Skip to content

build(deps): bump the cargo-major group across 1 directory with 3 updates - #257

Open
dependabot[bot] wants to merge 1 commit into
mainfrom
dependabot/cargo/cargo-major-a3206f95c4
Open

dependabot[bot] wants to merge 1 commit into
mainfrom
dependabot/cargo/cargo-major-a3206f95c4

Conversation

@dependabot

@dependabot dependabot Bot commented on behalf of github Nov 15, 2026

Copy link
Copy Markdown
Contributor

Bumps the cargo-major group with 3 updates in the / directory: rusqlite, sonora and zbus.

Updates rusqlite from 0.41.1 to 0.42.0

Release notes

Sourced from rusqlite's releases.

0.42.0

What's Changed

  • Bump bundled SQLite version to 3.54.1 #1920
  • Scope close large-error expectation to 64-bit cache builds #1922
  • Enable bundled SQLCipher on wasm32-unknown-unknown #1921
  • Fix panic with a zero-argument deterministic scalar function #1927
  • No check of PRAGMA name with SQLCipher #1924
  • Add bindings to sqlite3_key_v2 / sqlite3_rekey_v2 #1930
  • Tell sqlite that rarray is already ordered by rowid #1934
  • Fix clippy warnings #1940
  • Fix memory leak while registering a collation or function with invalid name #1942
  • detect_leaks #1943
  • Fix use-after-free in Session::apply / apply_strm #1941
  • Introduce RawValue #1933
  • Fix set_clientdata on OOM #1944
  • Bump bundled SQLCipher to 4.20.0 #1932

Full Changelog: rusqlite/rusqlite@v0.41.0...v0.42.0

Commits

Updates sonora from 0.2.0 to 0.3.0

Release notes

Sourced from sonora's releases.

v0.3.0

The port now tracks M145 plus 15 later upstream WebRTC changes, up to M156. cpp/NEWS lists them. Three of them (multi-channel defaults, shared comfort noise, joint coarse/refined filter choice) were already in 0.2.0.

Code written for 0.2.0 compiles unchanged. The version is 0.3.0 because of the behaviour changes below, which change the output for some callers.

Behaviour changes

  • Input volume controller: the default config is now upstream's production config (upstream d9b92fe1bc). With input_volume_controller enabled, recommended_stream_analog_level() targets a speech level in [-50, -12] dBFS instead of [-30, -18] dBFS, updates the volume from the speech level at most once every 100 frames, and also lowers it when clipping is predicted. The controller does not modify samples itself, but its recommendation reaches the output whenever it is applied: through capture_level_adjustment.analog_mic_gain_emulation, or when the caller passes it back with set_stream_analog_level(), where each volume change resets AGC2's adaptive digital level estimate and, with the echo canceller enabled, signals an echo path gain change to AEC3. So the output changes for callers that apply the recommendation.
  • Floating-point environment: like upstream, each process_* call now sets hardware flush-to-zero for its duration (MXCSR FTZ and DAZ on x86/x86_64 with SSE, FPCR.FZ on aarch64 with NEON). This targets the AEC3 slowdown on x86 while the render reference is silent (#34). Code that a call reaches, such as tracing subscribers and the global allocator, runs in that mode, and the output can differ by tiny amounts on near-silent input. Upstream sets the x86 bits only in clang builds, so on x86 a caller that replaces a GCC-built C++ library gets FTZ/DAZ where it had IEEE gradual underflow. Changing the floating-point environment is formally undefined behavior in Rust, and upstream's DenormalDisabler, nih-plug and the no_denormals crate also change it; sonora sets flush-to-zero only on the calling thread and restores that thread's setting after each call, but a thread started during a call can inherit flush-to-zero and keep it (on Linux, for example). Other targets do not flush and log "Denormal disabler unsupported" when the processor is built. The "Floating-point environment" section on AudioProcessing has the details.
  • Arithmetic on targets without native FMA (#37): on x86/x86_64 without the fma target feature (including the default x86_64 target), i686, armv7, wasm32 and riscv64, the biquad filters, the three-band filter bank and fft4g now use the C++ expressions instead of f32::mul_add, which compiled to an fmaf library call there. The pipeline is 1.1-2.2x faster on default x86_64, and output differs from 0.2.0 by small rounding amounts on those targets, closer to C++. AArch64 and x86 built with fma are unchanged. New: sonora_simd::NATIVE_FMA; sonora-fft now depends on sonora-simd.

