[Ns-developers] Status of Python bindings

[Gustavo Carneiro]

On 19/05/2008, Gustavo Carneiro <gjcarneiro at gmail.com> wrote:
>
> The PyBindGen API changes are more or less complete:
>
> https://code.launchpad.net/~gjc/pybindgen/trunk<https://code.launchpad.net/%7Egjc/pybindgen/trunk>
>
> If nothing else comes up API-wise, next steps are:
>
>  1. Generate NS-3 API definitions as a single Python file and make it work;


The above step is now done.

By "make it work" it means the bindings generate and compile without error,
and do not require (py)gccxml except for scanning.  But I have scanned the
current API and added it to the repository, for the time being, so that
anyone who wants to try only needs to install pybindgen.

The bindings are partially useful.  Simulator.Schedule methods are working.
ns3.Object constructors are working as well.  Callback support is not being
compiled in yet, so socket examples will not work.  Generic Attribute
handling is not working yet.

Anyway, my focus now will be to develop the bindings framework, not to
improve API coverage.  It's been a lot of work just getting this far, so any
help is appreciated.

 2. Split the above file into several files, and support local modification
> files;
>
>  3. Figure out issues related to splitting python bindings into a separate
> repository;
>
>  4. Document the process, adding examples;
>
>  5. Bring the code up to date to ns-3-dev (assuming it is API stable by
> then).
>
>
> On 17/05/2008, Mathieu Lacage <mathieu.lacage at sophia.inria.fr> wrote:
>>
>> On Sat, 2008-05-17 at 12:06 +0100, Gustavo Carneiro wrote:
>>
>> >
>> >         Well, I think you got it right the first time.  I won't cry
>> >         about it too much ;-)
>> >
>> > This didn't come out right.  Sorry, I've been sick the last few days
>> > and not in a good mood :(
>> >
>> > If you want to convince Mathieu that (returntype, name, params) is
>> > better, I'll change it back.  If you are unsure, as I am, then I would
>> > rather leave it as it is.  I really really hate API changes, in case
>> > you haven't noticed... :P
>>
>>
>> If you and craig agree, feel free to ignore me.
>>
>>
>> >         > Regarding returntype mandatory, fine by me; I'll keep it.
>> >
>> >
>> >         I think that was the real issue under discussion.
>> >
>> > About this, I was thinking of simplifying it slightly.  Instead of
>> > ReturnValue.new('void'), None would also work equivalently.  Does it
>> > sound OK?
>>
>>
>> It does sound ok.
>>
>>
>> Mathieu
>>
>>
>
>
> --
> Gustavo J. A. M. Carneiro
> INESC Porto, Telecommunications and Multimedia Unit
> "The universe is always one step beyond logic." -- Frank Herbert
>



-- 
Gustavo J. A. M. Carneiro
INESC Porto, Telecommunications and Multimedia Unit
"The universe is always one step beyond logic." -- Frank Herbert