Bug#524064: Regression: compile failure on valid C++ code

2009-04-14 Thread Bastian Blank
On Tue, Apr 14, 2009 at 09:19:49PM +0100, Alastair McKinstry wrote: > Ok. I'm not a C++ programmer, and would appreciate an explanation. The > standalone code is: I would try using "#include ". Bastian -- Lots of people drink from the wrong bottle sometimes. -- Edith Keeler, "

Bug#524064: Regression: compile failure on valid C++ code

2009-04-14 Thread Alastair McKinstry
Hi, Ok. I'm not a C++ programmer, and would appreciate an explanation. The standalone code is: #include #include #include using namespace std; class DateTime { public: DateTime(); virtual ~DateTime() {} operator tm() const; protected: long i; }; class Eps

Bug#524064: Regression: compile failure on valid C++ code

2009-04-14 Thread Bastian Blank
tags 524064 moreinfo thanks On Tue, Apr 14, 2009 at 05:17:59PM +0100, Alastair McKinstry wrote: > Relevant code: > 196 ostringstream out; > 197 tm convert = base_; > 198 locale loc(""); > 199 out.imbue(loc); > 200 const std::time_

Bug#524064: Regression: compile failure on valid C++ code

2009-04-14 Thread Alastair McKinstry
Package: g++-4.3 Version: 4.3.3-7 Severity: normal Compiling magics++, a new package for debian, I get: /bin/sh ../../libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I. -I. -I../../src-I../../src/MvObs -I../../src/terralib/kernel -I../../src -I../../src/basic -I../../src/common -I../..