This PR removes `#[doc(hidden)]` from types/functions in stable crates
(as defined in
[CrateSet.kt](ad520b080a/buildSrc/src/main/kotlin/CrateSet.kt (L19-L33))).
They are now `pub`, however, for those that are not intended to be used
directly, we preserve the original docs to be explicit about it.
After this PR, stable crates contain neither
<details>
<summary>#[doc(hidden)]</summary>
```
➜ smithy-rs git:(ysaito/remove-doc-hidden) rg -l '#\[doc\(hidden\)\]'
aws/rust-runtime/aws-http/src/user_agent.rs
CHANGELOG.next.toml
CHANGELOG.md
rust-runtime/aws-smithy-http/src/lib.rs
rust-runtime/aws-smithy-http/src/event_stream/sender.rs
rust-runtime/aws-smithy-http/src/event_stream/receiver.rs
codegen-core/src/main/kotlin/software/amazon/smithy/rust/codegen/core/smithy/generators/error/ErrorImplGenerator.kt (this is only for CodegenTarget.SERVER)
codegen-core/src/test/kotlin/software/amazon/smithy/rust/codegen/core/smithy/generators/StructureGeneratorTest.kt
design/src/server/instrumentation.md
design/src/rfcs/rfc0039_forward_compatible_errors.md
design/src/rfcs/rfc0032_better_constraint_violations.md
design/src/rfcs/rfc0026_client_crate_organization.md
rust-runtime/aws-smithy-http-server/src/macros.rs
rust-runtime/aws-smithy-http-server/src/routing/mod.rs
rust-runtime/aws-smithy-http-server/src/shape_id.rs
rust-runtime/aws-smithy-http-server/src/body.rs
rust-runtime/aws-smithy-http-server/src/lib.rs
rust-runtime/aws-smithy-http-server/src/plugin/mod.rs
codegen-server/src/main/kotlin/software/amazon/smithy/rust/codegen/server/smithy/generators/MapConstraintViolationGenerator.kt
codegen-server/src/main/kotlin/software/amazon/smithy/rust/codegen/server/smithy/generators/CollectionConstraintViolationGenerator.kt
```
</details>
nor
<details>
<summary>DocHidden</summary>
```
➜ smithy-rs git:(ysaito/remove-doc-hidden) rg -l 'DocHidden'
codegen-core/src/main/kotlin/software/amazon/smithy/rust/codegen/core/rustlang/RustType.kt
codegen-server/src/main/kotlin/software/amazon/smithy/rust/codegen/server/smithy/generators/ServerBuilderConstraintViolations.kt
```
</details>
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: Russell Cohen <rcoh@amazon.com>
## Motivation and Context
<!--- Why is this change required? What problem does it solve? -->
It implements Display trait For DateTime
<!--- If it fixes an open issue, please link to the issue here -->
https://github.com/smithy-lang/smithy-rs/issues/3161
## Description
<!--- Describe your changes in detail -->
I implemented Display trait for DateTime
## Testing
<!--- Please describe in detail how you tested your changes -->
<!--- Include details of your testing environment, and the tests you ran
to -->
I used this test
```rust
fn test_display_formatting() {
// Create a DateTime instance for testing
let datetime = DateTime {
seconds: 1636761600, // Example timestamp (replace with your actual timestamp)
subsecond_nanos: 123456789, // Example subsecond nanos (replace with your actual value)
};
// Expected RFC-3339 formatted string
let expected = "2021-11-13T00:00:00.123456789Z";
// Format the DateTime using Display trait
let formatted = format!("{}", datetime);
// Assert that the formatted string matches the expected result
assert_eq!(formatted, expected);
}
```
<!--- see how your change affects other areas of the code, etc. -->
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: Hakan Vardar <hakovardo123@gmail.com>
This PR removes deprecations throughout smithy-rs client and the SDK.
Some of these deprecations were ignored by the compiler (in the case
where `#[deprecated]` was used with `pub use`), so these will need to be
explicitly called out in the changelog.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: Russell Cohen <rcoh@amazon.com>
## Motivation and Context
aws-sdk-rust#960
## Description
Use checked_pow to avoid overflow in backoff computation
## Testing
- unit test
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This function isn't useful.
---
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
[See the upgrade guide for this feature to learn
more.](https://github.com/awslabs/aws-sdk-rust/discussions/956)
The breaking change is to the `MinimumThroughputBody::new` method.
## Motivation and Context
<!--- Why is this change required? What problem does it solve? -->
<!--- If it fixes an open issue, please link to the issue here -->
#1562
## Description
<!--- Describe your changes in detail -->
https://github.com/awslabs/aws-sdk-rust/discussions/956
## Testing
<!--- Please describe in detail how you tested your changes -->
<!--- Include details of your testing environment, and the tests you ran
to -->
<!--- see how your change affects other areas of the code, etc. -->
I added several tests
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: John DiSanti <jdisanti@amazon.com>
Co-authored-by: Russell Cohen <rcoh@amazon.com>
## Motivation and Context
Stable crates MUST only expose other stable crates.
## Description
This:
- fixes the remaining issues
- adds a lint tool to be sure we don't expose unstable crates by
accident in the future
## Testing
CI run
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
The previous names are just too verbose. This PR shortens them and tries
to make things clearer.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR implements
[RFC-39](https://github.com/smithy-lang/smithy-rs/blob/main/design/src/rfcs/rfc0039_forward_compatible_errors.md)
with a couple slight deviations:
- No `introspect` method is added since `Error` already implements
`ProvideErrorMetadata`.
- The same opaqueness and deprecation pointer is applied to the enum
unknown variant for consistency.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
See [rendered
RFC](df518bfb59/design/src/rfcs/rfc0039_behavior_major_versions.md)
## Description
This add `BehaviorMajorVersions` to the SDK and wires them in up and
down the stack.
## Testing
- [x] lots of ITs / UTs
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR moves `RequestId` into the aws-types stable crate.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
The `try_insert` and `try_append` methods on `Headers` would panic when
passing a non-ASCII value for the key, or an invalid UTF-8 value for the
value. This was due to them using the http crate's `from_static` method,
which panics on these invalid values. This PR makes them always use
`try_from` instead.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
For the same reason that the request wrapper types were created in
#3059, this PR establishes the response wrapper types and refactors
existing code to use them.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
- Fixes#3171
- Fixes#3155
- Fixes#3172
## Description
- Add `configReExport` to make it easy and consistent to re-export types
into the config module when they are used!
- Add this for a bunch of types and clean up some (now) dead code
- Re export `BuildError`
- Enable converting from `BuildError` into `SdkError`
There are probably more places this can be used, but I'd like to keep
this change small to facilitate easy backport to 0.57.
