# 0.1.31 (May 11, 2023)
This release of `tracing-core` fixes a bug that caused threads which
call `dispatcher::get_default` _before_ a global default subscriber is
set to never see the global default once it is set. In addition, it
includes improvements for instrumentation performance in some cases,
especially when using a global default dispatcher.
### Fixed
- Fixed incorrect thread-local caching of `Dispatch::none` if
`dispatcher::get_default` is called before
`dispatcher::set_global_default` (#2593)
### Changed
- Cloning a `Dispatch` that points at a global default subscriber no
longer requires an `Arc` reference count increment, improving
performance substantially (#2593)
- `dispatcher::get_default` no longer attempts to access a thread local
if the scoped dispatcher is not in use, improving performance when the
default dispatcher is global (#2593)
- Added `#[inline]` annotations called by the `event!` and `span!`
macros to reduce the size of macro-generated code and improve
recording performance (#2555)
Thanks to new contributor @ldm0 for contributing to this release!
## Motivation
Currently, when a call to `dispatcher::get_default` occurs, `tracing`
will check the thread-local default dispatcher first. If a thread-local
scoped default is set, it is returned. Otherwise, the thread will then
check the global default. If a global default is present, it is then
cached in the thread local, so that subsequent calls do not need to
check the global default.
Unfortunately, this behavior results in issues if the scoped default is
accessed (e.g. using `get_default` or creating a new span) *prior* to a
global default being set. When `get_default` runs for the first time and
there is no global default, a `none` dispatcher is cached as the
thread-local default. This means that the thread will now behave as
though its default dispatcher is `None` until the scoped default is
overridden, even if a global default is then set later. This is quite
bad, and results in issues such as #2587, #2436, and #2411.
## Solution
This branch makes several changes to remove the use of the thread-local
caching of the global default dispatcher, and to lessen the performance
impact of doing so.
On the `master` (v0.2.x) branch, we track the number of scoped
dispatchers currently active, and use it to determine whether or not to
check thread-local storage at all. This optimization was introduced in
PR #1017. This branch backports a similar change to `v0.1.x`.
In addition, #1017 also changes the dispatcher module to represent a
`Dispatch` internally using an enum of either an `Arc` in the case where
the dispatcher is scoped, or a `&'static dyn Subscriber + Send + Sync`
reference when the dispatcher is the global default. This makes cloning
and constructing the global default cheaper, and also allows us to
change the `None` dispatcher into a static singleton. That means that
the use of a `None` dispatcher no longer requires an allocation and arc
reference bump, an issue which was previously resolved by locally
caching a `None` dispatcher. A side benefit of this change is that
*cloning* a `Dispatch` is substantially cheaper when the dispatcher is a
global default, as it's just an `&'static` reference and no `Arc` bump
is necessary. This will also make cloning a `Span` cheaper when the
global default dispatcher is in use.
Finally, because the overhead of getting the global default is
substantially reduced, we are able to change the scoped default
dispatcher's behavior to remove the caching entirely. This means that
the category of bugs involving the local cache becoming stale is
resolved entirely.
Fixes#2587Fixes#2436Fixes#2411Closes#2592
## Performance Impact
This change results in a change in performance characteristics. Running
the benchmarks, we observe a significant improvement in performance in
most of the benchmarks that use the global default dispatcher, and a
noticeable decrease in performance for some benchmarks using the scoped
default. In my opinion, this performance change is acceptable, as a
global default dispatcher is the common case for most users, and is
generally expected to perform better than the scoped default. In
addition, resolving the variety of bugs that are caused by the local
caching of the default when using the scoped default dispatcher is worth
a performance cost when the scoped default is in use.
<details>
<summary>Benchmark results:</summary>
```
Running benches/baseline.rs (target/release/deps/baseline-9b70733ce49582d2)
comparison/relaxed_load time: [456.48 ps 456.55 ps 456.63 ps]
change: [+3.0281% +3.3135% +3.5664%] (p = 0.00 < 0.05)
Performance has regressed.
Found 10 outliers among 100 measurements (10.00%)
5 (5.00%) high mild
5 (5.00%) high severe
comparison/acquire_load time: [438.98 ps 439.32 ps 439.76 ps]
change: [-0.3725% -0.2092% -0.0614%] (p = 0.01 < 0.05)
Change within noise threshold.
Found 12 outliers among 100 measurements (12.00%)
2 (2.00%) high mild
10 (10.00%) high severe
comparison/log time: [227.05 ps 227.14 ps 227.27 ps]
change: [+3.1351% +3.2984% +3.4537%] (p = 0.00 < 0.05)
Performance has regressed.
Found 14 outliers among 100 measurements (14.00%)
5 (5.00%) high mild
9 (9.00%) high severe
```
```
Running benches/dispatch_get_clone.rs (target/release/deps/dispatch_get_clone-d4d6ca1f9895e432)
Dispatch::get_clone/none
time: [8.3974 ns 8.4004 ns 8.4039 ns]
change: [-22.870% -22.796% -22.728%] (p = 0.00 < 0.05)
Performance has improved.
