Visual FoxPro 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. Visual FoxPro

LOCAL lnSuccess
LOCAL loPdf
LOCAL loJson
LOCAL loJsonOut
LOCAL loSbRow
LOCAL lnCurrentRow
LOCAL lnRow
LOCAL j
LOCAL lnCount_j
LOCAL i
LOCAL lnCount_i

lnSuccess = 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 price_list.pdf

loPdf = CreateObject('Chilkat.Pdf')
lnSuccess = loPdf.LoadFile("qa_data/pdf/price_list.pdf")
IF (lnSuccess = 0) THEN
    ? loPdf.LastErrorText
    RELEASE loPdf
    CANCEL
ENDIF

loJson = CreateObject('Chilkat.JsonObject')
loJsonOut = CreateObject('Chilkat.JsonObject')

*  The "tables" operation returns the tables found on the selected pages.
*  Tables are recognized by their drawn cell borders.
lnSuccess = loPdf.ExtractText("tables",loJson,loJsonOut)
IF (lnSuccess = 0) THEN
    ? loPdf.LastErrorText
    RELEASE loPdf
    RELEASE loJson
    RELEASE loJsonOut
    CANCEL
ENDIF

? "Number of tables: " + STR(loJsonOut.IntOf("count"))

*  Used to build each row of output.
loSbRow = CreateObject('Chilkat.StringBuilder')

i = 0
lnCount_i = loJsonOut.SizeOfArray("tables")
DO WHILE i < lnCount_i
    loJsonOut.I = i
    ? "Table on page " + STR(loJsonOut.IntOf("tables[i].page")) + ": " + STR(loJsonOut.IntOf("tables[i].rows")) + " rows x " + STR(loJsonOut.IntOf("tables[i].columns")) + " columns, detected by " + loJsonOut.StringOf("tables[i].detectedBy")

    *  The cells are listed row by row.  Print each row as its cells separated by " | ".
    lnCurrentRow = -1
    loSbRow.SetString("")
    j = 0
    lnCount_j = loJsonOut.SizeOfArray("tables[i].cells")
    DO WHILE j < lnCount_j
        loJsonOut.J = j
        lnRow = loJsonOut.IntOf("tables[i].cells[j].row")
        IF (lnRow <> lnCurrentRow) THEN
            IF (lnCurrentRow >= 0) THEN
                ? loSbRow.GetAsString()
            ENDIF

            lnCurrentRow = lnRow
            loSbRow.SetString("  ")
        ELSE
            loSbRow.Append(" | ")
        ENDIF

        loSbRow.Append(loJsonOut.StringOf("tables[i].cells[j].text"))
        j = j + 1
    ENDDO
    ? loSbRow.GetAsString()
    i = i + 1
ENDDO

*  The cells of the first table as JSON:
loJsonOut.EmitCompact = 0
? loJsonOut.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",
*      ...
*    }
*  }

RELEASE loPdf
RELEASE loJson
RELEASE loJsonOut
RELEASE loSbRow