Intro to Dynamic Data

• This is chapter 14
  • This is the first part 14.1.
• We have been dealing with static memory.
  • All storage we currently use is allocated on the stack.
  • All storage we currently use is (sort of) known at compile time.
  • Think about those darned arrays.
  • All storate we currnetly use is managed by the compiler.
• A second way to deal with memory is to allocate it at compile time.
  • This is more flexiable, but requires more knowledge.
  • This is an attention to detial thing.
• Dynamic memory is allocated at run time
  • By the user
  • Off the heap.
• Dynamic memory must be managed by the user
  • In particular it must be freed when the user no longer needs it.
• Data
  • Pionters hold pointers to a specific type of data.
  • pointers are addresses, but never think of this again.
• Data types
  • datatype * identifier, * identifier, ...
  • nullptr is a non-pointer value, like string::npos
• Operations on pointers
  • new datatype returns a pointer to newly allocated memory large enough to hold an instance of the data type.
    • If it is a pointer to a class, the constructor is called.
  • new datatype[const size_t n] returns a pointer to newly allocated memory large enough to hold n instances of the data type.
  • delete identifier
  • delete [] i dentifier
    • Assumes identifier is pointing to allocated data.
    • Call the proper one (matches the new command)
    • do not call on nullptr.
• This is an ideal way to deal with an unknown size array.