|
|
Boost-Commit : |
Subject: [Boost-commit] svn:boost r78069 - sandbox/fixed_point/boost/fixed_point
From: vicente.botet_at_[hidden]
Date: 2012-04-18 14:06:25
Author: viboes
Date: 2012-04-18 14:06:24 EDT (Wed, 18 Apr 2012)
New Revision: 78069
URL: http://svn.boost.org/trac/boost/changeset/78069
Log:
FixedPoint: Added implicit/explicit conversion from builtins depending on the family
Text files modified:
sandbox/fixed_point/boost/fixed_point/number.hpp | 249 +++++++++++++++++++++++++++------------
1 files changed, 173 insertions(+), 76 deletions(-)
Modified: sandbox/fixed_point/boost/fixed_point/number.hpp
==============================================================================
--- sandbox/fixed_point/boost/fixed_point/number.hpp (original)
+++ sandbox/fixed_point/boost/fixed_point/number.hpp 2012-04-18 14:06:24 EDT (Wed, 18 Apr 2012)
@@ -24,6 +24,8 @@
#include <boost/mpl/assert.hpp>
#include <boost/type_traits/is_same.hpp>
#include <boost/type_traits/is_signed.hpp>
+#include <boost/type_traits/is_arithmetic.hpp>
+#include <boost/type_traits/is_integral.hpp>
#include <boost/utility/enable_if.hpp>
#include <boost/assert.hpp>
#include <boost/static_assert.hpp>
@@ -548,19 +550,23 @@
}
template <typename F>
- struct allows_explicit_conversion_from_fp: public is_same<typename F::conversion_from_fixed_point_type, conversion::explicitly>
+ struct allows_explicit_conversion_from_fp: public is_same<typename F::conversion_from_fixed_point_type,
+ conversion::explicitly>
{
};
template <typename F>
- struct allows_explicit_conversion_from_builtin: public is_same<typename F::conversion_from_builtin_type, conversion::explicitly>
+ struct allows_explicit_conversion_from_builtin: public is_same<typename F::conversion_from_builtin_type,
+ conversion::explicitly>
{
};
template <typename F>
- struct allows_implicit_conversion_from_fp: public is_same<typename F::conversion_from_fixed_point_type, conversion::implicitly>
+ struct allows_implicit_conversion_from_fp: public is_same<typename F::conversion_from_fixed_point_type,
+ conversion::implicitly>
{
};
template <typename F>
- struct allows_implicit_conversion_from_builtin: public is_same<typename F::conversion_from_builtin_type, conversion::implicitly>
+ struct allows_implicit_conversion_from_builtin: public is_same<typename F::conversion_from_builtin_type,
+ conversion::implicitly>
{
};
template <typename F>
@@ -770,13 +776,13 @@
typename To,
bool LE_Range= From::range_exp <= To::range_exp,
bool GE_Resolution= From::resolution_exp >= To::resolution_exp
- > struct number_cast;
+ > struct fxp_number_cast;
// LE_Range=true GE_Resolution=true
///////////////////////////////////
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
real_t<R1,P1,RP1,OP1,F1>,
real_t<R2,P2,RP2,OP2,F2>,
true, true >
@@ -793,7 +799,7 @@
};
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
ureal_t<R1,P1,RP1,OP1,F1>,
real_t<R2,P2,RP2,OP2,F2>,
true, true >
@@ -810,7 +816,7 @@
};
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
ureal_t<R1,P1,RP1,OP1,F1>,
ureal_t<R2,P2,RP2,OP2,F2>,
true, true >
@@ -828,7 +834,7 @@
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
real_t<R1,P1,RP1,OP1,F1>,
ureal_t<R2,P2,RP2,OP2,F2>,
true, true >
@@ -860,7 +866,7 @@
////////////////////////////////////
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
real_t<R1,P1,RP1,OP1,F1>,
real_t<R2,P2,RP2,OP2,F2>,
false, true >
@@ -894,7 +900,7 @@
};
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
real_t<R1,P1,RP1,OP1,F1>,
ureal_t<R2,P2,RP2,OP2,F2>,
false, true >
@@ -928,7 +934,7 @@
};
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
ureal_t<R1,P1,RP1,OP1,F1>,
ureal_t<R2,P2,RP2,OP2,F2>,
false, true >
@@ -957,7 +963,7 @@
};
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