Other changes

  • AudioProcessingBuilder::echo_canceller3_config sets the AEC3 tuning (#47, building on #41 by Roman Ernst), like C++ SetEchoCancellerConfig without a multichannel config. The configuration also applies to multichannel render, and an invalid one is clamped to valid ranges with a warning. EchoCanceller3Config is re-exported from sonora::config.
  • AGC2 no longer allocates in steady state (#47, building on #41 by Roman Ernst). sonora-agc2 adds Limiter::process_owned_signal, AdaptiveDigitalGainController::process_owned_frame, GainApplier::apply_gain_to_owned_signal and NoiseLevelEstimator::analyze_owned_frame; the existing methods keep their signatures.
  • Echo cancellation no longer allocates in steady state (#40, building on #39 by Roman Ernst): through the float and int16 APIs, with rate conversion, downmixing and stereo. sonora-aec3 adds FrameBlocker::insert_owned_sub_frame_and_extract_block; the existing method keeps its signature.
  • AEC3: ports the render buffer headroom fix for underruns (d460e60e19). Two further changes are internal and change no behaviour: the code that applies suppressor config updates (e10cd19640) is ported, but nothing in the port triggers it, and the code for three field-trial kill switches that the port never enabled is removed.
Changelog

Sourced from sonora's changelog.

0.3.0

The port now tracks M145 plus 15 later upstream WebRTC changes, up to M156. cpp/NEWS lists them. Three of them (multi-channel defaults, shared comfort noise, joint coarse/refined filter choice) were already in 0.2.0.

Code written for 0.2.0 compiles unchanged. The version is 0.3.0 because of the behaviour changes below, which change the output for some callers.

Behaviour changes

  • Input volume controller: the default config is now upstream's production config (upstream d9b92fe1bc). With input_volume_controller enabled, recommended_stream_analog_level() targets a speech level in [-50, -12] dBFS instead of [-30, -18] dBFS, updates the volume from the speech level at most once every 100 frames, and also lowers it when clipping is predicted. The controller does not modify samples itself, but its recommendation reaches the output whenever it is applied: through capture_level_adjustment.analog_mic_gain_emulation, or when the caller passes it back with set_stream_analog_level(), where each volume change resets AGC2's adaptive digital level estimate and, with the echo canceller enabled, signals an echo path gain change to AEC3. So the output changes for callers that apply the recommendation.
  • Floating-point environment: like upstream, each process_* call now sets hardware flush-to-zero for its duration (MXCSR FTZ and DAZ on x86/x86_64 with SSE, FPCR.FZ on aarch64 with NEON). This targets the AEC3 slowdown on x86 while the render reference is silent (#34). Code that a call reaches, such as tracing subscribers and the global allocator, runs in that mode, and the output can differ by tiny amounts on near-silent input. Upstream sets the x86 bits only in clang builds, so on x86 a caller that replaces a GCC-built C++ library gets FTZ/DAZ where it had IEEE gradual underflow. Changing the floating-point environment is formally undefined behavior in Rust, and upstream's DenormalDisabler, nih-plug and the no_denormals crate also change it; sonora sets flush-to-zero only on the calling thread and restores that thread's setting after each call, but a thread started during a call can inherit flush-to-zero and keep it (on Linux, for example). Other targets do not flush and log "Denormal disabler unsupported" when the processor is built. The "Floating-point environment" section on AudioProcessing has the details.
  • Arithmetic on targets without native FMA (#37): on x86/x86_64 without the fma target feature (including the default x86_64 target), i686, armv7, wasm32 and riscv64, the biquad filters, the three-band filter bank and fft4g now use the C++ expressions instead of f32::mul_add, which compiled to an fmaf library call there. The pipeline is 1.1-2.2x faster on default x86_64, and output differs from 0.2.0 by small rounding amounts on those targets, closer to C++. AArch64 and x86 built with fma are unchanged. New: sonora_simd::NATIVE_FMA; sonora-fft now depends on sonora-simd.

