diff --git a/ndnboost/numeric/conversion/detail/is_subranged.hpp b/ndnboost/numeric/conversion/detail/is_subranged.hpp
new file mode 100644
index 0000000..6f84872
--- /dev/null
+++ b/ndnboost/numeric/conversion/detail/is_subranged.hpp
@@ -0,0 +1,234 @@
+//  (c) Copyright Fernando Luis Cacciola Carballal 2000-2004
+//  Use, modification, and distribution is subject to the Boost Software
+//  License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+//  http://www.boost.org/LICENSE_1_0.txt)
+
+//  See library home page at http://www.boost.org/libs/numeric/conversion
+//
+// Contact the author at: fernando_cacciola@hotmail.com
+// 
+#ifndef BOOST_NUMERIC_CONVERSION_DETAIL_IS_SUBRANGED_FLC_12NOV2002_HPP
+#define BOOST_NUMERIC_CONVERSION_DETAIL_IS_SUBRANGED_FLC_12NOV2002_HPP
+
+#include "ndnboost/config.hpp"
+#include "ndnboost/limits.hpp"
+
+#include "ndnboost/mpl/int.hpp"
+#include "ndnboost/mpl/multiplies.hpp"
+#include "ndnboost/mpl/less.hpp"
+#include "ndnboost/mpl/equal_to.hpp"
+
+#include "ndnboost/type_traits/is_same.hpp"
+
+#include "ndnboost/numeric/conversion/detail/meta.hpp"
+#include "ndnboost/numeric/conversion/detail/int_float_mixture.hpp"
+#include "ndnboost/numeric/conversion/detail/sign_mixture.hpp"
+#include "ndnboost/numeric/conversion/detail/udt_builtin_mixture.hpp"
+
+namespace ndnboost { namespace numeric { namespace convdetail
+{
+  //---------------------------------------------------------------
+  // Implementations of the compile time predicate "T is subranged"
+  //---------------------------------------------------------------
+
+    // for integral to integral conversions
+    template<class T,class S>
+    struct subranged_Sig2Unsig
+    {
+      // Signed to unsigned conversions are 'subranged' because of possible loose
+      // of negative values.
+      typedef mpl::true_ type ;
+    } ;
+
+    // for unsigned integral to signed integral conversions
+    template<class T,class S>
+    struct subranged_Unsig2Sig
+    {
+       // IMPORTANT NOTE:
+       //
+       // This code assumes that signed/unsigned integral values are represented
+       // such that:
+       //
+       //  numeric_limits<signed T>::digits + 1 == numeric_limits<unsigned T>::digits
+       //
+       // The '+1' is required since numeric_limits<>::digits gives 1 bit less for signed integral types.
+       //
+       // This fact is used by the following logic:
+       //
+       //  if ( (numeric_limits<T>::digits+1) < (2*numeric_limits<S>::digits) )
+       //    then the conversion is subranged.
+       //
+
+       typedef mpl::int_< ::std::numeric_limits<S>::digits > S_digits ;
+       typedef mpl::int_< ::std::numeric_limits<T>::digits > T_digits ;
+
+       // T is signed, so take digits+1
+       typedef typename T_digits::next u_T_digits ;
+
+       typedef mpl::int_<2> Two ;
+
+       typedef typename mpl::multiplies<S_digits,Two>::type S_digits_times_2 ;
+
+       typedef typename mpl::less<u_T_digits,S_digits_times_2>::type type ;
+    } ;
+
+    // for integral to integral conversions of the same sign.
+    template<class T,class S>
+    struct subranged_SameSign
+    {
+       // An integral conversion of the same sign is subranged if digits(T) < digits(S).
+
+       typedef mpl::int_< ::std::numeric_limits<S>::digits > S_digits ;
+       typedef mpl::int_< ::std::numeric_limits<T>::digits > T_digits ;
+
+       typedef typename mpl::less<T_digits,S_digits>::type type ;
+    } ;
+
+    // for integral to float conversions
+    template<class T,class S>
+    struct subranged_Int2Float
+    {
+      typedef mpl::false_ type ;
+    } ;
+
+    // for float to integral conversions
+    template<class T,class S>
+    struct subranged_Float2Int
+    {
+      typedef mpl::true_ type ;
+    } ;
+
+    // for float to float conversions
+    template<class T,class S>
+    struct subranged_Float2Float
+    {
+      // If both T and S are floats,
+      // compare exponent bits and if they match, mantisa bits.
