B4X 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. B4X

Dim success As Boolean = 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

Dim pdf As ChilkatPdf
pdf.Initialize("pdf")
success = pdf.LoadFile("qa_data/pdf/price_list.pdf")
If success = False Then
    Log(pdf.LastErrorText)
    Return
End If


Dim json As ChilkatJsonObject
json.Initialize
Dim jsonOut As ChilkatJsonObject
jsonOut.Initialize

'  The "tables" operation returns the tables found on the selected pages.
'  Tables are recognized by their drawn cell borders.
success = pdf.ExtractText("tables", json, jsonOut)
If success = False Then
    Log(pdf.LastErrorText)
    Return
End If


Log("Number of tables: " & jsonOut.IntOf("count"))

'  Used to build each row of output.
Dim sbRow As ChilkatStringBuilder
sbRow.Initialize
Dim currentRow As Int
Dim row As Int
Dim j As Int
Dim count_j As Int

Dim i As Int = 0
Dim count_i As Int = jsonOut.SizeOfArray("tables")
Do While i < count_i
    jsonOut.I = i
    Log("Table on page " & jsonOut.IntOf("tables[i].page") & ": " & jsonOut.IntOf("tables[i].rows") & " rows x " & jsonOut.IntOf("tables[i].columns") & " columns, detected by " & jsonOut.StringOf("tables[i].detectedBy"))

    '  The cells are listed row by row.  Print each row as its cells separated by " | ".
    currentRow = -1
    sbRow.SetString("")
    j = 0
    count_j = jsonOut.SizeOfArray("tables[i].cells")
    Do While j < count_j
        jsonOut.J = j
        row = jsonOut.IntOf("tables[i].cells[j].row")
        If row <> currentRow Then
            If currentRow >= 0 Then
                Log(sbRow.GetAsString)
            End If

            currentRow = row
            sbRow.SetString("  ")
        Else
            sbRow.Append(" | ")
        End If

        sbRow.Append(jsonOut.StringOf("tables[i].cells[j].text"))
        j = j + 1
    Loop
    Log(sbRow.GetAsString)
    i = i + 1
Loop

'  The cells of the first table as JSON:
jsonOut.EmitCompact = False
Log(jsonOut.Emit)

'  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",
'      ...
'    }
'  }