Typedefs allow the programmer to create an alias for a data type, and use the aliased name instead of the actual type name. Typedef literally stands for, “type definition”.
To declare a typedef, simply use the typedef keyword, followed by the type to alias, followed by the alias name:
|
1 2 3 4 5 |
typedef double distance_t; // define distance_t as an alias for type double // The following two statements are equivalent: double howFar; distance_t howFar; |
By convention, typedef names are declared using a “_t” suffix. This helps indicate that the identifier represents a type, not a variable or function, and also helps prevent naming collisions with other identifiers.
Note that a typedef does not define a new type. Rather, it is simply an alias (another name) for an existing type. A typedef can be used interchangeably anywhere a regular type can be used.
Even though the following does not make sense semantically, it is valid C++:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 |
int main() { typedef long miles_t; typedef long speed_t; miles_t distance { 5 }; speed_t mhz { 3200 }; // The following is valid, because distance and mhz are both actually type long distance = mhz; return 0; } |
typedefs and type aliases follow the same scoping rules as variables. The typedefs miles_t and speed_t are only usable in the main() function. If they were placed in another function, main() wouldn’t be able to access them. If they were placed outside of main(), in the global scope, all functions would be able to access them.
However, typedefs have a few issues. First, it’s easy to forget whether the type name or type definition come first. Which is correct?
|
1 2 |
typedef distance_t double; // incorrect typedef double distance_t; // correct |
I can never remember.
Second, the syntax for typedefs gets ugly with more complex types, particularly function pointers (which we will cover in future lesson 10.9 -- Function Pointers):
Type aliases
To help address these issues, an improved syntax for typedefs has been introduced that mimics the way variables are declared. This syntax is called a type alias.
Given the following typedef:
|
1 |
typedef double distance_t; // define distance_t as an alias for type double |
This can be declared as the following type alias:
|
1 |
using distance_t = double; // define distance_t as an alias for type double |
The two are functionally equivalent.
Note that although the type alias syntax uses the “using” keyword, this is an overloaded meaning, and does not have anything to do with the using statements related to namespaces. As such, it is not affected by the rule of not using “using namespace”.
This type alias syntax is cleaner for more advanced typedefing cases, and should be preferred.
Using type aliases for legibility
One use for type aliases is to help with documentation and legibility. Data type names such as char, int, long, double, and bool are good for describing what type a function returns, but more often we want to know what purpose a return value serves.
For example, consider the following function:
|
1 |
int GradeTest(); |
We can see that the return value is an integer, but what does the integer mean? A letter grade? The number of questions missed? The student’s ID number? An error code? Who knows! Int does not tell us anything.
|
1 2 |
using testScore_t = int; testScore_t GradeTest(); |
However, using a return type of testScore_t makes it obvious that the function is returning a type that represents a test score.
Using type aliases for easier code maintenance
Type aliases also allow you to change the underlying type of an object without having to change lots of code. For example, if you were using a short to hold a student’s ID number, but then later decided you needed a long instead, you’d have to comb through lots of code and replace short with long. It would probably be difficult to figure out which shorts were being used to hold ID numbers and which were being used for other purposes.
However, with a type alias, all you have to do is change using studentID_t = short; to using studentID_t = long;. However, caution is still necessary when changing the type of a type alias to a type in a different type family (e.g. an integer to a floating point value, or vice versa)! The new type may have comparison or integer/floating point division issues, or other issues that the old type did not.
Using type aliases for platform independent coding
Another advantage of type aliases is that they can be used to hide platform specific details. On some platforms, an int is 2 bytes, and on others, it is 4 bytes. Thus, using int to store more than 2 bytes of information can be potentially dangerous when writing platform independent code.
Because char, short, int, and long give no indication of their size, it is fairly common for cross-platform programs to use type aliases to define aliases that include the type’s size in bits. For example, int8_t would be an 8-bit signed integer, int16_t a 16-bit signed integer, and int32_t a 32-bit signed integer. Using type aliases in this manner helps prevent mistakes and makes it more clear about what kind of assumptions have been made about the size of the variable.