ureal_t<R1,P1,RP1,OP1,F1>,
real_t<R2,P2,RP2,OP2,F2>,
false, true >
@@ -990,7 +996,7 @@
////////////////////////////////////
template <int R, int P1, typename RP1, typename OP1, typename F1,
int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
real_t<R,P1,RP1,OP1,F1>,
real_t<R,P2,RP2,OP2,F2>,
true, false >
@@ -1025,7 +1031,7 @@
};
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
real_t<R1,P1,RP1,OP1,F1>,
real_t<R2,P2,RP2,OP2,F2>,
true, false >
@@ -1046,7 +1052,7 @@
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
real_t<R1,P1,RP1,OP1,F1>,
ureal_t<R2,P2,RP2,OP2,F2>,
true, false >
@@ -1082,7 +1088,7 @@
////
template <int R, int P1, typename RP1, typename OP1, typename F1,
int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
ureal_t<R,P1,RP1,OP1,F1>,
real_t<R,P2,RP2,OP2,F2>,
true, false >
@@ -1111,7 +1117,7 @@
};
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
ureal_t<R1,P1,RP1,OP1,F1>,
real_t<R2,P2,RP2,OP2,F2>,
true, false >
@@ -1133,7 +1139,7 @@
template <int R, int P1, typename RP1, typename OP1, typename F1,
int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
ureal_t<R,P1,RP1,OP1,F1>,
ureal_t<R,P2,RP2,OP2,F2>,
true, false >
@@ -1162,7 +1168,7 @@
};
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
ureal_t<R1,P1,RP1,OP1,F1>,
ureal_t<R2,P2,RP2,OP2,F2>,
true, false >
@@ -1184,7 +1190,7 @@
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
real_t<R1,P1,RP1,OP1,F1>,
real_t<R2,P2,RP2,OP2,F2>,
false, false >
@@ -1219,7 +1225,7 @@
};
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
ureal_t<R1,P1,RP1,OP1,F1>,
real_t<R2,P2,RP2,OP2,F2>,
false, false >
@@ -1249,7 +1255,7 @@
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
real_t<R1,P1,RP1,OP1,F1>,
ureal_t<R2,P2,RP2,OP2,F2>,
false, false >
@@ -1286,7 +1292,7 @@
template <int R1, int P1, typename RP1, typename OP1, typename F1,
int R2, int P2, typename RP2, typename OP2, typename F2>
- struct number_cast<
+ struct fxp_number_cast<
ureal_t<R1,P1,RP1,OP1,F1>,
ureal_t<R2,P2,RP2,OP2,F2>,
false, false >
@@ -1319,6 +1325,100 @@
}
};
+
+ // arthm_number_cast
+ /////////////////////////////////////
+
+ template <
+ typename From,
+ typename To,
+ bool IsIntegral = is_integral<From>::value
+ >
+ struct arthm_number_cast;
+
+ template <
+ typename From,
+ int R2, int P2, typename RP2, typename OP2, typename F2
+ >
+ struct arthm_number_cast<From,real_t<R2,P2,RP2,OP2,F2>,true>
+ {
+ typedef real_t<R2,P2,RP2,OP2,F2> To;
+ typedef typename To::underlying_type underlying_type;
+
+ BOOST_CONSTEXPR To operator()(const From& rhs) const
+ {
+ return
+ To(index(detail::shift_left<-To::resolution_exp>(rhs)));
+ }
+ };
+ template <
+ typename From,
+ int R2, int P2, typename RP2, typename OP2, typename F2
+ >
+ struct arthm_number_cast<From,ureal_t<R2,P2,RP2,OP2,F2>,true>
+ {
+ typedef ureal_t<R2,P2,RP2,OP2,F2> To;
+ typedef typename To::underlying_type underlying_type;
+
+ BOOST_CONSTEXPR To operator()(const From& rhs) const
+ {
+ return
+ To(index(detail::shift_left<-To::resolution_exp>(rhs)));
+ }
+ };
+ template <
+ typename From,
+ int R2, int P2, typename RP2, typename OP2, typename F2
+ >
+ struct arthm_number_cast<From,real_t<R2,P2,RP2,OP2,F2>,false>
+ {
+
+ typedef real_t<R2,P2,RP2,OP2,F2> To;
+ typedef typename To::underlying_type underlying_type;
+
+ BOOST_CONSTEXPR To operator()(const From& rhs) const
+ {
+ return
+ To(index(To::classify(rhs)));
+ }
+
+ };
+ template <
+ typename From,
+ int R2, int P2, typename RP2, typename OP2, typename F2
+ >
+ struct arthm_number_cast<From,ureal_t<R2,P2,RP2,OP2,F2>,false>
+ {
+
+ typedef ureal_t<R2,P2,RP2,OP2,F2> To;
+ typedef typename To::underlying_type underlying_type;
+
+ BOOST_CONSTEXPR To operator()(const From& rhs) const
+ {
+ return
+ To(index(To::classify(rhs)));
+ }
+
+ };
+
