Re: UUID v7

> On 25 Nov 2024, at 22:53, Masahiko Sawada <sawada.mshk@gmail.com> wrote:
>
> In the following code, we use "defined(__darwin__) || defined(_MSC_VER)":
>
> +#if defined(__darwin__) || defined(_MSC_VER)
> +#define SUBMS_MINIMAL_STEP_BITS 10
> +#else
> +#define SUBMS_MINIMAL_STEP_BITS 12
> +#endif
> #define SUBMS_BITS     12
> -#define SUBMS_MINIMAL_STEP_NS ((NS_PER_MS / (1 << SUBMS_BITS)) + 1)
> +#define SUBMS_MINIMAL_STEP_NS ((NS_PER_MS / (1 <<
> SUBMS_MINIMAL_STEP_BITS)) + 1)
>
> on the other hand, we use "defined(__darwin__) || defined(WIN32)" here:
>
> +#if defined(__darwin__) || defined(WIN32)
> +       /*
> +        * On MacOS real time is truncted to microseconds. Thus, 2 least
> +        * significant are dependent on other time-specific bits, thus
> they do not
> +        * contribute to uniqueness. To make these bit random we mix in two bits
> +        * from CSPRNG.
> +        *
> +        * SUBMS_MINIMAL_STEP is chosen so that we still guarantee monotonicity
> +        * despite altering these bits.
> +        */
> +       uuid->data[7] = uuid->data[7] ^ (uuid->data[8] >> 6);
> +#endif
>
> Is there a reason for using different macros?

No, that's an oversight. We should mix these 2 bits if an only if SUBMS_MINIMAL_STEP_BITS=10.

<tldr>
In your review change_v33.patch you used WIN32, but it did not actually compile on Windows.
So on Saturday I squashed v33+change_v33.patch, and composed a message that I think we still should switch to _MSC_VER.
Andjust before sending I received your message with v36 where you used _MSC_VER :) 

I think this way:
_MSC_VER - native Windows without clock_gettime, we used gettimeofday() and 10 bits of sub-ms.
MinGW - we use clock_gettime() and 12 bits.
Darwin - we use clock_gettime() and 10 bits.
Anything else -  clock_gettime() and 12 bits.
</tldr>

>
> In get_real_time_ns_ascending(), we use _MSC_VER so we use
> clock_gettime() on MinGW.
>
>>
>> Sergey Prokhorenko just draw my attention to the new release of MariaDB [0]. They are doing very similar UUID v7
generationas we do [1]. 
>>
>
> Thank you for the references. It made me think that we can use the
> function name uuid_v7() rather than uuidv7().

I think it's a good idea if we will be kind of SQL-compatible.


Best regards, Andrey Borodin.