Other changes

  • AudioProcessingBuilder::echo_canceller3_config sets the AEC3 tuning (#47, building on #41 by Roman Ernst), like C++ SetEchoCancellerConfig without a multichannel config. The configuration also applies to multichannel render, and an invalid one is clamped to valid ranges with a warning. EchoCanceller3Config is re-exported from sonora::config.
  • AGC2 no longer allocates in steady state (#47, building on #41 by Roman Ernst). sonora-agc2 adds Limiter::process_owned_signal, AdaptiveDigitalGainController::process_owned_frame, GainApplier::apply_gain_to_owned_signal and NoiseLevelEstimator::analyze_owned_frame; the existing methods keep their signatures.
  • Echo cancellation no longer allocates in steady state (#40, building on #39 by Roman Ernst): through the float and int16 APIs, with rate conversion, downmixing and stereo. sonora-aec3 adds FrameBlocker::insert_owned_sub_frame_and_extract_block; the existing method keeps its signature.
  • AEC3: ports the render buffer headroom fix for underruns (d460e60e19). Two further changes are internal and change no behaviour: the code that applies suppressor config updates (e10cd19640) is ported, but nothing in the port triggers it, and the code for three field-trial kill switches that the port never enabled is removed.
Commits
  • 1632d47 chore: Release
  • c6bbfda Merge pull request #48 from dignifiedquire/release-prep-0.3.0
  • 315f65d chore: update lockfiles
  • 0f0f4cc docs(changelog): title the unreleased changes 0.3.0, behaviour changes first
  • ab39e59 docs(changelog): add the 0.2.0 section and mark 0.1.0 as released
  • 2db4315 Merge pull request #47 from dignifiedquire/aec3-config-agc2-followup
  • 00da3e8 docs(changelog): add the entries for #47
  • 44a95da fix(sonora): warn when a custom AEC3 config is clamped
  • 7a81a5f fix(sonora): preallocate every AGC2 channel buffer
  • 33f0a55 fix(agc2): keep the 0.2.0 signatures of the gain stages
  • Additional commits viewable in compare view

Updates zbus from 5.19.0 to 6.2.1

Release notes

Sourced from zbus's releases.

zbus-6.2.1

🐛 Fixed

  • Check message bodies without an if let guard.

✅ Testing

  • Benchmark taking a burst of signals off a connection's socket.
  • Benchmark checking a body of dicts of variants.

zbus_macros-6.2.1

No changes since 6.2.0: released alongside zbus 6.2.1, as the two are kept at the same version.

zbus-6.2.0

✨ Added

  • Add Connection::send_all. #2078
  • Add WriteHalf::send_messages.

♻️ Changed

  • Hand several buffers to each send of a runtime's socket.

🐛 Fixed

  • Limit nesting for each signature, and count dict entries. #2087
  • Reject a received message whose path field isn't an object path.
  • Accept references to unsized types as interface args again. #2076

⚙️ Other

  • 🦺 Reject a header with an unended string or a repeated field.
  • 🦺 Let a connection check the body of each message it receives.
  • 🦺 Add Body::validate, for a bus to check bodies with.
  • 🦺 Check a received message as a whole, not only its fields.

⚡️ Performance

  • Call the read half itself rather than through its Box.
  • Read ahead of each message on the sockets that zbus drives. #2085
  • Allocate a received message once, at its exact length.
  • Leave the maybe type of GVariant to the signature parser.
  • Add Message::with_sender, for a bus to set senders with.
  • Leave room for None in where a header field lies.
  • Parse message headers with winnow rather than serde. #2080
  • Read the header fields of a message without checking them again.

