Why don't win32 API functions have overloads and instead use Ex as suffix?

The win32 API has for example two methods StrFormatByteSize and StrFormatByteSizeEx. Even though both the methods symantically do the same thing and the Ex counter part only offers a new parameter to slightly change the behavior then couldn't they have two overloads of the same function?

Is it a limitation of c/c++ or wha开发者_如何学Pythont is the possible reason for such an awkward convention?

---

The Win32 API is a C (not C++) API. The C language doesn't support overloaded functions.

---

Complete aside: The Win32 API uses __stdcall-decorated functions, which include the number of bytes of parameters as part of the function name. __stdcall is not part of the C language, but Windows linkers have to know about it.

Microsoft could have used this to implement some kind of overloading, but (since a lot of languages don't understand overloading) that would limit the number of languages that could be used to program Windows.

---

The C language doesn't support function overloading at all.

---

Is it a limitation of c/c++ or what is the possible reason for such an awkward convention?

Yes, and the reason that C doesn't support overloading functions is because, name mangling(conversion of function names to be used by the linker) used for standard C doesn't account for its function parameters.
I.e. void func(int) in C gets mangled to _func so you cannot have func(int) and func(bool) together, as both will be converted to _func.

Whereas in C++, the mangled name for a function accounts for all its function parameters, but as name mangling in C++ was not standardized, the name mangling is compiler dependent.

One more thing to keep in mind is C++ doesn't consider the return parameter of the function in the mangled name. Hence, one cannot have overloaded functions as void func(int) and bool func(int) together.

--Samrat Patil

---

Microsoft never bothered.

Sure, a few people here say that C doesn't support overloading. That's irrelevant. The API already uses overloading, C-style. For instance, the StrFormatByteSize function you mentioned really has two overloads: LPSTR StrFormatByteSizeA(DWORD dw, LPSTR pszBuf, UINT cchBuf) and LPWSTR StrFormatByteSizeW( LONGLONG qdw, LPWSTR pszBuf, UINT cchBuf);. The problem with this mechanism is of course that it generalizes poorly to the various _Ex suffixes.

Microsoft could have added a header which provides StrFormatByteSize as two inlined C++ functions, instead of C macros. Had they done so, it would have been easy to add the third overload for the _Ex suffix. There's no such C++ header, though, and therefore no C++ overloads at all.