UmbracoExamine.PDF

PDF indexing support in UmbracoExamine using pdfSharp

Installation

Install with Nuget:

Install-Package UmbracoCms.UmbracoExamine.PDF

You will then have a new Examine index called "PDFIndex" available.

Multi index searchers

To use the MultiSearcher in V8, you can instantiate it when needed like:

using(var multiSearcher = new MultiIndexSearcher("MultiSearcher", new IIndex[] { 
    externalIndex, 
	pdfIndex 
}))
{
 ...
};

Or you can register a multi-index searcher with the ExamineManager on startup like:

[ComposeAfter(typeof(ExaminePdfComposer))] //this must execute after the ExaminePdfComposer composer
public class MyComposer : ComponentComposer<MyComponent>, IUserComposer
{
}

public class MyComponent : IComponent
{
    private readonly IExamineManager _examineManager;

    public MyComponent(IExamineManager examineManager)
    {
        _examineManager = examineManager;
    }

    public void Initialize()
    {
        //Get both the external and pdf index
        if (_examineManager.TryGetIndex(Constants.UmbracoIndexes.ExternalIndexName, out var externalIndex)
            && _examineManager.TryGetIndex(PdfIndexConstants.PdfIndexName, out var pdfIndex))
        {
            //register a multi searcher for both of them
            var multiSearcher = new MultiIndexSearcher("MultiSearcher", new IIndex[] { externalIndex, pdfIndex });
            _examineManager.AddSearcher(multiSearcher);
        }
    }

    public void Terminate() { }
}

With this approach, the multi searcher will show up in the Examine dashboard and it can be resolved from the ExamineManager like:

if (_examineManager.TryGetSearcher("MultiSearcher", out var searcher))
{
    //TODO: use the `searcher` to search
}

Known issues

The implementation of IPdfTextExtractor is PdfSharpTextExtractor in this library which uses PDFSharp to extract the bytes to convert to text but that implementation doesn't deal well with Unicode text which means when some PDF files are read, the result will be 'junk' strings.

It is certainly possible to replace the IPdfTextExtractor using your own composer like

composition.RegisterUnique<IPdfTextExtractor, MyCustomSharpTextExtractor>();

The iTextSharp library deals with Unicode in a better way but is a paid for license. If you wish to use iTextSharp or another PDF library you can easily swap out the IPdfTextExtractor with your own implementation.

Help wanted

If anyone knows how to update the PdfSharpTextExtractor implementation to better handle Unicode texts, please get in touch on the issue tracker or by submitting a PR.

Thanks!

License

The source code for UmbracoExamine.PDF is licensed under the MIT license.