Conduit Development (problems)
- From: Vesa Pentti <vpentti cc helsinki fi>
- To: gnome-pilot-list gnome org
- Cc: arto nissinen helsinki fi, kalle salo helsinki fi, tvihtari cs helsinki fi, hjraisan cc helsinki fi, vpentti cc helsinki fi
- Subject: Conduit Development (problems)
- Date: Thu, 24 Jul 2003 15:37:26 +0300
Hello,
I'm a member of a project group that is developing synchronization software for Gnome
Pilot and MultiSync. (This is a University project.) The objective of our software is
to be able to synchronize data between Palm Pilot and MultiSync, using Gnome Pilot.
(For those who are interested in MultiSync: http://multisync.sourceforge.net/)
It's clear that we need to develop a Gnome Pilot conduit as a part of our software.
But what kind of conduit? It's peculiar to MultiSync (it seems) that we can't have
access to the database that lies on the other side of MultiSync. Does that mean that
our conduit can't be a SyncAbs conduit? At least it seems to be quite impossible to
implement "for_each" signal or "match_record" signal when you can't directly access
the database.
We've been thinking about making a crude prototype version: we implement a custom
conduit, which, every time synchronization is required, simply reads the entire Palm
database and passes it on to a temporary data storage, from where the data is passed
on to MultiSync. I suppose this is possible and should not be too difficult to
implement. But is it sensible?
In our case, CopyToPilot functionality is especially interesting. Implementing
CopyToPilot without direct access to the other database seems somewhat impossible.
I don't have previous experience of synchronization software or stuff related to
synchronizing, so I'm a little confused. (Like the rest of the team, too)
All comments would be greatly appreciated.
Regards,
Vesa Pentti
E-mail: vpentti cc helsinki fi
University of Helsinki - Department of Computer Science
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]