Xbase++ Requires Chilkat v11.7.0+

Extract a PDF Table with Merged Cells

See more PDF Text Extraction Examples

Demonstrates how merged cells are reported by the "tables" operation: a header cell that spans several columns has colSpan greater than 1, and a row label that spans several rows has rowSpan greater than 1. Also shows the minRows and minColumns options. Xbase++

LOCAL nSuccess
LOCAL oPdf
LOCAL oJson
LOCAL oJsonOut
LOCAL nRowSpan
LOCAL nColSpan
LOCAL j
LOCAL nCount_j

nSuccess := 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 spans_table.pdf

oPdf := CreateObject("Chilkat.Pdf")
nSuccess := oPdf:LoadFile("qa_data/pdf/spans_table.pdf")
IF (nSuccess == 0)
    ? oPdf:LastErrorText
    oPdf:destroy()
    RETURN
ENDIF

oJson := CreateObject("Chilkat.JsonObject")
oJsonOut := CreateObject("Chilkat.JsonObject")

//  Grids smaller than minRows x minColumns are not reported as tables.  The defaults are 2 and 2.
//  Setting them higher skips small boxes that are not really tables.
oJson:UpdateInt("minRows", 3)
oJson:UpdateInt("minColumns", 3)

nSuccess := oPdf:ExtractText("tables", oJson, oJsonOut)
IF (nSuccess == 0)
    ? oPdf:LastErrorText
    oPdf:destroy()
    oJson:destroy()
    oJsonOut:destroy()
    RETURN
ENDIF

? "Tables found: " + Str(oJsonOut:IntOf("count"))
? Str(oJsonOut:IntOf("tables[0].rows")) + " rows x " + Str(oJsonOut:IntOf("tables[0].columns")) + " columns"

//  A merged cell is reported once, at its top-left position, with its spans.
//  The positions it covers have no cell of their own.

j := 0
nCount_j := oJsonOut:SizeOfArray("tables[0].cells")
DO WHILE j < nCount_j
    oJsonOut:J := j
    nRowSpan := oJsonOut:IntOf("tables[0].cells[j].rowSpan")
    nColSpan := oJsonOut:IntOf("tables[0].cells[j].colSpan")
    IF ((nRowSpan > 1) .OR. (nColSpan > 1))
        ? "row " + Str(oJsonOut:IntOf("tables[0].cells[j].row")) + ", col " + Str(oJsonOut:IntOf("tables[0].cells[j].col")) + ': "' + oJsonOut:StringOf("tables[0].cells[j].text") + '" spans ' + Str(nRowSpan) + " row(s) and " + Str(nColSpan) + " column(s)"
    ENDIF

    j := j + 1
ENDDO

//  Sample output:

//  Tables found: 1
//  6 rows x 6 columns
//  row 0, col 0: "Area" spans 2 row(s) and 1 column(s)
//  row 0, col 1: "City" spans 2 row(s) and 1 column(s)
//  row 0, col 2: "First Half" spans 1 row(s) and 2 column(s)
//  row 0, col 4: "Second Half" spans 1 row(s) and 2 column(s)
//  row 2, col 0: "East" spans 2 row(s) and 1 column(s)
//  row 4, col 0: "West" spans 2 row(s) and 1 column(s)

oPdf:destroy()
oJson:destroy()
oJsonOut:destroy()