[aerogear-dev] paging: different client platforms

[Summers Pittman]

On 01/11/2013 12:11 PM, Douglas Campos wrote:
> Yeah, mixing the context (state) with the actual data.

I like how simple and compact it makes things.  Also it lets UI controls 
be much simpler, and it doesn't make us rely on something like 
WeakHashMap to clean up old contexts and such.

IN either case, should the next() method block until Pipe.read()'s 
callback is called or should next consume a callback to be called after 
Pipe.read()'s callback is called?

> On 11/01/2013, at 15:06, Summers Pittman <supittma at redhat.com> wrote:
>
>> On 01/11/2013 11:55 AM, Douglas Campos wrote:
>>> What about returning a PagedList<T> (and using guava's ForwardingWhatever to decorate the list with the paging info)?
>>>
>> So something like PagedListExample?
>> https://gist.github.com/d835d4a56d4d25e4ca9a
>>> On 11/01/2013, at 14:14, Summers Pittman <supittma at redhat.com>
>>>   wrote:
>>>
>>>
>>>> On 01/11/2013 07:46 AM, Matthias Wessendorf wrote:
>>>>
>>>>> Of course, sending simple request parameters to the server is easy, in
>>>>> every library
>>>>>
>>>>> (using Dan's CURL client example)
>>>>>
>>>>> ```
>>>>> URL=
>>>>>
>>>>> "http://controllerdemo-danbev.rhcloud.com/aerogear-controller-demo"
>>>>>
>>>>>
>>>>> curl -i --header "Accept: application/json" "$URL/cars?page=0&perPage=4"
>>>>> ```
>>>>>
>>>>> In terms of an client API, that wants to scroll through the cars (like
>>>>> in this demo), I was wondering if some "PagingContext" (or
>>>>> "ResultSet") API is needed...
>>>>>
>>>>> // issues a request with ?page=0&perPage=4
>>>>> PagingContext pc = client.request(someCars...);
>>>>>
>>>>> // issues a request with ?page=0&perPage=4
>>>>> pc.next()
>>>>>
>>>>> // issues a request with ?page=1&perPage=4
>>>>> pc.next()
>>>>>
>>>>> // issues a request with ?page=0&perPage=4
>>>>> pc.back()
>>>>>
>>>>>
>>>>> NOT sure, but IMO a little higher level abstraction would be nice,
>>>>> instead of having the end users of the "paging API" applying the
>>>>> values for 'page' and 'perPage' manually.
>>>>>
>>>>> Any thoughts? (Not sure if this has been brought up before)
>>>>>
>>>>> -M
>>>>>
>>>>>
>>>> +1 for a higher level abstraction
>>>>
>>>> As far as how it will mesh with Pipe and such I am thinking
>>>>
>>>>
>>>> https://gist.github.com/4511860
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>> On Fri, Jan 11, 2013 at 12:39 PM, Matthias Wessendorf
>>>>>
>>>>> <matzew at apache.org>
>>>>>
>>>>>   wrote:
>>>>>
>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> was there already some discussion (and proposal) for paging APIs on the client ?
>>>>>>
>>>>>> It would be nice if the public client API semantics are somewhat
>>>>>> identical on the different platforms.
>>>>>> (I guess it's a MUST :))
>>>>>>
>>>>>> Any pointers to gists etc are welcome
>>>>>>
>>>>>> Thx
>>>>>>
>>>>>> --
>>>>>> Matthias Wessendorf
>>>>>>
>>>>>> blog:
>>>>>>
>>>>>> http://matthiaswessendorf.wordpress.com/
>>>>>>
>>>>>>
>>>>>> sessions:
>>>>>>
>>>>>> http://www.slideshare.net/mwessendorf
>>>>>>
>>>>>>
>>>>>> twitter:
>>>>>>
>>>>>> http://twitter.com/mwessendorf
>>>>>
>>>>>
>>>> _______________________________________________
>>>> aerogear-dev mailing list
>>>>
>>>> aerogear-dev at lists.jboss.org
>>>> https://lists.jboss.org/mailman/listinfo/aerogear-dev
>>> _______________________________________________
>>> aerogear-dev mailing list
>>>
>>> aerogear-dev at lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/aerogear-dev
>> _______________________________________________
>> aerogear-dev mailing list
>> aerogear-dev at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/aerogear-dev
>
> _______________________________________________
> aerogear-dev mailing list
> aerogear-dev at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/aerogear-dev