## Testing
CI
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR fixes issues introduced by moving the repository from
awslabs/smithy-rs to smithy-lang/smithy-rs.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
The tokio `StreamReader` implements `AsyncBufRead`, but we're returning
`AsyncRead` currently. If a user needs an `AsyncBufRead`, then they've
to wrap the returned value in tokio `BufReader` for no reason. Since
`ByteStream` doesn't implement `Stream` anymore, one has to either wrap
the returned value unnecessarily or write a wrapper similar to this
function.
See
https://github.com/smithy-lang/smithy-rs/pull/2983#discussion_r1380822997.
## Description
Simply changed the return type to say `impl AsyncBufRead`. Since
`AsyncBufRead` is a super-set of `AsyncRead`, this is not a breaking
change.
## Testing
The code example tests it.
## Checklist
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: Russell Cohen <rcoh@amazon.com>
## Motivation and Context
This PR updates our smithy-rs release tooling to prepare for handling
both stable and unstable crates. Merging this PR to `main` does not make
any of the runtime crates stable, instead we will just have a supporting
mechanism in place, ready to be turned on by implementing `TODO(GA)`.
## Description
At a high level, this PR will
- update the `Release smithy-rs` workflow UI to have an additional input
`Stable semantic version`:
- split an existing gradle property `smithy.rs.runtime.crate.version`
into `smithy.rs.runtime.crate.stable.version` and
`smithy.rs.runtime.crate.unstable.version`.
- make `publisher upgrade-runtime-crates-version` take a new, optional
(for backwards compatibility) command line argument `--stable-version`
to use the said gradle property
`smithy.rs.runtime.crate.stable.version`. This will allow the
`publisher` to set a value passed via a workflow input `Stable semantic
version` to `smithy.rs.runtime.crate.stable.version` in
`gradle.properties` (and the same goes for unstable crates).
- update `fixRuntimeCrateVersions` so that it fixes up runtime crate
versions based on properties `smithy.rs.runtime.crate.stable.version`
and `smithy.rs.runtime.crate.unstable.version`.
**NOTE**
There is a guard in place. When this PR gets merged and then we enter a
stable crate version in the `Stable semantic version` text box, it will
be overwritten by a version in `Unstable semantic version`, so that
1.x.y version will NOT be published to `crates.io` before GA and that
also replicates today's release workflow's behavior (only publishing
unstable crates). Just make sure we specify a 0.x.y in the `Untable
semantic version` text box because that does get shipped.
### What happens once `TODO(GA)` has been implemented?
Roughly, it will look like this. When we run the `Release smithy-rs`
workflow (not a dry-run), providing a stable version (say 1.0.0) and a
unstable version (say 0.57.0) in the workflow text boxes,
1. the workflow will create a new release branch
`smithy-rs-release-1.x.y`.
2. the workflow will set 1.0.0 to
`smithy.rs.runtime.crate.stable.version` and 0.57.0 to
`smithy.rs.runtime.crate.unstable.version` in `gradle.properties`.
3. for whatever smithy runtime crates whose package metadata says
```
[package.metadata.smithy-rs-release-tooling]
stable = true
```
their `Cargo.toml` will include `version = 1.0.0` (and `version =
0.57.0` if `smithy-rs-release-tooling` is not specified or if its value
is `stable = false`).
4. the workflow will publish smithy runtime crates accordingly to
`crates.io`.
5. releasing `aws-sdk-rust` subsequently will render `version = 1.0.0`
in `Cargo.toml` for stable AWS runtime crates (again as specified by
`smithy-rs-release-tooling`), will render `version = 1.1.0` for SDK
crates (we will not go into details here as to why it's not `1.0.0`),
and will publish those crates to `crates.io`.
## Testing
In a [separate
branch](https://github.com/awslabs/smithy-rs/tree/ysaito/stable-and-unstable-crates)
that implemented `TODO(GA)`, I verified that
- our internal release pipeline was executed without errors
- a generated AWS SDK had following manifests (showing excerpts from
arbitrary crates)
```
[package]
name = "aws-sdk-s3"
version = "1.1.0"
```
```
[package]
name = "aws-smithy-types"
version = "1.0.0"
```
```
[package]
name = "aws-smithy-http"
version = "0.57.0"
```
```
[package]
name = "aws-types"
version = "1.0.0"
```
```
[package]
name = "aws-config"
version = "1.0.0"
```
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: John DiSanti <jdisanti@amazon.com>
## Motivation and Context
Implements #3100
## Description
Currently, we expose `aws_smithy_http::event_stream::Receiver` in
generated SDKs, as shown in the following S3's example (see[ a generated
diff](https://d2luzm2xt3nokh.cloudfront.net/codegen-diff/cc303ab1a07693ab02d5ec4f06101b628d1dbabe/1aa6a8da7d2b7669ba3ab7179a1fd72aadc03162/aws-sdk-ignore-whitespace/index.html)
for
`tmp-codegen-diff/aws-sdk/sdk/s3/src/operation/select_object_content/_select_object_content_output.rs`):
```
pub struct SelectObjectContentOutput {
<p>The array of results.</p>
pub payload: ::aws_smithy_http::event_stream::Receiver<
crate::types::SelectObjectContentEventStream,
crate::types::error::SelectObjectContentEventStreamError,
>,
...
```
This PR wraps `Receiver` in a new-type, called `EventReceiver`, which
then supports `pub async fn recv` method whose signature is the same as
`aws_smithy_http::event_stream::Receiver::recv`.
## Testing
Relied on existing tests (e.g. `s3` and `transcribestreaming`
integration tests cover uses cases affected by this change).
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Fixes the following statement in
https://github.com/awslabs/smithy-rs/pull/3114
> A notable breaking change is the error type of the new recv method.
Previously, it was [SdkError<E,
RawMessage>>](https://docs.rs/aws-smithy-http/0.57.0/aws_smithy_http/event_stream/struct.Receiver.html#method.recv)
but it is now a BoxError. This means that upon an event receive error,
customers can only show what the error is, but not match on it and take
some actions.
so that the `recv` method on a new-type wrapper
`aws_smithy_http::event_stream::Receiver` can expose `SdkError<E,
RawMessage>>` in public API. It is the responsibility of the above PR to
move
[RawMessage](c7f0d5dc33/rust-runtime/aws-smithy-http/src/event_stream/receiver.rs (L92-L98))
from `aws-smithy-http` to `aws-smithy-types` but doing so first requires
`Message` to be moved to `aws-smithy-types`, hence this PR.
