Joachim Durchholz

Am 08.01.2017 um 19:17 schrieb Alexander Hartmaier: > SQL /is/ a DSL for relational databases, why would one invent one on top > of it? Because the DSL is not...

Full database portability is really difficult to achieve. Date-time types can have massively different granularity across databases (and I doubt that any two databases have the same set of available...

Am 09.01.2017 um 09:27 schrieb Alexander Hartmaier: > Date(Time) Perl 6 types should work transparent with the corresponding > RDBMS types. Losing precision on storage might happen Which is nonportable...

Test failure is unrelated to the change.

LANG=C eliminates internationalization. E.g. on my machine, `ls` does not respond with `total XXX` but with `insgesamt XXX`, making the script fail.

Not sure what that heuristics is doing in the first place (it's been quite a while since I even looked into cachecontrol). It's not fixing the bug that cachecontrol enters...

Just another data point from my side. CONTEXT: I am building a new REST service so legacy considerations do not affect me (not currently, I am very aware how important...

> Let's let this sit for a bit to see if others have feedback. Agreed. > Regardless, this should have an option to enable it. It should not be the...

I see that to allow a record builder to be both subclassable and fluent, it needs an additional type parameter for the setters' return type. This has ramifications for type...

Took me a bit to re-read the article and `Case.java`, but I have a somewhat clearer picture now indeed, and yes it's clear that `@Case` was intended to be something...