+ template <
+ typename From,
+ typename To,
+ bool IsArithmetic= is_arithmetic<From>::value
+ > struct number_cast;
+
+ template <
+ typename From,
+ typename To
+ > struct number_cast<From,To,true>: public arthm_number_cast<From,To>
+ {};
+
+ template <
+ typename From,
+ typename To
+ > struct number_cast<From,To,false>: public fxp_number_cast<From,To>
+ {};
+
template <typename T, typename U >
struct default_type_impl;
template <typename T>
@@ -1526,13 +1626,20 @@
>
{};
- template <typename S, typename T>
- struct is_implicitly_convertible :
+ template <typename S, typename T, bool IsArithmetic=is_arithmetic<S>::value >
+ struct is_implicitly_convertible;
+ template <typename S, typename T >
+ struct is_implicitly_convertible<S,T,false> :
mpl::and_<
allows_implicit_conversion_from_fp<T>,
mpl::not_<is_more_precisse<T,S> >
>
{};
+
+ template <typename S, typename T>
+ struct is_implicitly_convertible<S,T,true> :
+ allows_implicit_conversion_from_builtin<T>
+ {};
#endif
/**
@@ -1612,7 +1719,7 @@
real_t(real_t<R,P,RP,OP,F> const& rhs
, typename boost::enable_if<is_convertible<real_t<R,P,RP,OP,F>, real_t > >::type* = 0
)
- : value_(fixed_point::detail::number_cast<real_t<R,P,RP,OP,F>, real_t, true, true>()(rhs).count())
+ : value_(fixed_point::detail::fxp_number_cast<real_t<R,P,RP,OP,F>, real_t, true, true>()(rhs).count())
{
}
//! Implicit constructor from a ureal_t with no larger range and no better resolution
@@ -1620,7 +1727,7 @@
real_t(ureal_t<R,P,RP,OP,F> const& rhs
, typename boost::enable_if<is_convertible<ureal_t<R,P,RP,OP,F>, real_t > >::type* = 0
)
- : value_(fixed_point::detail::number_cast<ureal_t<R,P,RP,OP,F>, real_t, true, true>()(rhs).count())
+ : value_(fixed_point::detail::fxp_number_cast<ureal_t<R,P,RP,OP,F>, real_t, true, true>()(rhs).count())
{
}
@@ -1629,7 +1736,7 @@
explicit real_t(real_t<R,P,RP,OP,F> const& rhs
, typename boost::enable_if <is_explicitly_convertible<real_t<R,P,RP,OP,F>,real_t> >::type* = 0
)
- : value_(fixed_point::detail::number_cast<real_t<R,P,RP,OP,F>, real_t,
+ : value_(fixed_point::detail::fxp_number_cast<real_t<R,P,RP,OP,F>, real_t,
mpl::less_equal < mpl::int_<R>, mpl::int_<Range> >::value,
mpl::greater_equal < mpl::int_<P>, mpl::int_<Resolution> >::value
>()(rhs).count())
@@ -1640,7 +1747,7 @@
real_t(real_t<R,P,RP,OP,F> const& rhs
, typename boost::enable_if <is_implicitly_convertible<real_t<R,P,RP,OP,F>,real_t> >::type* = 0
)
- : value_(fixed_point::detail::number_cast<real_t<R,P,RP,OP,F>, real_t,
+ : value_(fixed_point::detail::fxp_number_cast<real_t<R,P,RP,OP,F>, real_t,
mpl::less_equal < mpl::int_<R>, mpl::int_<Range> >::value,
mpl::greater_equal < mpl::int_<P>, mpl::int_<Resolution> >::value
>()(rhs).count())
@@ -1651,7 +1758,7 @@
explicit real_t(ureal_t<R,P,RP,OP,F> const& rhs
, typename boost::enable_if <is_explicitly_convertible<ureal_t<R,P,RP,OP,F>,real_t> >::type* = 0
)
- : value_(fixed_point::detail::number_cast<ureal_t<R,P,RP,OP,F>, real_t,
+ : value_(fixed_point::detail::fxp_number_cast<ureal_t<R,P,RP,OP,F>, real_t,
mpl::less_equal < mpl::int_<R>, mpl::int_<Range> >::value,
mpl::greater_equal < mpl::int_<P>, mpl::int_<Resolution> >::value
>()(rhs).count())
@@ -1663,7 +1770,7 @@
real_t(ureal_t<R,P,RP,OP,F> const& rhs
, typename boost::enable_if <is_implicitly_convertible<ureal_t<R,P,RP,OP,F>,real_t> >::type* = 0
)
- : value_(fixed_point::detail::number_cast<ureal_t<R,P,RP,OP,F>, real_t,
+ : value_(fixed_point::detail::fxp_number_cast<ureal_t<R,P,RP,OP,F>, real_t,
mpl::less_equal < mpl::int_<R>, mpl::int_<Range> >::value,
mpl::greater_equal < mpl::int_<P>, mpl::int_<Resolution> >::value
>()(rhs).count())
@@ -1673,7 +1780,7 @@
//! implicit constructor from a real_t with larger range or better resolution when wrapped by a convert_tag.
