RE: Downloading BLOBS from Oracle DB

>-----Original Message-----
>From: enhydra-admin@xxxxxxxxxxx [mailto:enhydra-admin@xxxxxxxxxxx] On
Behalf Of Sreedhar.Gajula@xxxxxx
>Sent: Wednesday, January 22, 2003 2:07 AM
>To: enhydra@xxxxxxxxxxx
>Subject: Enhydra: Downloading BLOBS from Oracle DB



>Hi There, 

>I am trying to download a blob containing a MS word document from an
Oracle DB. 
>I was able to read the blob and create a .doc file on the server and
created an hyper link to open in the webserver. 
>I was wondering if there is a way to automatically create the doc file
on the fly without having to create the file on the >server. 
>
>BTW I was searching on google and found your message about the same
problem. 
>
>Appreciate any ideas on this. 
>
>Thanks
>
>Sreedhar Gajula 

Hi!
Last year in May, somebody ask something simular.
Here is my answer and the original question:

Hi!

I wrote a PO to retrieve a PDF or HTML - file dynamically from a
MySQL-Database and upload it to the client.

Here is the code:
========================================================================
=======
/*
 * Pdf
 *
 * (c) 2002 fellowhome.de GmbH
 *
 */

package fellowhome.presentation.pdfs;

import com.lutris.appserver.server.httpPresentation.*;

import fellowhome.FellowhomeException;
import fellowhome.data.objekt.ProjektDO;
import fellowhome.data.objekt.PrblobDO;
import fellowhome.data.objekt.PrblobQuery;

/**
 * <b>Title:</b>       Pdf - Präsentationsobjekt<br>
 * <b>Description:</b> PO zum Ausgeben von Pdfs<br>
 * <b>Company:</b>     fellowhome.de GmbH<br>
 * @author             Thorsten Hilker
 * @version            1.0
 */

public class Pdf implements HttpPresentation {

  // Methode die ausgeführt wird, wenn das PO aufgerufen wird
  public void run(HttpPresentationComms comms)
    throws Exception {

    // ProjektID aus der URL einlesen
    String projektID = comms.request.getParameter("projektID");

    if (projektID == null) {
      projektID = "";
    }

    // PDF-Typ und Name aus der URL einlesen
    String pdfTP = comms.request.getParameter("pdfTP");
    String pdfNA = comms.request.getParameter("pdfNA");

    // Das Objekt aus der Datenbank einlesen
    ProjektDO pDO = ProjektDO.createExisting(projektID);
    PrblobQuery pQuery = new PrblobQuery();
    pQuery.requireUniqueInstance();
    pQuery.setQueryPrid(pDO);
    pQuery.setQueryBltp(pdfTP);
    PrblobDO pbDO = pQuery.getNextDO();
    if (pbDO == null)
      throw new FellowhomeException(this, 2, "Das Dokument wurde nicht
gefunden!", "---");

    // Stream erzeugen
    if (pdfTP.equals("bpdf")) {
      comms.response.setContentType("application/pdf");
    }
    if (pdfTP.equals("bhtm")) {
      comms.response.setContentType("text/html");
    }
    comms.response.setContentLength(pbDO.getBdat().length);
    comms.response.setHeader("Content-Disposition",
                             "filename=" + '"' + pdfNA + '"');

    // Stream aufgrund des Typs füllen und senden
    HttpPresentationOutputStream stream =
comms.response.getOutputStream();
    stream.write(pbDO.getBdat());

    stream.flush();
  }
}
========================================================================
===
The Href to call this PO looks like this:

<A href="/dyn/pdfs/Pdf.po?projektID=176775&pdfTP=bpdf&pdfNA=Seeschloss
Hermsdorf.pdf" id="Bbpd" type="application/pdf"><IMG border="0"
height="11" src="/pics/link.gif" width="13">Baubeschreibung im
PDF-Format</A>

The parameters "projektID=176775&pdfTP=bpdf&pdfNA=Seeschloss
Hermsdorf.pdf id=Bbpd" are build dynamically by another PO which builds
the page which is shown in the browser.

If you want to upload wordfiles you have to use the mime-type
"application/msword" instead of "application/pdf" in the code and in the
Href.

Hope that helps.
If you have further questions let me know.

Greetings!
Thorsten

PS The comments in the code are in German, sorry :) 
PPS If you want to see how it works go to our website:
http://www.fellowhome.de/dyn/immosuche/ObjtHaus.po
then press "zu den Details"
then press "Baubeschreibung"
then press "Baubeschreibung im PDF-Format"

> -----Original Message-----
> From: enhydra-admin@xxxxxxxxxxx
> [mailto:enhydra-admin@xxxxxxxxxxx] On Behalf Of Dr Louis Coetzee
> Sent: Thursday, May 23, 2002 10:44 AM
> To: enhydra@xxxxxxxxxxx
> Cc: Moses Marivatse; Ilse Viviers
> Subject: Enhydra: File download from blobs
> 
> 
> Hi,
> 
> I have a number of word documents (or other unfriendly binary
> formats) stored 
> in a blob in a Oracle database.
> 
> I need to provide a HTML page providing a link (url) that the
> user can click 
> on to download this document to his/her own machine.
> 
> The one possibility I can think of is to read the blob and
> write it to disk 
> (e.q. /tmp/documentname) and then dynamically add a URL 
> reference to this file 
> on disk on the html page. The problem with this is that the 
> size of the /tmp 
> will grow the whole time.
> 
> Is there a way how I can provide a URL to the document,
> without writing it to 
> a tmp storage first ? I.e. access the blob directly in the database ?
> 
> 
> Well later
> 
> Louis
> 
> 
> _______________________________________________
> Enhydra mailing list
> Enhydra@xxxxxxxxxxx
> http://www.enhydra.org/mailman/listinfo.cgi/en> hydra
>