Re: [boost] RC_1_29_0 Win32 regression tests updated

From: David Abrahams (dave_at_[hidden])
Date: 2002-10-08 13:46:47

"Rozental, Gennadiy" <gennadiy.rozental_at_[hidden]> writes:

> 1. ublas tests named unclear and ublas is not a part of library name. Is it
> OK?
> 
> 2. Dave, what about your failures this morning?

Can't reproduce 'em. I'm not sure what the problem was, but when I
rebuilt everything, all was OK. Here was one backtrace

         MSVCP60D.DLL!std::basic_streambuf<char,std::char_traits<char> >::pptr()  Line 96 + 0xd	C++
         MSVCP60D.DLL!std::basic_stringbuf<char,std::char_traits<char>,std::allocator<char> >::str()  Line 34 + 0x3f	C++
         MSVCP60D.DLL!std::basic_ostringstream<char,std::char_traits<char>,std::allocator<char> >::str()  Line 239 + 0x1f	C++
         select_holder.exe!boost::test_toolbox::detail::wrapstrstream::str()  Line 51 + 0x14	C++
         select_holder.exe!boost::test_toolbox::detail::test_and_continue_impl(unsigned char predicate='
', const boost::test_toolbox::detail::wrapstrstream & message={...}, const char * file_name=0x00410540, int line_num=35, unsigned char add_fail_pass='
', boost::unit_test_framework::report_level loglevel=report_all_errors)  Line 131 + 0x4a	C++
         select_holder.exe!assert_same(boost::python::objects::value_holder<Base> * __formal=0x00000000, boost::python::objects::value_holder<Base> * __formal=0x00000000)  Line 35 + 0x4b	C++
         select_holder.exe!assert_holder(Base * __formal=0x00000000, Base * __formal=0x00000000, Base * __formal=0x00000000)  Line 43 + 0x1b	C++
         select_holder.exe!test_main(int __formal=1, int __formal=1)  Line 51 + 0xb	C++
         select_holder.exe!`anonymous namespace'::call_test_main()  Line 37 + 0x12	C++
         select_holder.exe!boost::unit_test_framework::function_test_case::do_run()  Line 135 + 0xd	C++
         select_holder.exe!boost::unit_test_framework::detail::unit_test_monitor::function()  Line 74	C++
         select_holder.exe!boost::detail::catch_signals(boost::execution_monitor & exmon={...}, int __formal=0)  Line 344	C++
         select_holder.exe!boost::execution_monitor::execute(int timeout=0)  Line 152 + 0xd	C++
         select_holder.exe!boost::unit_test_framework::detail::unit_test_monitor::execute_and_translate(int timeout=0)  Line 41	C++
         select_holder.exe!boost::unit_test_framework::test_case::run()  Line 69 + 0x97	C++
>	select_holder.exe!main(int argc=1, char * * argv=0x00324da8)  Line 77	C++
         select_holder.exe!mainCRTStartup()  Line 338 + 0x11	C
         kernel32.dll!77e7eb69() 	

BTW, are you installing some kind of SEH handler or using a catch(...)
block? I ask because when my test crashes, JIT debugging is inhibited,
which might not be the most useful behavior.

libs/python/test/module_tail.cpp shows a technique which will prevent
catch(...) from stopping structured exceptions in most cases.

-- 
           David Abrahams * Boost Consulting
dave_at_[hidden] * http://www.boost-consulting.com