Re: Evolution Feed Reader Support

Hi Andrew,

I had seen the brainread code quite a long time back. From what i 
remember, it was a sample code, made to prove how easy it was to have 
shell components implemented in Evolution.

Evolution currently has 4 [ rather 5 ], shell components : viz., 
mail,addressbook, calendar and tasks [ and another one being exchange ]. 
Addressbook, calendar and tasks use e-d-s as a backend store for their 
data types. Mailer has a separate store which it implements on its own 
using the camel apis. Exchange just morphs all these together.

For brainread data, i suppose you will have to follow the mailer way of 
implement your own data store. Probably you could use a xmldoc parser to 
store all the data in xml files under a particular directory in 
$HOME/.evolution.

Ofcourse, more thoughts can be added to this, but i hope that this 
clarifies the point that e-d-s is currently not the place where you want 
to store brain-read data, unless you add a new brain-read data-store 
implementation to e-d-s, which the existing shell component would treat 
as a backend to store the info.

HTH
-- Sarfraaz

Andrew Case wrote:

>Hey all,
>
>I'm working on the component previously known as "brainread" (from
>evolution 1.x).  It's a feed aggregator for evolution.  Forgive me if
>I'm off base at all here.  I've only been working on evolution a couple
>of months now and it's a very large system to try and learn.  So let me
>know if I'm completely off base.  Anyway, right now I'm just storing
>each feed's information in gconf, which is really hideous and is just a
>hack to get around the fact that EDS's backend isn't really designed to
>hold general data... to my understanding.   Anyway, I guess I want to
>know what people's opinions are on the best way to store this.  Should I
>really be looking at building a new backend into the eds architecture?
>Can an EBook or Calendar architecture store general data types?  I just
>have really little clue about EDS backends.  Anyway, the data I'll want
>to store eventually are things like: feed name, update interval, type of
>source, location of source, authentication information, etc.  Obviously
>gconf isn't the place to store this, and right now I'm just storing a
>huge concated string of all the sources(feeds) I'm using.  Anyway,
>whatever help being pointed in the right direction would help.
>
>Dev Screenshot: http://node7.org/files/user_files/darkfrog/evo_feeds.png
>
>--
>Andrew Case
>[email protected]
>http://www.node7.org/
>
>_______________________________________________
>Evolution-hackers mailing list
>[email protected]
>http://mail.gnome.org/mailman/listinfo/evolution-hackers
>  
>