Re: [Swig-user] Typemap for binary data

[William S F. <ws...@fu...>]

There is no magic, this is a multi-argument typemap. Just search the 
typemap library for LENGTH to find them and read about multi-argument 
typemaps - 
http://www.swig.org/Doc2.0/Typemaps.html#Typemaps_multi_argument_typemaps_patterns

William

On 17/07/13 03:16, Alexander Solovets wrote:
> Thanks Bob! I did not think about adding typemap separately for the
> second argument. Anyway I am still interested in understanding how
> exactly that 'const char *STRING, size_t LENGTH' works.
>
>
> On Wed, Jul 17, 2013 at 12:35 PM, Bob Hood <bh...@co...
> <mailto:bh...@co...>> wrote:
>
>     On 7/16/2013 6:19 PM, Alexander Solovets wrote:
>      > Greetings all!
>      >
>      > I am trying to wrap
>      >
>      > foo(const int *data, size_t len)
>      >
>      > in Python interface. I discovered that the following typemap can
>     be used
>      >
>      > %typemap(in) (const void *data, size_t len) = (const char
>     *STRING, size_t
>      > LENGTH)
>      >
>      > This works fine, but I want to impose additional check that "len" is
>      > divisible by sizeof(int). In order to do that I tried to find
>     definition of
>      > (const char *STRING, size_t LENGTH), but failed to do that. Is
>     this "magic"
>      > typemapping done internally in swig? I'm just trying to
>     understand how can I
>      > combine several parameters in one.
>
>     I may be using this incorrectly, but might a typecheck address this
>     for you?
>     Something like:
>
>          %typemap(check) size_t len {
>              if (($1 % sizeof(int)) != 0) {
>                  SWIG_exception(SWIG_ValueError,"Expected value not
>     divisible by
>     sizeof(int).");
>              }
>          }
>
>     probably with some casting, since "size_t" and "int" may not be the same
>     sizes, depending upon how you are building.
>
>     You might also want to give the variable a name something more
>     unique than
>     "len" just to make sure the typecheck doesn't apply to all "size_t len"
>     patterns, just those where you want to validate the division.
>
>     ------------------------------------------------------------------------------
>     See everything from the browser to the database with AppDynamics
>     Get end-to-end visibility with application monitoring from AppDynamics
>     Isolate bottlenecks and diagnose root cause in seconds.
>     Start your free trial of AppDynamics Pro today!
>     http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk
>     _______________________________________________
>     Swig-user mailing list
>     Swi...@li... <mailto:Swi...@li...>
>     https://lists.sourceforge.net/lists/listinfo/swig-user
>
>
>
>
> --
> С уважением, Александр
>
>
> ------------------------------------------------------------------------------
> See everything from the browser to the database with AppDynamics
> Get end-to-end visibility with application monitoring from AppDynamics
> Isolate bottlenecks and diagnose root cause in seconds.
> Start your free trial of AppDynamics Pro today!
> http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk
>
>
>
> _______________________________________________
> Swig-user mailing list
> Swi...@li...
> https://lists.sourceforge.net/lists/listinfo/swig-user
>