Found 10 outliers among 100 measurements (10.00%)
1 (1.00%) low severe
1 (1.00%) low mild
4 (4.00%) high mild
4 (4.00%) high severe
Dispatch::get_clone/scoped
time: [15.877 ns 15.959 ns 16.045 ns]
change: [+52.358% +52.943% +53.500%] (p = 0.00 < 0.05)
Performance has regressed.
Found 16 outliers among 100 measurements (16.00%)
2 (2.00%) high mild
14 (14.00%) high severe
Dispatch::get_clone/global
time: [8.3962 ns 8.4000 ns 8.4054 ns]
change: [-19.126% -18.961% -18.817%] (p = 0.00 < 0.05)
Performance has improved.
Found 15 outliers among 100 measurements (15.00%)
2 (2.00%) low severe
6 (6.00%) high mild
7 (7.00%) high severe
```
```
Running benches/dispatch_get_ref.rs (target/release/deps/dispatch_get_ref-6ce05749a0b1bf87)
Dispatch::get_ref/none time: [1.7551 ns 1.7564 ns 1.7579 ns]
change: [-51.858% -51.749% -51.644%] (p = 0.00 < 0.05)
Performance has improved.
Found 10 outliers among 100 measurements (10.00%)
3 (3.00%) low mild
2 (2.00%) high mild
5 (5.00%) high severe
Dispatch::get_ref/scoped
time: [3.6341 ns 3.6365 ns 3.6397 ns]
change: [-2.6892% -2.5955% -2.4968%] (p = 0.00 < 0.05)
Performance has improved.
Found 12 outliers among 100 measurements (12.00%)
5 (5.00%) high mild
7 (7.00%) high severe
Dispatch::get_ref/global
time: [1.7668 ns 1.7686 ns 1.7713 ns]
change: [-52.697% -52.647% -52.603%] (p = 0.00 < 0.05)
Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
2 (2.00%) high mild
5 (5.00%) high severe
```
```
Running benches/empty_span.rs (target/release/deps/empty_span-745c777d77b8b7ca)
empty_span/none time: [227.02 ps 227.10 ps 227.20 ps]
change: [-0.1729% -0.0705% +0.0495%] (p = 0.24 > 0.05)
No change in performance detected.
Found 10 outliers among 100 measurements (10.00%)
5 (5.00%) high mild
5 (5.00%) high severe
empty_span/scoped time: [218.51 ps 218.69 ps 218.90 ps]
change: [-0.7582% -0.6056% -0.4630%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 8 outliers among 100 measurements (8.00%)
5 (5.00%) high mild
3 (3.00%) high severe
empty_span/global time: [217.85 ps 218.15 ps 218.56 ps]
change: [-2.6528% -2.4341% -2.1602%] (p = 0.00 < 0.05)
Performance has improved.
Found 8 outliers among 100 measurements (8.00%)
4 (4.00%) high mild
4 (4.00%) high severe
empty_span/baseline_struct
time: [655.54 ps 656.09 ps 656.76 ps]
change: [-1.6595% -1.4125% -1.1776%] (p = 0.00 < 0.05)
Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
5 (5.00%) high mild
2 (2.00%) high severe
```
```
Running benches/enter_span.rs (target/release/deps/enter_span-7fc1c2a69c076475)
enter_span/none time: [0.0000 ps 0.0000 ps 0.0000 ps]
change: [-43.600% +6.5764% +109.38%] (p = 0.86 > 0.05)
No change in performance detected.
Found 14 outliers among 100 measurements (14.00%)
6 (6.00%) high mild
8 (8.00%) high severe
enter_span/scoped time: [2.6513 ns 2.6567 ns 2.6641 ns]
change: [+0.3121% +1.9504% +3.4648%] (p = 0.01 < 0.05)
Change within noise threshold.
Found 9 outliers among 100 measurements (9.00%)
2 (2.00%) high mild
7 (7.00%) high severe
enter_span/global time: [3.2108 ns 3.2160 ns 3.2220 ns]
change: [+25.963% +26.742% +27.434%] (p = 0.00 < 0.05)
Performance has regressed.
Found 2 outliers among 100 measurements (2.00%)
2 (2.00%) high mild
```
```
Running benches/event.rs (target/release/deps/event-6742eef6ebe07aa4)
event/none time: [227.04 ps 227.18 ps 227.41 ps]
change: [-1.6751% -1.5743% -1.4711%] (p = 0.00 < 0.05)
Performance has improved.
Found 13 outliers among 100 measurements (13.00%)
6 (6.00%) high mild
7 (7.00%) high severe
event/scoped time: [8.3849 ns 8.4335 ns 8.4888 ns]
change: [-3.4754% -3.0252% -2.6092%] (p = 0.00 < 0.05)
Performance has improved.
