Re: [paho-dev] MQTT-SN and MQTT embedded C client updates

[Roger Light <roger@xxxxxxxxxx>]

Hi Ian,

It sounds like we're in agreement. void * for the external interface,
something else like char * internally.

It's quite possible I've just assumed you were talking about the
developer API when you were actually just talking about what you were
doing internally - if that's the case then I apologise for causing
confusion.

Cheers,

Roger



On Tue, Aug 5, 2014 at 5:19 PM, Ian Craggs
<icraggs@xxxxxxxxxxxxxxxxxxxxxxx> wrote:
> Hi Roger,
>
> there are two sorts of buffer.  That supplied by the developer like payload,
> and those which are meant to be written to, or read from the network
> (typically parameters to socket read/write for MQTT). Void is probably the
> right type for the first - I'm not so sure about the second.
>
> Ian
>
>
> On 08/05/2014 04:20 PM, Roger Light wrote:
>>
>> Hi Ian,
>>
>> My view is that at the point the user (developer) provides it, the
>> payload is anonymous bytes of a particular length. We can then treat
>> is at as an array of chars. Easy for us and doesn't force unnecessary
>> casting on the developer.
>>
>> Cheers,
>>
>> Roger
>>
>> On Mon, Aug 4, 2014 at 12:01 PM, Ian Craggs
>> <icraggs@xxxxxxxxxxxxxxxxxxxxxxx> wrote:
>>>
>>> Hi Roger,
>>>
>>> I used unsigned char* rather than void* for the payload/buffers as I
>>> thought
>>> of them as byte arrays or similar, and to aid pointer arithmetic.  I'll
>>> consider that one.
>>>
>>>
>>> Ian
>>>
>>> On 08/02/2014 08:05 PM, Roger Light wrote:
>>>>
>>>> Hi Ian,
>>>>
>>>>> Packet ids have been changed to "unsigned short", and data buffers to
>>>>> "unsigned char".  Now the data types are also consistent between the
>>>>> MQTT
>>>>> and MQTT-SN libraries, which was another motivation.
>>>>>
>>>>> I'm not expecting any further data type changes unless a good reason
>>>>> appears.  I started using size_t for buffer sizes, but have stuck with
>>>>> ints
>>>>> to reduce dependencies on any headers - just in case some esoteric/old
>>>>> embedded compiler didn't like size_t.
>>>>
>>>> I'm inclined to agree with Frank - do it "properly" (use size_t) and
>>>> then fix the odd case where size_t isn't defined if it crops up.
>>>>
>>>> I'm also of the opinion that "void *" is a better data type for the
>>>> data buffers - less hassle for the end developer - but that's probably
>>>> less important.
>>>>
>>>> Cheers,
>>>>
>>>> Roger
>>>> _______________________________________________
>>>> paho-dev mailing list
>>>> paho-dev@xxxxxxxxxxx
>>>> To change your delivery options, retrieve your password, or unsubscribe
>>>> from this list, visit
>>>> https://dev.eclipse.org/mailman/listinfo/paho-dev
>>>
>>>
>>> --
>>> Ian Craggs
>>> icraggs@xxxxxxxxxx                 IBM United Kingdom
>>> Committer on Paho, Mosquitto
>>>
>>> _______________________________________________
>>> paho-dev mailing list
>>> paho-dev@xxxxxxxxxxx
>>> To change your delivery options, retrieve your password, or unsubscribe
>>> from
>>> this list, visit
>>> https://dev.eclipse.org/mailman/listinfo/paho-dev
>>
>> _______________________________________________
>> paho-dev mailing list
>> paho-dev@xxxxxxxxxxx
>> To change your delivery options, retrieve your password, or unsubscribe
>> from this list, visit
>> https://dev.eclipse.org/mailman/listinfo/paho-dev
>
>
> --
> Ian Craggs
> icraggs@xxxxxxxxxx                 IBM United Kingdom
> Committer on Paho, Mosquitto
>
> _______________________________________________
> paho-dev mailing list
> paho-dev@xxxxxxxxxxx
> To change your delivery options, retrieve your password, or unsubscribe from
> this list, visit
> https://dev.eclipse.org/mailman/listinfo/paho-dev