C++

Abstract

这篇笔记是《C++ Primer》英文版的书摘。中文版我已经读过两遍，但因长时间没有写 C++ 程序而淡忘。我阅读英文版作为复习，同时把一些自己还不太熟的点摘下来。

这篇笔记不适合初学者。

进度表

Current On: Chapter 12.

• Part I: The Basics
• Part II: The C++ Library
  • Ch 8: The IO Library
  • Ch 9: Sequential Containers
  • Ch 10: Generic Algorithms
  • Ch 11: Associative Containers
  • Ch 12: Dynamic Memory
• Part III: Tools for Class Authors
  • Ch 13: Copy Control
  • Ch 14: Overloaded Operations and Conversions
  • Ch 15: Object-Oriented Programming
  • Ch 16: Templates and Generic Programming
• Part IV: Advanced Topics
  • Chapter 17: Specialized Library Facilities
  • Chapter 18: Tools for Large Programs
  • Chapter 19: Specialized Tools and Techniques

Chapter 1 Getting Started

Somethings different from C:

• iostream library:
  • istream, ostream
  • cin, cout, cerr (stderr, not buffered), clog (stderr, buffered)
  • built-in (primitive) type, class type
  • :: scope operator

随手记点单词

• curly brace, open curly, close curly
• parentheses
• underscore
• directive

Part I: The Basics

Chapter 2 Variables and Basic Types

Key

• const Qualifier
• reference

Type Conversions

• If we assign an out-of-range value to an object of unsigned type, the result is the remainder of the value modulo the number of values the target type can hold (wrap around).
• If we assign an out-of-range value to an object of signed type, the result is undefined.

unsigned type

Assigning –1 to an 8-bit unsigned char gives that object the value 255.

Don't mix signed and unsigned types

Specifying the Type of a Literal

Character and character string literals prefix:

• u: char16_t (Unicode 16)
• U: char32_t (Unicode 32)
• L: wchar_t
• u8: char (utf-8, string literal only)

Integer literals suffixes:

• u: unsigned
• l: long
• ll: long long

Floating-point literals suffix:

• f: float
• l: long double

List Initialization (C++ 11)

Braced lists of initializers can now be used whenever we initialize an object and in some cases when we assign a new value to an object.

int units_sold = 0;
int units_sold = {0};
int units_sold{0};
int units_sold(0);

List initialization prevents narrowing conversions. In C++, a narrowing conversion is a potentially unsafe numeric conversion where the destination type may not be able to hold all the values of the source type.

Variables must be defined exactly once but can be declared many times.

Compound Types

More generally, a declaration is a base type followed by a list of declarators. Each declarator names a variable and gives the variable a type that is related to the base type.

A reference or pointer is part of a particular declarator and not part of the base type for the declaration.

• A reference defines an alternative name (alias) for an object.

  int &refVal = ival; // refVal refers to (is another name for) ival
  

  • There is no way to rebind a reference to refer to a different object. References must be initialized.
  • The type of a reference and the object to which the reference refers must match exactly.

  • A reference may be bound only to an object, not to a literal or to the result of a more general expression.

    reference to pointer

    int *p;
    int *&r = p;
    

    read from right to left

  • A reference is not an object. Hence, we may not have a pointer to a reference.

nullptr (C++ 11) is a literal that has a special type that can be converted to any other pointer type. Use it to initialize pointers. NULL is now defined in <cstdlib>.

We cannot use a void* to operate on the object it addresses—we don’t know that object’s type, and the type determines what operations we can perform on the object. Use a void* pointer to deal with memory as memory, rather than using the pointer to access the object stored in that memory.

const Qualifier

To share a const object among multiple files, you must define the variable as extern.

By Default, const Objects Are Local to a File

When a const object is initialized from a compile-time constant, the compiler will usually replace uses of the variable with its corresponding value during compilation.

To substitute the value for the variable, the compiler has to see the variable’s initializer. To support this usage, yet avoid multiple definitions of the same variable, const variables are defined as local to the file.

Sometimes we have a const variable that we want to share across multiple files but whose initializer is not a constant expression. We use the keyword extern on both its definition and declaration(s).

Initialize a reference to const from any expression that can be converted to the type of the reference.

We can bind a reference to const to a nonconst object, a literal, or a more general expression:

const int &r2 = 42; // ok: r1 is a reference to const
const int &r3 = r1 * 2; // ok: r3 is a reference to const

A Reference to const May Refer to an Object That Is Not const.

• Top-level const: an object itself is a const.
• Low-level const: the pointer points to a const object. appears in the base type of compound types such as pointers or references.

When we copy an object, top-level consts are ignored. Low-level const is never ignored.

int *const p1 = &i; // we can’t change the value of p1; const is top-level
const int ci = 42; // we cannot change ci; const is top-level
const int *p2 = &ci; // we can change p2; const is low-level

constexpr Variables

There are several contexts in the language that require constant expressions.

We can ask the compiler to verify that a variable is a constant expression by declaring the variable in a constexpr declaration. Variables declared as constexpr are implicitly const andmust be initialized by constant expressions.

constexpr int mf = 20; // 20 is a constant expression

The address of an object defined outside of any function is a constant expression (stored in fixed address), and so may be used to initialize a constexpr pointer.

constexpr imposes a top-level const on the objects it defines. A constexpr pointer may point to a const or a nonconst type.

Dealing with Types

• new way to define a type alias: using SI = Sales_item. (C++ 11)

Declarations that use type aliases that represent compound types and const can yield surprising results.

• auto (C++ 11): the type of the variable that we are defining will be deduced from the initializer we provide.
  • auto ordinarily ignores top-level consts.
  • If you want so, use const auto.

• decltype (C++ 11): analyzes the expression to determine its type but does not evaluate the expression.

  decltype(f()) sum = x; // sum has whatever type f returns
  

decltype will include top-level consts and references in its deduced type.

decltype is the only context in which a variable defined as a reference is not treated as a synonym for the object to which it refers.

decltype returns a reference type for expressions that yield objects that can stand on the left-hand side of the assignment

If p is a pointer to int, then decltype(*p) is int&, not int.

enclosing the name of a variable in parentheses affects the type returned by decltype

• decltype((variable)) (note, double parentheses) is always a reference type
• decltype(variable) is a reference type only if variable is a reference.

