Extract OLE Object (pdf) from Access DB

Question

We are upgrading/converting several old Access databases to MS-SQL. Many of these databases have OLE Object fields that store PDF files. I'm looking for a way to extract these files and store them in our SQL database. I've seen similar questions that answer how you might do this with image files (jpg, bmp, gif, etc) but I haven't found a way that works with PDF.

Answer 1

I finally got some code working for what I want it to do. The trick is determining what part is the OLE Header and removing it. Here is what is working for me (based on code found here)

    public static byte[] StripOleHeader(byte[] fileData)
    {
        const string START_BLOCK = "%PDF-1.3";
        int startPos = -1;

        Encoding u8 = Encoding.UTF7;
        string strEncoding = u8.GetString(fileData);

        if (strEncoding.IndexOf(START_BLOCK) != -1)
        {
            startPos = strEncoding.IndexOf(START_BLOCK);
        }

        if (startPos == -1)
        {
            throw new Exception("Could not find PDF Header");
        }

        byte[] retByte = new byte[fileData.LongLength - startPos];

        Array.Copy(fileData, startPos, retByte, 0, fileData.LongLength - startPos);

        return retByte;
    }

Note that this only works for PDF files.

Answer 2

OLEtoDisk

"This version saves the entire contents of a table containing OLE Objects to disk. Does NOT require the original application that served as the OLE server to insert the object. Supports all MS Office documents, PDF, All images inserted by MS Photo Editor, MS Paint, and Paint Shop Pro. Also supports extraction of PACKAGE class including original Filename. Contains function to produce a full Inventory of the OLE field including LINKED path and Filenames. Uses Structured Storage API's to read the actual contents of the field"

http://lebans.com/oletodisk.htm