[boost] Re: Constrained genericity at last? (any interest in anoperator traits/concept traits library?)

From: David Abrahams (dave_at_[hidden])
Date: 2004-09-20 08:22:02

Terje Slettebø <tslettebo_at_[hidden]> writes:

>> enable_if doesn't always improve error messages.  Sometimes you get a
>> long list of candidate functions when what you really wanted was a
>> message from inside just one of the overloads that indicates how
>> concept conformance was violated.

> How can you get a list of candidate functions, if enable_if excludes them
> from the overload set? Could you give a code example?

  #include <boost/utility/enable_if.hpp>
  #include <boost/type_traits/is_same.hpp>

  template <class T>
  typename boost::enable_if<boost::is_same<T,int>,int>::type
  f(T);

  template <class T>
  typename boost::enable_if<boost::is_same<T,long*>,int>::type
  f(T);

  int main()
  {
      f("foo");
  }

CWPro9:

  ### mwcc Compiler:
  #    File: foo.cpp
  # ----------------
  #      14:     f("foo");  
  #   Error:            ^
  #   function call 'f({lval} const char[4])' does not match
  #   'f<...>(__T0)'
  #   'f<...>(__T0)'

VC7.1:

  foo.cpp
  foo.cpp(14) : error C2893: Failed to specialize function template 'boost::enable_if<
          boost::is_same<T, long *>, int
      >::type f(T)'
  With the following template arguments:
  'const char *'
  foo.cpp(14) : error C2893: Failed to specialize function template 'boost::enable_if<
          boost::is_same<T, int>, int
      >::type f(T)'
  With the following template arguments:
  'const char *'

And even when there's no list as with GCC, "no match" is usually
less-useful than "here's what's wrong with the argument."

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com