Finally, thank you kind sir !

On Tue, Mar 12, 2024, at 16:29, Arthur O'Dwyer via Std-Proposals wrote:
> On Tue, Mar 12, 2024 at 7:15 AM Sebastian Wittmeier via Std-Proposals <std-proposals_at_[hidden]> wrote:
>> __
>> To be fair:
>>
>> How much different is it in power from a C23 #embed?
>>
>> Both feature that certain files, the ones which the implementation allows, can be fully (binarily) read at compile time.
>>
>> Differences:
>>
>> - the filepath can be created at runtime -> better hiding of the file read read; C23 probably supports macros for the filepath?
>>
>> - the contents can be processed by constexpr functions -> probably the same can be achieved by mixing C23 with C++ constexpr
>>
>> - the files could be read several times -> Denial of Service build; a long build can probably be achieved by other ways
>>
>
> The big one is the the filepath can be created at "runtime." Yes, #embed allows you to use macros — for example you can now write a quine as https://godbolt.org/z/hTj8jvcf8
> — but you can't actually smuggle in a fully constexpr-evaluated string like https://godbolt.org/z/v74GdjKse
>
> constexpr char myfilename[] = "test.txt";
> constexpr char data[] = {
> #embed (myfilename) // doesn't work
> };
>
> So #embed doesn't seem to be useful as a "constexpr filesystem," unfortunately.
>
> Similarly, AFAICT no vendor supports `_Pragma(mystring)` with mystring a constexpr-evaluated string; the argument must be literally a string literal. But IIUC, C++26 **is** getting `static_assert(false, mystring)` — not with char arrays, but at least with strings and string_views. Example: https://godbolt.org/z/vdjf9KP3E
>
> my $.02,
> –Arthur
> --
> Std-Proposals mailing list
> Std-Proposals_at_[hidden]
> https://lists.isocpp.org/mailman/listinfo.cgi/std-proposals
>