Delphi ActiveX Requires Chilkat v11.7.0+

Extract PDF Text to a StringBuilder and Save to a File

See more PDF Text Extraction Examples

Demonstrates ExtractTextSb, which writes the extracted text to a StringBuilder instead of returning it as a JSON string. Use it for large documents. The JSON output still receives the quality report. Delphi ActiveX

var
success: Integer;
pdf: TChilkatPdf;
json: TChilkatJsonObject;
sbText: TChilkatStringBuilder;
jsonOut: TChilkatJsonObject;

begin
success := 0;

//  This example requires the Chilkat API to have been previously unlocked.
//  See Global Unlock Sample for sample code.

//  The PDF used in this example is available at long_report.pdf

pdf := TChilkatPdf.Create(Self);
success := pdf.LoadFile('qa_data/pdf/long_report.pdf');
if (success = 0) then
  begin
    Memo1.Lines.Add(pdf.LastErrorText);
    Exit;
  end;

Memo1.Lines.Add('Number of pages: ' + IntToStr(pdf.NumPages));

//  ExtractTextSb takes the same operation and options as ExtractText, but the text
//  goes to a StringBuilder.  This avoids building a very large JSON string for
//  documents with many pages.
json := TChilkatJsonObject.Create(Self);
sbText := TChilkatStringBuilder.Create(Self);
jsonOut := TChilkatJsonObject.Create(Self);

success := pdf.ExtractTextSb('text',json.ControlInterface,sbText.ControlInterface,jsonOut.ControlInterface);
if (success = 0) then
  begin
    Memo1.Lines.Add(pdf.LastErrorText);
    Exit;
  end;

Memo1.Lines.Add('Extracted ' + IntToStr(sbText.Length) + ' characters.');

//  Save the text to a file.
success := sbText.WriteFile('qa_output/long_report.txt','utf-8',0);
if (success = 0) then
  begin
    Memo1.Lines.Add(sbText.LastErrorText);
    Exit;
  end;

//  The JSON output does not contain the text, but it still has the quality report.
Memo1.Lines.Add('Quality: ' + jsonOut.StringOf('quality.rating'));
Memo1.Lines.Add('Pages with no text (need OCR): [' + jsonOut.StringOf('quality.pagesNeedingOcr') + ']');

//  Sample output:

//  Number of pages: 3
//  Extracted 4522 characters.
//  Quality: good
//  Pages with no text (need OCR): []