Found 5 outliers among 100 measurements (5.00%)
3 (3.00%) high mild
2 (2.00%) high severe
event/scoped_recording time: [36.916 ns 37.022 ns 37.194 ns]
change: [+8.1054% +18.714% +30.381%] (p = 0.00 < 0.05)
Performance has regressed.
Found 10 outliers among 100 measurements (10.00%)
1 (1.00%) low mild
2 (2.00%) high mild
7 (7.00%) high severe
event/global time: [6.9694 ns 7.1677 ns 7.3469 ns]
change: [-23.407% -21.940% -20.398%] (p = 0.00 < 0.05)
Performance has improved.
```
```
Running benches/span_fields.rs (target/release/deps/span_fields-96dfd0a8a577dec6)
span_fields/none time: [3.5936 ns 3.6008 ns 3.6106 ns]
change: [+17.160% +17.776% +18.413%] (p = 0.00 < 0.05)
Performance has regressed.
Found 14 outliers among 100 measurements (14.00%)
8 (8.00%) high mild
6 (6.00%) high severe
span_fields/scoped time: [33.751 ns 33.765 ns 33.779 ns]
change: [+22.689% +22.873% +23.037%] (p = 0.00 < 0.05)
Performance has regressed.
Found 7 outliers among 100 measurements (7.00%)
1 (1.00%) low mild
4 (4.00%) high mild
2 (2.00%) high severe
span_fields/scoped_recording
time: [270.22 ns 270.55 ns 270.91 ns]
change: [+10.615% +10.827% +11.028%] (p = 0.00 < 0.05)
Performance has regressed.
Found 6 outliers among 100 measurements (6.00%)
6 (6.00%) high mild
span_fields/global time: [28.337 ns 28.428 ns 28.527 ns]
change: [+3.0582% +3.3355% +3.6278%] (p = 0.00 < 0.05)
Performance has regressed.
Found 13 outliers among 100 measurements (13.00%)
13 (13.00%) high mild
```
```
Running benches/span_no_fields.rs (target/release/deps/span_no_fields-f8c7d7a84f720442)
span_no_fields/none time: [1.5467 ns 1.5507 ns 1.5553 ns]
change: [+12.966% +13.206% +13.434%] (p = 0.00 < 0.05)
Performance has regressed.
Found 8 outliers among 100 measurements (8.00%)
7 (7.00%) high mild
1 (1.00%) high severe
span_no_fields/scoped time: [17.796 ns 17.810 ns 17.826 ns]
change: [+1.0381% +1.1673% +1.2914%] (p = 0.00 < 0.05)
Performance has regressed.
Found 12 outliers among 100 measurements (12.00%)
6 (6.00%) high mild
6 (6.00%) high severe
span_no_fields/scoped_recording
time: [30.397 ns 30.459 ns 30.524 ns]
change: [-0.8489% -0.6268% -0.3915%] (p = 0.00 < 0.05)
Change within noise threshold.
span_no_fields/global time: [12.747 ns 12.791 ns 12.844 ns]
change: [-27.930% -27.672% -27.386%] (p = 0.00 < 0.05)
Performance has improved.
```
```
Running benches/span_repeated.rs (target/release/deps/span_repeated-03bfaaf4ecd13d36)
span_repeated/none time: [699.28 ns 699.84 ns 700.53 ns]
change: [+2.4125% +2.6359% +2.8862%] (p = 0.00 < 0.05)
Performance has regressed.
Found 9 outliers among 100 measurements (9.00%)
7 (7.00%) high mild
2 (2.00%) high severe
span_repeated/scoped time: [2.5029 µs 2.5057 µs 2.5090 µs]
change: [+4.5095% +4.6605% +4.8122%] (p = 0.00 < 0.05)
Performance has regressed.
Found 16 outliers among 100 measurements (16.00%)
8 (8.00%) low mild
6 (6.00%) high mild
2 (2.00%) high severe
span_repeated/scoped_recording
time: [5.0509 µs 5.0535 µs 5.0566 µs]
change: [+0.7346% +1.0724% +1.3718%] (p = 0.00 < 0.05)
Change within noise threshold.
Found 13 outliers among 100 measurements (13.00%)
6 (6.00%) high mild
7 (7.00%) high severe
span_repeated/global time: [2.1264 µs 2.1272 µs 2.1282 µs]
change: [-11.213% -11.119% -11.031%] (p = 0.00 < 0.05)
Performance has improved.
