Summer of Code Proposal - Looking for a mentor

Nicola Gigante nicola.gigante at gmail.com
Mon Mar 9 08:11:17 GMT 2009


> I like this idea.
>
> Perhaps not part of your project, but I'd like to have a framework
> for dealing with different types of backups.
>
> eg. in korganizer there's a wish for no backups for calendar files;
> numbered backups for calendar files, and simple ~ based backups.
>
> eg. in kaddressbook, there is a rotating 7 day backup created.
> it would be nice for user selection of backup type for addressbook  
> data.

My idea included the ability to automatically make backups as an  
option for auto-saving (instead of doing auto-save on the working  
document itself), but it's hard to define kaddressbook a "document- 
based" application so I think my framework won't fit well with it.

> I don't see the point of taking this new framework that you're going  
> to come
> up with and port the kate part to that when it is already using a well
> defined set of interfaces (see KTextEditor) that were specifically  
> designed
> for the features of Kate and other text editors in mind.

Well I didn't talk about the Kate part actually. Instead I talk about  
the kate application, which has a lot of custom code for handling  
document operations, outside the kpart. Similar code can be found for  
example in kolourpaint, and in every document-oriented application (I  
think about okular but I didn't see its sources yet). I think this is  
a useless amount of duplication. Of course the Kate editor is also a  
lot complex and If its developers don't want to immediately use a new  
and young framework I can simply port another application. The point  
is to port something to show that the framework works (when it's  
finished).

Thanks :D




More information about the kde-core-devel mailing list