6.096 Introduction to C++ January 4, 2011 Massachusetts Institute of Technology Lecture 1 Notes: Introduction 1 Compiled Languages and C++ 1.1 Why Use a Language Like C++? At its core, a computer is just a processor with some memory, capable of running tiny instructions like “store 5 in memory location 23459.” Why would we express a program as a text file in a programming language, instead of writing processor instructions? The advantages: 1. Conciseness: programming languages allow us to express common sequences of commaand more concisely. C++ provides some especially powerful shorthands. 2. Maintainability: modifying codeiseasierwhenit entailsjust afewtext edits,instead of rearranging hundreds of processor instructions. C++ is object oriented (more on that in Lectures 7-8), which further improves maintainability. 3. Portability: differentprocessorsmakedifferentinstructionsavailable. Programswrittte as text can be translated into instructions for many different processors; one of C++’s strengths is that it can be used to write programs for nearly any processor. C++isahigh-level language: whenyouwriteaprograminit,theshorthandsaresufficiently expressivethatyoudon’t needtoworry aboutthedetailsofprocessorinstructions. C++does giveaccesstosomelower-levelfunctionality thanotherlanguages(e.g. memory addresses). 1.2 The Compilation Process Aprogramgoesfromtext files(or source files)to processor instructions as follows: Source File Object FileCompiler Source File Object FileCompiler ExecutableLinker Libraries Program in MemoryOS Object files are intermediate files that represent an incomplete copy of the program: eachsource file only expresses a piece of the program, so when it is compiled into an object file,the object file has some markers indicating which missing pieces it depends on. The linkertakes those object files and the compiled libraries of predefined code that they rely on, fills in all the gaps, and spits out the final program, which can then be run by the operating system(OS). The compiler andlinker arejust regularprograms. The stepinthe compilationprocessin which the compiler reads the file is called parsing. In C++, all these steps are performed ahead of time, before you start running a program. In some languages, they are done during the execution process, which takes time. This is one of the reasons C++ code runs far faster than code in many more recent languages. C++ actually adds an extra step to the compilation process: the code is run through a preprocessor, which applies some modifications to the source code, before being fed to the compiler. Thus, the modified diagram is: Source File Processed CodePreprocessor Object FileCompiler Source File Processed CodePreprocessor Object FileCompiler ExecutableLinker Libraries OS Program in Memory 1.3 General Notes on C++ C++ is immensely popular, particularly for applications that require speed and/or access to some low-level features. It was created in 1979 by Bjarne Stroustrup, at first as a set of extensions to the C programming language. C++ extends C; our first few lectures will basically be on the C parts of the language. Thoughyou can writegraphicalprogramsinC++,itis muchhairier andlessportablethan text-based(console)programs. We will be sticking to console programs in this course. Everything in C++ is case sensitive: someName is not the same as SomeName. 2 Hello World Inthetraditionofprogrammerseverywhere,we’ll usea “Hello,world!” programasanentry point into the basic features of C++. 2.1 The code 1 2 3 //A Hello World program # include 2 4 int main() {5 std::cout << "Hello, world!\n";67 return 0;8 }2.2 Tokens Tokens aretheminimalschunk ofprogramthathavemeaning tothecompiler –thesmallest meaningful symbols in the language. Our code displays all 6 kinds of tokens, though the usual use of operators is not present here: Token type Description/Purpose Examples Keywords Words withspecial meaningto the compiler int, double, for, auto Identifiers Names of things that are not built into the language cout, std, x, myFunction Literals Basic constant values whose value is specified directly in the source code "Hello, world!", 24.3, 0, ’c’ Operators Mathematical or logical operatiion +, -, &&, %, << Punctuation/Separators Punctuation defining the structure of a program { } ( ) , ; Whitespace Spaces of various sorts; ignoore by the compiler Spaces, tabs, newlines, commeent 2.3 Line-By-Line Explanation 1. //indicates that everything following it until the end of the line is a comment: it is ignoredby thecompiler. Anotherway towriteacommentistoputitbetween /* and */(e.g. x = 1 + /*sneaky comment here*/1;). A comment of this form may span multiple lines. Comments exist to explain non-obvious things going on in the code. Use them: document your code well! 2. Lines beginning with # are preprocessor commands, which usually change what code isactuallybeing compiled. #include tellsthepreprocessortodumpinthecontentsof another file, here the iostream file, which defines the procedures for input/output. 3 4. int main() {...}defines the code that should execute when the program starts up. The curly braces represent grouping of multiple commands into a block. More about this syntax in the next few lectures. 5. • cout << : This is the syntax for outputting some piece of text to the screen. We’ll discuss how it works in Lecture 9. • Namespaces: In C++, identifiers can be defined within a context – sort of a directory of names – called a namespace. When we want to access an identifier definedinanamespace,wetell thecompilertolookforitinthat namespaceusing the scope resolution operator (::). Here, we’re telling the compiler to look for cout inthe std namespace,inwhich many standardC++identifiersaredefined. A cleaner alternative is to add the following line below line 2: using namespace std;This line tells the compiler that it should look in the std namespace for any identifier we haven’t defined. If we do this, we can omit the std:: prefix when writing cout. This is the recommended practice. • Strings: Asequenceof characterssuch as Hello, world isknownasa string.A string that is specified explicitly in a program is a stringliteral. • Escape sequences: The \n indicatesa newline character. Itisanexampleof an escape sequence – a symbol used to represent a special character in a text literal. Here are all the C++ escape sequences which you can include in strings: Escape Sequence Represented Character \a System bell (beep sound) \b Backspace \f Formfeed (page break) \n Newline (line break) \r “Carriage return” (returns cursor to start of line) \t Tab \\ Backslash \’ Single quote character \" Double quote character \some integer x The character represented by x 7. return 0 indicatesthattheprogramshould tell theoperating systemithascompleted successfully. Thissyntax willbeexplainedinthecontext offunctions;fornow,just include it as the last line in the main block. 