SQL Server Requires Chilkat v11.7.0+
SQL Server
Search a PDF with a Regular Expression
See more PDF Text Extraction Examples
Demonstrates searching a PDF with a regular expression (regex: true), limiting the number of results (maxResults), and matching whole words only (wholeWord).Chilkat SQL Server Downloads
--
CREATE PROCEDURE ChilkatSample
AS
BEGIN
DECLARE @hr int
DECLARE @iTmp0 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 order_records.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/order_records.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
-- Find every order number. With regex set to true, the query is a regular expression.
EXEC sp_OAMethod @json, 'UpdateString', @success OUT, 'query', 'ORD-[0-9]{4}-[0-9]{4}'
EXEC sp_OAMethod @json, 'UpdateBool', @success OUT, 'regex', 1
-- We only want the matched text, so no context is needed.
EXEC sp_OAMethod @json, 'UpdateInt', @success OUT, 'context', 0
EXEC sp_OAMethod @pdf, 'ExtractText', @success OUT, 'search', @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 'Order numbers found: ' + @iTmp0
DECLARE @i int
SELECT @i = 0
DECLARE @count_i int
EXEC sp_OAMethod @jsonOut, 'SizeOfArray', @count_i OUT, 'matches'
WHILE @i < @count_i
BEGIN
EXEC sp_OASetProperty @jsonOut, 'I', @i
EXEC sp_OAMethod @jsonOut, 'StringOf', @sTmp0 OUT, 'matches[i].text'
PRINT @sTmp0
SELECT @i = @i + 1
END
-- Find the e-mail addresses, but stop after the first three.
EXEC sp_OAMethod @json, 'UpdateString', @success OUT, 'query', '[a-z0-9._]+@[a-z]+[.][a-z]+'
EXEC sp_OAMethod @json, 'UpdateInt', @success OUT, 'maxResults', 3
EXEC sp_OAMethod @pdf, 'ExtractText', @success OUT, 'search', @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 'First three e-mail addresses:'
SELECT @i = 0
EXEC sp_OAMethod @jsonOut, 'SizeOfArray', @count_i OUT, 'matches'
WHILE @i < @count_i
BEGIN
EXEC sp_OASetProperty @jsonOut, 'I', @i
EXEC sp_OAMethod @jsonOut, 'StringOf', @sTmp0 OUT, 'matches[i].text'
PRINT @sTmp0
SELECT @i = @i + 1
END
-- wholeWord: true matches only whole words. The word "cat" appears once in the PDF,
-- but it is also inside "catalog" and "category".
EXEC sp_OAMethod @json, 'Delete', @success OUT, 'maxResults'
EXEC sp_OAMethod @json, 'UpdateBool', @success OUT, 'regex', 0
EXEC sp_OAMethod @json, 'UpdateString', @success OUT, 'query', 'cat'
EXEC sp_OAMethod @pdf, 'ExtractText', @success OUT, 'search', @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 'Occurrences of cat (any): ' + @iTmp0
EXEC sp_OAMethod @json, 'UpdateBool', @success OUT, 'wholeWord', 1
EXEC sp_OAMethod @pdf, 'ExtractText', @success OUT, 'search', @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 'Occurrences of cat (whole word): ' + @iTmp0
-- Sample output:
-- Order numbers found: 9
-- ORD-2026-0301
-- ORD-2026-0302
-- ORD-2026-0303
-- ORD-2026-0304
-- ORD-2026-0305
-- ORD-2026-0306
-- ORD-2026-0307
-- ORD-2026-0308
-- ORD-2026-0307
-- First three e-mail addresses:
-- alice.moore@example.com
-- bkumar@example.org
-- carla@example.net
-- Occurrences of cat (any): 3
-- Occurrences of cat (whole word): 1
EXEC @hr = sp_OADestroy @pdf
EXEC @hr = sp_OADestroy @json
EXEC @hr = sp_OADestroy @jsonOut
END
GO