Comparing python and C++

Python and C++ have a number of similarities but also important differences that we will encounter when we apply our C++ knowledge to writing python programs.

Similarities

• Both are object-oriented languages, and use the . operator as the member-access operator. We will write classes and work with objects in the same way.

• Both have a robust standard library that provides many of the same features. For example. the python equivalent of std::vector is a list.

• The C++ std::format library was inspired by python, and the same formatting syntax works.

• Floating point math works the same way—both languages follow the same IEEE standard and have single- and double-precision math.

Differences

• python is an interpreted language

  • We run a python program by doing python program.py

  • C++ requires using the compiler (e.g. g++) to translate the source into machine-specific instructions

• python is dynamically typed—we don’t need to specify ahead of time what type of data a variable/object refers to, while all objects in C++ must have their datatypes specified ahead of time

  python

  # a floating point number
  x = 2.5
  
  # an integer
  a = 1
  

  C++

  double x{2.5};
  int a{1};
  

• python statements typically end at the newline.

  C++ uses a semicolon, ;, to terminate a statement

• python uses whitespace to denote code blocks for if, for, functions, etc.

  python

  if x > 0:
       print("x is positive")
  

  C++

  if (x > 0) {
      std::cout << x << std::endl;
  }
  

• integers behave differently in python, in 2 important ways

  • integer division results in a float, i.e. 1 / 2 is 0.5 in python

  • The size of an int (in bytes) expands as needed, so there is no overflow, i.e., we can store a = 123456789012345678901234567890

• python can be slow. Scientific libraries (like NumPy) and avoiding loops can help.

  C++ is generally fast if you use modern language features and compiler optimization.

• python scope works differently than C++.

  python

  x = 10
  if x > 0:
      y = x + 1
  
  # we can access y here
  

  C++

  int x{10};
  if (x > 0) {
      int y = x + 1;
  }
  
  // we cannot access y here
  

• python doesn’t have references and pointers. However, the standard python interpreter is written in C, so those concepts still exist behind the scenes.

  An important way this comes up is in passing objects to functions.

  • In python, we can think of objects as passed by reference, not value (see python data structures).

  • In C++, when we pass objects to a function, they are passed by value, unless we make the function argument a reference (add the &).