## Description
Moving `Message` to `aws-smithy-types` also requires moving the types it
depends upon, namely `Header`, `HederValue`, and `StrBytes`. However,
their serialization/deserialization methods (`read_from` and `write_to`)
remain in `aws-smithy-eventstream` and they are converted to free
functions, with the aim of moving things as minimally as possible.
`aws-smithy-eventstream` has fuzz testing. It previously implemented the
`Arbitrary` trait for moved types, but since they are now in
`aws-smithy-types`, we need to have newtypes for them to enable the
trait due to the orphan rules. A newly added module
`aws-smithy-eventstream::arbitrary` is exactly for that purpose.
## Testing
Relied on the existing tests.
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Fix examples
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR moves the HTTP types into the root of aws-smithy-runtime-api
since they're not client-specific, and the serializers/deserializers
will need to rely on them. It also refactors the headers and errors out
of the request module.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
https://docs.rs/crate/aws-smithy-types/0.57.0/builds/954806
Smithy-types docs failed to build.
1. Update CI to run a command more like docs.rs
2. Use auto_doc instead
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
Fixes#3112.
I opted to upgrade to the latest 0.17.5 in spite of the guidance in
[RFC-21](https://github.com/awslabs/smithy-rs/blob/main/design/src/rfcs/rfc0021_dependency_versions.md)
since this is a security critical dependency, and the maintainer has
aggressively yanked old versions in the past.
Note: ring 0.16 is still pulled in by rcgen as a dev dependency in
aws-smithy-http-server-python after these changes. The rcgen crate
hasn't upgraded yet.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
PR #3095 added a code-generated `${serviceName}Config` object on which
users can register layers and plugins. For example:
```rust
let config = PokemonServiceConfig::builder()
.layer(layers)
.http_plugin(authn_plugin)
.model_plugin(authz_plugin)
.build();
```
This PR makes it so that server decorators can inject methods on this
config builder object. These methods can apply arbitrary layers, HTTP
plugins, and/or model plugins. Moreover, the decorator can configure
whether invoking such method is required or not.
For example, a decorator can inject an `aws_auth` method that configures
some plugins using its input arguments. Missing invocation of this
method
will result in the config failing to build:
```rust
let _: SimpleServiceConfig<
// No layers have been applied.
tower::layer::util::Identity,
// One HTTP plugin has been applied.
PluginStack<IdentityPlugin, IdentityPlugin>,
// One model plugin has been applied.
PluginStack<IdentityPlugin, IdentityPlugin>,
> = SimpleServiceConfig::builder()
// This method has been injected in the config builder!
.aws_auth("a".repeat(69).to_owned(), 69)
// The method will apply one HTTP plugin and one model plugin,
// configuring them with the input arguments. Configuration can be
// declared to be fallible, in which case we get a `Result` we unwrap
// here.
.expect("failed to configure aws_auth")
.build()
// Since `aws_auth` has been marked as required, if the user misses
// invoking it, this would panic here.
.unwrap();
```
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This is a pared-down version of #2809.
This is the code-generated version of the
david-perez/smithy-rs-service-config#2 POC.
This introduces a code-generated `PokemonServiceConfig` object which is
provided to the service builder constructor via
`PokemonService::builder(config: PokemonServiceConfig)`. This will
displace the
current `builder_without_plugins` and `builder_with_plugins`,
deprecating them.
We will eventually remove them.
The motivation is to have a single place where to register plugins and
layers.
Smithy traits can leverage this object to inject methods that can apply
plugins
and layers. For example, an `@authorization` trait implementer could
vend a
smithy-rs decorator that pre-applied an authorization plugin inside the
`${serviceName}Config` object, possibly exposing a method to pass in any
runtime arguments needed to configure the middleware.
## Checklist
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
- small improvements / consistency updates to http wrapper
- fix bug where `.insert` was called but it should have been `append`
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Implements #3092
## Description
This PR moves the `connection` and `result` modules from
`aws-smithy-http` to `aws_smithy_runtime_api`. `SdkError` is primarily
used within the context of the orchestrator, hence
`aws_smithy_runtime_api` instead of `aws-smithy-types`.
Like the previous sibling PRs, type aliases for affected pub items are
left in `aws_smithy_http` for backwards compatibility. However, due to
lack of trait aliases, a trait `CreateUnhandledError` needs to be
consumed from `aws_smithy_runtime_api`.
This PR also updates existing places that consumed those moved types so
they now depend on `aws-smithy-runtime-api` to do so.
## Testing
Relied on the existing tests, which ensured no deprecated warnings of
using moved types from the old place were issued.
## Checklist
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Follow up on #3088
## Description
This PR reverts `ByteStream::read_with_body_0_4_from`,
`ByteStream::from_path_body_0_4`, and `ByteStream::from_file_body_0_4`
to the old names since from a customers' point of view, `FsBuilder`
should not mention anything about an `http-body` version at the API
level.
`FsBuilder` is currently an opt-in feature (with `rt-tokio` enabled) and
we make it tied to `http-body-0-4-x` by requiring `rt-tokio` including a
dependency on `http-body` version 0.4.x.
## Testing
Relied on the existing tests in CI
## Checklist
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
<!--- Why is this change required? What problem does it solve? -->
<!--- If it fixes an open issue, please link to the issue here -->
https://github.com/awslabs/smithy-rs/issues/1562
## Description
<!--- Describe your changes in detail -->
This change adds a new body wrapper: The minimum throughput limit
wrapper. It tracks the rate that data is being streamed from itself. If
that rate falls below some configurable limit, it emits an error instead
of the next chunk. This protects users from requests that start quickly
but then slow down considerably.
I'd like to get this merged and then figure out the
codegen/docs/examples/config part in a separate PR.
## Testing
<!--- Please describe in detail how you tested your changes -->
<!--- Include details of your testing environment, and the tests you ran
to -->
<!--- see how your change affects other areas of the code, etc. -->
Tests are included.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
https://github.com/awslabs/smithy-rs/pull/3076 missed type alias for
some types we moved to `aws-smithy-types`. This small PR will fix that.
## Testing
Relied on the existing tests in CI
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
The existing credentials provider was a DOS risk and didn't obey timeout
settings because it used `std::timeout::spawn` but relied on a
async-based timeout mechanism.
## Description
- Use `tokio::process` instead
- introduce new cargo feature
## Testing
- unit test
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Implements https://github.com/awslabs/smithy-rs/issues/3033
## Description
This PR hides behind cargo features the third-party types from
`http-body` and `hyper` crates that are used in`aws-smithy-types`'
public API. Customers need to opt-in by enabling a cargo feature
`http-body-0-4-x` in `aws-smithy-types` to create an `SdkBody` or
`ByteStream` using those third-party types. For more details, please see
[the upgrade
guide](https://github.com/awslabs/smithy-rs/discussions/3089).
