Woong Jun

I am not talking about a proposed text for the Standard, but about the text of the Standard. Even if I read the Dave Prosser's pseudo code after I wrote...

If you think the purpose of the proposals is for clarification, then they are actually trying to change the text that the committee _intentionally_ left ambiguous. If you think the...

> It seems they disregarded both the original intent ... The goal of the C90 Committee to write wording for preprocessing was _not_ to move Dave's algorithm into plain English...

Has so many corner cases that it seems too early to have a solid implementation.

Test cases: ```c #define foo1(...) __VA_OPT__(__VA_OPT__) #define foo2(...) start __VA_OPT__(__VA_OPT__()) end #define foo3(...) start __VA_OPT__(start __VA_OPT__(foo) end) end #define foo4(...) ## __VA_OPT__(test) #define foo5(...) test ## __VA_OPT__(test) #define foo6(...) __VA_OPT__...

`git diff --no-prefix` of an incomplete implementation: ```diff diff --git lib/mcr.c lib/mcr.c index 683c889..a3c80a8 100644 --- lib/mcr.c +++ lib/mcr.c @@ -510,13 +510,13 @@ static struct mtab *conflict(const char *chn) lex_t...

Need to consider issuing a warning for `42 + b? x: y` or `b? x: y + 42` is really worthwhile.

Decided not to support the "fix it" feature.

Thanks for your comment. `beluga` does not directly generate binary code; it currently generates assembly code and uses a system assembler and a linker to build binaries. Fortunately, modern Macs...

This issue is because the order in which an expression appers in code differs from that in which it is evaluated. For example, given this: ``` C for (p =...