I was looking at some open source code whereby I came across this line:
Is this faster than assigning it to another variable directly, like:Code:const wxString &text{ GetLine(i) };
Or is there something else I'm overlooking?Code:const wxString text{ GetLine(i) };
Thanks!
They have different semantics. First one creates a temporary and then creates a reference to it, extending the lifetime of the reference.
The second creates a temporary, then invokes copy constructor to copy object into the new variable text, then destructs the temporary.
The first one is possibly more efficient (not complex objects) if you don't need to modify the object.
Originally Posted by Adak
There could be return value optimisation though, in which case the second would be no less efficient than the first.
Look up a C++ Reference and learn How To Ask Questions The Smart Way
Assuming that GetLine returns by value, there really isn't much efficiency difference, because it's easy for a compiler to see that you're doing the same thing.
The second might be faster for global variables in rare cases, because in the first the type of the variable as seen by other compilation units is reference, which could potentially refer to another global variable, whereas in the second case it can't. So in that rare case it could prevent optimisations. But if it's found inside a function, then the initial value is visible to the optimizer so this is not a problem.
It is too clear and so it is hard to see.
A dunce once searched for fire with a lighted lantern.
Had he known what fire was,
He could have cooked his rice much sooner.