As can been seen from code changes, to reduce the surface area where we
need to feature-gate things, we have fused the
`aws_smithy_types::body::Inner::Streaming` enum variant into
`aws_smithy_types::body::Inner::Dyn` variant, thereby removing
`SdkBody::from_dyn`.
## Testing
Relied on existing tests in CI
## Checklist
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: Zelda Hessler <zhessler@amazon.com>
## Motivation and Context
- Fixes#3034
## Description
Because AuthSchemeOptions were being registered at the operation level,
there was no way for them to be overridden by customer-provided runtime
plugins. This moves them into a separate plugin that is added at
Client/Default priority.
## Testing
- new unit test
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR replaces the credentials cache with the new identity cache, and
adds config validation via the `SharedConfigValidator` runtime component
and `ValidateConfig` trait.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
# Motivation and Context
A follow-up on #3026
## Description
#3026 moved
- `aws_smithy_http::body::{BoxBody, Error, SdkBody}` to
`aws_smithy_types::body::{BoxBody, Error, SdkBody}`
- `aws_smithy_http::byte_stream::{AggregatedBytes, ByteStream,
error::Error}` to `aws_smithy_types::byte_stream::{AggregatedBytes,
ByteStream, error::Error}`
and also left "breadcrumbs", so that customers could still consume
updated types from `aws_smithy_http` after the move.
This PR turns breadcrumbs into deprecation messages (via
`#[deprecated(...)]`) and updates existing places that used to use moved
types from `aws_smithy_http` to directly depend on `aws_smithy_types`.
## Testing
Relied on tests in CI.
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Completes #3054 (a follow-up on #3032)
## Description
#3032 moved `BuildError` from `aws_smithy_http::operation::error` to
`aws_smithy_types::error::operation`. That PR also left "breadcrumbs",
so that customers could still consume `BuldError` from `aws_smithy_http`
after the move.
This PR turns breadcrumbs into deprecation messages (via
`#[deprecated(...)]`) and updates existing places that used to use moved
types from `aws_smithy_http` to `aws_smithy_types`.
## Testing
Relied on tests in CI.
## Checklist
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: Russell Cohen <rcoh@amazon.com>
## Motivation and Context
- Fixes https://github.com/awslabs/smithy-rs/issues/3043
As a follow up to #3072 this removes the old endpoint resolver
interfaces in favor of creating a per-service resolver trait.
This trait defines a `into_shared_resolver()` method which converts the
local trait into a global resolver that can be used with the
orchestrator.
## Description
<!--- Describe your changes in detail -->
## Testing
<!--- Please describe in detail how you tested your changes -->
<!--- Include details of your testing environment, and the tests you ran
to -->
<!--- see how your change affects other areas of the code, etc. -->
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
While implementing identity caching, I noticed we don't have a default
async sleep impl wired up for generic clients, which was causing caching
to panic in many tests since it needs a sleep impl for timeouts. I
likely need to figure out what to do other than panic if there's no
sleep impl, but that's a problem for a different PR.
This PR adds a default sleep impl to generic clients, and also
simplifies how default config works. Previously, the generated config
`Builder::build` method set all the defaults with a series of "if not
set, then set default" statements. In this PR, defaults are registered
via default ordered runtime plugins.
Additionally, I cleaned up the standard retry strategy:
- The `TokenBucketPartition` didn't appear to be used at all, so I
deleted it.
- `StandardRetryStrategy` was taking retry config at construction, which
means it isn't possible to config override the retry config unless the
strategy is recreated. It now doesn't take any config at all during
construction.
- The adaptive retry client rate limiter was created at construction
based on retry config at that point in time. This means config overrides
wouldn't work with it, so it is also no longer set up at construction
time.
- Removed some unused runtime plugins.
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR adds a `ResolveCachedIdentity` trait, and adapts the existing
LazyCredentialsCache implementation to it. It does NOT make this cache
used yet, as that will be done as a separate PR to keep code review
smaller.
Notable differences from the credentials cache:
- Supports multiple different identity types and identity resolvers
- Applies cache partitioning to the `SharedIdentityResolver` so that
operation config overrides of the identity resolver will result in the
correct identity being resolved.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
In preparation for inlining the old endpoint traits, I'm cleaning up the
endpoint resolver behavior.
## Description
1. The `Config` field of `service_runtime_plugin` was unused to I
deleted it
2. Store the endpoint resolver in `RuntimeComponents` instead of in
operation config. This allows a lot of complex logic around
`ConfigOverride` to be removed.
3. Move the default endpoint resolver into a runtime plugin.
4. Don't have a fallback implementation for `EndpointResolver`, allow
final construction of `RuntimeComponents` to fail
5. `ServiceRuntimePlugin` has been changed to `Defaults` so that it
doesn't overwrite settings set by the service config
## Testing
Well covered by existing UT / IT
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Takes care of the first part of
https://github.com/awslabs/smithy-rs/issues/3053 (the remaining part is
denoted as `TODO(runtimeCratesVersioningCleanup)` and until that and
https://github.com/awslabs/smithy-rs/issues/3033 are done the issue will
not be closed).
## Description
This PR moves from `aws-smithy-http` to `aws-smithy-types`:
- the `SdkBody` struct
- the `byte_stream` module
Within the origin crate, we leave "breadcrumbs" (i.e. reexports) for
existing customers of the items above so they are not broken by the
move.
We have just moved `SdkBody` to `aws-smithy-types` without renaming it
to see how it looks there. However, as
[TODO(naming)](https://github.com/awslabs/smithy-rs/pull/3026/files#diff-090c503b779024fdadb8ac97465c80438635df82e62c42b0d85b588a303d9a95R28)
says, we can choose to rename it to just `Body`. Curious to hear what
the reviewers think.
## Testing
Relied on the tests in CI.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: Russell Cohen <rcoh@amazon.com>
Agnostic in that it's generic over the operation it's applied to.
Figuring out how to write such a plugin can be fun but is not trivial.
I'm mostly adding this as a reference to refresh my memory the next time
I have to implement one of these.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR adds a `SsoTokenProvider` that loads session name cached SSO
tokens, and automatically refreshes them when possible. The
`SsoCredentialsProvider` is updated to use this new `SsoTokenProvider`
so that it is now compatible with the latest AWS CLI.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
- [ ] Upgrade guidance
- [x] changelog
To allow seamless adoption of the incoming `http` crate, we're updating
our client crates to use an intermediate type.
