Misplaced Pages

Revision as of 16:22, 27 April 2007

C++ (pronounced "see plus plus", /siː plʌs plʌs/) is a general-purpose, high-level programming language with low-level facilities. It is a statically typed free-form multi-paradigm language, supporting procedural programming, data abstraction, object-oriented programming, generic programming and RTTI. Since the 1990s, C++ has been one of the most popular commercial programming languages.

Bjarne Stroustrup developed C++ (originally named "C with Classes") in 1983 at Bell Labs as an enhancement to the C programming language. Enhancements started with the addition of classes, followed by, among other features, virtual functions, operator overloading, multiple inheritance, templates, and exception handling. The C++ programming language standard was ratified in 1998 as ISO/IEC 14882:1998, the current version of which is the 2003 version, ISO/IEC 14882:2003. A new version of the standard (known informally as C++0x) is being developed.

History

Stroustrup began work on C with Classes in 1979. The idea of creating a new language originated from Stroustrup's experience in programming for his Ph.D. thesis. Stroustrup found that Simula had features that were very helpful for large software development, but the language was too slow for practical use, while BCPL was fast but too low-level and unsuitable for large software development. When Stroustrup started working in Bell Labs, he had the problem of analyzing the UNIX kernel with respect to distributed computing. Remembering his Ph.D. experience, Stroustrup set out to enhance the C language with Simula-like features. C was chosen because it is general-purpose, fast, and portable. Besides C and Simula, some other languages which inspired him were ALGOL 68, Ada, CLU and ML. At first, the class, derived class, strong type checking, inlining, and default argument features were added to C via Cfront. The first commercial release occurred in October 1985.

In 1983, the name of the language was changed from C with Classes to C++. New features were added including virtual functions, function name and operator overloading, references, constants, user-controlled free-store memory control, improved type checking, and BCPL style single-line comments with two forward slashes (//). In 1985, the first edition of The C++ Programming Language was released, providing an important reference to the language, as there was not yet an official standard. In 1989, Release 2.0 of C++ was released. New features included multiple inheritance, abstract classes, static member functions, const member functions, and protected members. In 1990, The Annotated C++ Reference Manual was published. This work became the basis for the future standard. Late addition of features included templates, exceptions, namespaces, new casts, and a Boolean type.

As the C++ language evolved, a standard library also evolved with it. The first addition to the C++ standard library was the stream I/O library which provided facilities to replace the traditional C functions such as printf and scanf. Later, among the most significant additions to the standard library, was the Standard Template Library.

After years of work, a joint ANSI-ISO committee standardized C++ in 1998 (ISO/IEC 14882:1998). For some years after the official release of the standard in 1998, the committee processed defect reports, and published a corrected version of the C++ standard in 2003. In 2005, a technical report, called the "Library Technical Report 1" (often known as TR1 for short) was released. While not an official part of the standard, it gives a number of extensions to the standard library which are expected to be included in the next version of C++. Support for TR1 is growing in almost all currently maintained C++ compilers.

While the C++ language is royalty-free, the standard document itself is not freely available.

The name "C++"

This name is credited to Rick Mascitti (mid-1983) and was first used in December 1983. Earlier, during the research period, the developing language had been referred to as "new C", then "C with Classes". In computer science C++ is still referred to as a superstructure of C. The final name stems from C's "++" operator (which increments the value of a variable) and a common naming convention of using "+" to indicate an enhanced computer program. According to Stroustrup: "the name signifies the evolutionary nature of the changes from C". C+ was the name of an earlier, unrelated programming language.

Stroustrup addresses the origin of the name in Chapter 1 of his book, The C++ Programming Language, remarking that another interpretation of the C++ name could be seen from the appendix of George Orwell's Nineteen Eighty-Four. Of the three segments of the fictional language Newspeak, the "C vocabulary" is the one dedicated to technical terms and jargon. "Doubleplus" is the superlative modifier for Newspeak adjectives. Thus, "C++" might hold the meaning "most C-like" in Newspeak.

