Re: DB2: Generating epoch time

On 12-Nov-2015 10:14 -0600, rob wrote:

<<SNIP>>
[From HEX(GENERATE_UNIQUE())] I get
02144325069C758C0551A07001

<<SNIP>> that [epoch_time] UDF) is returning an epoch of
1447343795

Due to slight run time differences I can expect some variance but
not 14432... vs 14473...
<<SNIP>>

Quite likely, those first several _hexadecimal digits_ have nothing to do with the /date and time/, and instead have something to do with the /place/; i.e. representative of the /space/ portion of an _effective_ UUID that is produced by the SQL HEX(GENERATE_UNIQUE())

Plus, notably, that comparison improperly conflates the _Hexadecimal_ digit values seen from the GENERATE_UNIQUE data [conspicuously, the results capable of being displayed, effectively only per use of the HEX() scalar] and the _Decimal_ digit values of an INTEGER value. I am suggesting that there is a great probability, that the digits 144... being visible in both character strings of digits, is spectacularly coincidental; that on many other systems, those same digit positions would not also be 144... [even if derived at approximately the same date\time], and may even include some hexadecimal-only digits 'A' through 'F'. To compare hex to hex [though likely futile, as the *nix epoch surely is not what GENERATE_UNIQUE uses], similarly wrap the UDF invocation in HEX;
HEX(EPOCH_TIME())