Ruby 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. Ruby

require 'chilkat'

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 spans_table.pdf

pdf = Chilkat::CkPdf.new()
success = pdf.LoadFile("qa_data/pdf/spans_table.pdf")
if (success == false)
    print pdf.lastErrorText() + "\n";
    exit
end

json = Chilkat::CkJsonObject.new()
jsonOut = Chilkat::CkJsonObject.new()

#  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.
json.UpdateInt("minRows",3)
json.UpdateInt("minColumns",3)

success = pdf.ExtractText("tables",json,jsonOut)
if (success == false)
    print pdf.lastErrorText() + "\n";
    exit
end

print "Tables found: " + jsonOut.IntOf("count").to_s() + "\n";
print jsonOut.IntOf("tables[0].rows").to_s() + " rows x " + jsonOut.IntOf("tables[0].columns").to_s() + " columns" + "\n";

#  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
count_j = jsonOut.SizeOfArray("tables[0].cells")
while j < count_j
    jsonOut.put_J(j)
    rowSpan = jsonOut.IntOf("tables[0].cells[j].rowSpan")
    colSpan = jsonOut.IntOf("tables[0].cells[j].colSpan")
    if ((rowSpan > 1) or (colSpan > 1))
        print "row " + jsonOut.IntOf("tables[0].cells[j].row").to_s() + ", col " + jsonOut.IntOf("tables[0].cells[j].col").to_s() + ": \"" + jsonOut.stringOf("tables[0].cells[j].text") + 
            "\" spans " + rowSpan.to_s() + " row(s) and " + colSpan.to_s() + " column(s)" + "\n";
    end

    j = j + 1
end

#  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)