Thread: A "Naming Conventions" question

There are no naming conventions in C++, as such. Different entities (persons, companies, or even programming cultures) have conventions.

So, for example, the CamelCase convention you describe for classes is used by many companies, projects and cultures (e.g. QT, with a Q prefix, wxWidgets, with a wx prefix, or the Java culture), but it is not inherent to C++.

C++'s standard library is named using the K&R C culture, where mixed case is non-existent and upper case is reserved for macros and constants (with the very atypical exception of the FILE type). Therefore, it's string and runtime_error and copy_backwards. Note that C++, having come into existence in a time where memory was no longer measured in two-digit kilobyte numbers and where compilers thus allowed names longer than perhaps 10 characters, stopped using abbreviations for names and thus has no new names like atoi ("asciiz to integer"), strstr ("find string in string") or execvp ("load executable with a pointer to a variable number of arguments").

Originally Posted by medievalelks

Just be consistent with whatever you choose. And if you're editing someone else's code, follow their conventions.

Of course, when you're writing anything non-trivial, you'll inevitably get convention clashes in libraries. CamelCase, snake_case, and if you're unlucky enough to be doing Windows programming, the scourge of the programming world, convHungarianNotation.

So true...