Found 10 outliers among 100 measurements (10.00%)
5 (5.00%) high mild
5 (5.00%) high severe
```
</details>
## Motivation
When reading the docs I noticed some typo's and while fixing them I
noticed warnings from cargo doc
## Solution
Fixed the typos and the doc warnings in tracing
Co-authored-by: Neulichedl, Patrick - D0E05450 <Patrick.Neulichedl@dm-b2b.com>
# 0.1.38 (April 25th, 2023)
This `tracing` release changes the `Drop` implementation for
`Instrumented` `Future`s so that the attached `Span` is entered when
dropping the `Future`. This means that events emitted by the `Future`'s
`Drop` implementation will now be recorded within its `Span`. It also
adds `#[inline]` hints to methods called in the `event!` macro's
expansion, for an improvement in both binary size and performance.
Additionally, this release updates the `tracing-attributes` dependency
to [v0.1.24][attrs-0.1.24], which updates the [`syn`] dependency to
v2.x.x. `tracing-attributes` v0.1.24 also includes improvements to the
`#[instrument]` macro; see [the `tracing-attributes` 0.1.24 release
notes][attrs-0.1.24] for details.
### Added
- `Instrumented` futures will now enter the attached `Span` in their
`Drop` implementation, allowing events emitted when dropping the
future to occur within the span (https://github.com/tokio-rs/tracing/pull/2562)
- `#[inline]` attributes for methods called by the `event!` macros,
making generated code smaller (https://github.com/tokio-rs/tracing/pull/2555)
- **attributes**: `level` argument to `#[instrument(err)]` and
`#[instrument(ret)]` to override the level of the generated return
value event (https://github.com/tokio-rs/tracing/pull/2335)
- **attributes**: Improved compiler error message when `#[instrument]`
is added to a `const fn` (https://github.com/tokio-rs/tracing/pull/2418)
### Changed
- `tracing-attributes`: updated to [0.1.24][attrs-0.1.24]
- Removed unneeded `cfg-if` dependency (https://github.com/tokio-rs/tracing/pull/2553)
- **attributes**: Updated [`syn`] dependency to 2.0 (https://github.com/tokio-rs/tracing/pull/2516)
### Fixed
- **attributes**: Fix `clippy::unreachable` warnings in
`#[instrument]`-generated code (https://github.com/tokio-rs/tracing/pull/2356)
- **attributes**: Removed unused "visit" feature flag from `syn`
dependency (https://github.com/tokio-rs/tracing/pull/2530)
### Documented
- **attributes**: Documented default level for `#[instrument(err)]`
(https://github.com/tokio-rs/tracing/pull/2433)
- **attributes**: Improved documentation for levels in `#[instrument]`
(https://github.com/tokio-rs/tracing/pull/2350)
Thanks to @nitnelave, @jsgf, @Abhicodes-crypto, @LukeMathWalker,
@andrewpollack, @quad, @klensy, @davidpdrsn, @dbidwell94, @ldm0,
@NobodyXu, @ilsv, and @daxpedda for contributing to this release!
[`syn`]: https://crates.io/crates/syn
[attrs-0.1.24]:
https://github.com/tokio-rs/tracing/releases/tag/tracing-attributes-0.1.24
# 0.1.24 (April 24th, 2023)
This release of `tracing-attributes` adds support for passing an
optional `level` to the `err` and `ret` arguments to `#[instrument]`,
allowing the level of the generated return-value event to be overridden.
For example,
```rust
#[instrument(err(level = "info"))]
fn my_great_function() -> Result<(), &'static str> {
// ...
}
```
will emit an `INFO`-level event if the function returns an `Err`.
In addition, this release updates the [`syn`] dependency to v2.x.x.
### Added
- `level` argument to `err` and `ret` to override the level of the
generated return value event (#2335)
- Improved compiler error message when `#[instrument]` is added to a
`const fn` (#2418)
### Changed
- Updated `syn` dependency to 2.0 (#2516)
### Fixed
- Fix `clippy::unreachable` warnings in `#[instrument]`-generated code
(#2356)
- Removed unused "visit" feature flag from `syn` dependency (#2530)
### Documented
- Documented default level for `err` (#2433)
- Improved documentation for levels in `#[instrument]` (#2350)
Thanks to @nitnelave, @jsgf, @Abhicodes-crypto, @LukeMathWalker,
@andrewpollack, @quad, @klensy, @davidpdrsn, and @dbidwell94 for
contributign to this release!
[`syn`]: https://crates.io/crates/syn
# 0.3.17 (April 21, 2023)
This release of `tracing-subscriber` fixes a build error when using
`env-filter` with recent versions of the `regex` crate. It also
introduces several minor API improvements.
### Fixed
- **env-filter**: Add "unicode-case" and "unicode-perl" to the `regex`
dependency, fixing a build error with recent versions of `regex`
(#2566)
- A number of minor documentation typos and other fixes (#2384, #2378,
#2368, #2548)
### Added
- **filter**: Add `fmt::Display` impl for `filter::Targets` (#2343)
- **fmt**: Made `with_ansi(false)` no longer require the "ansi" feature,
so that ANSI formatting escapes can be disabled without requiring
ANSI-specific dependencies (#2532)
### Changed
- **fmt**: Dim targets in the `Compact` formatter, matching the default
formatter (#2409)
Thanks to @keepsimple1, @andrewhalle, @LeoniePhiline, @LukeMathWalker,
@howardjohn, @daxpedda, and @dbidwell94 for contributing to this
release!