When Rick Mascitti was questioned informally in 1992 about the naming, he indicated that it was given in a tongue-in-cheek spirit. He never thought that it would become the formal name of the language.

A common joke about the name is that in C++, when ++ is used as a postfix operator the variable is incremented only after its value has been used (hence, for this reason and to match the double-plus Newspeak usage described above, it should be ++C rather than C++).

Future development

C++ continues to evolve to meet future requirements. A new version of the C++ standard is currently being worked on, entitled C++0x denoting that it is expected to be released before 2010. Boost.org is a group working to make the most of C++ in its current form. They also advise the C++ standards committee on which features work well and which need improving. Current work indicates that C++ will continue to capitalize on its multi-paradigm nature. The work at Boost, for example, is greatly expanding C++'s functional and metaprogramming capabilities.

Philosophy

In The Design and Evolution of C++ (1994), Bjarne Stroustrup describes some rules that he uses for the design of C++. Knowing the rules helps to understand why C++ is the way it is. The following is a summary of the rules. Much more detail can be found in The Design and Evolution of C++.

• C++ is designed to be a statically typed, general-purpose language that is as efficient and portable as C
• C++ is designed to directly and comprehensively support multiple programming styles (procedural programming, data abstraction, object-oriented programming, and generic programming)
• C++ is designed to give the programmer choice, even if this makes it possible for the programmer to choose incorrectly
• C++ is designed to be as compatible with C as possible, therefore providing a smooth transition from C
• C++ avoids features that are platform specific or not general purpose
• C++ does not incur overhead for features that are not used
• C++ is designed to function without a sophisticated programming environment

"Inside the C++ Object Model" (Lippman, 1996) describes how compilers may convert C++ program statements into an in-memory layout. Compiler authors are free to implement the standard in their own manner.

Standard library

The 1998 C++ standard consists of two parts: the core language and the C++ standard library; the latter includes most of the Standard Template Library and a slightly modified version of the C standard library. Many C++ libraries exist which are not part of the standard, and, using external linkage, libraries can even be written in C.

The C++ standard library incorporates the C standard library with some small modifications to make it work better with the C++ language. Another large part of the C++ library is based on the Standard Template Library (STL). This provides such useful tools as containers (for example vectors and lists), iterators (generalized pointers) to provide these containers with array-like access and algorithms to perform operations such as searching and sorting. Furthermore (multi)maps (associative arrays) and (multi)sets are provided, all of which export compatible interfaces. Therefore it is possible, using templates, to write generic algorithms that work with any container or on any sequence defined by iterators. As in C, the features of the library are accessed by using the #include directive to include a standard header. C++ provides sixty-nine standard headers, of which nineteen are deprecated.

Using the standard library--for example, using std::vector or std::string instead of a C-style array--can help lead to safer and more scalable software.

The STL was originally a third-party library from HP and later SGI, before its incorporation into the C++ standard. The standard does not refer to it as "STL", as it is merely a part of the standard library, but many people still use that term to distinguish it from the rest of the library (input/output streams, internationalization, diagnostics, the C library subset, etc.).

Most C++ compilers provide an implementation of the C++ standard library, including the STL. Compiler-independent implementations of the STL, such as STLPort, also exist. Other projects also produce various custom implementations of the C++ standard library and the STL with various design goals.

Features introduced in C++

Compared to the C language, C++ introduced extra features, including declarations as statements, function-like casts, new/delete, bool, reference types, inline functions, default arguments, function overloading, namespaces, classes (including all class-related features such as inheritance, member functions, virtual functions, abstract classes, and constructors), operator overloading, templates, the :: operator, exception handling, and runtime type identification.

Contrary to popular belief, C++ did not introduce the const keyword first. Const was formally added to C shortly before it was adopted by C++.

C++ also performs more type checking than C in several cases (see "Incompatibility with C" below).

Comments starting with two slashes ("//") were originally part of C's predecessor, BCPL, and were reintroduced in C++.