随手记点单词

• variable-langth
• undefined behavior, implementation-defined behavior

Chapter 3 Strings, Vectors, and Arrays

Key

• using
• string
• vector
• Iterators

using Declaration

• A Separate using Declaration Is Required for Each Name

  using std::cout; using std::endl;
  

• Headers Should Not Contain using Declarations

String

Ways to Initialize a string:

string s1; // default initialization; s1 is the empty string
string s2(s1); // s2 is a copy of s1
string s3("value"); // s3 is a copy of the string literal
string s3 = "value"; // equivalent to s3("value")
...

Some operation:

is >> s; // Reads whitespace-separated string into s. Returns is.
getline(is, s); // Reads a line of input from is into s. Return is.
s.empty(); // Returns true if s is empty; otherwise returns false.
...

Stream is valid until it has hit end-of-file or an error.

Used as condition.

getline() don't save the newline character.

string::size_type Type

It is an unsigned type.

For example, if n is an int that holds a negative value, then s.size() < n will almost surely evaluate as true. It yields true because the negative value in n will convert to a large unsigned value.

string library lets us convert both character literals and character string literals to strings.

Range-Based for

for (declaration : expression)
    statement

• expression is an object of a type that represents a sequence
• declaration defines the variable that we’ll use to access the underlying elements in the sequence.

On each iteration, the variable in declaration is initialized from the value of the next element in expression.

decltype(s.size()) punct_cnt = 0;
// count the number of punctuation characters in s
for (auto c : s) // for every char in s
    if (ispunct(c)) // if the character is punctuation
        ++punct_cnt; // increment the punctuation counter
cout << punct_cnt
     << " punctuation characters in " << s << endl;

If we want to change the value of the characters in a string, wemust define the loop variable as a reference type

vector-container and class template

vector is a template, not a type. Types generated from vector must include the element type.

Ways to Initialize a vector:

vector<T> v1; // v1 is empty
vector<T> v2(v1); // v2 is a copy of v1
vector<T> v2 = v1; // equivalent to v2(v1)
vector<T> v3(n, val); // v3 has n elements with value val
vector<T> v4(n); // v4 has n copies of the value-initialized value of T
vector<T> v5{a, b, c...}; // v5 has as many elements as there are initializers
vector<T> v5 = {a, b, c...}; // equivalent to v5{a, b, c...}

Parentheses construct the object, curly braces list initialize the object.

Some operation:

v.push_back(t); // Adds a copy of t to the end of v. Returns void.

The body of a range for must not change the size of the sequence over which it is iterating.

vector<int>::size_type instead of vector::size_type

Iterators

All of the library containers have iterators, but only a few of them support the subscript operator.

A valid iterator either denotes an element or denotes a position one past the last element in a container.

Obtain an iterator:

auto b = v.begin(); // denotes the first element
auto b = v.cbegin(); // const iterator
auto c = cv.begin(); // also const iterator (vector is const)
auto e = v.end(); // denotes one past the last element

of-the-end iterator

If container is empty, begin and end iterator are equal.

Dereference an iterator to obtain the element denoted by an iterator.

*iter;
iter->mem;
++iter;
--iter;
iter1 == iter2;

C++ prefers != to > and <. Use it.

Iterator Types: vector<int>::iterator, vector<int>::const_iterator

Changes the size of a vector potentially invalidates all iterators into that vector.

Some iterator support all relational operators (such as vector and string):

+ - += -= > >= < <=

Subtraction of two iterators result in difference_type, which is signed.

iterator arithmetic

// text must be sorted
// beg and end will denote the range we’re searching
auto beg = text.begin(), end = text.end();
auto mid = text.begin() + (end - beg)/2; // original midpoint
// while there are still elements to look at and we haven’t yet found sought
while (mid != end && *mid != sought) {
    if (sought < *mid) // is the element we want in the first half?
        end = mid; // if so, adjust the range to ignore the second half
    else // the element we want is in the second half
        beg = mid + 1; // start looking with the element just after mid
    mid = beg + (end - beg)/2; // new midpoint
}

Complicated Array Declarations

No array of references. No vector of references.

Reference is not a type.

int *ptrs[10]; // ptrs is an array of ten pointers to int
int &refs[10] = /* ? */; // error: no arrays of references
int (*Parray)[10] = &arr; // Parray points to an array of ten ints
int (&arrRef)[10] = arr; // arrRef refers to an array of ten ints
int *(&arry)[10] = ptrs; // arry is a reference to an array of ten pointers

随手记点单词

• instantiation

Chapter 4 Expressions

Skipped.

Named Casts

A named cast has the following form:

cast-name<type>(expression);

• static_cast:
  • Any well-defined type conversion, other than those involving low-level const, can be requested using a static_cast.
  • useful to perform a conversion that the compiler will not generate automatically.
• dynamic_cast → see dynamic memory

• const_cast:

  const char *pc;
  char *p = const_cast<char*>(pc); // ok: but writing through p is undefined
  

  • A const_cast changes only a low-level const in its operand.
  • the compiler will no longer prevent us from writing to that object.

• reinterpret_cast:

  int *ip;
  char *pc = reinterpret_cast<char*>(ip);
  

  • performs a low-level reinterpretation of the bit pattern of its operands.

Avoid casts, especially reinterpret_cast and old-style casts (type) expr and type (expr).

Chapter 5 Statements

Skipped.

Chapter 6 Functions

Key

• Overloaded Functions
• Reference Parameters
• const Parameters and Arguments
• Return Types

• void f1(); implicit void parameter list.

In C, void no_args() declares a function that takes an unspecified (but not variable) number of parameters.

Arguments

• Pass by reference:

void reset(int &i) {
    i = 0;
}

Reference parameters that are not changed inside a function should be references to const.

Reference parameters let us effectively return multiple results.

• const arguments:

Just as in any other initialization, when we copy an argument to initialize a parameter, top-level consts are ignored. The following two declarations are equivalent:

void fcn(const int i) { /* fcn can read but not write to i */}
void fcn(int i) { /* .. . */} // error: redefines fcn(int)

The above rule is evident because arguments is passed by value except reference arguments which is low-level const.

