Simonas Kazlauskas

Should the static be defined in the `hit!` rather than `check!`?

6

Currently we define the `#[no_mangle]` static in the `check!` macro. This means that the test like this: ```rust fn safe_divide(dividend: u32, divisor: u32) -> u32 { if divisor == 0...

Add the usage instructions that are in README to top-level crate documentation

1

Those are useful and hard to find if you don’t read READMEs.

This commit uses version ranges to accurately specify the minimum version of the dependencies required for this library to work. These version bounds are verified in the CI with `-Zminimal-versions`...

Cache buffers, allow omitting span names, options for not formatting KVs all the time

1

With this plain `error!("banana")` can take as little as 200ns (~0 overhead) but it will also lead to no meaningful messages if you have `error!(someotherkey)`, and it will also drop...

Try to use `fiber` support to implement proper support for non-hierarchical tracing spans

1

Tracing spans don't need to be hierarchical, it is entirely possible to have spans looking like this ``` | -- S1 -- | | -- S2 -- | | --...

Add a way to query IP address through a specific interface

1

**Describe the bug** I have two uplinks in my system: `wan0` and `wan1`. `wan1` is wired, but is behind a CGNAT. `wan0` is a backup link to a sattelite, but...

[velbus] For examples to be fully-featured they should demonstrate use of Thing properties

5

See the `TODO` reference to this issue in the velbus module’s README. In short, there’s an example that reads along the lines of ``` Bridge velbus:bridge:1 [ port="COM1" ] {...

[Velbus] Unloading the bridge fails predictably

4

NB: this is not a request for somebody to fix the issue, I anticipate getting around to looking at it myself in near future. To reproduce this issue, I have...

Editing a r+ comment makes bors think the approval was done twice

2

If a comment with r+ has been edited then homu will emit > :bulb: This pull request was already approved, no need to approve it again. > > There's another...

r- force has a weird outcome

2

See https://github.com/rust-lang/rust/pull/33441#issuecomment-217312409. Namely, it should behave as if plain `r-` (i.e. cancel the build) without printing any messages. Currently it takes another PR on top of the queue and “forces”...