SQL Server 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. SQL Server

--
CREATE PROCEDURE ChilkatSample
AS
BEGIN
    DECLARE @hr int
    DECLARE @iTmp0 int
    DECLARE @iTmp1 int
    DECLARE @sTmp0 nvarchar(4000)
    DECLARE @success int
    SELECT @success = 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

    DECLARE @pdf int
    EXEC @hr = sp_OACreate 'Chilkat.Pdf', @pdf OUT
    IF @hr <> 0
    BEGIN
        PRINT 'Failed to create ActiveX component'
        RETURN
    END

    EXEC sp_OAMethod @pdf, 'LoadFile', @success OUT, 'qa_data/pdf/spans_table.pdf'
    IF @success = 0
      BEGIN
        EXEC sp_OAGetProperty @pdf, 'LastErrorText', @sTmp0 OUT
        PRINT @sTmp0
        EXEC @hr = sp_OADestroy @pdf
        RETURN
      END

    DECLARE @json int
    EXEC @hr = sp_OACreate 'Chilkat.JsonObject', @json OUT

    DECLARE @jsonOut int
    EXEC @hr = sp_OACreate 'Chilkat.JsonObject', @jsonOut OUT

    --  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.
    EXEC sp_OAMethod @json, 'UpdateInt', @success OUT, 'minRows', 3
    EXEC sp_OAMethod @json, 'UpdateInt', @success OUT, 'minColumns', 3

    EXEC sp_OAMethod @pdf, 'ExtractText', @success OUT, 'tables', @json, @jsonOut
    IF @success = 0
      BEGIN
        EXEC sp_OAGetProperty @pdf, 'LastErrorText', @sTmp0 OUT
        PRINT @sTmp0
        EXEC @hr = sp_OADestroy @pdf
        EXEC @hr = sp_OADestroy @json
        EXEC @hr = sp_OADestroy @jsonOut
        RETURN
      END


    EXEC sp_OAMethod @jsonOut, 'IntOf', @iTmp0 OUT, 'count'
    PRINT 'Tables found: ' + @iTmp0
    EXEC sp_OAMethod @jsonOut, 'IntOf', @iTmp0 OUT, 'tables[0].rows'

    EXEC sp_OAMethod @jsonOut, 'IntOf', @iTmp1 OUT, 'tables[0].columns'

    PRINT @iTmp0 + ' rows x ' + @iTmp1 + ' 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.
    DECLARE @rowSpan int

    DECLARE @colSpan int

    DECLARE @j int
    SELECT @j = 0
    DECLARE @count_j int
    EXEC sp_OAMethod @jsonOut, 'SizeOfArray', @count_j OUT, 'tables[0].cells'
    WHILE @j < @count_j
      BEGIN
        EXEC sp_OASetProperty @jsonOut, 'J', @j
        EXEC sp_OAMethod @jsonOut, 'IntOf', @rowSpan OUT, 'tables[0].cells[j].rowSpan'
        EXEC sp_OAMethod @jsonOut, 'IntOf', @colSpan OUT, 'tables[0].cells[j].colSpan'
        IF (@rowSpan > 1) or (@colSpan > 1)
          BEGIN

            EXEC sp_OAMethod @jsonOut, 'IntOf', @iTmp0 OUT, 'tables[0].cells[j].row'

            EXEC sp_OAMethod @jsonOut, 'IntOf', @iTmp1 OUT, 'tables[0].cells[j].col'

            EXEC sp_OAMethod @jsonOut, 'StringOf', @sTmp0 OUT, 'tables[0].cells[j].text'



            PRINT 'row ' + @iTmp0 + ', col ' + @iTmp1 + ': "' + @sTmp0 + '" spans ' + @rowSpan + ' row(s) and ' + @colSpan + ' column(s)'
          END
        SELECT @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)

    EXEC @hr = sp_OADestroy @pdf
    EXEC @hr = sp_OADestroy @json
    EXEC @hr = sp_OADestroy @jsonOut


END
GO