• Array arguments:

Cannot copy an array. Arrays is (usually) converted to pointer.

// despite appearances, these three declarations ofprintare equivalent
// each function has a single parameter of type constint*
void print(const int*);
void print(const int[]); // shows the intent that the function takes an array
void print(const int[10]); // dimension for documentation purposes (at best)

three common techniques used to manage pointer parameters

• Using a Marker to Specify the Extent of an Array
• Using the Standard Library Conventions: pass begin and end pointers.
• Explicitly Passing a Size Parameter

Array Reference Parameters

f(int (&arr)[10]);

• The parentheses around &arr are necessary.
• Because the size of an array is part of its type, it is safe to rely on the dimension in the body of the function.

Passing a Multidimensional Array

• The size of the second (and any subsequent) dimension is part of the element type and must be specified.

initializer_list Parameters (C++ 11)

an unknown number of arguments of a single type.

initializer_list header.

void error_msg(initializer_list<string> il) {
    for (auto beg = il.begin(); beg != il.end(); ++beg)
        cout << *beg << " ";
    cout << endl;
}

• the elements in an initializer_list are always const values.
• When we pass a sequence of values to an initializer_list parameter, we must enclose the sequence in curly brace.

Ellipsis Parameters

A C library facility named varargs.

• Only for types that are common to both C and C++. Objects of most class types are not copied properly when passed to an ellipsis parameter.
• Appear only as the last element in a parameter list.
• No type checking is done for the arguments that correspond to the ellipsis parameter.

Return Value

Values are returned in exactly the same way as variables and parameters are initialized: The return value is used to initialize a temporary at the call site, and that temporary is the result of the function call.

Never Return a Reference or Pointer to a Local Object

Here is a local temporary string.

return "Empty"; // WRONG: "Empty"is a local temporary string

Calls to functions that return references are lvalues; other return types yield rvalues.

char &get_val(string &str, string::size_type ix) {
    return str[ix]; // get_val assumes the given index is valid 
}
int main() {
    string s("a value"); 
    cout << s << endl; // prints a value 
    get_val(s, 0) = ’A’; // changes s[0] to A 
    cout << s << endl; // prints A value return 0;
}

• List Initializing the Return Value
• The main function may not call itself.

The form of a function that returns a pointer to an array is:

Type (*function(parameter_list))[dimension]

Think about follows

func(int);
(*func(int));
(*func(int))[10];
int (*func(int))[10];

• Trailing Return Type (C++ 11)

Useful for functions with complicated return types

A trailing return type follows the parameter list and is preceded by ->.

auto func(int i) -> int(*)[10];

Overloaded Functions

Functions that have the same name but different parameter lists and that appear in the same scope are overloaded.

• The main function may not be overloaded.
• A type alias does not create a new type. So not overloadable.
• top-level const is ignored. Can overload based on whether the parameter is a reference (or pointer) to the const or nonconst version of a given type.

const_cast is usually useful in overloaded functions:

string &shorterString(string &s1, string &s2)
{
    auto &r = shorterString(const_cast<const string&>(s1),
                            const_cast<const string&>(s2));
    return const_cast<string&>(r);
}

Inner scope name hides uses of that name declared in an outer scope:

string read();
void print(const string &);
void print(double); // overloads the printfunction v
void fooBar(int ival)
{
    bool read = false; // new scope: hides the outer declaration of read
    string s = read(); // error: readis a boolvariable, not a function
    // bad practice: usually it’s a bad idea to declare functions at local scope
    void print(int); // new scope: hides previous instances of print
    print("Value: "); // error: print(const string &)is hidden
    print(ival); // ok: print(int)is visible
    print(3.14);
}

In C++, name lookup happens before type checking.

Notice above all instances in outer scope is hidden.

Specialized Uses

• Default Arguments:

  • The default arguments are used for the trailing (right-most) arguments of a call.

  can omit only trailing arguments

• Default Argument Declarations

  string screen(sz, sz, char = ’*’);
  string screen(sz = 24, sz = 80, char); // ok: adds default arguments
  

  • it is legal to redeclare a function multiple times.
  • each parameter can have its default specified only once in a given scope.
  • any subsequent declaration can add a default only for a parameter that has not previously had a default specified.
  • defaults can be specified only if all parameters to the right already have defaults

• Default Argument Initializers

  • a default argument can be any expression that has a type that is convertible to the type of the parameter
• inline Functions Avoid Function Call Overhead
  • optimize small, straight-line functions that are called frequently
  • putting the keyword inline before the function’s return type
  • The compiler may choose to ignore this request.

• constexpr function is a function that can be used in a constant expression

  // scale(arg)is a constant expression if arg is a constant expression
  constexpr size_t scale(size_t cnt) { return new_sz() * cnt; }
  

  • The return type and the type of each parameter in a must be a literal type, and the function body must contain exactly one return statement.
  • A constexpr function is not required to return a constant expression.
  • A constexpr function is permitted to return a value that is not a constant.

Put inline and constexpr Functions in Header Files

All of the definitions of a given inline or constexpr must match exactly. As a result, inline and constexpr functions normally are defined in headers.

assert and NDEBUG

• __func__: compiler defines, the name of the function in which the call appears.
• __FILE__: preprocessor defines, string literal containing the name of the file
• __LINE__: preprocessor defines, integer literal containing the line number of the file
• __TIME__: preprocessor defines, string literal containing the time of translation
• __DATE__: preprocessor defines, string literal containing the date of translation

Function Matching

• candidate functions: same name, can be seen.
• viable functions: same number of parameters, each argument can be converted to the type of its corresponding parameter.

When a function has default arguments, a call may appear to have fewer arguments than it actually does.

Select the best match:

• The match for each argument is no worse than the match required by any other viable function.
• There is at least one argument for which the match is better than the match provided by any other viable function.

If after looking at each argument there is no single function that is preferable, then the call is in error. The compiler will complain that the call is ambiguous.

Casts should not be needed to call an overloaded function. The need for a cast suggests that the parameter sets are designed poorly.

