Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [paho-dev] next release?

Thanks for the patch Darren. The original WS client was written on the assumption that a single frame would contain a single MQTT packet - this was, at the time, the proposed convention for how MQTT should flow over WS. So not a bug as such, but a design decision.

In the last couple weeks, the OASIS TC who are busy standardising MQTT agreed that such alignment should not be required - so your bug is now 100% valid and absolutely should be fixed :)

Cheers,
Nick


On 23 September 2013 20:47, Darren Clark <dclark@xxxxxxxxxxxxx> wrote:
FYI, I submitted a bug and related patch (#413579) for a pretty significant issue in the _javascript_ client that it would be nice to get included. Basically the problem is that on subscribing to a pattern that contains many retained messages a single WebSocket packet will contain multiple MQTT packets, so clients get corrupt data on initial connection.

-Darren


On Mon, Sep 23, 2013 at 12:36 PM, Roger Light <roger@xxxxxxxxxx> wrote:
Hi Ian,

That all sounds pretty reasonable. I'm mostly lacking on the examples.
I've not got hudson tests running on eclipse infrastructure (it needs
Python >2.6 and at the time we were looking at getting a vps for
testing iirc) but do myself.

I presume we need to consider IP review status as well - I've just had
an email update with some queries for the Python client today.

Cheers,

Roger


On Sat, Sep 21, 2013 at 11:31 PM, Ian Craggs
<icraggs@xxxxxxxxxxxxxxxxxxxxxxx> wrote:
> Hi Roger,
>
> this is what I imagine:
>
>     - some documentation in a format appropriate to the language
>     - some samples, examples of use
>     - notes describing any missing MQTT function (a significant proportion
> of MQTT function ought to be supported)
>     - ideally a Hudson job that can run the regression tests
>
> Ian
>
>
>
> On 21/09/13 00:57, Roger Light wrote:
>>
>> Hi,
>>
>> What bar are we aiming for for a release? I'm quite happy with the
>> state of the Python code, although I do have a few fixes to push. It's
>> out there and being used.
>>
>> Cheers,
>>
>> Roger
>>
>> On Fri, Sep 20, 2013 at 11:59 PM, Ian Craggs
>> <icraggs@xxxxxxxxxxxxxxxxxxxxxxx> wrote:
>>>
>>> Hi Benjamin,
>>>
>>> I raised the question of the plan at the last Paho call, but
>>> unfortunately
>>> we didn't realize that you would be asking for a date for next week.
>>>
>>> The release will feature Java, C and _javascript_ for sure.   C++ could be
>>> included depending on what Frank feels about that.
>>>
>>> I'm sorry I can't help with a date right now.
>>>
>>> Ian
>>>
>>>
>>> On 20/09/13 18:24, Benjamin Cabé wrote:
>>>
>>> Hi list,
>>>
>>> I am putting together a set of slides for presenting all the Eclipse M2M
>>> projects at a conference next week and I wanted to indicate the date for
>>> the
>>> next release of Paho ; unfortunately I cannot find it. Could someone
>>> please
>>> indicate to me what is the next release and when it is planned? Also, the
>>> release would be for C (and C++ wrapper?) and Java, right?
>>> I understand the code base for the C and Java client is quite mature
>>> though
>>> and I will certainly highlight it in the talk though!
>>>
>>> Thanks!
>>> Benjamin--
>>>
>>>
>>>
>>> _______________________________________________
>>> paho-dev mailing list
>>> paho-dev@xxxxxxxxxxx
>>> https://dev.eclipse.org/mailman/listinfo/paho-dev
>>>
>>>
>>>
>>> _______________________________________________
>>> paho-dev mailing list
>>> paho-dev@xxxxxxxxxxx
>>> https://dev.eclipse.org/mailman/listinfo/paho-dev
>>>
>> _______________________________________________
>> paho-dev mailing list
>> paho-dev@xxxxxxxxxxx
>> https://dev.eclipse.org/mailman/listinfo/paho-dev
>
>
_______________________________________________
paho-dev mailing list
paho-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/paho-dev


_______________________________________________
paho-dev mailing list
paho-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/paho-dev



Back to the top