SQL Server Requires Chilkat v11.7.0+

Clipped, Off-Page and White-on-White Text in PDF Extraction

See more PDF Text Extraction Examples

Demonstrates the "clippedText", "offPageText" and "obscuredText" filters. Text that a reader never sees - hidden by a clipping path, outside the crop box, or drawn in the same color as the box beneath it - is excluded by default, and the quality report notes what was skipped. Each kind can be included. SQL Server

--
CREATE PROCEDURE ChilkatSample
AS
BEGIN
    DECLARE @hr 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 hidden_text.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/hidden_text.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

    --  The page has a visible paragraph plus three hidden lines: one drawn outside its clipping
    --  rectangle, one below the crop box, and one in white on a white box.
    EXEC sp_OAMethod @pdf, 'ExtractText', @success OUT, 'text', @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

    PRINT '---- default ----'
    EXEC sp_OAMethod @jsonOut, 'StringOf', @sTmp0 OUT, 'text'
    PRINT @sTmp0

    --  The quality report says what was left out.
    DECLARE @i int
    SELECT @i = 0
    DECLARE @count_i int
    EXEC sp_OAMethod @jsonOut, 'SizeOfArray', @count_i OUT, 'quality.pages[0].issues'
    WHILE @i < @count_i
      BEGIN
        EXEC sp_OASetProperty @jsonOut, 'I', @i

        EXEC sp_OAMethod @jsonOut, 'StringOf', @sTmp0 OUT, 'quality.pages[0].issues[i].code'
        PRINT 'issue: ' + @sTmp0
        SELECT @i = @i + 1
      END

    --  Include everything.
    EXEC sp_OAMethod @json, 'UpdateString', @success OUT, 'clippedText', 'include'
    EXEC sp_OAMethod @json, 'UpdateString', @success OUT, 'offPageText', 'include'
    EXEC sp_OAMethod @json, 'UpdateString', @success OUT, 'obscuredText', 'include'
    EXEC sp_OAMethod @pdf, 'ExtractText', @success OUT, 'text', @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

    PRINT '---- clippedText, offPageText, obscuredText: include ----'
    EXEC sp_OAMethod @jsonOut, 'StringOf', @sTmp0 OUT, 'text'
    PRINT @sTmp0

    --  Sample output:

    --  ---- default ----
    --  Visible Text
    --  
    --  This paragraph is visible. The page also carries three kinds of text a reader never sees.
    --  
    --  End of the visible text.
    --  issue: obscuredTextSkipped
    --  issue: clippedTextSkipped
    --  ---- clippedText, offPageText, obscuredText: include ----
    --  Visible Text
    --  
    --  This paragraph is visible. The page also carries three kinds of text a reader never sees.
    --  
    --  CLIPPED: this line is drawn outside its clipping rectangle.
    --  
    --  OBSCURED: this line is white on a white box.
    --  
    --  End of the visible text.
    --  
    --  OFFPAGE: this line sits below the crop box.

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


END
GO