The binder should keep around things that are bound, so that could be a
bug, please file an issue with a testcase. Best thing would be to file a PR
that adds to existing webidl test with something that breaks.

On Fri, Feb 19, 2016 at 12:34 PM, Alan deLespinasse <[email protected]
> wrote:

> This seems to still be a problem. In fact, not only classes, but ordinary
> functions bound with EMSCRIPTEN_BINDINGS seem to disappear when using -O2
> or higher. Has anyone else had this problem?
>
> I can't find a filed bug for it. I'm happy to file one if it is in fact a
> bug.
>
> On Thursday, May 7, 2015 at 5:30:13 AM UTC-4, Björn K. wrote:
>>
>> Sorry for not providing accurate enough reproduction-steps. I've set up a
>> repository at github: https://github.com/cee-dee/emscripten-tests
>> Additionally I've created two issues which describe exactly what to do in
>> order to see the problems.
>>
>> The directory "embind-test-01" contains the problem which I mentioned
>> here, the directory "webidl-test-01" contains a reduced test where I
>> experimented with the webidl-binding, where I at least partially succeeded,
>> but still have the problem that --closure 1 won't produce usable results.
>>
>> Best regards,
>> Björn
>>
>>
>> Am Mittwoch, 6. Mai 2015 20:54:25 UTC+2 schrieb Alon Zakai:
>>>
>>> I can't reproduce the problem on that testcase, but maybe we are doing
>>> things a little differently (it doesn't build by itself). Can you please
>>> file a github issue with complete steps to reproduce?
>>>
>>>
>>> On Tue, May 5, 2015 at 5:44 AM, Björn K. <[email protected]>
>>> wrote:
>>>
>>>> Hi there!
>>>>
>>>> When using the sample c++ code from
>>>>
>>>> https://kripken.github.io/emscripten-site/docs/porting/connecting_cpp_and_javascript/embind.html#classes
>>>> it just works as long as I do not use the optimization -O2.
>>>>
>>>> When I compile with that option, my browser always tells me "TypeError:
>>>> Module.MyClass is not a constructor". In
>>>> https://kripken.github.io/emscripten-site/docs/porting/connecting_cpp_and_javascript/Interacting-with-code.html#calling-compiled-c-functions-from-javascript-using-ccall-cwrap
>>>> it sais "At higher optimisation levels (-O2 and above), the *closure
>>>> compiler*
>>>> <https://kripken.github.io/emscripten-site/docs/site/glossary.html#term-closure-compiler>
>>>> runs and minifies (changes) function names. Exporting functions allows you
>>>> to continue to access them using the original name through the global
>>>> Module object.", so I tried using the option -s EXPORTED_FUNCTIONS,
>>>> but it seems to work only for functions, not classes.
>>>>
>>>> What am I doing wrong? Would it work to export global functions which
>>>> serve as constructors for my classes? Sounds like a hack to me...
>>>>
>>>> Best regards,
>>>> Björn
>>>>
>>>>
>>> --
> You received this message because you are subscribed to the Google Groups
> "emscripten-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"emscripten-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to