Several features of C++ were later adopted by C, including declarations in for loops, C++-style comments (using the // symbol), and inline, though the C99 definition of the inline keyword is not compatible with its C++ definition. However, C99 also introduced features that do not exist in C++, such as variadic macros and better handling of arrays as parameters; some C++ compilers may implement some of these features as extensions, but others are incompatible with existing C++ features.

A very common source of confusion is a subtle terminology issue: because of its derivation from C, in C++ the term object means memory area, just like in C, and not class instance, which is what it means in most other object oriented languages. For example, in both C and C++, the statement int i; defines an object of type int, that is the memory area where the value of the variable i will be stored on assignment.

Hello world program

The following is a Hello world program which uses the C++ standard library (not the Standard Template Library (STL)) stream facility to display a message then terminate.

#include <iostream> // provides std::cout
int main()
{
   std::cout << "Hello, world!" << std::endl;
   return 0;
}

Language features

Operators

Preprocessor

C++ is principally compiled in three phases: preprocessing, translation to object code, and linking (the two last phases are what is generally thought of as the "compilation" proper). In the first phase, preprocessing, preprocessor directives apply lexical transformations to the source code, which is then fed to the next compilation stage.

Preprocessor directives and macros

Also see preprocessor and C preprocessor

Pre-processing directives work by simple substitution of tokenized character sequences for other tokenized character sequences, according to user-defined rules. They typically perform macro substitution, inclusion of other files (by opposition to higher-order features such as inclusion of modules/packages/units/components), conditional compilation and/or conditional inclusion. For instance:

#include <iostream>

which includes (imports) all symbols from the standard library header file iostream. Another common example is

#define MY_ASSERT(x) assert(x)

Every time MY_ASSERT(x) appears in the source code it will be replaced with assert(x). This allows control over the use of assertions in a particular compilation unit.

In addition to the most common directives mentioned above, there are several additional preprocessor directives that control the flow of compilation, conditionally include or exclude code blocks, and control various other aspects of compilation.

Traditionally the preprocessor was used to define numerical constants, however the use of const is now preferred:

const long double PI = 3.1415926535897932384626433832795028841971693993751 ;

This provides stronger type checking and does not subvert the use of namespaces as the preprocessor does.

Templates

Templates are different from macros: while both of these compile-time language features can be used to produce conditional compilation, templates are not restricted to lexical substitution. Templates have an awareness of the semantics and type system of their companion language as well as all compile-time type definitions and can perform high-level operations including programmatic flow control based on evaluation of strictly type-checked parameters. Macros are capable of conditional control over compilation based on predetermined criteria but cannot instantiate new types, recurse or perform type evaluation and in effect are limited to pre-compilation text-substitution and text-inclusion/exclusion. In other words, macros can control compilation flow based on pre-defined symbols but cannot, unlike templates, independently instantiate new symbols. Templates are a tool for static polymorphism (see below) and generic programming. For example, a template replacing the common, but dangerous, macro #define max(x,y) ((x)>(y)?(x):(y)):

template <typename T>
T max(T x, T y)
{
    return x > y ? x : y ;
}

This can be found in the algorithm header as std::max().

In addition, templates are a compile time mechanism in C++ which is Turing-complete, meaning that any computation expressible by a computer program can be computed, in some form, by a template metaprogram prior to runtime.

Objects

C++ introduces some object-oriented (OO) features to C. It offers classes, which provide the four features commonly present in OO (and some non-OO) languages: abstraction, encapsulation, inheritance and polymorphism. Objects are instances of classes created at runtime. Think of the class as a template from which many different individual objects may be generated as a program runs.

Encapsulation

Encapsulation is the grouping together of data and functionality. C++ implements encapsulation by allowing all members of a class to be declared as either public, private, or protected. A public member of the class will be accessible to any function. A private member will only be accessible to functions that are members of that class and to functions and classes explicitly granted access permission by the class ("friends"). A protected member will be accessible to members of classes that inherit from the class in addition to the class itself and any friends.

