Old fashion free pdf indexing with Umbraco

PDF Indexer.cs

using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading.Tasks;
using System.Xml.Linq;
using Examine;
using Examine.LuceneEngine.Providers;
using Lucene.Net.Analysis;
using Microsoft.WindowsAzure.Storage.Auth;
using Microsoft.WindowsAzure.Storage.Blob;
using MyCustomer.Core.Infrastructure;
using MyCustomer.Core.Magazine;
using MyCustomer.Core.Queries;
using Directory = Lucene.Net.Store.Directory;

namespace MySite.Backoffice.Magazine
{
    public class MagazineIndexer : LuceneIndexer
    {
        private readonly Directory directory;
        private readonly IQueryHandler<MagazineQuery> queryHandler;

        private CloudBlobClient cloudBlobClient = new CloudBlobClient(
            new Uri("https://mycustomer.blob.core.windows.net"),
            new StorageCredentials(
                "storagename",
                "secret"
            )
        );

        private PDFParser pdfParser = new PDFParser();

        public MagazineIndexer()
        {
            queryHandler = Container.Instance.Resolve<IQueryHandler<MagazineQuery>>();
        }

        public MagazineIndexer(Directory directory, IQueryHandler<MagazineQuery> queryHandler, Analyzer analyzer, bool async)
            : base(
                new IndexCriteria(
                    Enumerable.Empty<IIndexField>(),
                    Enumerable.Empty<IIndexField>(),
                    Enumerable.Empty<string>(),
                    Enumerable.Empty<string>(),
                    null
                ),
                directory,
                analyzer,
                async
            )
        {
            this.directory = directory;
            this.queryHandler = queryHandler;
        }

        protected override void PerformIndexAll(string type)
        {
            var result = (IEnumerable<Magazine>)Task.Run(async () => await queryHandler.Execute(new MagazineQuery())).Result;
            var nodes = result.Select(x =>
            {
                var xElement = new XElement("url", x.PDF);
                xElement.Add(new XAttribute("id", Convert.ToInt32(x.Year) * 100 + Convert.ToInt32(x.Month)));
                return xElement;
            });
            AddNodesToIndex(nodes, "");
        }

        protected override void PerformIndexRebuild()
        {
            IndexAll("");
        }

        protected override Dictionary<string, string> GetDataToIndex(XElement node, string type)
        {
            try
            {
                var pdf = cloudBlobClient.GetBlobReferenceFromServer(new Uri("https://mycustomer.blob.core.windows.net" + node.Value));
                using (var stream = new MemoryStream())
                {
                    pdf.DownloadToStream(stream);
                    stream.Seek(0, SeekOrigin.Begin);
                    var text = pdfParser.GetTextFromAllPages(stream, (e) => { });
                    var value = node.Attribute("id").Value;
                    var edition = value.Substring(4, 2) + " / " + value.Substring(0, 4);
                    return new Dictionary<string, string>
                    {
                        {"type", "magasin" },
                        {"edition", edition },
                        {"url", node.Value },
                        {"body", text}
                    };
                }
            }
            catch
            {
                return new Dictionary<string, string>();
            }
        }
    }
}