template <int R, int P, typename RP, typename OP, typename F>
real_t(convert_tag<real_t<R,P,RP,OP,F> > rhs)
- : value_(fixed_point::detail::number_cast<real_t<R,P,RP,OP,F>, real_t,
+ : value_(fixed_point::detail::fxp_number_cast<real_t<R,P,RP,OP,F>, real_t,
mpl::less_equal < mpl::int_<R>, mpl::int_<Range> >::value,
mpl::greater_equal < mpl::int_<P>, mpl::int_<Resolution> >::value
>()(rhs.get()).count())
@@ -1682,7 +1789,7 @@
//! implicit constructor from a ureal_t with larger range or better resolution when wrapped by a convert_tag.
template <int R, int P, typename RP, typename OP, typename F>
real_t(convert_tag<ureal_t<R,P,RP,OP,F> > rhs)
- : value_(fixed_point::detail::number_cast<ureal_t<R,P,RP,OP,F>, real_t,
+ : value_(fixed_point::detail::fxp_number_cast<ureal_t<R,P,RP,OP,F>, real_t,
mpl::less_equal < mpl::int_<R>, mpl::int_<Range> >::value,
mpl::greater_equal < mpl::int_<P>, mpl::int_<Resolution> >::value
>()(rhs.get()).count())
@@ -1755,8 +1862,6 @@
template <typename FP>
static FP factor()
{
- // if (Resolution>=0) return FP(1 << Resolution);
- // else return FP(1)/(1 << -Resolution);
if (Resolution>=0) return FP(detail::shift_left<Resolution>(1));
else return FP(1)/(detail::shift_left<-Resolution>(1));
@@ -1852,23 +1957,16 @@
}
//! implicit conversion from int
-// real_t(int x
-// typename boost::enable_if<allows_implicit_conversion_from_builtin<real_t> >::type* = 0
-// ) : value_(detail::shift_left<-Resolution>(x))
-// {}
- //! explicit conversion from int
- explicit real_t(int x
- //, typename boost::disable_if<allows_implicit_conversion_from_builtin<real_t> >::type* = 0
- ) : value_(detail::shift_left<-Resolution>(x))
- {}
- //! implicit conversion from float
- explicit real_t(float x) : value_(classify(x))
- {}
- //! implicit conversion from double
- explicit real_t(double x) : value_(classify(x))
+ template <typename T>
+ real_t(T x,
+ typename boost::enable_if<is_implicitly_convertible<T, real_t> >::type* = 0
+ ) : value_(fixed_point::detail::arthm_number_cast<T, real_t>()(x).count())
{}
- //! implicit conversion from long double
- explicit real_t(long double x) : value_(classify(x))
+ //! explicit conversion from int
+ template <typename T>
+ explicit real_t(T x
+ , typename boost::disable_if<is_implicitly_convertible<T ,real_t> >::type* = 0
+ ) : value_(fixed_point::detail::arthm_number_cast<T, real_t>()(x).count())
{}
//! implicit conversion from int
@@ -2125,7 +2223,7 @@
ureal_t(ureal_t<R,P,RP,OP,F> const& rhs
, typename boost::enable_if<is_convertible<ureal_t<R,P,RP,OP,F>, ureal_t > >::type* = 0
)
- : value_(fixed_point::detail::number_cast<ureal_t<R,P,RP,OP,F>, ureal_t, true, true>()(rhs).count())
+ : value_(fixed_point::detail::fxp_number_cast<ureal_t<R,P,RP,OP,F>, ureal_t, true, true>()(rhs).count())
{
}
@@ -2134,7 +2232,7 @@
explicit ureal_t(ureal_t<R,P,RP,OP,F> const& rhs
, typename boost::enable_if<is_explicitly_convertible<ureal_t<R,P,RP,OP,F>, ureal_t > >::type* = 0
)
- : value_(fixed_point::detail::number_cast<ureal_t<R,P,RP,OP,F>, ureal_t,
+ : value_(fixed_point::detail::fxp_number_cast<ureal_t<R,P,RP,OP,F>, ureal_t,