The OO principle is that all and only the functions that can access the internal representation of a type should be encapsulated within the type definition. C++ supports this (via member functions and friend functions), but does not enforce it: the programmer can declare parts or all of the representation of a type to be public, and is also allowed to make public entities that are not part of the representation of the type. Because of this, C++ supports not just OO programming but other weaker decomposition paradigms, like modular programming.

It is generally considered good practice to make all data private or protected, and to make public only those functions that are part of a minimal interface for users of the class, that hides implementation details.

Inheritance

Inheritance allows one data type to acquire properties of other data types. Inheritance from a base class may be declared as public, protected, or private. This access specifier determines whether unrelated and derived classes can access the inherited public and protected members of the base class. Only public inheritance corresponds to what is usually meant by "inheritance". The other two forms are much less frequently used. If the access specifier is omitted, inheritance is assumed to be private for a class base and public for a struct base. Base classes may be declared as virtual; this is called virtual inheritance. Virtual inheritance ensures that only one instance of a base class exists in the inheritance graph, avoiding some of the ambiguity problems of multiple inheritance.

Multiple inheritance is a C++ feature sometimes considered controversial. Multiple inheritance allows a class to be derived from more than one base class; this can result in a complicated graph of inheritance relationships. For example, a "Flying Cat" class can inherit from both "Cat" and "Flying Mammal". Some other languages, such as C# or Java, accomplish something similar (although more limited) by allowing inheritance of multiple interfaces while restricting the number of base classes to one (interfaces, unlike classes, provide only declarations of member functions, no implementation or member data).

Polymorphism

Polymorphism (computer science) enables one common interface for many implementations, and for objects to act differently under different circumstances. C++ supports several kinds of static (compile-time) and dynamic (run-time) polymorphism. Compile-time polymorphism does not allow for certain run-time decisions, while run-time polymorphism typically incurs a performance penalty.

Static polymorphism

Function overloading

Function overloading allows programs to declare multiple functions with the same name. The functions are distinguished by the number and/or types of their formal parameters. Thus, the same function name can refer to different functions depending on the context in which it is used. The type returned by the function is not used to distinguish overloaded functions.

Operator overloading

Similarly, operator overloading allows programs to define certain operators (such as +, !=, <, or &) to result in a function call that depends on the types of the operands they are used on.

Class and function templates

Templates in C++ provide a sophisticated mechanism for writing generic, polymorphic code. In particular, through the Curiously Recurring Template Pattern it's possible to implement a form of static polymorphism that closely mimics the syntax for overriding virtual methods (a dynamic polymorphism technique described below). Since C++ templates are type-aware and Turing-complete they can also be used to let the compiler resolve recursive conditionals and generate substantial programs through template metaprogramming.

Dynamic polymorphism

Polymorphism through inheritance

Variable pointers (and references) of a base class type in C++ can refer to objects of any derived classes of that type in addition to objects exactly matching the variable type. This allows arrays and other kinds of containers to hold pointers to objects of differing types. Because assignment of values to variables usually occurs at run-time, this is necessarily a run-time phenomenon.

C++ also provides a dynamic_cast operator, which allows the program to safely attempt conversion of an object into an object of a more specific object type (as opposed to conversion to a more general type, which is always allowed). This feature relies on run-time type information (RTTI). Objects known to be of a certain specific type can also be cast to that type with static_cast, a purely compile-time construct which is faster and does not require RTTI.

Virtual member functions

Through virtual member functions, different objects that share a common base class may all support an operation in different ways. The member functions implemented by the derived class are said to override the same member functions of the base class. In contrast with function overloading, the parameters for a given member function are always exactly the same in number and type. Only the type of the object for which this method is called varies. In addition to standard member functions, operator overloads and destructors can also be virtual.