4 Notethat every statement endswith asemicolon(exceptpreprocessorcommandsandblocks using {}). Forgetting thesesemicolonsisacommonmistakeamong newC++programmers. 3 Basic Language Features So far our program doesn’t do very much. Let’s tweak it in various ways to demonstrate some more interesting constructs. 3.1 Values and Statements First, a few definitions: • Astatement isaunitof codethatdoessomething –abasicbuildingblock of aprogram. • An expression is a statement that has a value – for instance, a number, a string, the sum of two numbers, etc. 4+2, x-1, and "Hello, world!\n" are all expressions. Not every statement is an expression. It makes no sense to talk about the value of an #include statement,forinstance. 3.2 Operators Wecanperformarithmeticcalculationswith operators.Operatorsact onexpressionstoform a new expression. For example, we could replace "Hello, world!\n" with (4 + 2) /3, which would cause the program to print the number 2. In this case, the + operator acts on the expressions 4 and 2 (itsoperands). Operatortypes: • Mathematical: +, -, *, /, and parentheses have their usual mathematical meanings, including using -for negation. % (the modulus operator) takes the remainder of two numbers: 6%5 evaluatesto1. • Logical: used for “and,” “or,” and so on. More on those in the next lecture. • Bitwise: used tomanipulatethebinary representationsof numbers. Wewill notfocus onthese. 3.3 Data Types Every expression has a type – a formal description of what kind of data its value is. For instance,0 isaninteger,3.142 isafloating-point (decimal)number,and "Hello, world!\n" 5is a string value(a sequence of characters). Data ofdifferenttypestake adifferent amounts of memory to store. Here are the built-in datatypes we will use most often: Type Names Description Size Range char Single text character or small integer. Indicated with single quotes (’a’, ’3’). 1 byte signed: -128 to 127 unsigned: 0 to 255 int Larger integer. 4 bytes signed: -2147483648 to 2147483647 unsigned: 0 to 4294967295 bool Boolean (true/false). Indicaate with the keywords true and false. 1 byte Just true (1)or false (0). double “Doubly” precise floating point number. 8 bytes +/-1.7e +/-308 ( 15 digits) Notes on this table: • A signed integer is one that can represent a negative number; an unsigned integer will neverbeinterpreted asnegative,soit canrepresent awiderrangeofpositivenumbers. Most compilers assume signed if unspecified. • There are actually 3 integer types: short, int, and long, in non-decreasing order of size(int is usually a synonym for one of the other two). You generally don’t need to worry about which kind to use unless you’re worried about memory usage or you’re using reallyhugenumbers. Thesamegoesforthe3floatingpointtypes,float, double, and long double,which areinnon-decreasing orderofprecision(thereisusually some imprecision in representing real numbers on a computer). • The sizes/ranges for each type are not fully standardized; those shown above are the ones used on most 32-bit computers. An operation can only be performed on compatible types. You can add 34 and 3,butyou can’t take the remainder of an integer and a floating-point number. An operator also normally produces a value of the same type as its operands; thus, 1/4 evaluatesto 0 because with two integer operands, /truncates the result to an integer. To get 0.25, you’d need to write something like 1 /4.0. A text string, for reasons we will learn in Lecture 5, has the type char *. 64 Variables Wemight wanttogiveavalueanamesowecanrefertoitlater. Wedothisusing variables. A variable is a named location in memory. For example, say we wanted to use the value 4+2 multiple times. We might call it x and use it as follows: 1 # include 2 using namespace std ; 3 4 int main () { 5 int x; 6 x = 4 + 2; 7 cout << x /3 << ’ ’ << x * 2; 8 9 return 0; 10 } (Note how we can print a sequence ofvalues by “chaining” the << symbol.) The name of a variable is an identifier token. Identifiers may contain numbers, letters, and underscores( ), andmay not start with a number. Line 4 is the declaration of the variable x. We must tell the compiler what type x willbe so that it knows how much memory to reserve for it and what kinds of operations may be performed on it. Line 5 is the initialization of x, where we specify an initial value for it. This introduces a new operator: =, the assignment operator. We can also change the value of x later on in the code using this operator. We could replace lines 4 and 5 with a single statement that does both declaration and initialization: int x = 4 + 2;This form of declaration/initialization is cleaner, so it is to be preferred. 5 Input Now that we know how to give names to values, we can have the user of the program input values. Thisisdemonstratedinline6below: 7 1 # include 2 using namespace std ; 3 4 int main () { 5 int x; 6 cin >> x; 7 8 cout << x /3 << ’ ’ << x * 2; 9 10 return 0; 11 } Justas cout << isthesyntaxforoutputtingvalues, cin >> (line6)isthesyntaxforinputting values. Memory trick: ifyouhavetrouble remembering which way the anglebracketsgofor cout and cin, think of them as arrows pointing in the direction of data flow. cin representsthe terminal, with data flowing from it to your variables; cout likewise represents the terminal, and your data flows to it. 6 Debugging There are two kinds of errors you’ll run into when writing C++ programs: compilation errors and runtime errors.Compilationerrorsareproblemsraisedbythecompiler,generally resulting from violations of the syntax rules or misuse of types. These are often caused by typos and the like. Runtime errors are problems that you only spot when you run the program: you did specify a legal program, but it doesn’t do what you wanted it to. These are usually more tricky to catch, since the compiler won’t tell you about them. 8MIT OpenCourseWare http://ocw.mit.edu 6.096 Introduction to C++ January (IAP) 2011 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.