## Motivation
Missing features for the `regex` crate were causing build time errors
due to the the use of unicode characters in the regex without using
the proper features within the regex crate
## Solution
Add the missing feature flags.
Fixes#2565
Authored-by: Devin Bidwell <dbidwell@biddydev.com>
updated UI tests using TRYBUILD=overwrite with the latest stable version of Rust
## Motivation
UI tests are failing on the latest stable version of Rust
## Solution
Run `TRYBUILD=overwrite cargo test` to update the effected files.
## Motivation
Currently it is not possible to disable ANSI in `fmt::Subscriber`
without enabling the "ansi" crate feature. This makes it difficult for
users to implement interoperable settings that are controllable with
crate features without having to pull in the dependencies "ansi" does.
I hit this while writing an application with multiple logging options
set during compile-time and I wanted to cut down on dependencies if
possible.
## Solution
This changes `fmt::Subscriber::with_ansi()` to not require the "ansi"
feature flag. This way, `with_ansi(false)` can be called even when the
"ansi" feature is disabled. Calling `with_ansi(true)` when the "ansi"
feature is not enabled will panic in debug mode, or print a warning if
debug assertions are disabled.
Co-authored-by: daxpedda <daxpedda@gmail.com>
Co-authored-by: Eliza Weisman <eliza@buoyant.io>
The `tracing-subscriber` module `tests::support` included functionality
to mock a layer (via the `Layer` trait). This code depends on
some items from `tracing_mock::collector` which should otherwise not be
public.
This change moves the mocking functionality inside `tracing-mock` behind
a feature flag. Allowing the `Expect` enum and `MockHandle::new` from
`tracing_mock::collector` to be made `pub(crate)` instead of `pub`.
Since it's now used from two different modules, the `Expect` enum has
been moved to its own module.
This requires a lot of modifications to imports so that we're not doing
wildcard imports from another crate (i.e. in `tracing-subscriber`
importing wildcards from `tracing-mock`).
This PR is based on @hds' PR #2369, but modified to track renamings. I
also deleted all the doc comments temporarily because updating them was
a lot of work and I need to get a release of `tracing-subscriber` out
first.
Closes: #2359
## Motivation
Currently it is not possible to disable ANSI in `fmt::Subscriber`
without enabling the "ansi" crate feature. This makes it difficult for
users to implement interoperable settings that are controllable with
crate features without having to pull in the dependencies "ansi" does.
I hit this while writing an application with multiple logging options
set during compile-time and I wanted to cut down on dependencies if
possible.
## Solution
This changes `fmt::Subscriber::with_ansi()` to not require the "ansi"
feature flag. This way, `with_ansi(false)` can be called even when the
"ansi" feature is disabled. Calling `with_ansi(true)` when the "ansi"
feature is not enabled will panic in debug mode, or print a warning if
debug assertions are disabled.
Co-authored-by: Eliza Weisman <eliza@buoyant.io>
## Motivation
syn 2.0 is out!
## Solution
Update to syn 2.0 🚀
Co-authored-by: David Barsky <me@davidbarsky.com>
Co-authored-by: Eliza Weisman <eliza@buoyant.io>
Same reason as https://github.com/rust-lang/log/pull/536 :
`cfg_if` is only used in a single place and `tracing` is used by many
other crates, so even removing one dependency will be beneficial.
Remove dependency `cfg-if` and replace `cfg_if::cfg_if!` with a `const
fn get_max_level_inner() -> LevelFilter` and uses `if cfg!(...)` inside.
Using if in const function is stablised in
[1.46](https://github.com/rust-lang/rust/blob/master/RELEASES.md#version-1460-2020-08-27)
so this should work fine in msrv 1.56
Signed-off-by: Jiahao XU <Jiahao_XU@outlook.com>
; Conflicts:
; tracing/Cargo.toml
; tracing/src/level_filters.rs
This PR removes tracing-opentelemetry to a dedicated repo located at
https://github.com/tokio-rs/tracing-opentelemetry. (Note that at time of
writing this PR, the new repo has not be made public). We're moving
tracing-opentelemetry to a dedicated repository for the following
reasons:
1. opentelemetry's MSRV is higher than that of `tracing`'s.
2. more importantly, the main `tracing` repo is getting a bit unweildy
and it feels unreasonable to maintain backports for crates that
integrate with the larger tracing ecosystem.
(https://github.com/tokio-rs/tracing-opentelemetry does not have the
examples present in this repo; this will occur in a PR that will be
linked from _this_ PR.)
## Motivation
The current description for the default level of the `err` return value
event _strongly implies_ it's the same as the span. However, the
implementation actually defaults to `ERROR`.
