TextExtractionSample.cs :  » PDF » PDF-Clown » it » stefanochizzolini » clown » samples » C# / CSharp Open Source

Home
C# / CSharp Open Source
1.2.6.4 mono .net core
2.2.6.4 mono core
3.Aspect Oriented Frameworks
4.Bloggers
5.Build Systems
6.Business Application
7.Charting Reporting Tools
8.Chat Servers
9.Code Coverage Tools
10.Content Management Systems CMS
11.CRM ERP
12.Database
13.Development
14.Email
15.Forum
16.Game
17.GIS
18.GUI
19.IDEs
20.Installers Generators
21.Inversion of Control Dependency Injection
22.Issue Tracking
23.Logging Tools
24.Message
25.Mobile
26.Network Clients
27.Network Servers
28.Office
29.PDF
30.Persistence Frameworks
31.Portals
32.Profilers
33.Project Management
34.RSS RDF
35.Rule Engines
36.Script
37.Search Engines
38.Sound Audio
39.Source Control
40.SQL Clients
41.Template Engines
42.Testing
43.UML
44.Web Frameworks
45.Web Service
46.Web Testing
47.Wiki Engines
48.Windows Presentation Foundation
49.Workflows
50.XML Parsers
C# / C Sharp
C# / C Sharp by API
C# / CSharp Tutorial
C# / CSharp Open Source » PDF » PDF Clown 
PDF Clown » it » stefanochizzolini » clown » samples » TextExtractionSample.cs
using it.stefanochizzolini.clown.documents;
using it.stefanochizzolini.clown.documents.contents.objects;
using it.stefanochizzolini.clown.files;
using it.stefanochizzolini.clown.objects;

using System;
using System.Collections.Generic;
using System.Text;

namespace it.stefanochizzolini.clown.samples{
  /**
    <summary>This sample is a rough stub that demonstrates a basic way to extract text
    from a document.</summary>
    <remarks>
      <para>This implementation is definitely simplistic: its purpose is NOT to provide
      a real-life solution for PDF text mining; it lacks advanced features such as
      character encoding management, glyph position detection, dehyphenation,
      page-breaks handling and so on. Its purpose is to test the new content-stream parsing
      functionality.</para>
      <para>So, read my lips: this-is-just-a-toy (for now!).</para>
    </remarks>
  */
  public class TextExtractionSample
    : ISample
  {
    private static readonly Encoding ISO88591Encoding = Encoding.GetEncoding("iso-8859-1");

    public void Run(
      SampleLoader loader
      )
    {
      // (boilerplate user choice -- ignore it)
      string filePath = loader.GetPdfFileChoice("Please select a PDF file");

      // Open the PDF file!
      File file = new File(filePath);

      // Get the PDF document!
      Document document = file.Document;
      // Get the page collection!
      Pages pages = document.Pages;
//TODO:IMPL see PDF:1.6:5.9
      foreach(Page page in pages)
      {
        Console.WriteLine("\nScanning page " + (page.Index+1) + "...\n");

        Extract(page.Contents);

        Console.WriteLine("\nENTER (Scan next page)");
        Console.WriteLine("[Q] (End scanning)");
        Console.Write("Please select: ");
        try
        {
          string choice = Console.ReadLine();
          if(choice.ToUpper().Equals("Q")) // Quit.
            break;
        }
        catch{}
      }
    }

    private void Extract(
      IList<ContentObject> objects
      )
    {
      foreach(ContentObject obj in objects)
      {
        if(obj is Text)
        {
          foreach(Operation operation in ((Text)obj).Objects)
          {
            bool hit = false;
            string operator_ = operation.Operator;
            if(operator_.Equals("TJ")
              || operator_.Equals("Tj")
              || operator_.Equals("'")
              || operator_.Equals("''")
              )
            {
              foreach(PdfDirectObject operand in operation.Operands)
              {
                /*
                  NOTE: This algorythm doesn't deal with the character encoding
                  used to represent the font glyphs, which may be arbitrarily defined;
                  therefore, its text output has no direct relation with the actual
                  Unicode text it stays for (to say: you may see weird sequences of
                  symbols instead of the expected plain text).
                  PDF Clown will implement font decoding support in the next releases
                  through the documents.contents.fonts.Font.decode(byte[]) method.
                */
                if(operand is PdfString)
                {
                  hit = true;
                  Console.Write(ISO88591Encoding.GetString(((PdfString)operand).RawValue) + " "); // NOTE: This line outputs the internal text representation, NOT the actual plain text (decoding to do yet!).
                }
                else if(operand is PdfArray)
                {
                  foreach(PdfDirectObject item in ((PdfArray)operand))
                  {
                    if(item is PdfString)
                    {
                      hit = true;
                      Console.Write(ISO88591Encoding.GetString(((PdfString)item).RawValue) + " "); // NOTE: This line outputs the internal text representation, NOT the actual plain text (decoding to do yet!).
                    }
                  }
                }
              }
              if(hit)
              {Console.WriteLine();}
            }
          }
        }
        else if(obj is ContainerObject)
        {Extract(((ContainerObject)obj).Objects);}
      }
    }
  }
}
www.java2v.com | Contact Us
Copyright 2009 - 12 Demo Source and Support. All rights reserved.
All other trademarks are property of their respective owners.