Re: std::map example
William S Fulton <[email protected]> Wed, 31 Mar 2021 23:05:27 +0100
| Newsgroups | gmane.comp.programming.swig |
|---|---|
| Message-ID | <CANGqftCa8M2ubd8avS4uQOPC2D7zWxivHQFxg-k7ehtT7HAxRA@mail.gmail.com> |
Hi Ayman The CHANGES file has some useful info for dealing with this change. It's provided with the release as well as at http://swig.org/Release/README. Look for the entry dated 2019-02-14. William On Sat, 27 Mar 2021 at 17:18, Ayman Habib <[email protected]> wrote: > Hi all, > > I'm trying to upgrade the language bindings for our library (Java and > Python) from swig version 3.0.9 to 4.0.2 but I'm running into problem that > the generated Java code doesn't compile for a C++ class that is a subclass > of std::map<std::string, shared_ptr<SomeComplicatedClass>>. > > The generated Java class has the correct class hierarchy ( now a > subclass of java.util.AbstractMap<String, SomeComplicatedClass >) but the > generated Java class fails to compile because the methods internally refer > to the Iterator methods that have wrong signatures. Is there an example of > how to generate bindings for a subclass of std::map in Java and Python? > > I could take over the offending methods and replace them with ones that > compile and work by modifying the interface file(s) but I'd prefer to use > swig's native mechanisms if possible. > > Thank you for your help, > -Ayman > > _______________________________________________ > Swig-user mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/swig-user > _______________________________________________ Swig-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/swig-user