## Description
Update _client_ code to use `HttpRequest`. Some internal code still uses
`http::Request` but those usages are limited.
## Testing
CI
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR makes the orchestrator traits more idiomatic by making them
verbs.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
With `cargo-audit` v0.18.2, it reports the following vulnerabilities
when run against generated SDKs.
```
$ pwd
<smithy-rs root>/aws/sdk/build/aws-sdk
$ cargo generate-lockfile && cargo audit
...
Scanning Cargo.lock for vulnerabilities (339 crate dependencies)
Crate: atty
Version: 0.2.14
Warning: unsound
Title: Potential unaligned read
Date: 2021-07-04
ID: RUSTSEC-2021-0145
URL: https://rustsec.org/advisories/RUSTSEC-2021-0145
Dependency tree:
atty 0.2.14
├── env_logger 0.9.3
│ └── aws-credential-types 0.56.1
|
...
|___ criterion 0.4.0
```
Both of the latest `env_logger` and `criterion` removed their
dependencies on `atty`. This PR, therefore, updates a version of
`env_logger` and that of `criterion` within the `smithy-rs` codebase.
## Testing
Ran the above repro steps against SDKs generated from this branch. No
vulnerabilities detected.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR adds lifetimes to the `IdentityResolver`, `DnsResolver` (renamed
to `ResolveDns`), and `EndpointResolver` traits so that lifetime
gymnastics aren't needed when implementing those traits. For example,
`IdentityResolver::resolve_identity` takes `&ConfigBag` as an argument,
which means you have to pull things out of the ConfigBag outside of any
returned async block in order for the compiler to be satisfied. This
change removes that consideration and makes implementing these traits a
lot easier.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Takes care of the first part of
https://github.com/awslabs/smithy-rs/issues/3054 (the remaining part is
denoted as `TODO(runtimeCratesVersioningCleanup)` and until that's done
the issue will not be closed).
## Description
This PR moves from `aws-smithy-http::operation::error::{BuildError,
SerializationError}` to
`aws-smithy-types::error::operation::{BuildError, SerializationError}`.
Within the origin crate, we leave "breadcrumbs" (i.e. reexports) for
existing customers of the items above so they are not broken by the
move. The breadcrumps will be removed by the part two of the issue.
As part of the move, `aws_smithy_http::endpoint::EndpointPrefix::new`
now returns its `crate::endpoint::error::InvalidEndpointError` instead
of `crate::operation::error::BuildError` for two reasons:
- `BuildError` is now in a stable crate and for an `InvalidUri` variant
to be created, it needed to take a `http::uri::InvalidUri` from a
unstable crate, which we cannot expose in public API.
- The new error is more closely related to the endpoint business.
## Testing
Relied on the tests in CI.
## Checklist
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: John DiSanti <jdisanti@amazon.com>
This PR replaces the `Future` type-alias that was pointing to
`NowOrLater` with a new-type for each trait that returns a future.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
[Read the RFC here](https://github.com/awslabs/smithy-rs/pull/3018)
## Motivation and Context
<!--- Why is this change required? What problem does it solve? -->
<!--- If it fixes an open issue, please link to the issue here -->
#2417
## Description
<!--- Describe your changes in detail -->
Exactly what it says on the tin. I have a related RFC to publish that
goes into more depth.
## Testing
<!--- Please describe in detail how you tested your changes -->
<!--- Include details of your testing environment, and the tests you ran
to -->
<!--- see how your change affects other areas of the code, etc. -->
I wrote an integration test that ensures a custom retry classifier can
be set and is called.
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Implementation of the HTTP request wrapper type from RFC
## Description
RFC + implementation of HttpRequest
## Testing
- IT/UT
- Separate PR which uses this type in implementation of the SDK
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: John DiSanti <jdisanti@amazon.com>
This PR removes the remaining rust-runtime code to support middleware,
deprecates empty crates, and removes the remaining codegen references to
any of that code. In the interest of keeping code review easier, a
separate PR will finish addressing the remaining
`TODO(enableNewSmithyRuntime)` comments.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR removes the last usages of `DynConnector` and `HttpConnector`
from the `aws_smithy_client` crate with their counterparts in
aws-smithy-runtime-api and aws-smithy-runtime. It also introduces
`HttpClient` to make HTTP connector selection a smoother process, and
adds a runtime plugin that configures a default `HttpClient` so that
`Config` doesn't have to do that.
The `DnsService` from aws-config is also moved into aws-smithy-runtime
and refactored to be consistent with the other configurable traits in
the orchestrator since it will likely be used in the future for more
than just the ECS credentials provider.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
Issues fixed:
- `ConnectionPoisoningInterceptor` should use the
`read_after_deserialization` hook so that it's possible to poison
modeled transient errors.
- aws-config should enable connection poisoning on its IMDS/ECS/HTTP
clients.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
`AssumeRoleProvider` currently uses `ProviderConfig` as a source of
configuration, but that API is hard use and not intended for external
consumption.
This fixes the Assume Role issue but only for `AssumeRoleProvider`
## Description
Update the API (see changelog) to be more ergonomic and derive
configuration from `SdkConfig` instead.
## Testing
Existing tests + new unit tests
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This ports the direct uses of the `aws_smithy_client::Client` in
aws_config to the orchestrator.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Removes `futures_core::stream::Stream` from
`aws_smithy_http::byte_stream::ByteStream`
## Description
This PR is part of our ongoing effort,
https://github.com/awslabs/smithy-rs/issues/2413. We remove the
`futures_core::stream::Stream` trait from
`aws_smithy_http::byte_stream::ByteStream`. To continue support existing
places that relied upon `ByteStream` implementing the `Stream` trait, we
provide explicit `.next`, `.try_next`, and `.size_hints` methods on that
type. As part of it, a doc-hidden type `FuturesStreamCompatByteStream`,
which implements `Stream`, has been added to `aws_smithy_http` to cater
to those who need a type implementing `Stream` (see
[discussion](https://github.com/awslabs/smithy-rs/pull/2910#discussion_r1317112233)
why doc-hidden).
Another place we need to update is codegen responsible for rendering
stream payload serializer, and this affects the server. The regular
server and the python server have slightly different rendering
requirements, since the former uses
`aws_smithy_http::byte_stream::ByteStream` (that does _not_ implement
the `Stream` trait) and latter uses its own
[ByteStream](cb79a68e3c/rust-runtime/aws-smithy-http-server-python/src/types.rs (L343))
(that does implement `Stream`). We use
`ServerHttpBoundProtocolCustomization` to handle the difference:
`StreamPayloadSerializerCustomization` and
`PythonServerStreamPayloadSerializerCustomization`.
## Testing
No new behavior has been added, relied on the existing tests in CI.
## Checklist
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation
smithy-rs#1797
## Description
This PR adds Sigv4a support along with many tests that were ported from
the CRT Sigv4a project. **It does not include**:
- An E2E test. We should create one, but doing so requires us to set up
MRAP which is a good amount of work.
- A fuzz test. While I wrote and ran one on my machine, I found the fuzz
testing tool to be really difficult to use and it would take some work
to include fuzz testing in CI.
Additionally, test Sigv4a is annoying because Sigv4a signing is
non-deterministic. Because of this, the tests test signature generation
by first signing a request, and then by verifying that signature (_This
is how the CRT tests this feature_).
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: ysaito1001 <awsaito@amazon.com>
## Motivation and Context
Python middleware were unable to change the URI which makes it
impossible to change the route of an incoming request in Python.
## Description
URI has a #[setter] that Python side can use to change it.
## Testing
- Test has been added that ensures URI can be changed from python.
- If the URI is modified by a middleware in a local smithy-rs Python
server, a different route will be invoked.
## Checklist
- I've updated the changelog.next.toml
Co-authored-by: Fahad Zubair <fahadzub@amazon.com>
## Motivation and Context
Removes `futures_core::stream::Stream` from the `aws-smithy-async`
crate.
## Description
This PR is part of our ongoing effort,
https://github.com/awslabs/smithy-rs/issues/2413. We remove the
`futures_core::stream::Stream` trait from the public API in the
`aws-smithy-async` crate. While doing so, we compensate
- `FnStream` by providing the explicit `.next` and `.collect` methods to
let the previously working code continue working.
- `TryFlatMap` by making it return a new-type wrapper `PaginationStream`
to hide `FnStream` from those who use paginators.
With this change, the latest canary no longer uses
`tokio_stream::StreamExt`, since the paginator does not work in terms of
the `Stream` trait. Furthermore, `aws-sdk-rust` has been more than [3
releases since
release-2023-01-26](https://github.com/awslabs/aws-sdk-rust/releases),
so the module `release-2023-01-26` has been removed from
`canary-lambda`.
## Testing
No new tests added, but made sure the existing tests keep working with
the change.
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Closes#2831.
## Description
Fractional seconds will still be accepted during parsing, but not
emitted during serialization.
## Testing
```
./gradlew codegen-core:check codegen-client:check codegen-server:check codegen-client-test:check codegen-server-test:check
```
## Checklist
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: Russell Cohen <rcoh@amazon.com>
## Motivation and Context
<!--- Why is this change required? What problem does it solve? -->
<!--- If it fixes an open issue, please link to the issue here -->
smithy-rs#1767 aws-sdk-rust#536
## Description
<!--- Describe your changes in detail -->
This PR adds support for nullability i.e. much less unwraps will be
required when using the AWS SDK. For generic clients, this new behavior
can be enabled in codegen by setting `nullabilityCheckMode: "Client"` in
their codegen config:
```
"plugins": {
"rust-client-codegen": {
"codegen": {
"includeFluentClient": false,
"nullabilityCheckMode": "CLIENT_CAREFUL"
},
}
```
## Testing
<!--- Please describe in detail how you tested your changes -->
<!--- Include details of your testing environment, and the tests you ran
to -->
<!--- see how your change affects other areas of the code, etc. -->
Ran existing tests
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: John DiSanti <jdisanti@amazon.com>
Co-authored-by: Russell Cohen <rcoh@amazon.com>
## Motivation and Context
<!--- Why is this change required? What problem does it solve? -->
<!--- If it fixes an open issue, please link to the issue here -->
aws-sdk-rust#749
## Description
<!--- Describe your changes in detail -->
When wrapping an `SdkBody`, the resulting body is no longer signable.
Some services support this and others don't. This change adds a new
method for mapping an `SdkBody` called `map_readonly_body`. If called on
a body that is signable, it will produce a signable body. However, this
function should never when modifying the inner body is necessary, as
that will result in a signing error.
## Testing
<!--- Please describe in detail how you tested your changes -->
<!--- Include details of your testing environment, and the tests you ran
to -->
<!--- see how your change affects other areas of the code, etc. -->
Zelda wrote a test
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR adds a utility for making unmodeled requests with the
orchestrator, which is a prerequisite for getting aws-config completely
off of the old generic middleware client.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
Draft pull request pending merge of #2970
## Motivation and Context
- https://github.com/awslabs/aws-sdk-rust/issues/738
- https://github.com/awslabs/aws-sdk-rust/issues/858
<!--- Why is this change required? What problem does it solve? -->
<!--- If it fixes an open issue, please link to the issue here -->
## Description
<!--- Describe your changes in detail -->
This PR adds two additional classes of retries tested:
1. GO_AWAY: https://github.com/awslabs/aws-sdk-rust/issues/738
2. REFUSED_STREAM: https://github.com/awslabs/aws-sdk-rust/issues/858
I tested 1 by using the example helpfully provided. The fix for 2 is
untested and difficult to reproduce but since my fix for 1 worked, I'm
confident that we're detecting the correct error class here.
## Testing
I used the example provided by the customer and validated the H2 GO_AWAY
fix.
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Fixes https://github.com/awslabs/smithy-rs/issues/2926
## Description
This PR ports logic implemented in
https://github.com/awslabs/smithy-rs/pull/2603. Thankfully, even though
we did not port this at the time of the orchestrator launch, the
orchestrator has not logged sensitive bodies because we have never
logged response bodies in the orchestrator code.
The code changes in this PR
- now logs response bodies in `try_attempt`
- ports the logic from the previous PR in question to the orchestrator,
via an interceptor
Now, when credentials providers in `aws_config` need to say "I want to
redact a response body"
([example](2c27834f90/aws/rust-runtime/aws-config/src/http_credential_provider.rs (L48)))
when middleware is gone, they can pass an interceptor
`SensitiveOutputInterceptor` to `Config` of whatever clients they are
using.
## Testing
Depends on the existing tests.
Without the logic ported over the orchestrator and by logging response
bodies unconditionally in `try_attempt`, we got the following failures.
After we've ported the logic, they now pass.
```
default_provider::credentials::test::ecs_assume_role
default_provider::credentials::test::imds_assume_role
default_provider::credentials::test::sso_assume_role
default_provider::credentials::test::web_identity_token_env
default_provider::credentials::test::web_identity_token_profile
default_provider::credentials::test::web_identity_token_source_profile
profile::credentials::test::e2e_assume_role
profile::credentials::test::region_override
profile::credentials::test::retry_on_error
```
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: ysaito1001 <awsaito@amazon.com>
Co-authored-by: John DiSanti <jdisanti@amazon.com>
Debugging tests is easier when the order of arguments in assertions is
consistent across the board. With consistent order, you don't need to
look at the line of code to know which value is the correct value. Most
of the rest of the project uses expected/actual order, so this PR makes
the protocol tests also use that order.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR ports all the connectors from the `aws-smithy-client` crate into
`aws-smithy-runtime` implementing the new `HttpConnector` trait. The old
connectors are left in place for now, and follow up PRs will remove them
as well as revise the generated configs to take `HttpConnector` impls
rather than `DynConnector`.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: Zelda Hessler <zhessler@amazon.com>
## Motivation and Context
~~**The PR will be a draft until
https://github.com/awslabs/smithy-rs/pull/2675 is merged into `main`.
Once that's merged, this PR can remove the change to `Cargo.toml` in
`aws-smithy-client`.**~~
**Now that it's passed tests in CI, ysaito1001 will merge this PR once
it has passed the internal tests**
Fixes https://github.com/awslabs/smithy-rs/issues/2500.
## Description
This PR updates the base image for ci-build's Dockerfile as the latest
image for [amazonlinux](https://gallery.ecr.aws/amazonlinux/amazonlinux)
has been updated to 2023.
As part of it, the changes to Dockerfile include
- Fixing conflicting curl's libraries by specifying `--allowerasing`
Without it, we'd get an error as follows:
```
package curl-minimal-7.88.1-1.amzn2023.0.1.x86_64 conflicts with curl provided by curl-7.87.0-2.amzn2023.0.2.x86_64
```
- Adding perl explicitly as a new version of openssl now requires it
Without `perl`, we'd get an error as follows:
```
Step 24/48 : RUN cargo +${rust_nightly_version} -Z sparse-registry install cargo-deny --locked --version ${cargo_deny_version}
---> Running in 3c3431881cfa
...
error: failed to run custom build command for `openssl-sys v0.9.76`
...
--- stderr
Can't locate FindBin.pm in @INC (you may need to install the FindBin module) (@INC contains: /usr/local/lib64/perl5/5.32 /usr/local/share/perl5/5.32 /usr/lib64/perl5/vendor_perl /usr/share/perl5/vendor_perl /usr/lib64/perl5 /usr/share/perl5) at ./Configure line 15.
BEGIN failed--compilation aborted at ./Configure line 15.
thread 'main' panicked at '
```
Also, there is a change to `Cargo.toml` in `aws-smithy-client`:
- Forcing `aws-smithy-client` with `native-tls` feature to use a more
recent `openssl-sys` crate
Without it, `cargo minimal-versions check` would fail in
`aws-smithy-client` trying to build `openssl-sys` v0.9.39 where OpenSSL
in a container is more up-to-date, leading to a build failure (See [this
issue](https://github.com/sfackler/rust-openssl/issues/1724) for more
details)
Finally, updating CI for `Run Tests on Windows`
- Manually installing `openssl` as
[suggested](https://github.com/sfackler/rust-openssl/issues/1542#issuecomment-1399358351).
Without it, after introducing a more recent `openssl` 0.10.52
`dev-dependencies` in `aws-smithy-client`, we'd get this on Windows:
```
--- stderr
thread 'main' panicked at '
Could not find directory of OpenSSL installation, and this `-sys` crate cannot
proceed without this knowledge. If OpenSSL is installed and this crate had
trouble finding it, you can set the `OPENSSL_DIR` environment variable for the
compilation process.
Make sure you also have the development packages of openssl installed.
For example, `libssl-dev` on Ubuntu or `openssl-devel` on Fedora.
If you're in a situation where you think the directory *should* be found
automatically, please open a bug at https://github.com/sfackler/rust-openssl
and include information about your system as well as this message.
$HOST = x86_64-pc-windows-msvc
$TARGET = x86_64-pc-windows-msvc
openssl-sys = 0.9.90
```
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: Yuki Saito <awsaito@amazon.com>
Co-authored-by: Zelda Hessler <zhessler@amazon.com>
## Motivation and Context
Addresses 7 in #2413
## Description
This small PR removes uses of `once_cell` from public API. The
`http_versions` module in the `aws_smithy_http` crate was dead-code.
## Testing
Relies on existing tests in CI.
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: ysaito1001 <awsaito@amazon.com>
## Motivation and Context
Chrono deprecated an API
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
Debugging endpoint resolution can be difficult when the params being
used during resolution aren't logged. This PR adds that logging.
Example output below. It doesn't ordinarily have newlines and
indentation. I added those for clarity.
```
2023-08-22T00:27:54.363727Z DEBUG
invoke{service=s3 operation=ListObjectsV2}:try_op:try_attempt:
aws_smithy_runtime::client::orchestrator::endpoints: resolving endpoint
endpoint_params=EndpointResolverParams(TypeErasedBox[!Clone]:Params {
bucket: Some("test-bucket"),
region: Some("us-east-1"),
use_fips: false,
use_dual_stack: false,
endpoint: None,
force_path_style: false,
accelerate: false,
use_global_endpoint: false,
use_object_lambda_endpoint: None,
disable_access_points: None,
disable_multi_region_access_points: false,
use_arn_region: None
})
endpoint_prefix=None
```
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
Runtime plugins need to be able to wrap components configured in other
runtime components. For example, one runtime plugin should be able to
wrap the HTTP connector configured in another runtime plugin.
This PR makes this possible by:
- Introducing the ability to order runtime plugins within the
service/operation plugin "levels".
- Adding an argument to `RuntimePlugin::runtime_components` so that
implementations can reference components configured by previous plugins.
The `order` function has three separate order values: `Defaults`,
`Overrides`, and `NestedComponents`. The `Defaults` order is currently
unused, but will be used later when we refactor how defaults in config
work. Everything defaults to `Overrides` since most runtime plugins will
want to be in this slot. The `NestedComponents` order is specifically
for runtime plugins that want to create nested components, and runs at
the very end.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
**Will be merged to `smithy-rs-release-0.56.x` branch**
Addresses https://github.com/awslabs/aws-sdk-rust/issues/872
## Motivation and Context
With the introduction of the orchestrator, the core functions that live
in `aws-smithy-runtime/src/client/orchestrator.rs` were annotated by
`#instrument`. By default this emits logs at the `INFO` level, which
caused our customer to see the excessive amount of logging after
switching to the latest SDK creates `0.29`.
