Hey Scott!
it really is: (unix time 32b)6a2458e9 (msg CRC32)f2b1d9e9
Ah! I was thinking the last 8 characters were strftime's %N output, which for all time is restricted to 000000000-3b9ac9ff, which puts f2b1d9e9 way out of range. I should have checked more closely before guessing nanoseconds.
I've got to rethink the MSGID generation.
I am sticking with unixtime. However in my case I don't restrict to 8 characters so in 2106 it will automagically tack on an extra hex charater (nibble) which will be good until "Sun Aug 20 07:32:15 UTC 4147" before it tacks on another nibble bringing the total to 10 hex characters. I am not married to the nanosecond part but it does make an excellent method to ensure uniqueness for all time.
I'm on the fence about making the storage pointers 64b
Agreed. I prefer the method that doesn't limit the output to 8 and use unixtime. Only have to wait until "Sun Feb 7 06:28:16 UTC 2106" before it switches to 9 characters (36b). I can hardly wait. ;-)
Life is good,
Maurice
-o o- o- o- o- o- o- -o -o o- -o o-
(\ /) /) /) /) /) /) (\ (\ /) (\ /)
^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^
... Ne nanwuht ne byð yfel, ær mon wene þæt it yfel seo.
Nothing is misery unless one thinks it is misery.
--- GNU bash, version 5.3.9(1)-release (x86_64-pc-linux-gnu)
* Origin: Little Mikey's Brain - Ladysmith BC, Canada (1:153/7001)