using HtmlAgilityPack;

namespace AiAgent.Core.Helpers
{
    /// <summary>
    /// A helper class to extract URLs from HTML content.
    /// </summary>
    public static class HtmlUrlExtractor
    {
        /// <summary>
        /// Extracts all unique URLs from an HTML string.
        /// It finds URLs in the 'href' attributes of <a> and <link> tags,
        /// and in the 'src' attributes of <img>, <script>, and <iframe> tags.
        /// </summary>
        /// <param name="htmlContent">The HTML content as a string.</param>
        /// <returns>A List of unique URL strings found in the HTML.</returns>
        public static List<string> ExtractUrls(string htmlContent)
        {
            if (string.IsNullOrEmpty(htmlContent))
                return new List<string>();

            var htmlDoc = new HtmlDocument();
            htmlDoc.LoadHtml(htmlContent);

            // Using a HashSet to automatically handle uniqueness of URLs.
            var urls = new HashSet<string>();

            // Select nodes with 'href' attribute (typically <a> and <link> tags)
            var hrefNodes = htmlDoc.DocumentNode.SelectNodes("//*[@href]");
            if (hrefNodes != null)
            {
                foreach (var node in hrefNodes)
                {
                    var href = node.GetAttributeValue("href", string.Empty);
                    if (!string.IsNullOrEmpty(href))
                        urls.Add(href);
                }
            }

            // Select nodes with 'src' attribute (typically <img>, <script>, <iframe> tags)
            var srcNodes = htmlDoc.DocumentNode.SelectNodes("//*[@src]");
            if (srcNodes != null)
            {
                foreach (var node in srcNodes)
                {
                    var src = node.GetAttributeValue("src", string.Empty);
                    if (!string.IsNullOrEmpty(src))
                        urls.Add(src);
                }
            }

            return urls.ToList();
        }
    }
}