/// <summary>
/// Extracts a text from a PDF file.
/// </summary>
/// <param name="fileName">The full path to the pdf file.</param>
/// <param name="success">Indicate if operation was successfull.</param>
/// <returns>The extracted text.</returns>
internal static String ExtractText(String fileName, out bool success)
{
String result = String.Empty;
PdfReader reader = null;
success = false;
try
{
reader = new PdfReader(fileName);
PdfReaderContentParser parser = new PdfReaderContentParser(reader);
for (int page = 1; page <= reader.NumberOfPages; page++)
{
SimpleTextExtractionStrategy strategy = parser.ProcessContent(page, new SimpleTextExtractionStrategy());
result += strategy.GetResultantText();
}
success = true;
return result;
}
catch (Exception)
{
return String.Empty;
}
finally
{
if (reader != null)
{
reader.Close();
}
}
}
public List <string> GetPdfPagesContent(Stream pdfStream)
{
List <string> allPDFPagesContent = new List <string>();
PdfReader reader = new PdfReader(pdfStream);
PdfReaderContentParser parser = new PdfReaderContentParser(reader);
for (int i = 1; i <= reader.NumberOfPages; i++)
{
strategy = parser.ProcessContent(i, new SimpleTextExtractionStrategy());
allPDFPagesContent.Add(strategy.GetResultantText());
}
return(allPDFPagesContent);
}
/// <summary>
/// Extract The Text Content From Pdf
/// </summary>
/// <param name="sourcePdf">The Pdf in base64 string format</param>
/// <returns>A string of lines separated by \n </returns>
public static List <String> ExtractTextPaged(byte[] sourcePdf)
{
List <String> _textExtractedPaged = new List <string>();
using (PdfDocument _sourcePdfDocument = new PdfDocument(new PdfReader(new MemoryStream(sourcePdf))))
{
for (int i = 1; i < _sourcePdfDocument.GetNumberOfPages(); i++)
{
FilteredEventListener listener = new FilteredEventListener();
SimpleTextExtractionStrategy extractionStrategy = listener.AttachEventListener(new SimpleTextExtractionStrategy());
new PdfCanvasProcessor(listener).ProcessPageContent(_sourcePdfDocument.GetPage(i));
_textExtractedPaged.Add(extractionStrategy.GetResultantText());
}
}
return(_textExtractedPaged);
}
//converting pdf to a txt file
public static void getTextFromPdf(String dest)
{
PdfDocument pdfDoc = new PdfDocument(new PdfReader(SRC));
SimpleTextExtractionStrategy strategy = new SimpleTextExtractionStrategy();
PdfCanvasProcessor parser = new PdfCanvasProcessor(strategy);
for (var i = 1; i <= pdfDoc.GetNumberOfPages(); i++)
{
parser.ProcessPageContent(pdfDoc.GetPage(i));
byte[] array = Encoding.UTF8.GetBytes(strategy.GetResultantText());
using (FileStream stream = new FileStream(dest, FileMode.OpenOrCreate))
{
stream.Write(array, 0, array.Length);
}
}
}
public static void Main(string[] args)
{
Contract.Assert(args.Any(), "No arguments passed.");
var path = args.First();
var extractedText = string.Empty;
// monsters are on page 208 - 253
using (var pdfReader = new PdfReader(path))
{
var stringBuilder = new StringBuilder();
var pdfDocument = new PdfDocument(pdfReader);
var listener = new FilteredEventListener();
CustomFilter customFilter = null;
SimpleTextExtractionStrategy extractionStrategy = null;
PdfCanvasProcessor pdfProcessor = null;
for (int i = 208; i <= 253; i++)
{
Console.WriteLine(@"Reading page {0}.", i);
var page = pdfDocument.GetPage(i);
if (pdfProcessor == null)
{
customFilter = new CustomFilter(page.GetArtBox());
extractionStrategy = listener.AttachEventListener(new SimpleTextExtractionStrategy(), customFilter);
pdfProcessor = new PdfCanvasProcessor(listener);
}
pdfProcessor.ProcessPageContent(page);
stringBuilder.AppendLine(extractionStrategy.GetResultantText());
if (Program.TextWithWidth.Last.Value.Equals(Program.TextWithWidth.Last.Previous.Value))
{
Debugger.Break();
}
}
extractedText = stringBuilder.ToString();
}
//m_parser = new TextToMonsterDataParser(extractedText);
//var monsters = m_parser.GetMonsterData();
Console.WriteLine("Read all text.");
Console.ReadLine();
if (System.IO.File.Exists(@"AllLines.txt"))
{
System.IO.File.Delete(@"AllLines.txt");
}
if (System.IO.File.Exists(@"EntireText.txt"))
{
System.IO.File.Delete(@"EntireText.txt");
}
Console.WriteLine("Printing text.");
System.IO.File.WriteAllText(@"EntireText.txt", extractedText);
Console.WriteLine("Found {0} duplicates.", TextWithWidth.Count(n => n.String.Equals("die’")));
Console.WriteLine("Printing lines.");
