* Apply formatting
* Apply no block comments rule
* Apply OmitExplicitReturns
* Apple OnlyOneTrailingClosureArgument
* Apply NoAssignmentInExpressions
* Fix up DontRepeatTypeInStaticProperties lint errors
* Apply `OrderedImports`
* Apply `ReplaceForEachWithForLoop`
* format file
* Enable the formatting pipeline
* Adopt `AmbiguousTrailingClosureOverload`
* Fix license header
* Fix format check
* Fix `EndOfLineComment`
* Fix CI
* Adapt CI script to check if changes when running formatting
* Separate lint and format into to steps
* Fix format
* Adopt `UseEarlyExits`
* Revert "Adopt `UseEarlyExits`"
This reverts commit d1ac5bbe12.
# Motivation
Adding a handler via the regular `pipeline.addHandler` APIs can lead to transferring the handler over isolation regions. To avoid running into `Sendable` warnings here we can use the `syncOperations` of the pipeline to avoid transferring the handlers.
# Modification
This PR is mostly changing code to avoid transferring handlers to and from the event loops in tests. This should be a no-op mostly.
Co-authored-by: Cory Benfield <lukasa@apple.com>
Motivation:
As we've largely completed our move to split out our core abstractions,
we now have an opportunity to clean up our dependencies and imports. We
should arrange for everything to only import NIO if it actually needs
it, and to correctly express dependencies on NIOCore and NIOEmbedded
where they exist.
We aren't yet splitting out tests that only test functionality in
NIOCore, that will follow in a separate patch.
Modifications:
- Fixed up imports
- Made sure our protocols only require NIOCore.
Result:
Better expression of dependencies.
Co-authored-by: George Barnett <gbarnett@apple.com>
Motivation:
In some cases it may be possible to write the (usually small) web socket
frame header to the buffer provided by the user. If we do this we can
avoid an extra pipeline traversal for the small write. This is only
going to be a performance win in cases where we can avoid a
copy-on-write operation on the buffer, but if we can then it's a useful
small win to achieve.
Modifications:
- Refactored the WebSocketFrameEncoder to potentially prepend the frame
header.
- Added a missing @inlinable attribute.
- Tests.
Result:
Potentially improved performance.
Motivation:
The WebSocketFrameEncoder naively allocated a new block to write the
frame header into every time it wrote. This is excessive: in many cases
it would be able to re-use the same buffer as last time.
Modifications:
- Attempt to re-use the buffer we used for the last header.
Result:
Fewer allocations in some applications.
Motivation:
Spotted a couple of issues with the documentation and fixed them.
Modifications:
- explicitly marked all `public class`es as `public final class` to get
consistency in the Jazzy documentation.
- removed `public` extension methods of the `internal struct
PriorityQueue` which Jazzy showed in the docs.
- improved the largely missing `EmbeddedChannel` and `EmbeddedEventLoop`
documentation.
- clear up lies in the B2MD docs
- add some other missing docs
Result:
better docs makes happier users
Motivation:
readInbound/Outbound will soon throw errors if the type isn't right.
Modifications:
prepare tests for throwing readIn/Outbound
Result:
@ianpartridge's PR should pretty much merge after this.
Motivation:
- `ChannelPipeline.add(name:handler:...)` had a strange order of arguments
- `remove(handler:)` and `remove(ctx:)` both remove `ChannelHandler`s
but they read like they remove different things
So let's just fix the argument order and name them `addHandler` and
`removeHandler` making clear what they do.
Modifications:
- rename all `ChannelPipeline.add(name:handler:...)`s to `ChannelPipeline.addHandler(_:name:...)`
- rename all `ChannelPipeline.remove(...)`s to `ChannelPipeline.removeHandler(...)`
Result:
more readable and consistent code
Motivation:
ByteBuffer methods like `set(string:)` never felt very Swift-like and
also didn't look the same as their counterparts like `getString(...)`.
Modifications:
- rename all `ByteBuffer.set/write(<type>:,...)` methods to
`ByteBuffer.set/write<Type>(...)`
- polyfill the old spellings in `_NIO1APIShims`
Result:
code more Swift-like
Remove EmbeddedChannel restriction to IOData
Motivation:
You should be able to read any type from the outbound buffer.
Modifications:
Change Embedded channel outboundBuffer and pendingOutboundBuffer to store NIOAny instead of IOData and refactor test code accordingly.
Result:
- Users can test what went through the pipeline without serializing everything and using "if case let" syntax everywhere.
- fixes#551
Motivation:
EmbeddedChannel just treated all `write`s as flushed. That's not a good
idea as all real channels obviously only actually write the bytes when
flushed. This changes that behaviour for `EmbeddedChannel`.
Modifications:
- make `write` only write to a buffer
- make `flush` then actually pretend to write the bytes
Result:
- EmbeddedChannel behaves more realisticly
Motivation:
Websockets is a major protocol in use on the web today, and is
particularly valuable in applications that use asynchronous I/O
as it allows servers to keep connections open for long periods of
time for fully-duplex communication.
Users of NIO should be able to build websocket clients and servers
without too much difficulty.
Modifications:
Provided a WebsocketFrameEncoder and Decoder that can serialize and
deserialize Websocket frames.
Result:
Easier use of websockets.
Franz Busch
Cory Benfield
Johannes Weiss
JovanMilenkovic
Cory Benfield