+
+      typedef mpl::int_< ::std::numeric_limits<S>::digits > S_mantisa ;
+      typedef mpl::int_< ::std::numeric_limits<T>::digits > T_mantisa ;
+
+      typedef mpl::int_< ::std::numeric_limits<S>::max_exponent > S_exponent ;
+      typedef mpl::int_< ::std::numeric_limits<T>::max_exponent > T_exponent ;
+
+      typedef typename mpl::less<T_exponent,S_exponent>::type T_smaller_exponent ;
+
+      typedef typename mpl::equal_to<T_exponent,S_exponent>::type equal_exponents ;
+
+      typedef mpl::less<T_mantisa,S_mantisa> T_smaller_mantisa ;
+
+      typedef mpl::eval_if<equal_exponents,T_smaller_mantisa,mpl::false_> not_bigger_exponent_case ;
+
+      typedef typename
+        mpl::eval_if<T_smaller_exponent,mpl::true_,not_bigger_exponent_case>::type
+          type ;
+    } ;
+
+    // for Udt to built-in conversions
+    template<class T,class S>
+    struct subranged_Udt2BuiltIn
+    {
+      typedef mpl::true_ type ;
+    } ;
+
+    // for built-in to Udt conversions
+    template<class T,class S>
+    struct subranged_BuiltIn2Udt
+    {
+      typedef mpl::false_ type ;
+    } ;
+
+    // for Udt to Udt conversions
+    template<class T,class S>
+    struct subranged_Udt2Udt
+    {
+      typedef mpl::false_ type ;
+    } ;
+
+  //-------------------------------------------------------------------
+  // Selectors for the implementations of the subranged predicate
+  //-------------------------------------------------------------------
+
+    template<class T,class S>
+    struct get_subranged_Int2Int
+    {
+      typedef subranged_SameSign<T,S>  Sig2Sig     ;
+      typedef subranged_Sig2Unsig<T,S> Sig2Unsig   ;
+      typedef subranged_Unsig2Sig<T,S> Unsig2Sig   ;
+      typedef Sig2Sig                  Unsig2Unsig ;
+
+      typedef typename get_sign_mixture<T,S>::type sign_mixture ;
+
+      typedef typename
+        for_sign_mixture<sign_mixture, Sig2Sig, Sig2Unsig, Unsig2Sig, Unsig2Unsig>::type
+           type ;
+    } ;
+
+    template<class T,class S>
+    struct get_subranged_BuiltIn2BuiltIn
+    {
+      typedef get_subranged_Int2Int<T,S> Int2IntQ ;
+
+      typedef subranged_Int2Float  <T,S> Int2Float   ;
+      typedef subranged_Float2Int  <T,S> Float2Int   ;
+      typedef subranged_Float2Float<T,S> Float2Float ;
+
+      typedef mpl::identity<Int2Float  > Int2FloatQ   ;
+      typedef mpl::identity<Float2Int  > Float2IntQ   ;
+      typedef mpl::identity<Float2Float> Float2FloatQ ;
+
+      typedef typename get_int_float_mixture<T,S>::type int_float_mixture ;
+
+      typedef for_int_float_mixture<int_float_mixture, Int2IntQ, Int2FloatQ, Float2IntQ, Float2FloatQ> for_ ;
+
+      typedef typename for_::type selected ;
+
+      typedef typename selected::type type ;
+    } ;
+
+    template<class T,class S>
+    struct get_subranged
+    {
+      typedef get_subranged_BuiltIn2BuiltIn<T,S> BuiltIn2BuiltInQ ;
+
+      typedef subranged_BuiltIn2Udt<T,S> BuiltIn2Udt ;
+      typedef subranged_Udt2BuiltIn<T,S> Udt2BuiltIn ;
+      typedef subranged_Udt2Udt<T,S>     Udt2Udt ;
+
+      typedef mpl::identity<BuiltIn2Udt> BuiltIn2UdtQ ;
+      typedef mpl::identity<Udt2BuiltIn> Udt2BuiltInQ ;
+      typedef mpl::identity<Udt2Udt    > Udt2UdtQ     ;
+
+      typedef typename get_udt_builtin_mixture<T,S>::type udt_builtin_mixture ;
+      
+      typedef typename
+        for_udt_builtin_mixture<udt_builtin_mixture, BuiltIn2BuiltInQ, BuiltIn2UdtQ, Udt2BuiltInQ, Udt2UdtQ>::type
+          selected ;
+
+      typedef typename selected::type selected2 ;
+ 
+      typedef typename selected2::type type ;
+    } ;
+
+
+  //-------------------------------------------------------------------
+  // Top level implementation selector.
+  //-------------------------------------------------------------------
+  template<class T, class S>
+  struct get_is_subranged
+  {
+    typedef get_subranged<T,S>         non_trivial_case ;
+    typedef mpl::identity<mpl::false_> trivial_case ;
+
+    typedef is_same<T,S> is_trivial ;
+   
+    typedef typename mpl::if_<is_trivial,trivial_case,non_trivial_case>::type selected ;
+    
+    typedef typename selected::type type ;
+  } ;
+
+} } } // namespace ndnboost::numeric::convdetail
+
+#endif
+
+