By virtue of inherited objects being polymorphic, it may not be possible for the compiler to determine the type of the object at compile time. The decision is therefore put off until runtime, and is called dynamic dispatch. In this way, the most specific implementation of the function is called, according to the actual run-time type of the object. In C++, this is commonly done using virtual function tables. This may sometimes be bypassed by prepending a fully qualified class name before the function call, but calls to virtual functions are in general always resolved at run time.

An example

#include <iostream>
class Bird                 // the "generic" base class
{
public:
  virtual void OutputName() {std::cout << "a bird";}
  virtual ~Bird() {}
};
class Swan : public Bird   // Swan derives from Bird
{ 
public:
  void OutputName() {std::cout << "a swan";} // overrides virtual function
};
int main()
{
  Swan mySwan ;            // Creates a swan.
  Bird* myBird = &mySwan;  // Declares a pointer to a generic Bird,
                           // and sets it pointing to a newly created Swan.
  myBird->OutputName();    // This will output "a swan", not "a bird".
  return 0;
}

This example program makes use of virtual functions, polymorphism, and inheritance to derive new, more specific objects from a base class. In this case, the base class is a Bird, and the more specific Swan is made.

Parsing and processing C++ source code

It is relatively difficult to write a good C++ parser with classic parsing algorithms such as LALR(1) (see ). This is partly because the C++ grammar is not LALR. Because of this, there are very few tools for analyzing or performing non-trivial transformations (e.g., refactoring) of existing code. One way to handle this difficulty is to choose a different syntax, such as Significantly Prettier and Easier C++ Syntax which is LALR(1) parseable.

Parsing (in the literal sense of producing a syntax tree) is not the most difficult problem in building a C++ processing tool. Such tools must also have the same understanding of the meaning of the identifiers in the program as a compiler might have. Practical systems for processing C++ must then not only parse the source text, but be able to resolve for each identifier precisely which identifier definition applies, e.g., correctly handle C++'s complex scoping rules, but must also be able to determine the type of each identifer and expression.

Finally, a practical C++ processing tool must be able to handle the variety of C++ dialects used in practice (GNU, Microsoft, ...), but also implement appropriate analyzers, source code transformers, and regenerate source text. Combining advanced parsing algorithms such as GLR with symbol table construction and program transformation machinery can enable the construction of arbitrary C++ tools.

Problems and controversies

Traditionally, producing a reasonably standards-compliant C++ compiler has proven to be a difficult task for compiler vendors in general. For many years, different C++ compilers implemented the C++ language to different levels of compliance to the standard, and their implementations varied widely in some areas such as partial template specialization. Recent releases of most popular C++ compilers support almost all of the C++ 1998 standard.

One particular point of contention is the export keyword, intended to allow template definitions to be separated from their declarations. The first compiler to implement export was Comeau C++, in early 2003 (5 years after the release of the standard); in 2004, the beta compiler of Borland C++ Builder X was also released with export. Both of these compilers are based on the EDG C++ front end. It should also be noted that many C++ books provide example code using the keyword export (for example, Beginning ANSI C++ by Ivor Horton) which will not compile in most compilers, but there is no reference to the problem with the keyword export mentioned. Other compilers such as GCC do not support it at all. Herb Sutter, secretary of the C++ standards committee, recommended that export be removed from future versions of the C++ standard Template:PDFlink, but finally the decision was made to leave it in the C++ standard .

In order to give compiler vendors greater freedom, the C++ standards committee decided not to dictate the implementation of name mangling, exception handling, and other implementation-specific features. The downside of this decision is that object code produced by different compilers are expected to be incompatible. There are, however, third party standards for particular machines or operating systems which attempt to standardize compilers on those platforms (for example C++ ABI); some compilers adopt a secondary standard for these items.

Modern critics of the language raise several points. First, since C++ is based on and largely compatible with C, it inherits most of the criticisms levelled at that language. Taken as a whole C++ has a large feature set, including all of C, plus a large set of its own additions, in part leading to criticisms of being a "bloated" and complicated language, especially for embedded systems due to features such as exceptions and RTTI which add to code size. However, every compiler allows the developer to disable exceptions if desired. Bjarne Stroustrup also points out that resultant executables don't support these claims of bloat: "I have even seen the C++ version of the 'hello world' program smaller than the C version." The Embedded C++ standard was specified to deal with part of this, but it received criticism for leaving out useful parts of the language that incur no runtime penalty. Because of its large featureset it can be quite difficult to fully master C++, leading to programmers often bringing unnecessarily advanced or complicated solutions to simple problems.

