Joseph F. Ryan wrote:
I'm really curious as to this myself. I didn't even know an
"include" existed; its not in the imcc docs.
Its in docs/macros.pod, though this file is not mentioned in the main doca AFAIK.
I don't have a docs/macros.pod. I guess my version of imcc is old (-:
... I only bring this up because around a month ago, I started working on use/include/inline semantics for languages/perl6. Then I got mono and I stopped. However, I'm better now and I'd really like to know of how much I have is redundant, and if .include can help resolve some of the symbol name conflicts that I was having.
.include doesn't help here. If you have symbol name conflicts (which?)
For instance, in the interests of speed, it would be nice if perl6 builtins written in perl6 would be pre-compiled to imcc, so that they could just be tacked on at the end of the .imcc file (or even .include-d, now that I know about that.) However, if these are precompiled, then their local symbols will numbered starting at the same place (for instance, L_1, L_2, etc), and you get a conflict when this file is included. So, I had to mangle.
you can do several things:
1) for variables
- use .namespace
This seems like it will really help; one question I have is:
Does .namespace work with .subs? As in, does it help support "local subs?" Or is this wishful thinking?
- mangle them - us uniq names, as P6C does
I'm using these 2 methods to avoid conflicts now. Maybe I can take out much of the mangling with the help of .namespace.
Thanks for the reply,
Joseph F. Ryan [EMAIL PROTECTED]
