On 2013-02-20 11:55, Stefan Baur wrote:
So as result we will have a PDF File on server and it not a problem it all to download it to client system.
Yes, it is a problem. Sorry, but it is a horrible design decision to convert data back and forth, causing disk IO and CPU cycles when there's no good reason to do so.
Well, actually there is: Relieabilty and Maintainability. My solution has 3 little steps.
- Create pdf on the server (easy to check and not to difficult too debug)
- Copy file to client (easy to check, and can even be verified)
- Print pdf. This is also something quite easy to get running. Just exec "someprog --print $FILE"
While with your solution there is a multitude of things that can go wrong. It's really hard to find out whether it's a server side, client side, x2go-client (ssh forwarding), driver compability or whatever issue. The data gets piped through a lot of systems and you need quite a bit of knowledge to debug that. Also you need to handle connection losses - something that is not too easy either as you don't have direct control over the system having to cope with that. Further on reliability and maintainability is much more important than a little bit of performance. The spooler-dir also allows you to easily do things like adding "print@clinet"-button to OpenOffice (or any other scriptable app) by exporting the current document as pdf to the spool-dir.
Cheers Morty
-- Dipl.-Ing. Moritz 'Morty' Struebe (Wissenschaftlicher Mitarbeiter) Lehrstuhl für Informatik 4 (Verteilte Systeme und Betriebssysteme) Friedrich-Alexander-Universität Erlangen-Nürnberg Martensstr. 1 91058 Erlangen
Tel : +49 9131 85-25419 Fax : +49 9131 85-28732 eMail : struebe@informatik.uni-erlangen.de WWW : http://www4.informatik.uni-erlangen.de/~morty