PDF-to-Image Renderer for .NET C# wrapper for poppler tools

var pdfToText = new PdfToTextConverter();
pdfToText.CustomArgs = " -bbox-layout -eol dos "; 
var textWithLayoutMetadata = pdfToText.GenerateText("test.pdf");

• LT-version nuget doesn't include poppler binaries for Windows (DLL size is only ~64kb).
• A license key is required by LT-version (no trial mode).
• When a .NET app is deployed on Linux, the wrapper tries to locate poppler tools in /usr/local/bin or /usr/bin. If poppler executables (pdftoppm, pdfinfo etc) use another location you need to set "ToolPath" property explicitly. On Windows, there is no standard location for poppler tools so "ToolPath" needs to be initialized if NReco.PdfRenderer.LT is used on Windows.

• Ubuntu/Debian: sudo apt-get install -y poppler-utils
• Other OS: use latest poppler tools release

void ProcessFileFromUrl(string url, Action<Stream> handler) {
  using (var httpClient = new System.Net.Http.HttpClient()) {
    using (var responseStream = httpClient.GetStreamAsync(url).GetAwaiter().GetResult()) {
      handler(responseStream);
    }
  }
}

var pdfToImage = new PdfToImageConverter();

var pdfUrl = "https://www.adobe.com/support/products/enterprise/knowledgecenter/media/c4611_sample_explain.pdf";
ProcessFileFromUrl(pdfUrl, (pdfInputStream) => {
  pdfToImage.GenerateImage(pdfInputStream, 1, ImageFormat.Jpeg, "c:\\temp\\test.jpg");
});

var pdfInfo = new PdfInfo();
var testPdfInfo = pdfInfo.GetPdfInfo(@"test.pdf");
Console.WriteLine($"Pages: {testPdfInfo.Pages}");
Console.WriteLine($"Page size: {testPdfInfo.PageSize.PageFormat}");

var fileNames = pdfToImage.GenerateImages(@"test.pdf", 
	3, 7,  // render from pages 3-7
	ImageFormat.Png, "outputFolder");

If PDF text is not rendered most likely you need to use poppler encoding data (it is not included into distribution by default). You can verify that by handling "LogReceived" event; typical errors are smth like Missing language pack for 'Adobe-Korea1' mapping, Missing language pack for 'Adobe-Japan1' mapping, Missing language pack for 'Adobe-CNS1' mapping.

In case of .NET Core on Linux it is enough to install "poppler-data" package. For Windows environment:

1. download an archive with "poppler-data": https://poppler.freedesktop.org/poppler-data-0.4.12.tar.gz and extract poppler-data-0.4.12 folder (it should contain folders "cidToUnicode", "cMap", "nameToUnicode", "unicodeMap")
2. prepare special folder for poppler binaries and poppler-data files. Let's assume this is <your_asp_net_app_folder>\App_Data\PdfRenderer
3. copy everything from poppler-data-0.4.12 folder to <your_asp_net_app_folder>\App_Data\PdfRenderer\share\poppler
4. in C# code, change default path for poppler binaries to use your special folder:

   var pdfToImage = new PdfToImageConverter();
   pdfToImage.ToolPath = "<your_asp_net_app_folder>\App_Data\PdfRenderer\poppler\bin";
   

5. now PdfRenderer should render PDFs with non-Unicode characters correctly

It is possible to use NReco.PdfRenderer in Azure Function under the following conditions:

• because of Azure Functions hosting environment System.Drawing API is not supported. This means that you should avoid usage of API methods that return System.Drawing.Image object or use NReco.PdfRenderer.LT nuget.
• in trial mode component applies a watermark and this causes an exception for the same reason (System.Drawing API cannot be used). You can request a demo key for testing purposes if you want to ensure that everything works fine.
• by default wrapper extracts poppler tools binaries into the folder with .NET Core app. It is read-only in case of Azure Functions, so you need to specify another location:

  var pdfToImg = new NReco.PdfRenderer.PdfToImageConverter();
  pdfToImg.ToolPath = Path.Combine(Path.GetTempPath(), "NRecoPdfRenderer");