Conversion ranks:

• exact match:
  • identical type
  • array to pointer
  • top-level const
• const conversion
• promotion: integral promotion, floating-point promotion
• arithmetic or pointer conversion
• class-type conversion

Pointers to Functions

Pointers to Overloaded Functions must match one of the overloaded functions exactly.

Returning a Pointer to Function better use type alias. Try to read this:

int (*f1(int))(int*, int);
auto f1(int) -> int (*)(int*, int);

Using auto or decltype for Function Pointer Types:

string::size_type sumLength(const string&, const string&);
string::size_type largerLength(const string&, const string&); // depending on the value of its string parameter,
// getFcn returns a pointer to sumLengthor to largerLength
decltype(sumLength) *getFcn(const string &);

The only tricky part in declaring getFcn is to remember that when we apply decltype to a function, it returns a function type, not a pointer to function type. We must add a * to indicate that we are returning a pointer, not a function.

Chapter 7 Classes

Key

• this
• Constructors
• Static Class Members

this

Member functions access the object on which they were called through an extra, implicit parameter named this. When we call a member function, this is initialized with the address of the object on which the function was invoked.

this is top-level const by default.

By default, the type of this is a const pointer to the nonconst version of the class type. Thus we cannot bind this to a const object. A const following the parameter list indicates that this is a pointer to const. Member functions that use const in this way are const member functions.

You can think of :

std::string isbn() const { return this->bookNo; }

as:

std::string Sales_data::isbn(const Sales_data, *const this)
{ return this->isbn; }

But the latter is not legal because we may not explicitly define the this pointer ourselves.

Return this object: use reference to this.

Sales_data& Sales_data::combine(const Sales_data &rhs) // return reference
{ return *this; } // return the object on which the function was called

Return reference!

Members

Class is itself a scope. The definitions of the member functions of a class are nested inside the scope of the class itself.

When we define a member function outside the class body, the member’s definition must match its declaration. The name of a member defined outside the class must include the name of the class of which it is a member.

Use the scope operator ::.

double Sales_data::avg_price() const { 
    if (units_sold) 
        return revenue/units_sold;
    else 
        return 0; 
}

Constructors

• Constructors have the same name as the class.
• Constructors have no return type.
• A class can have multiple constructors.
• Constructors may not be declared as const.

When we create a const object of a class type, the object does not assume its "constness" until after the constructor completes the object’s initialization. Thus, constructors can write to const objects during their construction.

Defining outside the class body:

Sales_data::Sales_data(std::istream &is) {
    read(is, *this); // read will read a transaction from is into this object
}

Synthesized Default Constructor

The default constructor is one that takes no arguments. The compiler-generated constructor is known as the synthesized default constructor.

• If there is an in-class initializer , use it to initialize the member.
• Otherwise, default-initialize the member.

The compiler generates a default constructor automatically only if a class declares no constructors.

For some classes, the synthesized default constructor does the wrong thing.

1. Classes that have members of built-in or compound type should ordinarily either initialize those members inside the class or define their own version of the default constructor. Otherwise, users could create objects with members that have undefined value.
2. A class has a member that has a class type, and that class doesn’t have a default constructor, then the compiler can’t initialize that member.

Often we are defining default constructor only because we want to provide other constructors as well as the default constructor

Sales_data() = default; // we already provide initializers for the data members with built-in type.

A constructor that supplies default arguments for all its parameters also defines the default constructor.

Constructor Initializer List

It is usually best for a constructor to use an in-class initializer if one exists and gives the member the correct value.

If your compiler does not support in-class initializers, your default constructor should use the constructor initializer list (described immediately following) to initialize every member of the class.

Sales_data(const std::string &s, unsigned n, double p): bookNo(s), units_sold(n), revenue(p*n) { }

When a member is omitted from the constructor initializer list, it is implicitly initialized using the same process as is used by the synthesized default constructor.

By the time the body of the constructor begins executing, initialization is complete. Our only chance to initialize const or reference datamembers is in the constructor initializer. A class type member that does not have a default constructor also must be initialized.

Members are initialized in the order in which they appear in the class definition.

It is a good idea to write constructor initializers in the same order as the members are declared. Moreover, when possible, avoid using members to initialize other members.

Delegating Constructors

In a delegating constructor, the member initializer list has a single entry that is the name of the class itself.

// remaining constructors all delegate to another constructor
Sales_data(): Sales_data("", 0, 0) {}
Sales_data(std::string s): Sales_data(s, 0,0) {}
Sales_data(std::istream &is): Sales_data()
    { read(is, *this); }

When a constructor delegates to another constructor, the constructor initializer list and function body of the delegated-to constructor are both executed.

The Role of the Default Constructor

Default initialization happens:

• nonstatic variables or Arrays at block scope without initializers.
• a class that itself has members of class type uses the synthesized default constructor
• members of class type are not explicitly initialized in a constructor initializer list

Value initialization happens:

• array initialization when we provide fewer initializers than the size of the array
• local static object without an initializer
• request value initialization by writing an expressions of the form T() where T is the name of a type

It is a common mistake among programmers new to C++ to try to declare an object initialized with the default constructor as follows:

Sales_data obj(); // oops! declares a function, not an object 
Sales_data obj2; // ok: obj2is an object, not a function

Implicit Class-Type Conversions

Every constructor that can be called with a single argument defines an implicit conversion to a class type.

Only One Class-Type Conversion Is Allowed

// error: requires two user-defined conversions: 
// (1) convert "9-999-99999-9"to string
// (2) convert that (temporary) stringto Sales_data 
item.combine("9-999-99999-9");

We can prevent the use of a constructor in a context that requires an implicit conversion by declaring the constructor as explicit. The explicit keyword is meaningful only on constructors that can be called with a single argument. The explicit keyword is used only on the constructor declaration inside the class.

Implicit conversions happens when we use the copy form of initialization (with an =).

explicit Constructors Can Be Used Only for Direct Initialization.

• string constructor that takes a single parameter of type const char* is not explicit.
• vector constructor that takes a size is explicit.

Copy, Assignment, and Destruction

If we do not define these operations, the compiler will synthesize them for us. Ordinarily, the versions that the compiler generates for us execute by copying, assigning, or destroying each member of the object.

In particular, the synthesized versions are unlikely to work correctly for classes that allocate resources that reside outside the class objects themselves.

