pw_status

pw_status provides features for communicating the result of an operation. The classes in pw_status are used extensively throughout Pigweed.

pw::Status

The primary feature of pw_status is the pw::Status class. pw::Status (pw_status/status.h) is a simple, zero-overhead status object that wraps a status code.

pw::Status uses Google’s standard status codes (see the Google APIs repository). These codes are used extensively in Google projects including Abseil (status/status.h ) and gRPC (doc/statuscodes.md).

An OK Status is created by the pw::OkStatus function or by the default Status constructor. Non-OK Status is created with a static member function that corresponds with the status code.

// Ok (gRPC code "OK") does not indicate an error; this value is returned on
// success. It is typical to check for this value before proceeding on any
// given call across an API or RPC boundary. To check this value, use the
// `status.ok()` member function rather than inspecting the raw code.
//
// OkStatus() is provided as a free function, rather than a static member
// function like the error statuses to avoid conflicts with the ok() member
// function. Status::Ok() would be too similar to Status::ok().
pw::OkStatus()

// Cancelled (gRPC code "CANCELLED") indicates the operation was cancelled,
// typically by the caller.
pw::Status::Cancelled()

// Unknown (gRPC code "UNKNOWN") indicates an unknown error occurred. In
// general, more specific errors should be raised, if possible. Errors raised
// by APIs that do not return enough error information may be converted to
// this error.
pw::Status::Unknown()

// InvalidArgument (gRPC code "INVALID_ARGUMENT") indicates the caller
// specified an invalid argument, such a malformed filename. Note that such
// errors should be narrowly limited to indicate to the invalid nature of the
// arguments themselves. Errors with validly formed arguments that may cause
// errors with the state of the receiving system should be denoted with
// `FailedPrecondition` instead.
pw::Status::InvalidArgument()

// DeadlineExceeded (gRPC code "DEADLINE_EXCEEDED") indicates a deadline
// expired before the operation could complete. For operations that may change
// state within a system, this error may be returned even if the operation has
// completed successfully. For example, a successful response from a server
// could have been delayed long enough for the deadline to expire.
pw::Status::DeadlineExceeded()

// NotFound (gRPC code "NOT_FOUND") indicates some requested entity (such as
// a file or directory) was not found.
//
// `NotFound` is useful if a request should be denied for an entire class of
// users, such as during a gradual feature rollout or undocumented allow list.
// If, instead, a request should be denied for specific sets of users, such as
// through user-based access control, use `PermissionDenied` instead.
pw::Status::NotFound()

// AlreadyExists (gRPC code "ALREADY_EXISTS") indicates the entity that a
// caller attempted to create (such as file or directory) is already present.
pw::Status::AlreadyExists()

// PermissionDenied (gRPC code "PERMISSION_DENIED") indicates that the caller
// does not have permission to execute the specified operation. Note that this
// error is different than an error due to an *un*authenticated user. This
// error code does not imply the request is valid or the requested entity
// exists or satisfies any other pre-conditions.
//
// `PermissionDenied` must not be used for rejections caused by exhausting
// some resource. Instead, use `ResourceExhausted` for those errors.
// `PermissionDenied` must not be used if the caller cannot be identified.
// Instead, use `Unauthenticated` for those errors.
pw::Status::PermissionDenied()

// ResourceExhausted (gRPC code "RESOURCE_EXHAUSTED") indicates some resource
// has been exhausted, perhaps a per-user quota, or perhaps the entire file
// system is out of space.
pw::Status::ResourceExhausted()

// FailedPrecondition (gRPC code "FAILED_PRECONDITION") indicates that the
// operation was rejected because the system is not in a state required for
// the operation's execution. For example, a directory to be deleted may be
// non-empty, an "rmdir" operation is applied to a non-directory, etc.
//
// Some guidelines that may help a service implementer in deciding between
// `FailedPrecondition`, `Aborted`, and `Unavailable`:
//
//  (a) Use `Unavailable` if the client can retry just the failing call.
//  (b) Use `Aborted` if the client should retry at a higher transaction
//      level (such as when a client-specified test-and-set fails, indicating
//      the client should restart a read-modify-write sequence).
//  (c) Use `FailedPrecondition` if the client should not retry until
//      the system state has been explicitly fixed. For example, if an "rmdir"
//      fails because the directory is non-empty, `FailedPrecondition`
//      should be returned since the client should not retry unless
//      the files are deleted from the directory.
pw::Status::FailedPrecondition()

// Aborted (gRPC code "ABORTED") indicates the operation was aborted,
// typically due to a concurrency issue such as a sequencer check failure or a
// failed transaction.
//
// See the guidelines above for deciding between `FailedPrecondition`,
// `Aborted`, and `Unavailable`.
pw::Status::Aborted()

// OutOfRange (gRPC code "OUT_OF_RANGE") indicates the operation was
// attempted past the valid range, such as seeking or reading past an
// end-of-file.
//
// Unlike `InvalidArgument`, this error indicates a problem that may
// be fixed if the system state changes. For example, a 32-bit file
// system will generate `InvalidArgument` if asked to read at an
// offset that is not in the range [0,2^32-1], but it will generate
// `OutOfRange` if asked to read from an offset past the current
// file size.
//
// There is a fair bit of overlap between `FailedPrecondition` and
// `OutOfRange`.  We recommend using `OutOfRange` (the more specific
// error) when it applies so that callers who are iterating through
// a space can easily look for an `OutOfRange` error to detect when
// they are done.
pw::Status::OutOfRange()

