language}"
    
    if [ ! -d "$lang_dir" ]; then
        return
    fi
    
    log_info "Processing $content_type/$language"
    
    # Start with empty array
    echo "[]" > "/tmp/${content_type}_${language}.json"
    
    # Process each markdown file
    local count=0
    find "$lang_dir" -name "*.md" -type f | while read -r md_file; do
        log_info "  Processing: $(basename "$md_file")"
        
        # Generate entry and add to array
        local entry=$(process_markdown_file "$md_file" "$content_type" "$language")
        echo "$entry" > "/tmp/current_entry.json"
        
        # Add to main array
        jq '. += [inputs]' "/tmp/${content_type}_${language}.json" "/tmp/current_entry.json" > "/tmp/${content_type}_${language}_new.json"
        mv "/tmp/${content_type}_${language}_new.json" "/tmp/${content_type}_${language}.json"
        
        count=$((count + 1))
    done
    
    # Create final index file
    local index_file="${lang_dir}/index.json"
    local entries=$(cat "/tmp/${content_type}_${language}.json")
    
    cat > "$index_file" << EOF
{
  "$content_type": $entries,
  "language": "$language"
}
EOF
    
    local final_count=$(jq length "/tmp/${content_type}_${language}.json")
    log_success "Created $content_type/$language index with $final_count entries"
    
    # Cleanup temp files
    rm -f "/tmp/${content_type}_${language}.json" "/tmp/current_entry.json"
}

# Main execution
log_info "Populating content indices..."

# Check for jq
if ! command -v jq &> /dev/null; then
    echo "❌ jq is required but not installed"
    exit 1
fi

# Process blog content
process_content_lang "blog" "en"
process_content_lang "blog" "es"

# Process recipes content  
process_content_lang "recipes" "en"
process_content_lang "recipes" "es"

log_success "Content indices populated successfully!"
log_info "Run 'just dev' to rebuild and embed the new indices."