C# Pdf to Text with image placeholder

Question

I have a batch of PDFs that I want to convert to Text. It's easy to get text with something like this from iTextSharp:

PdfTextExtractor.GetTextFromPage(reader, pageNumber);

It's easy to get Images using this answer (or similar answers in the thread).

What I can't figure out easily... is how to interleave image placeholders in the text.

Given a PDF, a page # and GetTextFromPage I expect the output to be:

line 1
line 2
line 3

When I'd like it to be (Where 1.1 means page 1, image 1... Page 1, image 2):

line 1
[1.1]
line 2
[1.2]
line 3

Is there a way to get an "image placeholder" for iTextSharp, PdfSharp or anything similar? I'd like a GetTextAndPlaceHoldersFromPage method (or similar).

PS: Hrm... it's not letting me tag iTextSHARP - not iText. C# not Java.

Are you looking for something like the solution for iText/Java presented in [this answer](http://stackoverflow.com/a/28087521/1729265)? It should be easily translated to iTextSharp/C#. — mkl, Jun 28 '16 at 09:07
@mkl updated explanation to more closely match that question. Thought about expanding a bit like it last night. That question is basically the same, except I'm working in C# with iTextSharp (although I'm not averse to moving to other libraries like PdfSharp if need be). — WernerCD, Jun 28 '16 at 14:49
Thus, one could create what you want by porting the iText/Java solution from that old question to iTextSharp/C#. That shouldn't be too difficult... — mkl, Jun 28 '16 at 15:04
@mkl yeah, I'll be trying that if someone doesn't beat me too it... never ported Java before, but doesn't look too terribly difficult as the two language are close (yet different) — WernerCD, Jun 28 '16 at 15:06
@mkl I got it working :) Thanks to that question and another. — WernerCD, Jun 28 '16 at 20:24
@mkl thanks again. Many breakthroughs in my code. Much simplification. Much happiness. I can no work on OTHER parts of my dastardly plans. — WernerCD, Jun 29 '16 at 17:44

score 1 · Answer 1 · edited May 23 '17 at 11:52

C# Pdf to Text with image placeholder
https://stackoverflow.com/a/28087521/
https://stackoverflow.com/a/33697745/

Although this doesn't have the exact layout mentioned in my question (Since that was a simplified version of what I really wanted anyways), it does have the starting parts as listed by the second note (translated from iText Java)... with extra information pulled from the third note (Some of the reflection used in Java didn't seem to work in C#, so that info came from #3).

Working from this, I'm able to get a List of Strings representing lines in the PDF (all pages, instead of just page 1)... with text added where images should be (Huzzah!). ByteArrayToFile extension method added for flavor (Although I didn't include other parts/extensions that may break a copy/paste usages of this code).

I've also been able to greatly simplify other parts of my process and gut half of the garbage I had working before. Huzzah!!! Thanks @Mkl

internal class Program
{
    public static void Main(string[] args)
    {
        var dir = Settings.TestDirectory;
        var file = Settings.TestFile;

        Log.Info($"File to Process: {file.FullName}");

        using (var reader = new PdfReader(file.FullName))
        {
            var parser = new PdfReaderContentParser(reader);
            var listener = new SimpleMixedExtractionStrategy(file, dir);
            parser.ProcessContent(1, listener);
            var x = listener.GetResultantText().Split('\n');
        }
    }
}

public class SimpleMixedExtractionStrategy : LocationTextExtractionStrategy
{
    public static readonly ILog Log = LogManager.GetLogger(MethodBase.GetCurrentMethod().DeclaringType);
    public DirectoryInfo OutputPath { get; }
    public FileInfo OutputFile { get; }

    private static readonly LineSegment UNIT_LINE = new LineSegment(new Vector(0, 0, 1), new Vector(1, 0, 1));
    private int _counter;

    public SimpleMixedExtractionStrategy(FileInfo outputFile, DirectoryInfo outputPath)
    {
        OutputPath = outputPath;
        OutputFile = outputFile;
    }

    public override void RenderImage(ImageRenderInfo renderInfo)
    {
        try
        {
            var image = renderInfo.GetImage();
            if (image == null) return;
            var number = _counter++;

            var imageFile = new FileInfo($"{OutputFile.FullName}-{number}.{image.GetFileType()}");
            imageFile.ByteArrayToFile(image.GetImageAsBytes());

            var segment = UNIT_LINE.TransformBy(renderInfo.GetImageCTM());
            var location = new TextChunk("[" + imageFile + "]", segment.GetStartPoint(), segment.GetEndPoint(), 0f);
            var locationalResultField = typeof(LocationTextExtractionStrategy).GetField("locationalResult", BindingFlags.NonPublic | BindingFlags.Instance);
            var LocationalResults = (List<TextChunk>)locationalResultField.GetValue(this);
            LocationalResults.Add(location);
        }
        catch (Exception ex)
        {
            Log.Debug($"{ex.Message}");
            Log.Verbose($"{ex.StackTrace}");
        }
    }
}

public static class ByteArrayExtensions
{
    public static bool ByteArrayToFile(this FileInfo fileName, byte[] byteArray)
    {
        try
        {
            // Open file for reading
            var fileStream = new FileStream(fileName.FullName, FileMode.Create, FileAccess.Write);

            // Writes a block of bytes to this stream using data from a byte array.
            fileStream.Write(byteArray, 0, byteArray.Length);

            // close file stream
            fileStream.Close();

            return true;
        }
        catch (Exception exception)
        {
            // Error
            Log.Error($"Exception caught in process: {exception.Message}", exception);
        }

        // error occured, return false
        return false;
    }
}

C# Pdf to Text with image placeholder

1 Answers1