> As far as I can recall, all the code I've seen which tries to get
> under the hood of stdio could be satisifed with a single call: a
> call which, when applied to a readable FILE pointer, returns the
> number of bytes currently buffered, that is, the number of bytes
> readable before stdio tries to make a read() call.

Actually, one of the worst examples here is MH, which roots around
quite extensively in the stdio buffer.  I've tried three times to
convert MH to use the abstract stdio interface, and each time I have
given up in disgust.

I sort of suspect that the original poster was trying to port MH.  I
wound up converting it to use the NetBSD stdio internals. :/