Boost logo

Boost-Commit :

Subject: [Boost-commit] svn:boost r61203 - trunk/libs/config/test
From: nielsdekker_at_[hidden]
Date: 2010-04-11 11:27:32


Author: niels_dekker
Date: 2010-04-11 11:27:32 EDT (Sun, 11 Apr 2010)
New Revision: 61203
URL: http://svn.boost.org/trac/boost/changeset/61203

Log:
Added link to a new value-initialization compiler bug report: http://qc.embarcadero.com/wc/qcmain.aspx?d=83751
Text files modified:
   trunk/libs/config/test/boost_no_com_value_init.ipp | 11 ++++++++++-
   1 files changed, 10 insertions(+), 1 deletions(-)

Modified: trunk/libs/config/test/boost_no_com_value_init.ipp
==============================================================================
--- trunk/libs/config/test/boost_no_com_value_init.ipp (original)
+++ trunk/libs/config/test/boost_no_com_value_init.ipp 2010-04-11 11:27:32 EDT (Sun, 11 Apr 2010)
@@ -72,7 +72,7 @@
     return arg.data == 0;
   }
 
- // Equivalent to the struct TData from CodeGear bug report 51854
+ // Equivalent to the struct TData from CodeGear bug report 51854,
   // "Value-initialization: POD struct should be zero-initialized",
   // reported by me (Niels Dekker, LKEB) in 2007:
   // http://qc.embarcadero.com/wc/qcmain.aspx?d=51854
@@ -372,6 +372,15 @@
   public:
     value_initializer()
     :
+ // Note: CodeGear/Borland may produce a warning, W8039, for each data member
+ // whose type is an array type, saying "Constructor initializer list ignored".
+ // If it does, it probably won't value-initialize those arrays, as reported
+ // by me (Niels Dekker, LKEB) in 2010, report 83751, "Value-initialization:
+ // arrays should have each element value-initialized",
+ // http://qc.embarcadero.com/wc/qcmain.aspx?d=83751
+ // On the other hand, Microsoft Visual C++ may produce warnings of type C4351,
+ // saying "new behavior: elements of array '...' will be default initialized",
+ // which is actually the right behavior!
     int_struct(),
     m_enum_holder(),
     m_enum_holder_array(),


Boost-Commit list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk