|  | Declaration changes meaning | | |
| | | Guest |  |
|  Posted: Fri Aug 22, 2008 6:44 am Post subject: Re: Undiagnosable rule paradox (was Re: Declaration changes |  |
| |  | |
On Aug 10, 2:01 am, Alberto Ganesh Barbati <AlbertoBarb...@libero.it>
wrote:
| Quote: | Now, because R is not a diagnosable rule (Fact #2), the program does not
violate any diagnosable rules, nor syntax rules, nor the ODR, so
according to Fact #3, the program is well-formed! However, R itself says
that the program is ill-formed and we have a contradiction because the
two concepts of well- and ill-formedness are mutually exclusive
according to fact #4.
How do we reconcile this? Is the wording defective? I would say yes, but
I'm not quite sure about where defect is. Maybe we should just remove
the adjective "diagnosable" from 1.3.15.
|
I'd say "rule R" is worded well because you're talking about the
differences between machine and visual inspections. The rule rightly
outlines an expectation too idealistic to impose on current-day
compilers and implies that one should self-critique for such
violations until it's plausible for mandatory compiler inclusion. If
you were to restrict the standard to only those parts that a compiler
will implement then every "no diagnostic required" should be removed.
If the compiler were responsible for enforcing every letter of the
standard then no one but compiler maintainers would read it.
Kevin P. Barry
--
[ See LINK for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ] |
| |
|
|