Access Control and Encapsulation

access specifiers to enforce encapsulation:

• public: members are accessible to all parts of the program.
• private: members are accessible to the member functions of the class.

The only difference between struct and class is the default access level. If we use the struct keyword, the members defined before the first access specifier are public; if we use class, then the members are private.

Friends

A class can allow another class or function to access its nonpublic members by making that class or function a friend.

Friend declarations may appear anywhere inside a class definition:

class Sales_data { // friend declarations for nonmember Sales_dataoperations added
    friend Sales_data add(const Sales_data&, const Sales_data&);
    friend std::istream &read(std::istream&, Sales_data&);
    friend std::ostream &print(std::ostream&, const Sales_data&);
}

A friend declaration only specifies access. It is not a general declaration of the function.

Additional Class Features

• Defining a Type Member
• member functions defined inside the class are automatically inline
• mutable Data Members is never const, a const member function may change a mutable member.
• Initializers for Data Members of Class Type (C++ 11). When we provide an in-class initializer, we must do so following an = sign or inside braces.

Overloading Based on const

Screen &display(std::ostream &os) 
    { do_display(os); return *this; }
const Screen &display(std::ostream &os) const 
    { do_display(os); return *this; }
void do_display(std::ostream &os) const {os << contents;}

USE PRIVATE UTILITY FUNCTIONS FOR COMMON CODE

Forward Declaration

After a declaration and before a definition is seen, the type Screen is an incomplete type—it’s known that Screen is a class type but not known what members that type contains.

We can use an incomplete type in only limited ways: We can define pointers or references to such types, and we can declare (but not define) functions that use an incomplete type as a parameter or return type.

• Friendship between Classes
• Making A Member Function a Friend
  • First, define the Window_mgr class, which declares, but cannot define, clear. Screen must be declared before clear can use the members of Screen.
  • Next, define class Screen, including a friend declaration for clear.
  • Finally, define clear, which can now refer to the members in Screen.

Good practice: don’t use a member name for a parameter or other local variable

Aggregate Classes and Literal Classes

An aggregate class gives users direct access to its members and has special initialization syntax. A class is an aggregate if:

• All of its data members are public
• It does not define any constructors
• It has no in-class initializers
• It has no base classes or virtual functions

We can initialize the data members of an aggregate class by providing a braced list of member initializers.

An aggregate Class whose data members are all of literal type is a literal class. A nonaggregate class, that meets the following restrictions, is also a literal class.

Although constructors can’t be const, constructors in a literal class can be constexpr functions. Indeed, a literal class must provide at least one constexpr constructor.

static Class Members

Members that are associated with the class, rather than with individual objects of the class type.

The static members of a class exist outside any object. Objects do not contain data associated with static data members.

r = Account::rate(); // access a staticmember using the scope operator
r = ac1.rate(); // through an Account object or reference
r = ac2->rate(); // through a pointer to an Account object

They are not initialized by the class’ constructors. We may not initialize a static member inside the class. We must define and initialize each static data member outside the class body.

double Account::interestRate = initRate();

Even if a const static data member is initialized in the class body, that member ordinarily should be defined outside the class definition with no initializer.

static Members Can Be Used in Ways Ordinary Members Can’t. A static data member can have incomplete type. We can use a static member as a default argument.

Similarly, static member functions are not bound to any object; they do not have a this pointer. As a result, static member functions may not be declared as const, and we maynot refer to this in the body of a static member.

随手记点单词

• encapsulation

Part II: The C++ Library

Abstract

Central to the library are a number of container classes and a family of generic algorithms that let us write programs that are succinct and efficient.

Chapter 8 The IO Library

Headers:

• iostream: istream, wistream, ostream, wostream, iostream, wiostream
• fstream: ifstream, wifstream, ofstream, wofstream, fstream, wfstream
• sstream: istringstream, wistringstream, ostringstream, wostringstream, stringstream, wstringstream

The names of the widecharacter versions begin with a w.

The library lets us ignore the differences among these different kinds of streams by using inheritance.

Example

The types ifstream and istringstream inherit from istream.Thus, we can use objects of type ifstream or istringstream as if they were istream objects.

No Copy or Assign for IO Objects

We can read from or write to a stream only when it is in a non-error state.

Interrogating the State of a Stream

• iostate Type
• Functions: .eof(), .fail(), .bad(), .good().
• Recover: .clear(), .clear(flags), .setstate(flags), .rdstate().

// turns off failbit and badbit but all other bits unchanged 
cin.clear(cin.rdstate() & ~cin.failbit & ~cin.badbit);

Each output stream manages a buffer. Several conditions that cause the buffer to be flushed:

• The program completes normally.
• the buffer can become full.
• using a manipulator such as endl.
• use the unitbuf manipulator to set the stream’s internal state to empty the buffer after each output operation.
• An output stream might be tied to another stream. In this case, the output stream is flushed whenever the stream to which it is tied is read or written. By default, cin and cerr are both tied to cout. Hence, reading cin or writing to cerr flushes the buffer in cout.

When an input stream is tied to an output stream, any attempt to read the input stream will first flush the buffer associated with the output stream.

Interactive systems usually should tie their input stream to their output stream. Doing so means that all output, which might include prompts to the user, will be written before attempting to read the input.

• tie(): return the ostream to which this istream is tied.
• tie(&ostream): take a pointer to ostream and tie to this ostream, return the old tie.

Each stream can be tied to at most one stream at a time. However, multiple streams can tie themselves to the same ostream.

manipulators

• endl: newline, flush
• flush: flush
• ends: null, flush
• unitbuf, nounitbuF: flush or not.

File Stream

fstream fstrm(s);
fstream fstrm(s, mode);
fstrm.open(s);
fstrm.open(s, mode);
fstrm.close();
fstrm.is_open();

Calling open on a file stream that is already open will fail and set failbit. Subsequent attempts to use that file stream will fail.

When an fstream object is destroyed, close is called automatically.