## Solution
This PR documents that, so future generations don't have to chase down
the truth, like I did. 😉
There are new warnings as errors reported by clippy in Rust 1.67.0.
This are causing builds to fail, e.g.:
https://github.com/tokio-rs/tracing/actions/runs/4027112923/jobs/6922513360
In both cases they are reports of lifetimes that can be elided. This
change removes the unnecessary lifetime annotations to make clippy
happy.
## Motivation
The `#[instrument]` macro cannot be used on `const fn`s, because the
generated code will perform runtime tracing behavior. However, when
adding the attribute to a `const fn`, the compiler errors generated
currently are somewhat unclear (see #2414). It would be better if we
generated a less verbose error that simply states that `#[instrument]`
is not supported on `const fn`s.
## Solution
This branch changes the `#[instrument]` macro to detect when the
annotated function is a `const fn`, and emit a simpler, more descritpive
error message. The new error simply states that the `#[instrument]`
attribute cannot be used on `const fn`s, and should be much less
confusing to the user.
Fixes#2414
## Motivation
There is a small wording mistake in the tracing-subscriber docs
that can be fixed quickly and easily.
## Solution
Two duplicate words were removed.
## Motivation
The worker thread name in non blocking mode is always "tracing-appender".
It can be convenient to quickly identify the appender threads for
audit reasons or affinity pinning.
## Solution
This patch adds a new setter to the builder and propagates the info to
the thread initialization.
Closes#2364
This `doc_cfg` attribute's `cfg` part has multiple predicates without an
`all`, which iis what's breaking the netlify build. I'm...kind of
surprised this ever succeeded, since the cfg is malformed...
This branch adds documentation and tests noting that the `#[instrument]`
macro accepts `tracing::Level` directly. Using `tracing::Level` directly
allows for IDE autocomplete and earlier detection of typos.
The documentation for tracing-attributes was also rewritten to remove
the usage of the second-person perspective, making it more consistent
with the rest of tracing's documentation.
Co-authored-by: David Barsky <me@davidbarsky.com>
; Conflicts:
; tracing-attributes/Cargo.toml
; tracing-attributes/src/lib.rs
This branch adds the ability to override the level of the events
generated by the `ret` and `err` arguments to `#[instrument]`. An
overridden level can be specified with:
```rust
```
```rust
```
and so on.
This syntax is fully backwards compatible with existing uses of the
attribute.
In addition, some refactoring was done to how levels are parsed and how
the tokens for a specified level is generated.
Fixes#2330
; Conflicts:
; tracing-attributes/src/lib.rs
## Motivation
There's currently a `fmt::Display` impl for `EnvFilter` that emits an
equiovalent filter string that can be parsed back into an `EnvFilter`,
but the `Targets` filter does not have a `fmt::Display` impl. We ought
to have one, especially to make using `Targets` with `clap` v4.0 easier.
## Solution
This branch adds a `fmt::Display` impl for `filter::Targets`. The
implementation is pretty straightforward.
I also added tests that a `Targets`' `fmt::Display` output can be parsed
back into a filter that's equivalent to the original.
Currently, the Netlify docs are failing because of a warning that one of
the lints we explicitly enable is no longer a warning on nightly but
becoming a hard error. This is because the docs are built on nightly
with `-D warnings`, which denies all warnings...which seems not ideal.
For now, let's not remove that lint from the `deny` list, as it's still
a valid lint on stable. Instead, this PR explicitly allows that lint on
the docs build.
## Motivation
Clippy check fails in recent CI runs in v0.1.x branch PRs, for example
this run:
https://github.com/tokio-rs/tracing/actions/runs/4641107803/jobs/8215263838
Relevant error logs:
```
error: lint `const_err` has been removed: converted into hard error, see issue #71800 <https://github.com/rust-lang/rust/issues/71800> for more information
--> tracing-core/src/lib.rs:132:5
|
132 | const_err,
| ^^^^^^^^^
|
= note: `-D renamed-and-removed-lints` implied by `-D warnings`
error: deref which would be done by auto-deref
--> tracing-core/src/dispatcher.rs:371:26
|
371 | return f(&*entered.current());
| ^^^^^^^^^^^^^^^^^^^ help: try this: `&entered.current()`
|
= help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#explicit_auto_deref
= note: `-D clippy::explicit-auto-deref` implied by `-D warnings`
error: deref which would be done by auto-deref
--> tracing-core/src/dispatcher.rs:393:20
|
393 | Some(f(&*entered.current()))
| ^^^^^^^^^^^^^^^^^^^ help: try this: `&entered.current()`
|
= help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#explicit_auto_deref
```
## Solution
Fix the warnings based on the suggestions for Clippy.
