roger@procase.UUCP (Roger H. Scott) (04/12/90)
The inability of AT&T's cfront to deal with "forward references" to inline functions is purely an artifact of its simplistic one-pass design. Nothing in the *language* interferes with handling such inlines "correctly". Handling of inlines, in general, in cfront is a mess and AT&T knows it. Given that cfront is an implementation of the C++ language that is teetering on the brink of collapsing under its own weight, I don't see why this news group spends so much time worrying about what cfront does and doesn't do. Perhaps part of the reason is that AT&T needs to make more of an effort to distinguish C++ language issues from cfront implementation issues.