I managed to trace a very nasty Python interpreter segfault to an
allocation failure here. Adding this after the tp_new call:
if (PyErr_Occurred()) {
PyErr_Print();
}
results in output of 'TypeError: object() takes no parameters', followed
by a segfault that takes down the Python interpeter.
The 'object' constructor doesn't seem to be suitable for instantiating
SWIG shadow instances in this way, so simply use the constructor
function in the PyTypeObject 'tp_new' slot of data->newargs.
The 'if (inst)' check after this doesn't hurt in as much as it prevented
a segfault immediately after this failed allocation, but it doesn't help
much since the null pointer dereference will probably happen sooner or
later anyway.
Advantages:
- it avoids the swig user having to jump through hoops to get print to
work as expected when redefining repr/str slots.
- typing the name of a variable on the python prompt now prints the
result of a (possibly redefined) repr, without the swig user having to
do any extra work.
- when redefining repr, the swig user doesn't necessarily have to
redefine str as it will call the redefined repr
- the behaviour is exactly the same as without the -builtin option while
requiring no extra work by the user (aside from adding the
%feature("python:slot...) statements of course)
- the patch simplifies pyrun.swg a tiny bit.
Disadvantage:
- default str() will give different (but clearer?) output on swigged
classes compared to unpatched swig
SF Bug #326
- Don't mark a method as METH_O if it has compactdefaultargs
- In SWIG_Python_UnpackTuple, allow for a non-tuple 'args'.
Added compatibility for python versions 2.3 and 2.4. These are only
partially supported; inheriting from wrapped types looks problematic.
Versions older that 2.3 are unlikely ever to work.
git-svn-id: https://swig.svn.sourceforge.net/svnroot/swig/branches/szager-python-builtin@12590 626c5289-ae23-0410-ae9c-e8d60b6d4f22
- Fixed naming conventions; SwigPyBuiltin is used a lot
- Removed use of std::vector
- builtin.swg isn't included if -builtin isn't specified
- Changed many feature names to use a "python:" prefix
- Eliminated static vars in std_pair.i
- Eliminated C++-style comments (//)
- Enabled autodoc and docstring with -builtin
- Fixed non-ansi generated C code
- Detect and complain if two incompatible swig modules are loaded
- Removed argcargvtest_runme3.py, and fixed argcargvtest_runme.py
so that 2to3 handles it better
- Removed anonymous namespaces
- Eliminated builtin_init typemaps; consolidated functionality into
SWIG_Python_NewPointerObj
- Eliminate printf warnings from %U conversion character by switching
to %S, which works just as well
- Fixed li_std_set_runme.py for python3, which returns set members in
a different order from python2
git-svn-id: https://swig.svn.sourceforge.net/svnroot/swig/branches/szager-python-builtin@12562 626c5289-ae23-0410-ae9c-e8d60b6d4f22
A partial fix for the problem with overloading, varargs,
and fastunpack: with this fix, the wrappers will compile.
However, they still contain faulty logic and unreachable
code in the dispatch function.
The comprehensive fix would have to be in overload.cxx.
git-svn-id: https://swig.svn.sourceforge.net/svnroot/swig/branches/szager-python-builtin@12465 626c5289-ae23-0410-ae9c-e8d60b6d4f22
- Throwing wrapped types as exceptions is unsupported.
- Reverse comparison operators (e.g., __radd__) aren't supported.
Rationalized destructors.
Finished std::map implementation. Required fixes to typecheck for
SWIGTYPE* const&.
Need a little special handling of the swig_type_info for SwigPyObject
when multiple modules are loaded.
Fall back to SwigPyObject_richcompare if there's no operator overload.
"memberget" and "memberset" attrs are applied strangely; work around
them.
Added 'this' attribute.
git-svn-id: https://swig.svn.sourceforge.net/svnroot/swig/branches/szager-python-builtin@12415 626c5289-ae23-0410-ae9c-e8d60b6d4f22
Refactored some type initialization out of SWIG_init.
Use __all__ attribute of module to define public interface.
This is necessary to make available symbols starting with '_'.
Now dying on li_boost_shared_ptr. Looks like it's gonna be ugly.
git-svn-id: https://swig.svn.sourceforge.net/svnroot/swig/branches/szager-python-builtin@12373 626c5289-ae23-0410-ae9c-e8d60b6d4f22
Apply operator features to both the operator name, and the renamed
"__*__" method. That's the only way to hit all corners.
Added support for %pythonnondynamic. I believe this implementation
is more correct than the existing implementation, but I'm still
waiting for an adjudication on the behavior of the python_nondynamic
test.
Current list of unsupported features that require minor tweaks
to the test suite:
- 'this' member variable is obsolete.
- No support for reversible operator overloads (e.g., __radd__). You
can still support this:
a = MyString("foo")
b = "bar"
c = a + b
... but you can't do this:
a = "foo"
b = MyString("bar")
c = a + b
With the tweaks, the test suite now fails on python_nondynamic.
git-svn-id: https://swig.svn.sourceforge.net/svnroot/swig/branches/szager-python-builtin@12353 626c5289-ae23-0410-ae9c-e8d60b6d4f22
li_std_pair_extra test fails, because the _runme.py uses the
secret 'this' member variable, which doesn't exist when the
-builtin option is used. Seems like a flaw in the test.
Test suite now fails on li_std_string_extra, because static
member variables are not fully implemented.
git-svn-id: https://swig.svn.sourceforge.net/svnroot/swig/branches/szager-python-builtin@12351 626c5289-ae23-0410-ae9c-e8d60b6d4f22
