search

solodon4 / Mach7

Functional programming style pattern-matching library for C++
Other
1.28k stars 79 forks source link

Cannot switch on poly and variant in same file #52

Closed akrzemi1 closed 8 years ago

akrzemi1 commented 8 years ago

It seams it is currently impossible to declare two type-switch statements in one file, where one is "traditional" (vtable-based) and the other is xtl-based. I need to include two headers: type_switchN-patterns.hpp and type_switchN-patterns-xtl.hpp which contain conflicting definitions.

(For the same reason, I guess I would be unable to do a switch on two arguments, where one argument is a polymorphic type, and the other is boost::variant.)

I realize, there are plans for unification. But for the time being, would it be possible to provide a way for the two headers to coexist? Define symbols in different namespaces, #undef the macros?

solodon4 commented 8 years ago

You don't need to include both headers, just include type_switchN-patterns-xtl.hpp it works also with regular polymorphism. XTL makes the subclassing in C++ to be a special case of its generalized subtyping relation. If it doesn't work with regular class hierarchies, please file bugs with concrete examples.

akrzemi1 commented 8 years ago

Thanks. It works!

But given that it works, does the information in the README.md (known-bugs-and-limitations) still make sense?

The most recent experimentation and the header you are prpobably looking for is type_switchN-patterns.hpp.

Is <mach7/type_switchN-patterns-xtl.hpp> not superior by now? Or is the former faster?

solodon4 commented 8 years ago

You are right, type_switchN-patterns.hpp was an intermediary step on the way to Match statement with multiple subjects. The first one was doing just type switching on multiple arguments (type_switchN.hpp), then the second one expanded it to the declaration syntax that was supported in match.hpp but will also be made obsolete (type_switchN-decl.hpp) and then the type_switchN-patterns.hpp was the actual extension to support the normal patterns. type_switchN-pattern-xtl.hpp was an extension of it. During header reorg some of these headers didn't even make it to mach7 folder, because they were used in just a couple of tests.

The file name is not final yet, ultimately I'd like to have a single header but it might be two: the current match.hpp and this one. Match statements in match.hpp support other ADT encoding schemes, which are not easily generalizable to multiple arguments. The type_switchN-patterns-xtl.hpp or whatever it will be renamed to will be the main header as it has the widest functionality, but I'd probably have to keep match.hpp as well.