File modes:

• in: read
• out: write
• app: append
• ate: seek to the end of the stream immediately after opening the file
• trunc: truncate the file, only when out is set
• binary: binary mode

String Stream

sstream strm(s);
strm.str(); // return a copy of the string
strm.str(s); // copy s into strm

An istringstream is often used when we have some work to do on an entire line.

string line, word; // will hold a line and word from input, respectively
vector<PersonInfo> people; // will hold all the records from the input
// read the input a line at a time until cinhits end-of-file (or another error)
while (getline(cin, line)) {
    PersonInfo info; // create an object to hold this record’s data
    istringstream record(line); // bind recordto the linewe just read
    record >> info.name; // read the name
    while (record >> word) // read the phone numbers
        info.phones.push_back(word); // and store them
    people.push_back(info); // append this record to people
}

An ostringstream is useful when we need to build up our output a little at a time but do not want to print the output until later.

for(const auto &entry : people) { // for each entry in people
    ostringstream formatted, badNums; // objects created on each loop
    for (const auto &nums : entry.phones) { // for each number
        if (!valid(nums)) {
            badNums << " " << nums; // string in badNums
        } else
            // “writes” to formatted’s string
            formatted << " " << format(nums);
    }
    if (badNums.str().empty()) // there were no bad numbers
        os << entry.name << " " // print the name
            << formatted.str() << endl; // and reformatted numbers
    else // otherwise, print the name and bad numbers
        cerr << "input error: " << entry.name
            << " invalid number(s) " << badNums.str() << endl;
}

Chapter 9 Sequential Containers

Abstract

• vector
  • random access
  • fast insert/delete at the end
• deque
  • random access
  • fast insert/delete at front or back
• list: doubly linked list
  • fast insert and delete at any point in the sequence
  • bidirectional sequential access
  • substantial memory overhead
• forward_list: singly linked list
  • fast insert and delete at any point in the sequence
  • sequential access in one direction
  • no size operation
  • substantial memory overhead
• array: fixed-size
  • cannot remove elements
• string: similar to vector

Each container is defined in a header file with the same name as the type.

Container Operations

Contatiner Operations

• Type Aliases: useful in generic programming
  • iterator
  • const_iterator
  • size_type
  • difference_type
  • value_type
  • reference
  • const_reference
• construction
  • C c;
  • C c1(c2);
  • C c(b, e);
  • C c{a, b, c};
• assignment and swap
  • c1 = c2
  • a.swap(b)
  • swap(a, b)
• Add/Remove Elements
  • c.insert(args)
  • c.emplace(inits)
  • c.erase(args)
  • c.clear()
• Equality and Relational Operators
  • ==, !=, <, <=, >, >=
• Obtain Iterators
  • c.begin(), c.end()
  • c.cbegin(), c.cend()
• Additional Members of Reversible Containers
  • reverse_iterator
  • const_reverse_iterator
  • c.rbegin(), c.rend()
  • c.crbegin(), c.crend()

Iterator Ranges

An iterator range is denoted by a pair of iterators each of which refers to an element, or to one past the last element, in the same container.

This element range is called a left-inclusive interval. The standard mathematical notation for such a range is

\[ [begin, end) \]

Three convient properties:

• If begin equals end, the range is empty
• If begin is not equal to end, there is at least one element in the range, and begin refers to the first element in that range
• We can increment begin some number of times until begin == end

while (begin != end) { 
    *begin = val; // ok: range isn’t empty so begindenotes an element 
    ++begin; // advance the iterator to get the next element 
}

Initializing a Container as a Copy of Another Container:

• To create a container as a copy of another container, the container and element types must match.
• When we pass iterators, there is no requirement that the container types be identical. as long as it is possible to convert.
  • As usual, the iterators mark the first and one past the last element to be copied

the size of a library array is part of its type.

Library array is:

• Default initialized
• If there are fewer initializers than the size of the array, any remaining elements are value initialized.

Although we cannot copy or assign objects of built-in array types, there is no such restriction on array.

Assign

Abstract

sqe.assign(b, e);
sqe.assign(il);
seq.assign(n, t);

Assignment related operations invalidate iterators, references, and pointers into the left-hand container. Aside from string they remain valid after a swap,and (excepting arrays) the containers to which they refer are swapped.

• assignment operator requires that the left-hand and right-hand operands have the same type. It copies all the elements from the right-hand operand into the lefthand operand.
• assign lets us assign from a different but compatible type, or assign from a subsequence of a container. replaces all the elements in the left-hand container with (copies of) the elements specified by its arguments.

list<string> names;
vector<const char*> oldstyle;
names = oldstyle; // error: container types don’t match
// ok: can convert from const char* to string
names.assign(oldstyle.cbegin(), oldstyle.cend());

Because the existing elements are replaced, the iterators passed to assign must not refer to the container on which assign is called.

Swap

The swap operation exchanges the contents of two containers of the same type. After the call to swap, the elements in the two containers are interchanged.

Excepting array, swapping two containers is guaranteed to be fast— the elements themselves are not swapped; internal data structures are swapped. guaranteed to run in constant time.

with the exception of string, iterators, references, and pointers into the containers are not invalidated. They refer to the same elements as they did before the swap. However, after the swap, those elements are in a different container.

Relational Operators

Comparing two containers performs a pairwise comparison of the elements. These operators work similarly to the string relationals

Relational Operators Use Their Element’s Relational Operator. We can use a relational operator to compare two containers only if the appropriate comparison operator is defined for the element type.

Sequential Container Operations

Container Operations May Invalidate Iterators

Refer to book.

Add

• push_back, emplace_back
  • aside from array and forward_list
• push_front, emplace_front
  • not valid for vector, array or string

Specified Point:

c.insert(p, t);
c.emplace(p, args);
c.insert(p, n, t);
c.insert(p, b, e);
c.insert(p, il);
// all insert before p
// return iterator to first element inserted

Note it returns an iterator to the first element inserted. Try to understand this:

list<string> lst;
auto iter = lst.begin();
while (cin >> word)
    iter = lst.insert(iter, word); // same as calling push_front

Adding elements to a vector, string, or deque potentially invalidates all existing iterators, references, and pointers into the container.

Access

• front, back return reference to elements
• c[n], c.at(n) return reference to elements
  • at throws out_of_range exception
  • c[n] is undefined if n >= c.size()

Don't call front or back on an empty container.

