doxygen batch file

git-svn-id: svn://svn.code.sf.net/p/magicseteditor/code/trunk@3 0fc631ac-6414-0410-93d0-97cfa31319b6
This commit is contained in:
twanvl
2006-10-01 15:16:44 +00:00
parent ddfb1a5089
commit a9b6c73407
7 changed files with 780 additions and 0 deletions
+123
View File
@@ -0,0 +1,123 @@
/** \mainpage
This is the documentation of the Magic Set Editor (MSE) source code, automatically generated using doxygen.
<h2>Structure</h2>
The MSE source code is subdivided into several directories, with the following meaning:
- util: Utility functions and classes, stuff that would work equally well in another project
- gfx: Graphics related functions, again mostly independent of MSE.
This directory contains algorithms for image blending, scaling, and bezier curve functions.
- data: Data structures, like sets, cards, symbols, etc.
These data structures are documented in the <a href="http://magicseteditor.sourceforge.net/extending">'Extending MSE'</a> section of the official documentation.
- data/action: Actions that can be applied to those data structures.
- gui: Graphical User Interface
- resource: Resource files used (icons, cursors, etc.)
<h2>Libraries/dependencies</h2>
MSE depends on the following libraries:
- <a href="http://wxwidgets.org">wxWidgets</a> for the GUI.
- <a href="http://boost.org">boost</a>, just for shared_ptr and some preprocessor tricks.
Additional tools (not needed for building MSE) also depend on:
- <a href="http://doxygen.org">doxygen</a> for generating the documentation.
- Perl for small utility scripts
<h2>Coding style</h2>
MSE uses the following coding style:
@code
class ClassName {
public:
void someFunction();
private:
int someMember;
};
@endcode
The exception to this rule are wxWidgets functions, which LookLikeThis; and classes, which look like wxSomeClass.
As well as standard library functions.
<h2>Macro and template tricks</h2>
The source code uses several macro/preprocessor and template tricks to make the code more readable.
<h3>Smart pointers</h3>
MSE makes extensive use of boost::shared_ptr. To make the code more readable there are typedefs for these pointer types, using a suffix P.
These are defined using a macro:
@code
DECLARE_POINTER_TYPE(MyClass);
MyClassP someObject; // the same as boost::shared_ptr<MyClass> someObject
@endcode
To create new shared_ptrs the function new_shared# can be used (where # is the number of arguments):
@code
MyClassP someObject;
someObject = new_shared2<MyClass>(arg1, arg2);
@endcode
Implemented in: util/smart_ptr.hpp
<h3>Iterating</h3>
To iterate over containers the FOR_EACH macro is used:
@code
vector<CardP> cards;
FOR_EACH(card, cards) {
doSomething(card);
}
@endcode
Is equivalent to:
@code
vector<CardP> cards;
for(vector<CardP>::iterator it = cards.begin() ; it != cards.end() ; ++it) {
CardP& card = *it;
doSomething(card);
}
@endcode
The iterators are completely hidden!
There are several veriations to this macro, for using const iterators (FOR_EACH_CONST), iterating in reverse (FOR_EACH_REVERSE),
for iterating over two collections in parallel (FOR_EACH_2), and for getting access to the iterator (FOR_EACH_IT).
Each of these macros require that the collection type has been declared using:
@code
DECLARE_COLLECTION_TYPE(CardP);
@endcode
This allows the calling of TYPEOF(cards) to evaluate to vector<CardP>.
Implemented in: util/for_each.hpp
<h3>Reflection</h3>
The io (input/output) system is based on reflection.
For a class to support reflection the following must be declared:
@code
class SomeClass {
int member1, member2;
DECLARE_REFLECTION();
};
@endcode
Then in a source file the members of the class have to be specified:
@code
IMPLEMENT_REFLECTION(SomeClass) {
REFLECT(member1);
REFLECT_N("another_name", member2);
}
@endcode
Simlairly for enumerations (a declaration is not necessary):
@code
IMPLEMENT_REFLECTION_ENUM(MyEnum) {
VALUE(value_of_enum_1); // the default value
VALUE(value_of_enum_2);
}
@endcode
Implemented in: util/reflect.hpp
*/