Most vexing parse

The most vexing parse is an extremely counterintuitive disambiguation of code in C++.

Example with classes

An example is:

class Timer {
 public:
  Timer();
};

class TimeKeeper {
 public:
  TimeKeeper(const Timer& t);

  int get_time();
};

int main() {
  TimeKeeper time_keeper(Timer());
  return time_keeper.get_time();
}

The line

  TimeKeeper time_keeper(Timer());

could be disambiguated either as

1. a variable definition for variable time_keeper of class TimeKeeper, taking an anonymous instance of class Timer or
2. a function declaration for a function time_keeper which returns an object of type TimeKeeper and takes a single (unnamed) argument which is a function returning type Timer (and taking no input). (See Function object#In C and C++)

Most programmers expect the first, but the C++ standard requires it to be interpreted as the second.

For example, g++ gives the following error message:

$ g++ -c time_keeper.cc
time_keeper.cc: In function ‘int main()’:
time_keeper.cc:15: error: request for member ‘get_time’ in ‘time_keeper’, which is of non-class type ‘TimeKeeper ()(Timer (*)())’

Clang++ provides a useful warning:

$ clang++ time_keeper.cc
time_keeper.cc:14:25: warning: parentheses were disambiguated as a function declarator
  TimeKeeper time_keeper(Timer());
                        ^~~~~~~~~

One way to force the compiler to consider this as a variable definition is to add an extra pair of parentheses:

  TimeKeeper time_keeper( (Timer()) );

Simpler example

An even simpler example appears when a functional cast is intended to convert an expression for initializing a variable or passing to a constructor parameter

void f(double adouble) {
  int i(int(adouble));
}

In this case, the parentheses around "adouble" are superfluous and the declaration of "i" is again a function declaration equivalent to the following

// takes an integer and returns an integer
int i(int adouble);

To disambiguate this in favour of a variable declaration, the same technique can be used as for the first case above. Another solution is to use the cast notation or to use a named cast

// declares a variable called 'i'
int i(static_cast<int>(adouble));

References

• Scott Meyers, "Effective STL", ISBN 0201749629
• http://www.informit.com/guides/content.aspx?g=cplusplus&seqNum=439
• http://blog.llvm.org/2010/04/amazing-feats-of-clang-error-recovery.html
• http://arstechnica.com/civis/viewtopic.php?f=20&t=767929