✅ Testing

  • Hold deserializing to the nesting that Body::validate allows.
  • Benchmark parsing and checking short signatures.
  • Fuzz the parser of message headers against serde.
  • Benchmark decoding a received message.

zbus_macros-6.2.0

🐛 Fixed

  • Accept setter args with lifetime parameters again.
  • Accept references to unsized types as interface args again. #2076

... (truncated)

Commits
  • 69816bc 🔖 zb,zm: Bump versions and generate the changelogs
  • 4551af5 Merge pull request #2102 from zeenix/msrv-fix
  • 626dd37 👷 CI: Check every feature with the MSRV
  • 8e635fd 🐛 zb: Check message bodies without an if let guard
  • ae80453 Merge pull request #2101 from z-galaxy/renovate/ubuntu-26.x
  • f56f481 👷 CI: Take only rustfmt from nightly Rust
  • d2b7151 👷 CI: Build the walltime benchmarks in a container of Ubuntu 22.04
  • 7efcf62 Merge pull request #2099 from z-galaxy/renovate/actions-upload-artifact-7.x
  • 4948586 Merge pull request #2098 from z-galaxy/renovate/actions-download-artifact-8.x
  • f5e63f7 ⬆️ Update dependency ubuntu to v26
  • Additional commits viewable in compare view

You can trigger a rebase of this PR by commenting @dependabot rebase.


Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

  • @dependabot rebase will rebase this PR
  • @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
  • @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
  • @dependabot ignore <dependency name> major version will close this group update PR and stop Dependabot creating any more for the specific dependency's major version (unless you unignore this specific dependency's major version or upgrade to it yourself)
  • @dependabot ignore <dependency name> minor version will close this group update PR and stop Dependabot creating any more for the specific dependency's minor version (unless you unignore this specific dependency's minor version or upgrade to it yourself)
  • @dependabot ignore <dependency name> will close this group update PR and stop Dependabot creating any more for the specific dependency (unless you unignore this specific dependency or upgrade to it yourself)
  • @dependabot unignore <dependency name> will remove all of the ignore conditions of the specified dependency
  • @dependabot unignore <dependency name> <ignore condition> will remove the ignore condition of the specified dependency and ignore conditions

…ates

Bumps the cargo-major group with 3 updates in the / directory: [rusqlite](https://github.com/rusqlite/rusqlite), [sonora](https://github.com/dignifiedquire/sonora) and [zbus](https://github.com/z-galaxy/zbus).


Updates `rusqlite` from 0.41.1 to 0.42.0
- [Release notes](https://github.com/rusqlite/rusqlite/releases)
- [Changelog](https://github.com/rusqlite/rusqlite/blob/master/Changelog.md)
- [Commits](rusqlite/rusqlite@v0.41.1...v0.42.0)

Updates `sonora` from 0.2.0 to 0.3.0
- [Release notes](https://github.com/dignifiedquire/sonora/releases)
- [Changelog](https://github.com/dignifiedquire/sonora/blob/main/CHANGELOG.md)
- [Commits](dignifiedquire/sonora@sonora-v0.2.0...sonora-v0.3.0)

Updates `zbus` from 5.19.0 to 6.2.1
- [Release notes](https://github.com/z-galaxy/zbus/releases)
- [Changelog](https://github.com/z-galaxy/zbus/blob/main/release-plz.toml)
- [Commits](z-galaxy/zbus@zbus-5.19.0...zbus-6.2.1)

---
updated-dependencies:
- dependency-name: rusqlite
  dependency-version: 0.42.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: cargo-major
- dependency-name: sonora
  dependency-version: 0.3.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: cargo-major
- dependency-name: zbus
  dependency-version: 6.2.1
  dependency-type: direct:production
  update-type: version-update:semver-major
  dependency-group: cargo-major
...

Signed-off-by: dependabot[bot] <[email protected]>
@dependabot
dependabot Bot requested a review from sunerpy as a code owner November 15, 2026 05:10
@dependabot dependabot Bot added dependencies Pull requests that update a dependency file rust Pull requests that update rust code labels Nov 15, 2026

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

dependencies Pull requests that update a dependency file rust Pull requests that update rust code

Projects

None yet

Development

Successfully merging this pull request may close these issues.

0 participants