In order to make sure each aliased type resolves to a type of the right size, type aliases of this kind are typically used in conjunction with preprocessor directives:
|
1 2 3 4 5 6 7 8 9 |
#ifdef INT_2_BYTES using int8_t = char; using int16_t = int; using int32_t = long; #else using int8_t = char; using int16_t = short; using int32_t = int; #endif |
On machines where integers are only 2 bytes, INT_2_BYTES can be #defined, and the program will be compiled with the top set of type aliases. On machines where integers are 4 bytes, leaving INT_2_BYTES undefined will cause the bottom set of type aliases to be used. In this way, int8_t will resolve to a 1 byte integer, int16_t will resolve to a 2 bytes integer, and int32_t will resolve to a 4 byte integer using the combination of char, short, int, and long that is appropriate for the machine the program is being compiled on.
This is exactly how the fixed width integers, like std::int8_t (covered in lesson 4.6 -- Fixed-width integers and size_t), are defined!
This is also where the issue with int8_t being treated as a char comes from -- std::int8_t is a type alias of char, and thus is just an alias for a char rather than being a unique type. As a result:
|
1 2 3 4 5 6 7 8 9 10 |
#include <cstdint> // for fixed-width integers #include <iostream> int main() { std::int8_t i{ 97 }; // int8_t is actually a type alias for signed char std::cout << i; return 0; } |
This program prints:
a
not 97, because std::cout prints char as an ASCII character, not a number.
Using type aliases to make complex types simple
Although we have only dealt with simple data types so far, in advanced C++, you could see a variable and function declared like this:
|
1 2 3 4 5 6 |
std::vector<std::pair<std::string, int> > pairlist; bool hasDuplicates(std::vector<std::pair<std::string, int> > pairlist) { // some code here } |
Typing std::vector<std::pair<std::string, int> > everywhere you need to use that type can get cumbersome. It’s much easier to use a type alias:
|
1 2 3 4 5 6 7 8 |
using pairlist_t = std::vector<std::pair<std::string, int> >; // make pairlist_t an alias for this crazy type pairlist_t pairlist; // instantiate a pairlist_t variable bool hasDuplicates(pairlist_t pairlist) // use pairlist_t in a function parameter { // some code here } |
Much better! Now we only have to type “pairlist_t” instead of std::vector<std::pair<std::string, int> >.
Don’t worry if you don’t know what std::vector, std::pair, or all these crazy angle brackets are yet. The only thing you really need to understand here is that type aliases allow you to take complex types and give them a simple name, which makes those types easier to work with and understand.
Best practice
Favor type aliases over typedefs, and use them liberally to document the meaning of your types.
Quiz time
Question #1
Given the following function prototype:
|
1 |
int printData(); |
Convert the int return value to a type alias named error_t. Include both the type alias statement and the updated function prototype.
 6.14 -- The auto keyword |
 Index |
 6.12 -- Using statements |
Hi Nascar,
Where to declare type alias, on the beginning of the file or when needed like inside the function or block of code?
In the smallest scope possible. If you it in multiple functions, at the beginning of the file (After the includes).
Maybe to put this as a useful tip? :)
Thanks :)
Thanks for another great lesson! Very well explained
Type aliases seem to be incredibly useful. Is it a C++ only feature? I don't recall seeing it in any other programming languages (but then again, this is the furthest that I've gotten into learning a programming language).
Section "Using type aliases for platform independent coding":
> This is exactly how the fixed width integers (like int8_t) that were introduced in C++11
You might want to remove "that were introduced in C++11" as per [your comment](https://www.learncpp.com/cpp-tutorial/the-auto-keyword/#comment-460178).
P.S. have you considered adding Markdown functionality to comments? That would make comments look much prettier :D
Many languages have type aliases, they're very prominent in C++ code that uses templates. Templates would be horrible to work with if it wasn't for type aliases.
