doc/named-param.dox
author alpar
Thu, 19 Jan 2006 16:56:39 +0000
changeset 1900 b16ca599472f
parent 1709 a323456bf7c8
child 1929 84d87d6024af
permissions -rw-r--r--
Fix bug #18: bug in LpSolverBase::Col operator!= and ::Row operator!=
alpar@1624
     1
/*!
alpar@954
     2
alpar@953
     3
\page named-param Named Parameters
alpar@953
     4
alpar@1624
     5
\section named-func-param Named Function Parameters
alpar@955
     6
alpar@1536
     7
C++ makes it possible to use default parameter values when calling a
alpar@1536
     8
function. In such a case we do not have to give value for parameters,
alpar@1536
     9
the program will use the default ones.  Unfortunately sometimes this
alpar@1536
    10
is not enough. If we do not want to give values for all the
alpar@1536
    11
parameters, only for some of them we come across problems, because an
alpar@1536
    12
arbitrary set of parameters cannot be omitted. On the other hand
alpar@1536
    13
parameters have a fixed order in the head of the function.  C++ can
alpar@1536
    14
apply the default values only in the back of the order, if we do not
alpar@1536
    15
give other value for them.  So we can not give the function for
alpar@1536
    16
example the value of the first, and the third parameter, expecting
alpar@1536
    17
that the program will aplly the default value for the second
alpar@1536
    18
parameter.  However sometimes we would like to use some functinos
alpar@1536
    19
exactly in this way. With a crafty trick and with some little
alpar@1536
    20
inconvenience this is possible. We have implemented this little trick
alpar@1536
    21
as an example below.
hegyi@1141
    22
hegyi@1141
    23
\code
alpar@1713
    24
class namedFn 
hegyi@1141
    25
{
hegyi@1141
    26
  int _id;
hegyi@1141
    27
  double _val;
hegyi@1141
    28
  int _dim;
hegyi@1141
    29
  
hegyi@1141
    30
  public:
alpar@1713
    31
  namedFn() : _id(0), _val(1), _dim(2) {}
alpar@1713
    32
  namedFn& id(int p)     { _id  = p ; return *this; }
alpar@1713
    33
  namedFn& val(double p) { _val = p ; return *this; }
alpar@1713
    34
  namedFn& dim(int p)    { _dim = p ; return *this; }
hegyi@1141
    35
hegyi@1141
    36
  run() {
hegyi@1141
    37
    printf("Here is the function itself.");
hegyi@1141
    38
  }
hegyi@1141
    39
};
hegyi@1141
    40
\endcode
hegyi@1141
    41
hegyi@1141
    42
hegyi@1141
    43
The usage is the following.
hegyi@1141
    44
alpar@1713
    45
We have to define a class, let's call it \c namedFn.  Let us assume that
alpar@1713
    46
we would like to use a parameter, called \c X. In the \c namedFn class we
alpar@1624
    47
have to define an \c _X attribute, and a function \c X. The function
alpar@1624
    48
expects a parameter with the type of \c _X, and sets the value of
alpar@1624
    49
\c _X. After setting the value the function returns the class itself. The
alpar@1624
    50
class also have to have a function, called for example <tt>run()</tt>, we have
alpar@1536
    51
to implement here the original function itself. The constructor of the
alpar@1624
    52
class have to give all the attributes like \c _X the default values of
alpar@1536
    53
them.
hegyi@1141
    54
alpar@1536
    55
If we instantiate this class, the default values will be set for the
hegyi@1619
    56
attributes (originally the parameters), initially. If we call function
alpar@1624
    57
\c X, we get a class with the modified parameter value of
alpar@1713
    58
\c X. Therefore we can modify any parameter-value, independently from the
alpar@1624
    59
order. To run the algorithm we have to call the <tt>run()</tt> function at the
alpar@1536
    60
end of the row.
hegyi@1141
    61
alpar@1624
    62
Example:
alpar@1624
    63
\code
alpar@1713
    64
namedFn().id(3).val(2).run();
alpar@1624
    65
\endcode
alpar@955
    66
alpar@1713
    67
\note Although it is a class, namedFn is used pretty much like as it were
alpar@1713
    68
a function. That it why it is called namedFn and not \c NamedFn.
alpar@1713
    69
alpar@1713
    70
\note In fact, the final <tt>.run()</tt> could be made unnecessary if the
alpar@1713
    71
actual function code were put in the destructor instead. This however would make
alpar@1713
    72
hard to implement functions with return values, and would also make the
alpar@1713
    73
implementation of \ref named-templ-func-param "named template parameters"
alpar@1713
    74
very problematic. Therefore, by convention, <tt>.run()</tt> is used
alpar@1713
    75
to explicitly execute function having named parameters in Lemon. 
alpar@1713
    76
alpar@1713
    77
alpar@955
    78
\section traits-classes Traits Classes
alpar@955
    79
alpar@1536
    80
The procedure above can also be applied when defining classes. In this
alpar@1536
    81
case the type of the attributes can be changed.  Initially we have to
alpar@1536
    82
define a class with the default attribute types. This is the so called
alpar@1536
    83
Traits Class. Later on the types of these attributes can be changed,
alpar@1536
    84
as described below. In our software \ref lemon::DijkstraDefaultTraits is an
alpar@1536
    85
example of how a traits class looks like.
hegyi@1141
    86
alpar@955
    87
\section named-templ-param Named Class Template Parameters
alpar@954
    88
alpar@1536
    89
If we would like to change the type of an attribute in a class that
alpar@1536
    90
was instantiated by using a traits class as a template parameter, and
alpar@1536
    91
the class contains named parameters, we do not have to reinstantiate
alpar@1536
    92
the class with new traits class. Instead of that, adaptor classes can
alpar@1536
    93
be used like in the following cases.
hegyi@1141
    94
alpar@954
    95
\code
deba@1709
    96
Dijkstra<>::SetPredNodeMap<NullMap<Node,Node> >::Create
alpar@954
    97
\endcode
hegyi@1141
    98
alpar@954
    99
It can also be used in conjunction with other named template
alpar@954
   100
parameters in arbitrary order.
hegyi@1141
   101
alpar@954
   102
\code
deba@1709
   103
Dijkstra<>::SetDistMap<MyMap>::SetPredMap<NullMap<Node,Edge> >::Create
alpar@954
   104
\endcode
alpar@954
   105
alpar@1536
   106
The result will be an instantiated Dijkstra class, in which the
alpar@1536
   107
DistMap and the PredMap is modified.
hegyi@1141
   108
alpar@1624
   109
\section named-templ-func-param Named Function Template Parameters
alpar@955
   110
alpar@1536
   111
If the class has so called wizard functions, the new class with the
alpar@1536
   112
modified tpye of attributes can be returned by the appropriate wizard
alpar@1536
   113
function. The usage of these wizard functions is the following:
alpar@953
   114
alpar@953
   115
*/