WIN-25.1: natPlainSocketImpl.cc [UPDATED]

[Adam Megacz]

Bryce McKinlay <bryce@waitaki.otago.ac.nz> writes:
> >-  int r = _Jv_recv (::recv, fnum, (void *) bytes, count, 0);
> >+  int r = _Jv_recv ((int (*)(unsigned int, char*, int, int))(::recv),
> >+		    fnum, (void *) bytes, count, 0);
> >   if (r == 0)
> >     return -1;
> >   if (java::lang::Thread::interrupted())

> I don't like the look of this bit ;-(. The _Jv_recv thing seems to break 
> one platform or another every time someone changes it. Then again I 
> always thought it was a bit of a hack, perhaps it would be better to 
> just do some configure tests after all. 

I don't like it either. Why do we need it? I was just trying to go
with the flow.


> Why did you need to add the casting?

../../../gcc/libjava/java/net/natPlainSocketImpl.cc:582: invalid conversion 
   from `int (*)(unsigned int, char*, int, int)' to `int (*)(unsigned int, char*, int, int)'

"Sorry, sir, we can't accept your US Dollars. Please come back and pay
 us with US Dollars"

  - a

-- 
"If I put copyrighted material into the 'chewy nougat center', I can
 claim the crunchy chocolate coating is an 'Access Control
 Mechanism'."                                     --lynx_user_abroad