Erase

• pop_back()
  • not valid for forward_list
• pop_front()
  • not valid for vector, array or string

c.erase(p);
c.earse(b, e);
c.clear();
// returns an iterator to the element after the one deleted or the off-the-end iterator if p denotes the last element.

Removing elements anywhere but the beginning or end of a deque invalidates all iterators, references, and pointers. Iterators, references, and pointers to elements after the erasure point in a vector or string are invalidated.

Resize

c.resize(n);
c.resize(n, t);

If resize shrinks the container, then iterators, references, and pointers to the deleted elements are invalidated.

forward_list

lst.before_begin();
lst.cbefore_begin();
lst.insert_after(p, t);
lst.insert_after(p, n, t);
lst.insert_after(p, b, e);
lst.insert_after(p, il);
emplace_after(p, args);
lst.erase_after(p); // Returns an iterator to the element after the one deleted
lst.erase_after(b, e); // from the one after the iterator b up to but not including the one denoted by e

When we add or remove elements in a forward_list, wehaveto keep track of two iterators—one to the element we’re checking and one to that element’s predecessor.

vector and string

part of the implementation leaks into its interface:

• Container Size Management
  • c.shrink_to_fit(), no guarantee to reduce capacity
  • c.capacity()
  • c.reserve(n), will never shrink

Understand the difference between size and capacity.

Also notice that resize() member do not change the capacity of the container, just change number of elements.

String supports addtional operations:

• Construct:

  string s(cp, n);
  string s(s2, pos2);
  string s(s2, pos2, len2);
  

• Substr: cpp string s2 = s.substr(pos, len); string s3 = s.substr(pos); In addition to the versions of insert and erase that take iterators, string provides versions that take an index. The string library also provides versions of insert and assign that take Cstyle character arrays.

The string class defines two additional members, append and replace,that can change the contents of a string.

• Search:

  s.find(args);
  s.rfind(args);
  s.find_first_of(args);
  s.find_last_not_of(args);
  ...
  

args can be:

• cp, pos, n: Look for the first n characters in the array pointed to by the pointer cp. Start looking at position pos in s. No default for pos or n.
• c, pos: Look for the character c starting at position pos in s. pos defaults to 0.

If there is no match, the function returns a staticmember named string::npos.

• compare function also has many overloaded versions.

• Numeric Conversions

  to_string(val);
  stoi(s, p, b);
  stol(s, p, b);
  stoul(s, p, b);
  stof(s, p);
  stod(s, p);
  ...
  

Container Adaptors

Three sequential container adaptors: stack, queue,and priority_queue. A container adaptor takes an existing container type and makes it act like a different type.

stack<int> stk(deq); // copies elements from deq into stk

By default:

• both stack and queue are implemented in terms of deque
• a priority_queue is implemented on a vector.

// empty stackimplemented on top ofvector
stack<string, vector<string>> str_stk;

stack

s.pop();
s.push(item);
s.emplace(args);
s.top();

queue and priority_queue

q.pop();
q.push(item);
q.emplace(args);
q.front();
q.back(); // only for queue
q.top(); // only for priority_queue

Chapter 10 Generic Algorithms

Abstract

• Header Files: <algorithm> <numeric><functional><iterator>

• Algorithms:

  find, count, accumulate, equal
  fill, fill_n, copy, replace, replace_copy
  back_inserter(container)
  sort, unique
  stable_sort, partition
  find_if
  for_each
  transform
  count_if
  unique_copy
  replace, replace_if, replace_copy, replace_copy_if
  

• Container-Specifc Algorithm as members: list and forward_list

  sort, merge, remove, reverse, unique
  splice
  

• Lambda expression and Bind() function:

  [capture list](parameter list) -> return type {function body};
  []{return 42;};
  bind(callable, arg_list);
  ref(); cref();
  vector<int> vec(in_iter, eof);
  

• Iterator:

  • adaptor: insert, stream, reverse, move
  • back_inserter, front_inserter, inserter
  • (template) istream_iterator, ostream_iterator, default initialize: end iterator value

  • rbegin, rend, crbegin, crend, .base() adjacent positions

    Understand left-inclusive range and asymmetric

  • categorize by operations required by algorithms: input, output, forward, bidirectional, random-access

    • single pass: may invalidate all other iterators into the stream
    • random-access: subscript iter[n] synonym for *(iter + n)
    • some auxiliary function: distance()

• Misc:

  • C-style string equal
  • container capacity, reserve, resize operation
  • unique overwrites but not reorder

Chapter 11 Associative Containers

Two primary types:

• map key-value pairs (array)
• set keys

Eight associative containers:

• map or set
• unique or multi
• ordered or unordered (often Hash)

Common feature:

• bidirectional iterators
• initialize as copy of another container (vector for example)

Key constraints:

• ordered: must define a way to compare elements, strict weak ordering <

  Example of comparison function

  multiset<Sales_data, decltype(compareIsbn) *> bookstore(compareIsbn);
  

  Function pointer type. Supply the function to the constructor.

Aliases:

key_type
mapped_type
value_type // pair for map, key is const

Iterator:

• keys are const

  • For example, both iterator and const_iterator for set are read-only.

  // notice the function const arguments
  multiset<Sales_data, bool (*)(const Sales_data &, const Sales_data &>::iterator it = bookstore.begin();
  

• ascending order

!!! note "In general, we do not use the generic algorithms with the associative containers"

Insert:

• insert returns a pair of an iterator and a bool indicating whether the insertion took place.

  auto ret = word_count.insert({word, 1});
  if (!ret.second)
      ++ret.first->second;
  

• For multi containers, don't return bool because insert always success.

Erase：

• key_type argument version return the number of elements removed

must not be equal to the end iterator

Find:

• find, count
• lower_bound, upper_bound, equal_range

Unordered Containers

• use == operator, hash function
• map hash to a bucket
• object hash<key_type> to generate hash value
  • built-in types and string, smart pointer has hash function

customized hash function see Chap.16

Pair

{v1, v2}; // easiest after c++11
make_pair(v1, v2);
p1 < p2; // compare first, then second
p1 == p2; // respectively equal
return pair<string, int>(v1, v2);

Map

• m[k] returns value
  • if key is not present, new element will be inserted (value initialized)
  • type returned differ from dereferencing iterator
• m.at(k) will throw out_of_range exception

Can't be used on const map

Only map and unordered_map has [] operator and at function

• range for fetch template type pair, which has first and second members

Chapter 12 Dynamic Memory

We've used automatic (saved in stack memory) and static (saved in static memory) objects. C++ lets us allocate objects dynamically

Dynamically allocated objects have a lifetime that is independent of where they are created; they exist until they are explicitly freed. They are stored in heap memory.

