Delphi DLL Requires Chilkat v11.7.0+

Extract Tables from a PDF

See more PDF Text Extraction Examples

Demonstrates the "tables" operation, which returns the tables detected on the pages as rows and cells. Each cell has its row and column (counted from 0), its row and column span, and its text. Delphi DLL

var
success: Boolean;
pdf: HCkPdf;
json: HCkJsonObject;
jsonOut: HCkJsonObject;
sbRow: HCkStringBuilder;
currentRow: Integer;
row: Integer;
j: Integer;
count_j: Integer;
i: Integer;
count_i: Integer;

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 price_list.pdf

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

json := CkJsonObject_Create();
jsonOut := CkJsonObject_Create();

//  The "tables" operation returns the tables found on the selected pages.
//  Tables are recognized by their drawn cell borders.
success := CkPdf_ExtractText(pdf,'tables',json,jsonOut);
if (success = False) then
  begin
    Memo1.Lines.Add(CkPdf__lastErrorText(pdf));
    Exit;
  end;

Memo1.Lines.Add('Number of tables: ' + IntToStr(CkJsonObject_IntOf(jsonOut,'count')));

//  Used to build each row of output.
sbRow := CkStringBuilder_Create();

i := 0;
count_i := CkJsonObject_SizeOfArray(jsonOut,'tables');
while i < count_i do
  begin
    CkJsonObject_putI(jsonOut,i);
    Memo1.Lines.Add('Table on page ' + IntToStr(CkJsonObject_IntOf(jsonOut,'tables[i].page')) + ': ' + IntToStr(CkJsonObject_IntOf(jsonOut,'tables[i].rows')) + ' rows x '
         + IntToStr(CkJsonObject_IntOf(jsonOut,'tables[i].columns')) + ' columns, detected by ' + CkJsonObject__stringOf(jsonOut,'tables[i].detectedBy'));

    //  The cells are listed row by row.  Print each row as its cells separated by " | ".
    currentRow := -1;
    CkStringBuilder_SetString(sbRow,'');
    j := 0;
    count_j := CkJsonObject_SizeOfArray(jsonOut,'tables[i].cells');
    while j < count_j do
      begin
        CkJsonObject_putJ(jsonOut,j);
        row := CkJsonObject_IntOf(jsonOut,'tables[i].cells[j].row');
        if (row <> currentRow) then
          begin
            if (currentRow >= 0) then
              begin
                Memo1.Lines.Add(CkStringBuilder__getAsString(sbRow));
              end;
            currentRow := row;
            CkStringBuilder_SetString(sbRow,'  ');
          end
        else
          begin
            CkStringBuilder_Append(sbRow,' | ');
          end;
        CkStringBuilder_Append(sbRow,CkJsonObject__stringOf(jsonOut,'tables[i].cells[j].text'));
        j := j + 1;
      end;

    Memo1.Lines.Add(CkStringBuilder__getAsString(sbRow));
    i := i + 1;
  end;

//  The cells of the first table as JSON:
CkJsonObject_putEmitCompact(jsonOut,False);
Memo1.Lines.Add(CkJsonObject__emit(jsonOut));

//  Sample output:

//  Number of tables: 1
//  Table on page 1: 8 rows x 5 columns, detected by ruled
//    Code | Product | Unit | Price | In Stock
//    HT-101 | Claw hammer, 16 oz | each | $18.00 | 24
//    HT-102 | Adjustable wrench, 10 in | each | $14.50 | 31
//    HT-103 | Screwdriver set, 6 piece | set | $22.00 | 12
//    HT-104 | Tape measure, 25 ft | each | $9.75 | 40
//    HT-105 | Utility knife | each | $6.20 | 58
//    HT-106 | Pliers, needle nose | each | $11.00 | 19
//    HT-107 | Hand saw, 20 in | each | $21.40 | 8
//  {
//    "count": 1,
//    "tables": [
//      {
//        "page": 1,
//        "detectedBy": "ruled",
//        "rows": 8,
//        "columns": 5,
//        "cells": [
//          {
//            "row": 0,
//            "col": 0,
//            "rowSpan": 1,
//            "colSpan": 1,
//            "text": "Code"
//          },
//          {
//            "row": 0,
//            "col": 1,
//            "rowSpan": 1,
//            "colSpan": 1,
//            "text": "Product"
//          },
//          ...
//        ]
//      }
//    ],
//    "quality": {
//      "rating": "good",
//      ...
//    }
//  }

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