Re: [boost] Trouble with statistic_tests.cpp and gcc-3.2/3.3

From: Gabriel Dos Reis (gdr_at_[hidden])
Date: 2002-10-31 08:30:19

Paolo Carlini <pcarlini_at_[hidden]> writes:

| Gabriel Dos Reis wrote:
| 
| >| According to the analysis of Nathan Sidwell and Gabriel Dos Reis this is 
| >| due to a missing workaround for the currently non conforming 
| >| libstdc++-v3 <limits> headers.
| >
| >The analysis concerns floating-point types.  Not integer types (which
| >are fine).  Or is there something in the above I'm missing?
| >
| I Gaby. First blush you right. However, this is the exact problem I'm 
| seeing and after Nathan's
| analysis (mentioning doubles and infinity, see audit trail) we all 
| agreed, that the problem
| has to do with numeric_limits<double>::infinity...

My point is that numeric_limits<> is correctly set for integer types;
so if you're seeing the same problem there, then it ought to be
something different.  

As for floating point type, GCC didn't provide fulll support until
3.3.  So, I'm having hard time to see where you're driving at.
Before relying on the value of numeric_limits<double>::infinity, you
ought to check for the value of numeric_limits<double>::has_infinity.

-- Gaby