This option, if used, has not had any effect on Python 3 code since commit a863d3 9 years ago.
I think we can assume that it is not needed for Python 3.
Running the examples and test-suite (Python 2) doesn't change the code
paths with and without -safecstrings because only SWIG_OLDOBJ and SWIG_NEWOBJ
are used in the typemaps and the following code is thus unaltered by -safecstrings
(which sets SWIG_PYTHON_SAFE_CSTRINGS):
%#if defined(SWIG_PYTHON_SAFE_CSTRINGS)
if (*alloc != SWIG_OLDOBJ)
%#else
if (*alloc == SWIG_NEWOBJ)
%#endif
{
*cptr = %new_copy_array(cstr, len + 1, char);
*alloc = SWIG_NEWOBJ;
printf("safe strings: %s\n", *cptr ? *cptr : "NULLSTRING");
} else {
*cptr = cstr;
*alloc = SWIG_OLDOBJ;
}
Note: nosafecstrings was also the default and -O didn't actually change this.
A custom implementation for Py_None was implemented in SWIG_Py_None().
This was used by default on Windows only. It isn't clear why this
was done just for Windows. Now Py_None is the real Py_None on all
operating systems.
This option runs the equivalent init code in Python C code instead of pure Python code.
The init code adds the C++ pointer into the 'this' variable.
If the variable already exists, it appends it. This
seems to only happen in directors and multiple inheritance, see
PyMulti in director_basic_runme.py.
We only still default to generate a no-op __del__ method for theoretical
compatibility with old code, and 4.0.0 is a good time to make a cut-off
should such code really still exist.
On the flip-side, the presence of a __del__ method seems to be able to
prevent garbage collection from kicking in (see #1215), so it's
definitely desirable to get rid of it when there's nothing for it to do.
Conflicts:
Source/Modules/python.cxx
This is a cherry-pick and merge from the patch in #1261
What SWIG calls "modern" classes are supported by Python 2.3 and up
which means they're supported by all the Python versions we aim to
support in 4.0.0.
Conflicts:
Source/Modules/python.cxx
This is a cherry-pick and merge from the patch in #1261
There were only needed to support Python < 2.2, and we now require at
least Python 2.6.
Conflicts:
.travis.yml
Examples/test-suite/python/autodoc_runme.py
Source/Modules/python.cxx
This is a cherry-pick and merge from patch in #1261
Activated if a user sets SWIG_JAVA_USE_THREAD_NAME.
Implementations provided for linux/android/macos/unix.
A user's implementation will be used if SWIG_JAVA_GET_THREAD_NAME is
defined. It must implement the function:
namespace Swig {
SWIGINTERN int GetThreadName(char *name, size_t len);
}
Otherwise the native thread name gets clobbered with "Thread-XX", which
makes debugging hard. This issue seems to be specific to the Android JVM.
Someone proposed a similar fix to SWIG:
https://github.com/swig/swig/pull/831
but abandoned it.
Since we are mixing `new` and `override` a class instance may have multiple methods with the same name. in case of director_basic test one method comes from `MyClassMiddle` (one that participates in dynamic dispatch) and one comes from `MyClassEnd` (one with `new` keyword). Therefore all methods have to be checked.
C# does not treat `virtual` methods of child classes as overriding (ass opposed to c++). In order to override a method it must have `override` specified.
Previous version of this method treated `virtual void foo()` or `void foo()` in a subclass as methods that override virtual method of parent class. This resulted in `SwigDirectorConnect()` creating delegates and connecting them to a native class. Presence of such methods caused needless roundtrip through managed layer while in the end correct native function was called.
This is the old code flow:
```cpp
void SwigDirector_MyClass::nonOverride() {
if (!swig_callbacknonOverride) { // 0. swig_callbacknonOverride is set
MyClass::nonOverride();
return;
} else {
swig_callbacknonOverride(); // 1. this is called because wrapper mistakenly assumes overriding
}
}
SWIGEXPORT void SWIGSTDCALL CSharp_director_basicNamespace_MyClass_nonOverrideSwigExplicitMyClass(void * jarg1) {
MyClass *arg1 = (MyClass *) 0 ;
arg1 = (MyClass *)jarg1;
(arg1)->MyClass::nonOverride(); // 5. Correct method called in the end
}
```
```cs
private void SwigDirectornonOverride() {
nonOverride(); // 2.
}
public virtual void nonOverride() {
if (SwigDerivedClassHasMethod("nonOverride", swigMethodTypes4)) // 3. This returns `true`
director_basicPINVOKE.MyClass_nonOverrideSwigExplicitMyClass(swigCPtr); // 4. Native method of director class called explicitly
else
director_basicPINVOKE.MyClass_nonOverride(swigCPtr);
}
```
This is the new code flow:
```cpp
void SwigDirector_MyClass::nonOverride() {
if (!swig_callbacknonOverride) { // 0. swig_callbacknonOverride is not set
MyClass::nonOverride(); // 1. Calls correct method immediately
return;
} else {
swig_callbacknonOverride();
}
}
```
Add runtime test for scenario described in #1323.
Expand testing to cover a mix of override/virtual method overloading.
Remove testSwigDerivedClassHasMethod() test - not fixed yet.
Fix when using -builtin and wrapping std::map, std::set, std::unordered_map or
std::unordered_set to ensure __contains__ is called. This is a wrapper for the STL
container's find method. Without it, Python will do its own slower sequence search.
Found via `codespell -q 3 -L "uint,bae,objext,cmo,goin,struc,ois,upto"`
whitespaces were unintentionally fixed due to my editors settings.
Rebased patch #1327