using (System.IO.StreamWriter file = new System.IO.StreamWriter(@"AllLines.txt", false))
{
var node = TextWithWidth.First;
while (node != null)
{
file.WriteLine("{0}({1})", node.Value.String, node.Value.Width);
node = node.Next;
}
}
Console.ReadLine();
}
static void Main()
{
Console.OutputEncoding = Encoding.UTF8;
char[] nepLat = { 'c', 'k', 'r', 'v' };
char[] nepCir = { 'ц', 'к', 'р', 'в' };
string[] pojmoviOrg = File.ReadAllLines("pojmovi.txt");
string[] pojmovi = (string[])pojmoviOrg.Clone();
List <int>[] stranice = new List <int> [pojmovi.Length];
for (int i = 0; i < stranice.Length; i++)
{
stranice[i] = new List <int>();
}
PdfDocument pdfDocument = new PdfDocument(new PdfReader("knjiga.pdf"));
var pageNumbers = pdfDocument.GetNumberOfPages();
for (int pageIndex = 1; pageIndex <= pageNumbers; pageIndex++)
{
Console.Title = (pageIndex / (float)pageNumbers).ToString("P");
SimpleTextExtractionStrategy strategy = new SimpleTextExtractionStrategy();
new PdfCanvasProcessor(strategy).ProcessPageContent(pdfDocument.GetPage(pageIndex));
string page = strategy.GetResultantText().RemoveWrap();
for (int p = 0; p < pojmovi.Length; p++)
{
pojmovi[p] = pojmovi[p].Replace('-', ' '); //briše crte
pojmovi[p] = new string(pojmovi[p].Where(c => char.IsLetter(c) || c == ' ' || c == '.' || c == '(').ToArray()).Trim(); //ostavlja samo slova, razmake i tačke
string[] podpojmovi = pojmovi[p].Split('(');
foreach (string podpojam in podpojmovi)
{
List <string> pojam = podpojam.Trim().Split(' ').ToList(); //deli pojam na reči
List <string> pojamNep = new List <string>(); //lista reči bez nepostojanog a
List <int> brNep = new List <int>(); //lista brojeva reči bez nepostojanog a
for (int i = 0; i < pojam.Count; i++)
{
if (pojam[i].Length > 3 && pojam[i].Length < 9 && (nepLat.Contains(pojam[i].Last()) || nepCir.Contains(pojam[i].Last())) && (pojam[i][pojam[i].Length - 2] == 'a' || pojam[i][pojam[i].Length - 2] == 'а'))
{
pojamNep.Add(pojam[i].Substring(0, pojam[i].Length - 2));
brNep.Add(i);
}
if ((pojam.Count > 1 || pojam[i].Length > 3) && pojam[i].Substring(Math.Max(0, pojam[i].Length - 2)) != "ић")
{
pojam[i] = pojam[i].Remove(pojam[i].Length - (pojam[i].Length > 8 ? 2 : 1)); //briše poslednje slovo sa reči ako pojam ima više od jedne reči ili reč ima više od 3 slova
}
if (pojam[i].Length < 3)
{
pojam.RemoveAt(i);
i--;
}
}
int minIma = Math.Max(pojam.Count == 1 ? 1 : 2, (int)Math.Round(pojam.Count * 0.58)); //minimalan broj reči da bi pojam bio nađen
int brBezNep = pojam.Count;
pojam.AddRange(pojamNep);
for (int r = 0; r < pojam.Count; r++) //za svaku reč u pojmu
{
if (page.Contains(pojam[r])) //ako stranica sadrži reč
{
int where = page.IndexOf(pojam[r]); //gde reč počinje na stranici
while (where != -1 && where < page.Length) //dokle god ima još takvih reči na stranici
{
int ima = 1; //trenutno nađeno reči
List <string> pojamBez = new List <string>(pojam);
pojamBez.RemoveAt(r); //lista bez nađene reči
if (r < brBezNep && brNep.IndexOf(r) != -1)
{
pojamBez.RemoveAt(brBezNep + brNep.IndexOf(r) - 1); //lista bez nađene reči bez nepostojanog a
}
else if (r >= brBezNep)
{
pojamBez.RemoveAt(brNep[r - brBezNep]); //lista bez nađene reči sa nepostojanim a
}
int min = Math.Max(0, where - (int)Math.Round(podpojam.Length * 1.5)); //početak opsega je početak reči minus dužina pojma
int max = Math.Min(page.Length - min, (int)Math.Round(podpojam.Length * 3.5)); //dužina opsega je tri puta veća od dužine pojma
foreach (string reč in pojamBez) //za svaku ostalu reč
{
if (page.Substring(min, max).Contains(reč)) //da li se druga reč nalazi u opsegu prve reči
{
ima++; //nađena druga reč
}
if (ima >= minIma)
{
break; //nađen pojam
}
}
if (ima >= minIma && !stranice[p].Contains(pageIndex) && (pageIndex < 406 || pageIndex > 416) && pageIndex != 5) //ako je pojam nađen i ako već nije nađen na ovoj stranici
{
stranice[p].Add(pageIndex); //dodaj broj stranice na spisak
break; //ne traži više tu reč na stranici
}
int newWhere = page.Substring(where + pojam[r].Length).IndexOf(pojam[r]);
where = newWhere != -1 ? newWhere + where + 1 : -1; //traži da li ima još takve reči na stranici
}
}
}
}
}
}
var file = new StreamWriter("registar.txt");
for (int i = 0; i < pojmoviOrg.Length; i++)
{
file.Write(pojmoviOrg[i] + " –");
foreach (int stranica in stranice[i])
{
file.Write(" " + stranica + (stranica != stranice[i].Last() ? "," : "."));
}
file.Write("\n");
}
file.Close();
pdfDocument.Close();
}