11:14 <FlibberTGibbet> howdy. is anyone else seeing issues with crystal apt updates from the opensuse.org repos?

22:12 <oprypin> Cindy, IO::ByteFormat::SystemEndian reflects the compiler target because Crystal supports exactly 0 little-endian systems

22:12 <oprypin> FlibberTGibbet, I've seen discussions on Discord, sadly you'd probably have to join that

22:13 <Cindy> oprypin: exactly 0 little-endian systems?

22:14 <oprypin> Cindy, yes. sorry did you not understand me or did i not understand you? 😅

22:15 <Cindy> so if it supported 0 little-endian systems, then should IO::ByteFormat::SystemEndian be set to BigEndian?

22:15 <oprypin> you can see here, in many cases Crystal has *theoretical* support for little-endian systems https://github.com/crystal-lang/crystal/blob/3a839961e4ebefe1f6d0a940cf8b02da2d60a489/src/io/byte_format.cr#L130

22:16 <oprypin> there are platform-specific macros such as https://github.com/crystal-lang/crystal/blob/3a839961e4ebefe1f6d0a940cf8b02da2d60a489/src/va_list.cr#L21

22:17 <oprypin> Cindy, oh i'm sorry yes I confused the two. it supports 0 big-endian systems

22:17 <Cindy> at first i thought you were insane

22:18 <Cindy> i'm glad it was just a mistake

22:18 <oprypin> anyway.. to add support, someone could write ` {% if flag?(:powerpc) %} alias SystemEndian = BigEndian`. but it's useless because `:powerpc` is never set, there is no support for it

22:18 <Cindy> also architecture != endian anymore

22:18 <Cindy> some architectures are biendian

22:20 <Cindy> like in crystal's aarch64 port, ARM is biendian, so technically it is big-endian aswell as little-endian

22:20 <Cindy> but most ARM CPUs run in little-endian

22:22 <Cindy> but still bi-endian architectures are an important detail to miss if you make flags simply on architectures alone

22:23 <Cindy> which is why i thought there could be a flag?(:littleendian)