Delphi DLL 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 DLL

var
success: Boolean;
pdf: HCkPdf;
json: HCkJsonObject;
sbText: HCkStringBuilder;
jsonOut: HCkJsonObject;

begin
success := False;

//  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 := CkPdf_Create();
success := CkPdf_LoadFile(pdf,'qa_data/pdf/long_report.pdf');
if (success = False) then
  begin
    Memo1.Lines.Add(CkPdf__lastErrorText(pdf));
    Exit;
  end;

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

//  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 := CkJsonObject_Create();
sbText := CkStringBuilder_Create();
jsonOut := CkJsonObject_Create();

success := CkPdf_ExtractTextSb(pdf,'text',json,sbText,jsonOut);
if (success = False) then
  begin
    Memo1.Lines.Add(CkPdf__lastErrorText(pdf));
    Exit;
  end;

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

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

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

//  Sample output:

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

CkPdf_Dispose(pdf);
CkJsonObject_Dispose(json);
CkStringBuilder_Dispose(sbText);
CkJsonObject_Dispose(jsonOut);