SQL Server Requires Chilkat v11.7.0+

Extract Right-to-Left (Hebrew, Arabic) Text in Logical Order

See more PDF Text Extraction Examples

Demonstrates the "bidi" clean-up option. PDFs draw right-to-left text in visual order (the order of the glyphs on the page); by default the extracted text is reordered into logical order, so that it reads and compares correctly. Set "bidi" to false to get the glyphs in the order they are drawn. 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 hebrew_line.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/hebrew_line.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 line of Hebrew with a Latin word (PDF) and a number (2026) embedded in it.
    --  In logical order the line reads: kovetz PDF mishnat 2026 muchan ("the PDF file from 2026 is ready"),
    --  with the Hebrew words in reading order and "PDF" and "2026" kept left to right.
    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 (bidi: true, logical order) ----'
    EXEC sp_OAMethod @jsonOut, 'StringOf', @sTmp0 OUT, 'text'
    PRINT @sTmp0

    --  Without reordering, the Hebrew words come out reversed and in the opposite order, as drawn.
    EXEC sp_OAMethod @json, 'UpdateBool', @success OUT, 'bidi', 0
    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 '---- bidi: false (as drawn) ----'
    EXEC sp_OAMethod @jsonOut, 'StringOf', @sTmp0 OUT, 'text'
    PRINT @sTmp0

    --  Sample output (the Hebrew is shown here transliterated):

    --  ---- default (bidi: true, logical order) ----
    --  Right-to-Left Text
    --  
    --  The line below is Hebrew with a Latin word and a number embedded in it:
    --  
    --  kovetz PDF mishnat 2026 muchan
    --  ---- bidi: false (as drawn) ----
    --  Right-to-Left Text
    --  
    --  The line below is Hebrew with a Latin word and a number embedded in it:
    --  
    --  nachum 2026 tanshim PDF ztevok    (the words in reverse order, each Hebrew word with its letters reversed)

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


END
GO