Clemens Winter
Clemens Winter
Final pass for merging query results is done by a single thread. Parallelizing this would give large speed up for queries with high cardinality group bys.
Add support for streaming subresults between merge operators. Would give significant speedups for queries that perform high cardinality group bys.
LocustDB should have support for entropy coding data using Huffman or ANS. [This](https://github.com/Cyan4973/FiniteStateEntropy) seems to be the leading implementation, but there don't actually seem to be any Rust bindings for...
PoC for fixing https://github.com/softprops/shiplift/issues/295. Works well enough as a workaround for my use case, but the current implementation still has some issues which I would like to get feedback on...
When querying logs for containers with a tty, the logs stream is not multiplexed via headers but just returns the raw output: https://github.com/docker/cli/blob/86e1f04b5f115fb0b4bbd51e0e4a68233072d24b/vendor/github.com/docker/docker/client/container_logs.go#L19 The current logs implementation will parse random...
## What did you implement: Fixes return type of network `inspect` function. Closes: https://github.com/softprops/shiplift/issues/293 ## How did you verify your change: Tested manually, before change `inspect` returns parse error, after...
Calling the network `inspect` function returns an error like `missing field `rx_dropped` at line 1 column 1032`. This is because the `inspect` function returns the `NetworkInfo` type when it should...
The `imagebuild` example is much slower than the docker cli. For a relatively small build (docker cli reports 25.15MB of build context) which is fully cached, the docker cli takes...
I don't have a minimal repro yet, but the pyo3 crate seems to be recompiled on most builds which significantly increases compilation times. I'm on pyo3 version 0.13.2, rustc version...