# 0.3.16 (October 6, 2022)
This release of `tracing-subscriber` fixes a regression introduced in
[v0.3.15][subscriber-0.3.15] where `Option::None`'s `Layer`
implementation would set the max level hint to `OFF`. In addition, it
adds several new APIs, including the `Filter::event_enabled` method for
filtering events based on fields values, and the ability to log internal
errors that occur when writing a log line.
This release also replaces the dependency on the unmaintained
[`ansi-term`] crate with the [`nu-ansi-term`] crate, resolving an
*informational* security advisory ([RUSTSEC-2021-0139] for
[`ansi-term`]'s maintainance status. This increases the minimum
supported Rust version (MSRV) to Rust 1.50+, although the crate should
still compile for the previous MSRV of Rust 1.49+ when the `ansi`
feature is not enabled.
### Fixed
- **layer**: `Option::None`'s `Layer` impl always setting the
`max_level_hint` to `LevelFilter::OFF` (#2321)
- Compilation with `-Z minimal versions` (#2246)
- **env-filter**: Clarify that disabled level warnings are emitted by
`tracing-subscriber` (#2285)
### Added
- **fmt**: Log internal errors to `stderr` if writing a log line fails
(#2102)
- **fmt**: `FmtLayer::log_internal_errors` and
`FmtSubscriber::log_internal_errors` methods for configuring whether
internal writer errors are printed to `stderr` (#2102)
- **fmt**: `#[must_use]` attributes on builders to warn if a
`Subscriber` is configured but not set as the default subscriber
(#2239)
- **filter**: `Filter::event_enabled` method for filtering an event
based on its fields (#2245, #2251)
- **filter**: `Targets::default_level` accessor (#2242)
### Changed
- **ansi**: Replaced dependency on unmaintained `ansi-term` crate with
`nu-ansi-term` ((#2287, fixes informational advisory
[RUSTSEC-2021-0139])
- `tracing-core`: updated to [0.1.30][core-0.1.30]
- Minimum Supported Rust Version (MSRV) increased to Rust 1.50+ (when
the `ansi`) feature flag is enabled (#2287)
### Documented
- **fmt**: Correct inaccuracies in `fmt::init` documentation (#2224)
- **filter**: Fix incorrect doc link in `filter::Not` combinator
(#2249)
Thanks to new contributors @cgbur, @DesmondWillowbrook, @RalfJung, and
@poliorcetics, as well as returning contributors @CAD97, @connec,
@jswrenn, @guswynn, and @bryangarza, for contributing to this release!
[nu-ansi-term]: https://github.com/nushell/nu-ansi-term
[ansi_term]: https://github.com/ogham/rust-ansi-term
[RUSTSEC-2021-0139]: https://rustsec.org/advisories/RUSTSEC-2021-0139.html
[core-0.1.30]: https://github.com/tokio-rs/tracing/releases/tag/tracing-core-0.1.30
[subscriber-0.3.15]: https://github.com/tokio-rs/tracing/releases/tag/tracing-subscriber-0.3.15
This reverts commit a0824d398aa2511de28371d30dda9203360a6cf5 (PR #2247).
As discussed in [this comment][1], the implementation for `Arc`s may
cause subtly incorrect behavior if actually used, due to the `&mut self`
receiver of the `LookupSpan::register_filter` method, since the `Arc`
cannot be mutably borrowed if any clones of it exist.
The APIs added in PRs #2269 and #2293 offer an alternative solution to
the same problems this change was intended to solve, and --- since this
change hasn't been published yet --- it can safely be reverted.
[1]:
https://giethub.com/tokio-rs/tracing/pull/2247#issuecomment-1199924876
# 0.1.23 (October 6, 2022)
This release of `tracing-attributes` fixes a bug where compiler
diagnostic spans for type errors in `#[instrument]`ed `async fn`s have
the location of the `#[instrument]` attribute rather than the location
of the actual error, and a bug where inner attributes in
`#[instrument]`ed functions would cause a compiler error.
### Fixed
- Fix incorrect handling of inner attributes in `#[instrument]`ed
functions ([#2307])
- Add fake return to improve spans generated for type errors in `async
fn`s ([#2270])
- Updated `syn` dependency to fix compilation with `-Z minimal-versions`
([#2246])
Thanks to new contributors @compiler-errors and @e-nomem, as well as
@CAD97, for contributing to this release!
[#2307]: https://github.com/tokio-rs/tracing/pull/2307
[#2270]: https://github.com/tokio-rs/tracing/pull/2270
[#2246]: https://github.com/tokio-rs/tracing/pull/2246
## Motivation
When the `instrument` attribute is used on a function with inner
attributes, the proc macro generates code above the attributes within
the function block that causes compilation errors. These should be
parsed out separately and handled.
Fixes#2294
## Solution
I updated `MaybeItemFn` and `MaybeItemFnRef` to so they hold both the
outer and inner attributes for the instrumented function and updated the
codegen to inlcude them in the appropriate locations.