C++11 note removed, thanks
I'll look into markdown comments, but making that happen might be too much effort
Thanks :)
*off topic*
what is that cartoon?
It's a Traflorkian from Rick and Morty :)
https://rickandmorty.fandom.com/wiki/Traflorkians
One use for type aliases is to help with documentation and legibility.
what documentation means ?
To remember the order of a typedef:
typedef double distance_t;// _t = trailing edge
Regards,
Patrick
Hi,
I was wondering why 'short' in the top set and 'long' in the bottom set weren't defined!
At the top `short` is either 8 or 16 bits, but we already have types with those sizes, so we don't need `short`.
At the bottom, `long` would be 32 and 64 bits. We already have `int` for 32 bits, and the example doesn't cover 64 bits, so we don't need `long`.
You initialised the variable i using (). Is that a mistake or I am understanding wrong?
Since it should be a function if it has () and not a variable.
It is another way of initializing variables in C++, called direct initialization.
See the link below for more information:
https://www.learncpp.com/cpp-tutorial/variable-assignment-and-initialization/
@sami is right, this is the direct initialization syntax. As per best practice, I updated the lesson to use list initialization instead.
I was wondering why the following code is working: as you mentioned, 'If they were placed outside of main(), in the global scope, all functions would be able to access them.' But I defined exactly the same type alias!
Line 13 overrides `testScore_t` in main, ie. `main` can no longer see the global `testScore_t`. The aliases could refer to different types.
Thank you!
Hello,could you tell me how below code dinstinguishes whether a pc defines int as 4 bits or 2 bits ?:
no matter what, but whatever I think about this block of code alone cannot do that.then could you tell me what else has to be done in addition to this ?
It is useful to mention that the same typedef can appear multiple times in a single file.
Hi! Fantastic website so far, really helping me get through this lockdown haha!
Just a quick question regarding the last section. When I enter the complex type "std::vector<std::pair<std::string, int> >" it results in a semantic error "Implicit instantiation of undefined template 'std::__1::vector<std::__1::pair<std::__1::basic_string<char>, int>, std::__1::allocator<std::__1::pair<std::__1::basic_string<char>, int> > >'". How would I fix this?
Can you provide an example that you think should compile, but doesn't? Telling from the error message alone, you might be missing one of the includes
Adding #include <vector> has fixed the problem, thanks!
Do typedefs/aliases need to be #included in header files wherever you want to use them? Might be good to add a comment to this chapter :)
Everything you want to use has to be included. Aliases aren't any different from everything else in this concern.
Is there any way to declare an alias for a data members? For example, a derived class declares an alias for a member in its base class, without using any pointer or reference, so that no additional memory is used.
No. You can add a function that returns a reference to the member of the parent.
Your situation sounds like you have public data members. Try to avoid that. You might be allowing the user of your class to break it.
If I am only using my computer, which defines int as 4 bytes, and I compile the code into an exe file, then I send the exe to a friend whose computer defines int as 2 bytes, will there be any problems? From my understanding on how it works, there wouldn’t be any problems.
So do I only need to do:
if I was working on the same project on different computers (i.e. the project could be a library like a game engine that a person would release to work for anyone to program with, if I’m working with others who have different computers on the same project, compiling it on different os, etc.)?
Use the fixed-width types from lesson 4.6.
Types don't change after compilation. If you compiled the code with an int being 16 bits wide, nothings going to change that. The program will use a 16 bit int on your friend's computer as well (If your friend's computer can execute the program).
Ok, thanks!
Hey, thanks for the very thorough tutorials.
I tried to define a type alias like this
This results in "error: expected type-specifier before 'static'".
Does anybody know why this does not work?
Can keywords like "static" or "constexpr" even be used in a type alias definition?
> Can keywords like "static" or "constexpr" even be used in a type alias definition?
No, they can't. That would makes declarations very confusing.