mpl::less_equal < mpl::int_<R>, mpl::int_<Range> >::value,
mpl::greater_equal < mpl::int_<P>, mpl::int_<Resolution> >::value
>()(rhs).count())
@@ -2146,7 +2244,7 @@
ureal_t(ureal_t<R,P,RP,OP,F> const& rhs
, typename boost::enable_if<is_implicitly_convertible<ureal_t<R,P,RP,OP,F>, ureal_t > >::type* = 0
)
- : value_(fixed_point::detail::number_cast<ureal_t<R,P,RP,OP,F>, ureal_t,
+ : value_(fixed_point::detail::fxp_number_cast<ureal_t<R,P,RP,OP,F>, ureal_t,
mpl::less_equal < mpl::int_<R>, mpl::int_<Range> >::value,
mpl::greater_equal < mpl::int_<P>, mpl::int_<Resolution> >::value
>()(rhs).count())
@@ -2156,7 +2254,7 @@
//! implicit constructor from a ureal_t with larger range or better resolution
template <int R, int P, typename RP, typename OP, typename F>
ureal_t(convert_tag<ureal_t<R,P,RP,OP,F> > rhs)
- : value_(fixed_point::detail::number_cast<ureal_t<R,P,RP,OP,F>, ureal_t,
+ : value_(fixed_point::detail::fxp_number_cast<ureal_t<R,P,RP,OP,F>, ureal_t,
mpl::less_equal < mpl::int_<R>, mpl::int_<Range> >::value,
mpl::greater_equal < mpl::int_<P>, mpl::int_<Resolution> >::value
>()(rhs.get()).count())
@@ -2252,17 +2350,16 @@
}
//! implicit conversion from int
- explicit ureal_t(unsigned int x) : value_(detail::shift_left<-Resolution>(x))
- {}
-
- //! implicit conversion from float
- explicit ureal_t(float x) : value_(classify(x))
- {}
- //! implicit conversion from double
- explicit ureal_t(double x) : value_(classify(x))
+ template <typename T>
+ ureal_t(T x,
+ typename boost::enable_if<is_implicitly_convertible<T, ureal_t> >::type* = 0
+ ) : value_(fixed_point::detail::arthm_number_cast<T, ureal_t>()(x).count())
{}
- //! implicit conversion from long double
- explicit ureal_t(long double x) : value_(classify(x))
+ //! explicit conversion from int
+ template <typename T>
+ explicit ureal_t(T x
+ , typename boost::disable_if<is_implicitly_convertible<T, ureal_t> >::type* = 0
+ ) : value_(fixed_point::detail::arthm_number_cast<T, ureal_t>()(x).count())
{}
//! implicit conversion from int
@@ -2669,16 +2766,16 @@
* signed + signed -> signed.
* @Returns <c>RT(index(RT(lhs).count()+RT(rhs).count())</c>.
*/
-// template <typename T, int R1, int P1, typename RP1, typename OP1, typename F1,
-// int R2, int P2, typename RP2, typename OP2, typename F2>
-// inline
-// T
-// add(real_t<R1,P1,RP1,OP1,F1> const& lhs, real_t<R2,P2,RP2,OP2,F2> const& rhs)
-// {
-// typedef T result_type;
-//
-// return result_type(index(result_type(lhs).count()+result_type(rhs).count()));
-// }
+ // template <typename T, int R1, int P1, typename RP1, typename OP1, typename F1,
+ // int R2, int P2, typename RP2, typename OP2, typename F2>
+ // inline
+ // T
+ // add(real_t<R1,P1,RP1,OP1,F1> const& lhs, real_t<R2,P2,RP2,OP2,F2> const& rhs)
+ // {
+ // typedef T result_type;
+ //
+ // return result_type(index(result_type(lhs).count()+result_type(rhs).count()));
+ // }
/**
* signed + signed -> signed.
* @Returns <c>RT(index(RT(lhs).count()+RT(rhs).count())</c>.
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