[Cosmo-dev] atom media creation/update

[Travis Vachon]

Yeah, I think that's probably correct, I think I conflated the stuff  
you mentioned about changing the subscription and preference formats  
with the fact that you said we'd be changing some details of the item  
create/save process. Sorry about that, it was definitely largely my  
fault for not reading your response to my first response closely enough!

Now that I'm more on the same page, yeah, that doesn't sound like too  
much work on my end. That said, I'm not sure this will be a huge  
performance boost on our side, since the atom wrapper isn't really  
that big at the moment. Assuming large numbers of users I could  
imagine this being a win for the server if processing Atom is  
resource intensive. If it is a significant server win I think it's  
definitely worth doing sooner (which might still mean after preview)  
rather than later, but I would guess that it won't make the webui  
experience that much faster.

-Travis

On Jun 11, 2007, at 12:18 PM, Brian Moseley wrote:

> On 6/11/07, Travis Vachon <travis at osafoundation.org> wrote:
>
>> That's probably true, but every time estimate I've made for this
>> process has turned out to be dreadfully optimistic. Given that we're
>> talking about changing the XML wrapper that the browser is in charge
>> of interpreting, and given how unpredictable and difficult to debug
>> browser differences often are, I'd say this is about right.
>
> i think you didn't understand the proposal. it  deals with how the
> client sends creates and updates (ie write requests), not with how it
> reads responses (which is where the browser has to interpret xml).
>
> looking at the itemToAtomEntry function in eim.js, it seems that you'd
> just change it to this:
>
>
>    itemToAtomEntry: function (object){
>         return dojo.json.serialize(this.objectToRecordSet(object));
>    },
>
> then in Atom.js's saveItem and createitem, you'd set r.contentType to
> "application/eim+json", and voila, you're done.
>
>> Bug 8720, though I think where we left it was with me doing the CMP
>> work. Things got a little confused because of the data format switch,
>> so we should probably go back and rethink whether this one makes it
>> on the schedule :)
>
> ah, i'm not cc'd on that bug, which explains why i didn't know
> anything about it.
> _______________________________________________
> cosmo-dev mailing list
> cosmo-dev at lists.osafoundation.org
> http://lists.osafoundation.org/mailman/listinfo/cosmo-dev