Re: Development /w Evolution Conduits

Tom Billiet <[email protected]> Tue, 05 Aug 2008 17:30:39 +0200
Newsgroups gmane.comp.gnome.apps.pilot
Message-ID <[email protected]>

----- Original Message -----
From: David Wade Hagar
Time: 02-08-08 01:11
> I would love to help out with the development of the Evolution conduits.
> I'm not the best programmer, but I'll throw down with trying to get it
> to work.
> 
> I've noticed that its already a known issue about the address book
> fields, but additionally, there is lack of support in the calendar for
> the Location field.
> 
> Also, categories in contacts don't sync either.
> 

About the categories in contacts. I've added category support for the 
calendar lately, and have been looking to add it for contacts too. 
Unfortunately where it wasn't very hard for the calendar, it is hard for 
the contacts.
The problem lies in the fact that the categories in contact are added in 
the ContactsDB, which was added on OS5. Currently gpilot is only using 
the AddressDB, the old version of the database. It's actually only a 
transparent mapping of the new ContactsDB to the old format, but it 
lacks a lot of fields.
I've been looking into it, but encountered the following problems:
- gpilot has a list of conduits he can handle, and launches the 
appropriate handler. Now the addressbook handler is launched if the 
AddressDB is found, and therefor you cannot access the ContactsDB in 
that handler. A modification to gpilot is needed that will check first 
if ContactsDB is available, then launch a handler for that. If not, use 
AddressDB and launch a handler for that format.
- The functions currently used to read/write data to evolution are 
actually wrapper functions to provide easy access. But unfortunately 
these functions are far from complete, and a lot of fields (I think 
birthday was one of them) is not available trough this api. The only way 
to do that I think is to start using the LDIFF format from evolution 
directly, but that requires a complete refactoring of the contacts handler.

So I've started a little on it, but dropped it because of the lack of 
time and the amount of work it would require.

But if you have the time to work on it, I would say: go for it.

Kind regards,
Tom