While C++ is more complex than some other programming languages, Bjarne Stroustrup points out that "The programming world is far more complex today than it was 30 years ago, and modern programming languages reflect that." The ISO standard of the C++ language is about 310 pages (excluding library). For comparison, the C programming language's is about 160 pages, even though it was designed more than 15 years prior and doesn't consider Object Oriented Programming. Furthermore, C#'s ECMA language definition document is about 440 pages.

C++ is also sometimes compared unfavorably with single-paradigm object-oriented languages such as Java, on the basis that it allows programmers to "mix and match" object-oriented and procedural programming, rather than strictly enforcing a single paradigm. This is part of a wider debate on the relative merits of the two programming styles.

Incompatibility with C

C++ is often considered to be a superset of C, but this is not strictly true. Most C code can easily be made to compile correctly in C++, but there are a few differences that cause some valid C code to be invalid in C++, or to behave differently in C++.

Perhaps the most commonly encountered difference is that C allows implicit conversion from void* to other pointer types, but C++ does not. So, the following is valid C code:

int *i = malloc(sizeof(int) * 5);     /* Implicit conversion from void* to int* */

but to make it work in both C and C++ one would need to use an explicit cast:

int *i = (int *) malloc(sizeof(int) * 5);

Another common portability issue is that C++ defines many new keywords, such as new and class, that may be used as identifiers (e.g. variable names) in a C program.

Some incompatibilities have been removed by the latest (C99) C standard, which now supports C++ features such as // comments and mixed declarations and code. However, C99 introduced a number of new features that conflict with C++ (such as variable-length arrays, native complex-number types, and compound literals), so the languages may be diverging more than they are converging.

In order to intermix C and C++ code, any C code which is to be called from/used in C++ must be put in extern "C" { /* C code */ }.

It should be noted that for C++-only code, the use of C++ cast operators, including static_cast, dynamic_cast, const_cast and reinterpret_cast, is preferred over the C-style casting semantics. e.g.:

int *i = static_cast<int*>(malloc(sizeof(int) * 5));

See also

More information

• C++ structures and classes
• List of C++ compilers and integrated development environments
• Operators in C and C++
• C++0x

Miscellaneous

• Boost C++ Libraries - Selection of peer-reviewed C++ libraries designed to work with (and possibly become part of) the standard C++ library
• CodeSynthesis XSD - Open-source XML data binding framework for C++
• OpenC++
• Name mangling
• Template metaprogramming
• Comparison of Java and C++
• Comparison of programming languages
• Aard - Programming tool to check memory use

References

1. "Bjarne Stroustrup's FAQ - When was C++ invented?". Retrieved 2006-05-30.
2. While this is a legal example, it should be noted that the use of "raw" pointers as in this example, along with the related use of explicit dynamic memory allocation via new and delete, is a common source of error and confusion in C++. It is much safer to do all dynamic memory management inside a class that manages these resources via constructor and destructor calls, to prevent such problems.
3. Herb Sutter (2003-04-15). "C++ Conformance Roundup". Dr. Dobb's Journal. Retrieved 2006-05-30. {{cite web}}: Check date values in: |date= (help)
4. "Minutes of J16 Meeting No. 36/WG21 Meeting No. 31, April 7-11, 2003". 2003-04-25. Retrieved 2006-09-04.
5. "C++ ABI". Retrieved 2006-05-30.
6. Why is the code generated for the "Hello world" program ten times larger for C++ than for C?
7. Why is C++ so BIG?

• C++
• Class-based programming languages
• Curly bracket programming languages
• Multi-paradigm programming languages