Struct tracing_core::metadata::Level
source · [−]pub struct Level(_);
Expand description
Describes the level of verbosity of a span or event.
Comparing Levels
Level
implements the PartialOrd
and Ord
traits, allowing two
Level
s to be compared to determine which is considered more or less
verbose. Levels which are more verbose are considered “greater than” levels
which are less verbose, with Level::ERROR
considered the lowest, and
Level::TRACE
considered the highest.
For example:
use tracing_core::Level;
assert!(Level::TRACE > Level::DEBUG);
assert!(Level::ERROR < Level::WARN);
assert!(Level::INFO <= Level::DEBUG);
assert_eq!(Level::TRACE, Level::TRACE);
Filtering
Level
s are typically used to implement filtering that determines which
spans and events are enabled. Depending on the use case, more or less
verbose diagnostics may be desired. For example, when running in
development, DEBUG
-level traces may be enabled by default. When running in
production, only INFO
-level and lower traces might be enabled. Libraries
may include very verbose diagnostics at the DEBUG
and/or TRACE
levels.
Applications using those libraries typically chose to ignore those traces. However, when
debugging an issue involving said libraries, it may be useful to temporarily
enable the more verbose traces.
The LevelFilter
type is provided to enable filtering traces by
verbosity. Level
s can be compared against LevelFilter
s, and
LevelFilter
has a variant for each Level
, which compares analogously
to that level. In addition, LevelFilter
adds a LevelFilter::OFF
variant, which is considered “less verbose” than every other `Level. This is
intended to allow filters to completely disable tracing in a particular context.
For example:
use tracing_core::{Level, LevelFilter};
assert!(LevelFilter::OFF < Level::TRACE);
assert!(LevelFilter::TRACE > Level::DEBUG);
assert!(LevelFilter::ERROR < Level::WARN);
assert!(LevelFilter::INFO <= Level::DEBUG);
assert!(LevelFilter::INFO >= Level::INFO);
Examples
Below is a simple example of how a Subscriber
could implement filtering through
a LevelFilter
. When a span or event is recorded, the Subscriber::enabled
method
compares the span or event’s Level
against the configured LevelFilter
.
The optional Subscriber::max_level_hint
method can also be implemented to allow spans
and events above a maximum verbosity level to be skipped more efficiently,
often improving performance in short-lived programs.
use tracing_core::{span, Event, Level, LevelFilter, Subscriber, Metadata};
#[derive(Debug)]
pub struct MySubscriber {
/// The most verbose level that this subscriber will enable.
max_level: LevelFilter,
// ...
}
impl MySubscriber {
/// Returns a new `MySubscriber` which will record spans and events up to
/// `max_level`.
pub fn with_max_level(max_level: LevelFilter) -> Self {
Self {
max_level,
// ...
}
}
}
impl Subscriber for MySubscriber {
fn enabled(&self, meta: &Metadata<'_>) -> bool {
// A span or event is enabled if it is at or below the configured
// maximum level.
meta.level() <= &self.max_level
}
// This optional method returns the most verbose level that this
// subscriber will enable. Although implementing this method is not
// *required*, it permits additional optimizations when it is provided,
// allowing spans and events above the max level to be skipped
// more efficiently.
fn max_level_hint(&self) -> Option<LevelFilter> {
Some(self.max_level)
}
// Implement the rest of the subscriber...
fn new_span(&self, span: &span::Attributes<'_>) -> span::Id {
// ...
}
fn event(&self, event: &Event<'_>) {
// ...
}
// ...
}
It is worth noting that the tracing-subscriber
crate provides additional
APIs for performing more sophisticated filtering, such as
enabling different levels based on which module or crate a span or event is
recorded in.
Implementations
The “trace” level.
Designates very low priority, often extremely verbose, information.
Trait Implementations
type Err = ParseLevelError
type Err = ParseLevelError
The associated error which can be returned from parsing.
This method returns an ordering between self
and other
values if one exists. Read more
This method tests less than (for self
and other
) and is used by the <
operator. Read more
This method tests less than or equal to (for self
and other
) and is used by the <=
operator. Read more
This method tests greater than (for self
and other
) and is used by the >
operator. Read more
This method returns an ordering between self
and other
values if one exists. Read more
This method tests less than (for self
and other
) and is used by the <
operator. Read more
This method tests less than or equal to (for self
and other
) and is used by the <=
operator. Read more
This method tests greater than (for self
and other
) and is used by the >
operator. Read more
This method returns an ordering between self
and other
values if one exists. Read more
This method tests less than (for self
and other
) and is used by the <
operator. Read more
This method tests less than or equal to (for self
and other
) and is used by the <=
operator. Read more
This method tests greater than (for self
and other
) and is used by the >
operator. Read more
This method tests greater than or equal to (for self
and other
) and is used by the >=
operator. Read more