Library provides two smart pointer types that manage dynamic objects. A smart pointer acts like a regular pointer with the important exception that it automatically deletes the object to which it points.

Common operations:

p.get(); // return a pointer
swap(p, q);
p.swap(q);
shared_ptr<T> p(q, d);

shared_ptr

shared_ptr<string> p1;
auto p6 = make_shared<vector<string>>();
auto q(p);
p.unique();
p.use_count();

It's a template.

Allows multiple pointers to refer to the same object. Think of a shared_ptr as if it has an associated counter.

• count is incremented when
  • copied
  • used as right-hand operand of assignment
  • pass to or
  • return from a function by value.
• decremented when
  • assign new value to
  • destroyed.

Once a shared_ptr’s counter goes to zero, the shared_ptr automatically frees the object that it manages.

• Cannot convert ordinary Pointers to shared_ptrs implicitly. Use initializer: shared_ptr<int> p2(new int(42));.

initialize a smart pointer must point to dynamic memory because use delete to free the associated object

Use make_shared function to allocate and use dynamic memory.

• It returens a shared_ptr.
• It uses its arguments to construct an object of the given type.
  • Value initialization if no arguments.
• Use auto to avoid type name.

Value Initialization

Value Initialization is similar to default initialization. In the case of built-in types the difference is significant; a value-initialized object of built-in type has a well-defined value but a default-initialized object does not.

We can usually omit the value and supply only a size. In this case the library creates a value-initialized element initializer for us. This library-generated value is used to initialize each element in the container. The value of the element initializer depends on the type of the elements stored in the vector.

If the vector holds elements of a built-in type, such as int, then the element initializer has a value of 0. If the elements are of a class type, such as string, then the element initializer is itself default initialized.

Memory Leak: shared_ptrs in a container

If you put shared_ptrs in a container, and you subsequently need to use some, but not all, of the elements, remember to erase the elements you no longer need.

shared_ptr is usually used to allow multiple objects to share the same state (refer to same object when copied).

StrBlob

class StrBlob { 
public:
    typedef std::vector<std::string>::size_type size_type; 
    StrBlob();
    StrBlob(std::initializer_list<std::string> il); 
    size_type size() const { return data->size(); } 
    bool empty() const { return data->empty(); } 
    void push_back(const std::string &t) {data->push_back(t);} 
    void pop_back(); 
    std::string& front(); 
    std::string& back();
private:
    std::shared_ptr<std::vector<std::string>> data; 
    void check(size_type i, const std::string &msg) const;
};

Uses the default versions of the operations that copy, assign, and destroy objects of its type.

Direct manage memory

int *pi = new int;
int *pi = new int(1024);
vector<int> *pv = new vector<int>{0,1,2,3,4,5,6,7,8,9};
string *ps = new string(); // value initialized
string *ps1 = new string; // default initialized
auto p1 = new auto(obj);
const string *pcs = new const string;
int *p1 = new int; // ifallocation fails, new throws std::bad_alloc
int *p2 = new (nothrow) int; // if allocation fails, new returns a null pointer

Two operators: new, delete.

• new returns a pointer to allocated memory.
  • can use direct initialization, traditional construction (parentheses) or list initialization.

Caller is responsible for deleting the memory.

assign nullptr to the pointer after we use delete

Provides Only Limited Protection: There can be several pointers that point to the same memory.

Part III: Tools for Class Authors

Chapter 13 Copy Control

Five special member functions: copy constructor, copy-assignment operator, move constructor, move-assignment op- erator, and destructor.

• copy and move constructor: initialize new object from another object of the same type
• copy and move assignment: assign an object of a class type to another object of that same class type

Copy Constructor

its first parameter is a reference to the class type and any additional parameters have default values.

Foo(const Foo&);

Copy initialization

string dots(10, ’.’); // direct initialization
string s(dots); // direct initialization
string s2 = dots; // copy initialization
string null_book = "9-999-99999-9"; // copy initialization
string nines = string(100, ’9’); // copy initialization

When copy initialization happens and that copy initialization requires either the copy constructor or the move constructor.

• Pass an object as an argument
• Return
• Brace initialize
• library containers allocate objects

Chapter 14 Overloaded Operations and Conversions

Chapter 15 Object-Oriented Programming

Chapter 16 Templates and Generic Programming

Part IV: Advanced Topics

Chapter 17 Specialized Library Facilities

Chapter 18 Tools for Large Programs

Exception Handling

• throw expression
• try block
• catch clause
• exception class
  • what member function: return C-style string

Functions Are Exited during the Search for a Handler

If no appropriate catch is found, execution is transferred to a library function named terminate. The behavior of that function is system dependent but is guaranteed to stop further execution of the program.

Standard Exception Classes

Default initialization:

• exception
• bad_alloc
• bad_cast
• bad_typeid
• bad_exception

Cannot be default initialized:

• runtime_error
  • range_error
  • overflow_error
  • underflow_error
• logic_error
  • domain_error
  • invalid_argument
  • length_error
  • out_of_range

Stack Unwind

Because destructors may be invoked during stack unwinding, they should never throw exceptions that the destructor itself does not handle.

All of the standard library types guarantee that their destructors will not raise an exception.

Function try Blocks and Constructors

An exception might occur while processing a constructor initializer.

template <typename T>
Blob<T>::Blob(std::initializer_list<T> il) try :
data(std::make_shared<std::vector<T>>(il)) {
/*empty body*/
} catch(const std::bad_alloc &e) { handle_out_of_memory(e); }

Chapter 19 Specialized Tools and Techniques

零碎技巧

• 被换行符困扰？使用getline函数，或者在>>后面加上ws。
  • 注意 getline 不会跳过前导空白字符。

while (map_file >> key && getline(map_file, value))