I couldn't preserve the existing implementation of
`From<&'_ ItemFn> for MaybeItemFnRef<'_, Box<Block>>`, because it is
now necessary to separate the inner and outer attributes of the
`ItemFn` into two separate `Vec`s. That implementation was replaced
with a `From<ItemFn> for MaybeItemFn`, which uses `Iterator::partition`
to separate out the inner and outer attributes.
Co-authored-by: Eliza Weisman <eliza@buoyant.io>
# 0.1.30 (October 6, 2022)
This release of `tracing-core` adds a new `on_register_dispatch` method
to the `Subscriber` trait to allow the `Subscriber` to perform
initialization after being registered as a `Dispatch`, and a
`WeakDispatch` type to allow a `Subscriber` to store its own `Dispatch`
without creating reference count cycles.
### Added
- `Subscriber::on_register_dispatch` method ([#2269])
- `WeakDispatch` type and `Dispatch::downgrade()` function ([#2293])
Thanks to @jswrenn for contributing to this release!
[#2269]: https://github.com/tokio-rs/tracing/pull/2269
[#2293]: https://github.com/tokio-rs/tracing/pull/2293
## Motivation
This fixes a discrepancy in the `tracing-opentelemetry` docs for the new
MetricsLayer. The docs refer to `value.` as a way to collect discrete
data point, but this doesn't match any of the prefix constant mentioned
in the same file.
```rust
const METRIC_PREFIX_MONOTONIC_COUNTER: &str = "monotonic_counter.";
const METRIC_PREFIX_COUNTER: &str = "counter.";
const METRIC_PREFIX_HISTOGRAM: &str = "histogram.";
```
## Solution
This fixes the documentation and test by referring to `histogram.`
instead of `value.`.
## Motivation
Currently, when using the `Layer` impl for `Option<S: Layer<...>>`, the
`Layer::max_level_hint` returns `Some(LevelFilter::OFF)`. This was
intended to allow totally disabling output in the case where a
`Subscriber` is composed entirely of `None` `Layer`s. However, when
other `Layer`s *do* exist but return `None` from their `max_level_hint`
implementations to indicate that they don't care what the max level is,
the presence of a single `None` layer will globally disable everything,
which is not the wanted behavior.
Fixes#2265
## Solution
This branch introduces a special downcast marker that can be used to
detect when a `Layer` in a `Layered` is `None`. This allows the
`pick_level_hint` method to short-circuit when a `Layer` implementation
which is `None` returns `Some(LevelFilter::OFF)` from its
`max_level_hint` if the other half of the `Layered` is not `None`. The
tests I added should be pretty thorough!
Additionally, the downcast marker is special-cased in the `reload`
`Layer`. Normally, this `Layer` doesn't support downcasting, but it can
in the case of the special marker value.
Co-authored-by: Eliza Weisman <eliza@buoyant.io>
## Motivation
`tracing-appender` does not have `Rotation` based on size yet. Also, it
doesn't have the feature of keeping the most recent `N` log files
I believe the second feature is more easy to implement, and also will
partially solve the `Rotation` based on size problem. Because people may
choose `hourly` or `daily` rotation based on their needs, and put an
extra boundary of `keep the last 5 files` for example. Of course it
won't handle all the edge cases for `Rotation` based on size. But it
will cover most of the scenarios. And also, it is a good feature to have
on its own :)
## Solution
Introduce another field called `max_files: Option<usize>` to the `Inner`
of `RollingFileAppender` struct. I managed to did not touch any of the
existing functions, so it **WON'T BE A BREAKING CHANGE**. Yay :)
The solution is, whenever the rotation should happen, the
`refresh_writer()` is called. So I embed the following logic into that
function:
1- check the log folder and detect the log files 2- if there are more
log files than the `max_files` amount 3- store the filenames in a
vector, and sort them by their dates (dates are already present in the
filename) 4- keep deleting the oldest ones, till we have desired amount
of log files in the log folder
P.S. this PR was opened before, but got closed since it would be easier
for the maintainers to target `master` branch instead of `v0.1.x` Also,
@CBenoit contributed to this PR, it would be great to give credit to him
:)
Co-authored-by: Benoît Cortier <bcortier@proton.me>
Co-authored-by: Eliza Weisman <eliza@buoyant.io>
Allows `Subscriber`s and `Layer`s to stash their own `Dispatch` without
causing a memory leak.
## Motivation
Resolves a shortcoming of #2269: that it's impossible for `Subscriber`s
or `Layer`s to stash a copy of their own `Dispatch` without creating a
reference cycle that would prevent them from being dropped.
## Solution
Introduces `WeakDispatch` (analogous to `std::sync::Weak`) that holds a
weak pointer to a `Subscriber`. `WeakDispatch` can be created via
`Dispatch::downgrade`, and can be transformed into a `Dispatch` via
`WeakDispatch::upgrade`.
Co-authored-by: Eliza Weisman <eliza@buoyant.io>
