lmi-commits
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[lmi-commits] [lmi] master afed971 3/3: Remove support for cygwin-native


From: Greg Chicares
Subject: [lmi-commits] [lmi] master afed971 3/3: Remove support for cygwin-native build [391]
Date: Sat, 22 Apr 2017 16:52:25 -0400 (EDT)

branch: master
commit afed971f4496cfb091d9fa5dd924de932e152f7e
Author: Gregory W. Chicares <address@hidden>
Commit: Gregory W. Chicares <address@hidden>

    Remove support for cygwin-native build [391]
    
    Removed outmoded workarounds for an ancient version of newlib, and
    directives that would have included a makefile that was expunged before
    the lmi epoch.
---
 config.hpp            |  6 ------
 numeric_io_traits.hpp | 13 -------------
 2 files changed, 19 deletions(-)

diff --git a/config.hpp b/config.hpp
index 2af2251..5027632 100644
--- a/config.hpp
+++ b/config.hpp
@@ -202,12 +202,6 @@ namespace fs = boost::filesystem;
 #       undef  OK_TO_INCLUDE_CONFIG_MING323_HPP
 #   endif // MinGW gcc 3.2.3+ .
 
-#   if defined __CYGWIN__ && defined __GNUC__ && 30203 <= LMI_GCC_VERSION
-#       define OK_TO_INCLUDE_CONFIG_CYG323_HPP
-#       include "config_cyg323.hpp"
-#       undef  OK_TO_INCLUDE_CONFIG_CYG323_HPP
-#   endif // Cygwin gcc 3.2.3+ .
-
 #   if defined LMI_COMO_WITH_MINGW
 #       define OK_TO_INCLUDE_CONFIG_COMO_WITH_MINGW_HPP
 #       include "config_como_mingw.hpp"
diff --git a/numeric_io_traits.hpp b/numeric_io_traits.hpp
index bb2509d..c47d2ae 100644
--- a/numeric_io_traits.hpp
+++ b/numeric_io_traits.hpp
@@ -80,21 +80,8 @@ inline int floating_point_decimals(T t)
         {
         return 0;
         }
-// TODO ?? As this is written on 2005-11-03, cygwin lacks fabsl().
-// It would be far better to write replacements for this and other
-// such functions in one unit-tested module, and use them here as
-// well as in 'round_to.hpp'.
-#if !defined __CYGWIN__
     long double z = std::numeric_limits<T>::epsilon() * fabsl(t);
     return std::max(0, static_cast<int>(-log10l(z)));
-#else  // defined __CYGWIN__
-    long double z = std::numeric_limits<T>::epsilon() * t;
-    if(t < 0.0)
-        {
-        z = -z;
-        }
-    return std::max(0, static_cast<int>(-log10(z)));
-#endif // defined __CYGWIN__
 }
 
 /// Simplify a formatted floating-point number.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]