[Swig-devel] [ swig-Bugs-2115237 ] const/non const typemaps

[SourceForge.net <no...@so...>]

Bugs item #2115237, was opened at 2008-09-17 01:32
Message generated for change (Comment added) made by wsfulton
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101645&aid=2115237&group_id=1645

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
>Status: Closed
>Resolution: Invalid
Priority: 5
Private: No
Submitted By: Mark Gossage (mgossage)
Assigned to: Nobody/Anonymous (nobody)
Summary: const/non const typemaps

Initial Comment:
Given the interface

%inline %{
struct T{};
%}

%typemap(in) const T& " //TODO const T& in"
%typemap(in,numinputs=0) T& " //TODO T& in"
%typemap(argout) T& " //TODO T& argout"

%inline %{
void fn1(const T&){}
void fn2(T&){}
%}

Checking the wrappered code for many languages (Lua,Python,Java). 
The fn2() wrapper has 
  //TODO T& in
  fn2(*arg1);
  //TODO T& argout
As expected, but the fn1() has
  //TODO const T& in
  fn1((T const &)*arg1);
  //TODO T& argout
Which means it matches the T& argout.
It can be fixed with a extra typemap
%typemap(argout) const T& ""

So the bug is: should SWIG be matching a non const typemap to a const value?
It doesn't happen the other way around (if you define the const T& typemap only)


----------------------------------------------------------------------

>Comment By: William Fulton (wsfulton)
Date: 2009-11-14 15:19

Message:
Hi Mark, This is expected behaviour and documented in the "Pattern matching
rules" in the Typemaps chapter of the docs.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101645&aid=2115237&group_id=1645