## Description
This PR reduces the log level to `DEBUG` wherever we use `#instrument`
throughout the file.
## Testing
No new tests have been added as part of this PR. [An
enhancement](https://github.com/awslabs/smithy-rs/issues/2932) has been
filed to check logging volume.
Here is a quick check that all instances in the file specify `trace`:
```
✗ rg "instrument"
src/client/orchestrator.rs
34:use tracing::{debug, debug_span, instrument, trace, Instrument};
160: .instrument(debug_span!("invoke", service = %service_name, operation = %operation_name))
167:#[instrument(skip_all, level = "debug")]
186:#[instrument(skip_all, level = "debug")]
319:#[instrument(skip_all, level = "debug")]
389: .instrument(debug_span!("read_body"))
398: .instrument(debug_span!("deserialization"))
407:#[instrument(skip_all, level = "debug")]
419:#[instrument(skip_all, level = "debug")]
```
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: ysaito1001 <awsaito@amazon.com>
## Motivation and Context
This PR implements a short-term solution for aws-sdk-rust#864 while a
long-term solution is worked out.
## Testing
- Tested manually against S3.
- Added DVR tests.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
- #2087
## Description
- remove lingering usages of SystemTime::now()
- ensure they don't return by adding a clippy.toml
## Testing
- CI
- Ran the webassembly example
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: John DiSanti <jdisanti@amazon.com>
This PR merges the benchmark and integration tests from `aws/sra-test`
into the SDK integration tests, and updates the tests so that they
compile and pass.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR adds missing documentation to the `aws-smith-runtime` crate, and
moves the `ServiceClockSkewInterceptor` into the `aws-runtime` crate.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR creates new-types for `Input`, `Output`, and `Error` so that the
type system can enforce that an output isn't given to an input and vice
versa. It also removes `TypedBox` since that isn't really needed
anymore.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR adds a `name` function to the `Interceptor` trait so that
interceptor errors are easier to narrow down.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
- #2087
test-util pulled in Hyper but that's not strictly necessary
## Description
Split out wiremock as it's own features and do some other cleanup.
## Testing
CI
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
When `TimeSource` was created, we didn't carefully track down all the
places and left a bit of existing implementation to limit the scope of
the change.
## Description
This removes the public (doc hidden) Testing code from
aws-credential-types and our other test time sources instead.
## Testing
- IT/UT
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: John DiSanti <jdisanti@amazon.com>
This PR finishes documenting `aws-smithy-runtime-api` and also:
- Adds `Send + Sync` bounds to `Interceptor`.
- Moves `Interceptors` into `aws-smithy-runtime`.
- Expands the more complicated macros in the interceptor context
wrappers.
- Makes the `OrchestratorError` an informative error according to
[RFC-22](https://github.com/awslabs/smithy-rs/blob/main/design/src/rfcs/rfc0022_error_context_and_compatibility.md).
- Renames `ConnectorError::is_other` to `as_other` and adds an
equivalent `is_other` that returns a boolean.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR makes some progress towards documenting the
`aws-smithy-runtime-api` crate, as well as some additional work to make
it more stable:
- Places the `client` module behind a `client` feature so that it will
be possible to add a `server` module/feature in the future.
- Deletes `ConfigBagAccessors`.
- Renames auth types to reflect changes in the internal spec.
- Moves `RequestAttempts` into the `client::retries` module.
- Moves several types that were in floating around in
`client::orchestrator` to their own modules.
- Renames `Connector` to `HttpConnector`.
- Changes `DynResponseDeserializer` to `SharedResponseDeserializer` for
consistency with serialization, and also so that it could be moved into
runtime components or config in the future (since those need to
implement `Clone`).
- Updates the identity and auth design doc.
- Updates READMEs and crate-level documentation.
- Fixes most, but not all, the missing documentation in the crate.
- Hides the builder macros.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
The current module location is misleading, since you never want to run
the layer as a plugin: plugins always run after routing, and the health
check should be enacted before routing. Examples have been corrected,
and a test has been added.
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
- Simplify no-op code in `as_nanos`
- Add more proptests
## Testing
Ran the tests
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR improves the docs on the aws-smithy-types `config_bag` and
`type_erasure` modules.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR changes the default runtime mode to orchestrator for generated
clients and the AWS SDK.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Addresses item 8 in #2413
## Description
This PR removes the third party types as follows:
- removes `SegmentedBuf` from public API in `aws-smithy-http` (used when
the feature `event-stream` was enabled)
## Testing
- [x] Passed tests in CI
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: ysaito1001 <awsaito@amazon.com>
This PR fixes Timestream in the orchestrator implementation.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
Addresses item 1, 3, and 9 in #2413
## Description
This PR removes the third party types as follows:
- removes `InvalidHeaderValue` from public API in `aws-http`
- removes `SendError` from public API in `aws-smithy-async`
- removes `xmlparser` from public API in `aws-smithy-xml`
Those types were exposed to public APIs primarily due to the
implementation of traits, e.g. `From` and `Iterator`. Those
implementations are used internally (such as XML deserialization and
converting an error type) so we should be able to hide them within
crates.
## Testing
- [x] Passed tests in CI
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [x] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: ysaito1001 <awsaito@amazon.com>
During the orchestrator implementation, a bug was introduced that always
enabled the `test-util` feature on `aws-smithy-runtime` in the generated
crates. This led to several pieces of non-test code relying on test-only
code, specifically around `SystemTime` implementing `TimeSource`.
This PR fixes that issue, and also fixes another issue where the
`RuntimeComponentsBuilder` was being initialized without a name for the
service config.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR fixes a panic that would result from cloning a cloneable
`TypeErasedBox` twice.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
This PR moves all the "runtime components", pieces that are core to the
operation of the orchestrator, into a separate `RuntimeComponents` type
for the orchestrator to reference directly.
The reason for this is so that these core components cannot be changed
by interceptors while the orchestrator is executing a request.
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
## Motivation and Context
There is a bug in cloneable layer that allows non-clone types to enter
## Description
Remove `DerefMut` and resolve the consequences
## Testing
- new doc test
- existing orchestrator CI
## Checklist
<!--- If a checkbox below is not applicable, then please DELETE it
rather than leaving it unchecked -->
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the
smithy-rs codegen or runtime crates
- [ ] I have updated `CHANGELOG.next.toml` if I made changes to the AWS
SDK, generated SDK code, or SDK runtime crates
----
_By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice._
---------
Co-authored-by: John DiSanti <jdisanti@amazon.com>
ysaito1001