1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307
//! An experimental new error-handling library. Guide-style introduction
//! is available [here](https://boats.gitlab.io/failure/).
//!
//! The primary items exported by this library are:
//!
//! - `Fail`: a new trait for custom error types in Rust.
//! - `Error`: a wrapper around `Fail` types to make it easy to coalesce them
//! at higher levels.
//!
//! As a general rule, library authors should create their own error types and
//! implement `Fail` for them, whereas application authors should primarily
//! deal with the `Error` type. There are exceptions to this rule, though, in
//! both directions, and users should do whatever seems most appropriate to
//! their situation.
//!
//! ## Backtraces
//!
//! Backtraces are disabled by default. To turn backtraces on, enable
//! the `backtrace` Cargo feature and set the `RUST_BACKTRACE` environment
//! variable to a non-zero value (this also enables backtraces for panics).
//! Use the `RUST_FAILURE_BACKTRACE` variable to enable or disable backtraces
//! for `failure` specifically.
#![cfg_attr(not(feature = "std"), no_std)]
#![deny(missing_docs)]
#![deny(warnings)]
#![cfg_attr(feature = "small-error", feature(extern_types, allocator_api))]
macro_rules! with_std { ($($i:item)*) => ($(#[cfg(feature = "std")]$i)*) }
macro_rules! without_std { ($($i:item)*) => ($(#[cfg(not(feature = "std"))]$i)*) }
// Re-export libcore using an alias so that the macros can work without
// requiring `extern crate core` downstream.
#[doc(hidden)]
pub extern crate core as _core;
mod as_fail;
mod backtrace;
#[cfg(feature = "std")]
mod box_std;
mod compat;
mod context;
mod result_ext;
use core::any::TypeId;
use core::fmt::{Debug, Display};
pub use as_fail::AsFail;
pub use backtrace::Backtrace;
pub use compat::Compat;
pub use context::Context;
pub use result_ext::ResultExt;
#[cfg(feature = "failure_derive")]
#[allow(unused_imports)]
#[macro_use]
extern crate failure_derive;
#[cfg(feature = "failure_derive")]
#[doc(hidden)]
pub use failure_derive::*;
with_std! {
extern crate core;
mod sync_failure;
pub use sync_failure::SyncFailure;
mod error;
use std::error::Error as StdError;
pub use error::Error;
/// A common result with an `Error`.
pub type Fallible<T> = Result<T, Error>;
mod macros;
mod error_message;
pub use error_message::err_msg;
}
/// The `Fail` trait.
///
/// Implementors of this trait are called 'failures'.
///
/// All error types should implement `Fail`, which provides a baseline of
/// functionality that they all share.
///
/// `Fail` has no required methods, but it does require that your type
/// implement several other traits:
///
/// - `Display`: to print a user-friendly representation of the error.
/// - `Debug`: to print a verbose, developer-focused representation of the
/// error.
/// - `Send + Sync`: Your error type is required to be safe to transfer to and
/// reference from another thread
///
/// Additionally, all failures must be `'static`. This enables downcasting.
///
/// `Fail` provides several methods with default implementations. Two of these
/// may be appropriate to override depending on the definition of your
/// particular failure: the `cause` and `backtrace` methods.
///
/// The `failure_derive` crate provides a way to derive the `Fail` trait for
/// your type. Additionally, all types that already implement
/// `std::error::Error`, and are also `Send`, `Sync`, and `'static`, implement
/// `Fail` by a blanket impl.
pub trait Fail: Display + Debug + Send + Sync + 'static {
/// Returns the "name" of the error.
///
/// This is typically the type name. Not all errors will implement
/// this. This method is expected to be most useful in situations
/// where errors need to be reported to external instrumentation systems
/// such as crash reporters.
fn name(&self) -> Option<&str> {
None
}
/// Returns a reference to the underlying cause of this failure, if it
/// is an error that wraps other errors.
///
/// Returns `None` if this failure does not have another error as its
/// underlying cause. By default, this returns `None`.
///
/// This should **never** return a reference to `self`, but only return
/// `Some` when it can return a **different** failure. Users may loop
/// over the cause chain, and returning `self` would result in an infinite
/// loop.
fn cause(&self) -> Option<&dyn Fail> {
None
}
/// Returns a reference to the `Backtrace` carried by this failure, if it
/// carries one.
///
/// Returns `None` if this failure does not carry a backtrace. By
/// default, this returns `None`.
fn backtrace(&self) -> Option<&Backtrace> {
None
}
/// Provides context for this failure.
