> On 17 Nov 2025, at 12:08, Jakub Jelinek <[email protected]> wrote:
>
> On Mon, Nov 17, 2025 at 04:45:44PM +0530, Jason Merrill wrote:
>> On 11/16/25 4:28 AM, Jakub Jelinek wrote:
>>> I've tried to test a patch to switch -std=gnu++17 C++ default
>>> to -std=gnu++20 (will post momentarily), but ran into various problems
>>> during GCC bootstraps, our codebase isn't fully C++20 ready.
>>>
>>> The most common problems are arithmetic or bitwise operations
>>> between enumerators of different enum types (or between enumerator
>>> and floating point in the testsuite), ambiguous overloaded
>>> operator == because of forgotten const qualification of const inside
>>> of the argument and then libcody being largely stuck in C++ and incompatible
>>> with C++20 which introduced char8_t type and uses it for u8 literals.
>>>
>>> The following patch fixes various issues I've run into, for libcody
>>> this patch just makes sure code including cody.hh can be compiled
>>> with -std=gnu++20, libcody itself I have a tweak in the other patch.
>>
>> Frankly I'm not sure what the point of libcody's u8 dance is; UTF-8 for all
>> the ASCII characters that it uses S2C(u8"x") for is the same byte as the 'x'
>> plain character literal. I'd be inclined to strip that all out.
>
> It isn't the same for -fexec-charset=IBM1047 (or other non-ASCII execution
> charsets), so perhaps the intent is to communicate using UTF-8 (or ASCII?)
> instead of using arbitrary other character set.
> Detail::S2C is from I understand just a workaround for C++11 and one can use
> (char) u8'x' instead for C++14 and later (perhaps conditionally based on
> feature test macros).
> I don't know what other projects use libcody.
> If it is just GCC, perhaps it is time to update it to minimum C++14 like
> rest of GCC and the workarounds can go.
> But for C++20 another thing is that I think we'd need to add some extra
> overloads for const char8_t * etc. and just cast to const char. Plus
> there are some uses of std::string created from u8 literals, dunno if it
> should use std::u8string in that case or have casts to const char * first.
There was a plan at one point to implement P1394 on clang which would
have used it, but that did not (yet) happen and I’m no longer working on
clang modules - so someone else would need to take it on (I don’t expect
it to happen any time soon unless the Apple folks decide to do it).
FWIW: I would expect Nathan to respond to any query about libcody - he’s
around - just not so active on C++.
Iain
>
> Jakub
>