// Unimplemented (gRPC code "UNIMPLEMENTED") indicates the operation is not
// implemented or supported in this service. In this case, the operation
// should not be re-attempted.
pw::Status::Unimplemented()

// Internal (gRPC code "INTERNAL") indicates an internal error has occurred
// and some invariants expected by the underlying system have not been
// satisfied. This error code is reserved for serious errors.
pw::Status::Internal()

// Unavailable (gRPC code "UNAVAILABLE") indicates the service is currently
// unavailable and that this is most likely a transient condition. An error
// such as this can be corrected by retrying with a backoff scheme. Note that
// it is not always safe to retry non-idempotent operations.
//
// See the guidelines above for deciding between `FailedPrecondition`,
// `Aborted`, and `Unavailable`.
pw::Status::Unavailable()

// DataLoss (gRPC code "DATA_LOSS") indicates that unrecoverable data loss or
// corruption has occurred. As this error is serious, proper alerting should
// be attached to errors such as this.
pw::Status::DataLoss()

// Unauthenticated (gRPC code "UNAUTHENTICATED") indicates that the request
// does not have valid authentication credentials for the operation. Correct
// the authentication and try again.
pw::Status::Unauthenticated()

Note

Status enumerations are also supported for Python and Typescript.

Tracking the first error encountered

In some contexts it is useful to track the first error encountered while allowing execution to continue. Manually writing out if statements to check and then assign quickly becomes verbose, and doesn’t explicitly highlight the intended behavior of “latching” to the first error.

Status overall_status;
for (Sector& sector : sectors) {
  Status erase_status = sector.Erase();
  if (!overall_status.ok()) {
    overall_status = erase_status;
  }

  if (erase_status.ok()) {
    Status header_write_status = sector.WriteHeader();
    if (!overall_status.ok()) {
      overall_status = header_write_status;
    }
  }
}
return overall_status;

pw::Status has an Update() helper function that does exactly this to reduce visual clutter and succinctly highlight the intended behavior.

Status overall_status;
for (Sector& sector : sectors) {
  Status erase_status = sector.Erase();
  overall_status.Update(erase_status);

  if (erase_status.ok()) {
    overall_status.Update(sector.WriteHeader());
  }
}
return overall_status;

Unused result warnings

If the PW_STATUS_CFG_CHECK_IF_USED option is enabled, pw::Status objects returned from function calls must be used or it is a compilation error. To silence these warnings call IgnoreError() on the returned status object.

PW_STATUS_CFG_CHECK_IF_USED defaults to off. Pigweed compiles with this option enabled, but projects that use Pigweed will need to be updated to compile with this option. After all projects have migrated, unused result warnings will be enabled unconditionally.

C compatibility

pw_status provides the C-compatible pw_Status enum for the status codes. For ease of use, pw::Status implicitly converts to and from pw_Status. However, the pw_Status enum should never be used in C++; instead use the Status class.

The values of the pw_Status enum are all-caps and prefixed with PW_STATUS_. For example, PW_STATUS_DATA_LOSS corresponds with the C++ Status::DataLoss().

StatusWithSize

pw::StatusWithSize (pw_status/status_with_size.h) is a convenient, efficient class for reporting a status along with an unsigned integer value. It is similar to the pw::Result<T> class, but it stores both a size and a status, regardless of the status value, and only supports a limited range (27 bits).

pw::StatusWithSize values may be created with functions similar to pw::Status. For example,

// An OK StatusWithSize with a size of 123.
StatusWithSize(123)

// A NOT_FOUND StatusWithSize with a size of 0.
StatusWithSize::NotFound()

// A RESOURCE_EXHAUSTED StatusWithSize with a size of 10.
StatusWithSize::ResourceExhausted(10)

PW_TRY

PW_TRY (pw_status/try.h) is a convenient set of macros for working with Status and StatusWithSize objects in functions that return Status or StatusWithSize. The PW_TRY and PW_TRY_WITH_SIZE macros call a function and do an early return if the function’s return status is not ok.

Example:

Status PwTryExample() {
  PW_TRY(FunctionThatReturnsStatus());
  PW_TRY(FunctionThatReturnsStatusWithSize());

  // Do something, only executed if both functions above return OK.
}

StatusWithSize PwTryWithSizeExample() {
  PW_TRY_WITH_SIZE(FunctionThatReturnsStatus());
  PW_TRY_WITH_SIZE(FunctionThatReturnsStatusWithSize());

  // Do something, only executed if both functions above return OK.
}

PW_TRY_ASSIGN is for working with StatusWithSize objects in in functions that return Status. It is similar to PW_TRY with the addition of assigning the size from the StatusWithSize on ok.

Status PwTryAssignExample() {
  size_t size_value
  PW_TRY_ASSIGN(size_value, FunctionThatReturnsStatusWithSize());

  // Do something that uses size_value. size_value is only assigned and this
  // following code executed if the PW_TRY_ASSIGN function above returns OK.
}

Compatibility

C++14

Zephyr

To enable pw_status for Zephyr add CONFIG_PIGWEED_STATUS=y to the project’s configuration.