Francesco Mazzoli

@DaniilSNikulin that is excellent, thank you! I'll try it out. It would be great to "mainline" this.

@DaniilSNikulin are there any caveats that @nmoehrle might be worried about? If not, could you submit a PR? I think this would be a big boon to people wishing to...

The streaming support is already `conduit`-indipendent, see https://github.com/fpco/store/blob/master/src/Data/Store/Streaming.hs . CC @kantp

Actually, I think you're proposing to _change_ the streaming interface. Reopening

@ygale I think you're proposing two things: 1. Split the conduit support in another package 2. Make the conduit-indipendent streaming API different For what concerns 1, I prefer fewer packages...

Most likely due to https://ghc.haskell.org/trac/ghc/ticket/5642

I've mitigated this in e9aa2be52fa9e8c37b74f06522507604bd135d85 . This became necessary because tests were failing while building docker images.

@mgsloan I think that this means that the smallcheck submodule is not needed anymore.

OK, I figured out why this is: the script just grabs `ID_NET_NAME_PATH`, but if `ID_NET_NAME_ONBOARD` is present that will be picked instead. In fact, the linked blog post lists how...

Oh just to be clear: my `/boot-*` partitions are _not_ on ZFS. So I'm not entirely sure how the ZFS filesystem is involved apart from it being used to write...