///
/// This can provide additional information about this error, appropriate
/// to the semantics of the current layer. That is, if you have a
/// lower-level error, such as an IO error, you can provide additional context
/// about what that error means in the context of your function. This
/// gives users of this function more information about what has gone
/// wrong.
///
/// This takes any type that implements `Display`, as well as
/// `Send`/`Sync`/`'static`. In practice, this means it can take a `String`
/// or a string literal, or another failure, or some other custom context-carrying
/// type.
fn context<D>(self, context: D) -> Context<D>
where
D: Display + Send + Sync + 'static,
Self: Sized,
{
Context::with_err(context, self)
}
/// Wraps this failure in a compatibility wrapper that implements
/// `std::error::Error`.
///
/// This allows failures to be compatible with older crates that
/// expect types that implement the `Error` trait from `std::error`.
fn compat(self) -> Compat<Self>
where
Self: Sized,
{
Compat { error: self }
}
#[doc(hidden)]
#[deprecated(since = "0.1.2", note = "please use the 'iter_chain()' method instead")]
fn causes(&self) -> Causes
where
Self: Sized,
{
Causes { fail: Some(self) }
}
#[doc(hidden)]
#[deprecated(
since = "0.1.2",
note = "please use the 'find_root_cause()' method instead"
)]
fn root_cause(&self) -> &dyn Fail
where
Self: Sized,
{
find_root_cause(self)
}
#[doc(hidden)]
fn __private_get_type_id__(&self) -> TypeId {
TypeId::of::<Self>()
}
}
impl dyn Fail {
/// Attempts to downcast this failure to a concrete type by reference.
///
/// If the underlying error is not of type `T`, this will return `None`.
pub fn downcast_ref<T: Fail>(&self) -> Option<&T> {
if self.__private_get_type_id__() == TypeId::of::<T>() {
unsafe { Some(&*(self as *const dyn Fail as *const T)) }
} else {
None
}
}
/// Attempts to downcast this failure to a concrete type by mutable
/// reference.
///
/// If the underlying error is not of type `T`, this will return `None`.
pub fn downcast_mut<T: Fail>(&mut self) -> Option<&mut T> {
if self.__private_get_type_id__() == TypeId::of::<T>() {
unsafe { Some(&mut *(self as *mut dyn Fail as *mut T)) }
} else {
None
}
}
/// Returns the "root cause" of this `Fail` - the last value in the
/// cause chain which does not return an underlying `cause`.
///
/// If this type does not have a cause, `self` is returned, because
/// it is its own root cause.
///
/// This is equivalent to iterating over `iter_causes()` and taking
/// the last item.
pub fn find_root_cause(&self) -> &dyn Fail {
find_root_cause(self)
}
/// Returns a iterator over the causes of this `Fail` with the cause
/// of this fail as the first item and the `root_cause` as the final item.
///
/// Use `iter_chain` to also include the fail itself.
pub fn iter_causes(&self) -> Causes {
Causes { fail: self.cause() }
}
/// Returns a iterator over all fails up the chain from the current
/// as the first item up to the `root_cause` as the final item.
///
/// This means that the chain also includes the fail itself which
/// means that it does *not* start with `cause`. To skip the outermost
/// fail use `iter_causes` instead.
pub fn iter_chain(&self) -> Causes {
Causes { fail: Some(self) }
}
/// Deprecated alias to `find_root_cause`.
#[deprecated(
since = "0.1.2",
note = "please use the 'find_root_cause()' method instead"
)]
pub fn root_cause(&self) -> &dyn Fail {
find_root_cause(self)
}
/// Deprecated alias to `iter_chain`.
#[deprecated(since = "0.1.2", note = "please use the 'iter_chain()' method instead")]
pub fn causes(&self) -> Causes {
Causes { fail: Some(self) }
}
}
#[cfg(feature = "std")]
impl<E: StdError + Send + Sync + 'static> Fail for E {}
#[cfg(feature = "std")]
impl Fail for Box<dyn Fail> {
fn cause(&self) -> Option<&dyn Fail> {
(**self).cause()
}
fn backtrace(&self) -> Option<&Backtrace> {
(**self).backtrace()
}
}
/// A iterator over the causes of a `Fail`
pub struct Causes<'f> {
fail: Option<&'f dyn Fail>,
}
impl<'f> Iterator for Causes<'f> {
type Item = &'f dyn Fail;
fn next(&mut self) -> Option<&'f dyn Fail> {
self.fail.map(|fail| {
self.fail = fail.cause();
fail
})
}
}
fn find_root_cause(mut fail: &dyn Fail) -> &dyn Fail {
while let Some(cause) = fail.cause() {
fail = cause;
}
fail
}