Bulldozer00's Blog

For grins, I perused a few books written by several C++ mentors that I respect and admire. I was interested in the naming conventions that they personally use when writing code. Here are the results.

Scott Meyers (Effective C++):

class names – class PhoneBook {};

member function names – void addPhoneNumber(const std::string& pn);

member attribute names – std::string theAddress;

Note: no annotation to distinguish class member attributes from local function variables.

Bjarne Stroustrup (The C++ Programming Language):

“I consider the CamelCodingStyle of identifiers “pug ugly” and strongly prefer underscore_style as cleaner and inherently more readable, and many people agree. On the other hand, many reasonable people disagree.“

class names – class Phone_book {};

member function names – void add_phone_number(const std::string& pn);

member attribute names – std::string the_address;

Note: no annotation to distinguish class member attributes from local function variables.

Herb Sutter and Andrei Alexandrescu (C++ Coding Standards):

class names – class PhoneBook {};

member function names – void AddPhoneNumber(const std::string& pn);

member attribute names – std::string theAddress_;

Note: they use post-underscores to distinguish class member attributes from local function variables (int clientName_;  //is a class member)

When I’m not constrained by a specific naming standard, I use this one:

class names – class PhoneBook {};

member function names – void add_phone_number(const std::string& pn);

member attribute names – std::string the_address_;

Note: Like Herb & Andrei, I use post-underscores to distinguish class member attributes from local function variables (int client_name_; //is a class member).