Jan Teske

I'd like to try my hands on heap sort.

The traits in the prototype have meanwhile been updated to also include this associated type. The two implementations are still sufficiently different, so I would suggest you take a look...

The STM32F3-DISCOVERY, which is used to run all the examples in this repo, also has a CCM not accessible by DMA (which is why we place the stack in normal...

Ah, yes, we kind of ignore circular transfers in the README. That's mostly because we've been focusing on the buffer type and using one-shot transfers in the examples makes things...

@thalesfragoso I think it's still true that we can't just allow `B` to be anything (or even just anything that impls `Dma{Read,WriteBuffer}`), due to the alignment issue you mentioned. So...

I've updated the DMA PR for the stm32f3xx-hal to use the proposed buffer traits. Once that's merged, we can hopefully get some real-world experience using them. https://github.com/stm32-rs/stm32f3xx-hal/pull/86

I'm also currently tripping over this. @parasyte's suggestions are helpful but it complains about the missing `test` crate for me too. In case that helps anyone, my settings for VS...

Just FYI, I'm also stumbling over this currently trying to use the `TestClient` in an `async` test function (with `pytest-asyncio`). Minimal test case: ```python import pytest from starlette.responses import HTMLResponse...

Something that came up in a discussion with @vmarcos: The `mz_sources` table currently contains both Materialize-external sources (Kafka, Postgres), and introspection sources. Those are fundamentally different things so it might...

@joacoc `mz_records_per_dataflow_global` is an introspection view, i.e. a view that depends on compute introspection sources (whereas `mz_indexes` is a system table). As such it can only return information from a...