Skip to content

feat(server)!: convert server.rs internals + on_disconnected to ServerError#1244

Open
Greg Lamberson (glamberson) wants to merge 3 commits intoDevolutions:masterfrom
lamco-admin:feat/server-typed-error-server
Open

feat(server)!: convert server.rs internals + on_disconnected to ServerError#1244
Greg Lamberson (glamberson) wants to merge 3 commits intoDevolutions:masterfrom
lamco-admin:feat/server-typed-error-server

Conversation

@glamberson
Copy link
Copy Markdown
Contributor

Summary

Third step of the staged migration started in #1242 and continued in #1243. Combines the on_disconnected signature change with the server.rs internal site conversion since both touch anyhow-flowing code; doing them together avoids a temporary anyhow ↔ ServerError two-way bridge during the intermediate state.

Public API change

 fn on_disconnected(
     &mut self,
     peer: SocketAddr,
     duration: Duration,
-    error: Option<&anyhow::Error>,
+    error: Option<&ServerError>,
 ) -> PostConnectionAction { ... }

This is a breaking change for handler implementations of ConnectionHandler. Pre-1.0, and per Marc-Andre Lureau (@elmarco)'s note on #1209: "I am ok with breaking API at this point :)".

Internal changes

  • The two run_inner / run_connection_inner wrapper functions introduced in feat(server)!: introduce typed ServerError on the public API boundary #1242 are folded back into run / run_connection. The accept loop calls the public method directly; result.as_ref().err() now feeds the new ServerError-typed parameter naturally.
  • ~25 .context() / bail! sites in run, run_connection, accept_finalize, handle_io_channel_data, handle_x224, handle_input_backlog, and the encode_share_data_pdu / deactivate_all helpers replaced with typed ServerError variants. Pattern alignment with ConnectorErrorExt:
    • EncodeError sources → ServerError::encode
    • DecodeError sources → ServerError::decode
    • std::io::Error sources → ServerError::io
    • Option<channel> with .ok_or_elseServerError::channel
    • bail!("Fastpath output not supported!")ServerError::unsupported
    • everything else → ServerError::custom with a static context.

What this PR does NOT touch

Stacking note

Branch is stacked on feat/server-typed-error-internal (#1243), which is in turn stacked on feat/server-typed-error (#1242). Rebase order on landing:

  1. feat(server)!: introduce typed ServerError on the public API boundary #1242 → master
  2. feat(server): convert encoder/helper/echo internals from anyhow to ServerError #1243 → master (rebase)
  3. feat(server)!: convert server.rs internals + on_disconnected to ServerError #1244 (this PR) → master (rebase)

Coordinate with #1239

The ConnectionHandler trait was just extended in #1239 (SuppressOutput / RefreshRectangle / FrameAcknowledge handlers). If this PR lands first, #1239 needs a trivial rebase to the new trait shape (the new methods stay; only on_disconnected's signature changes around them). If #1239 lands first, this PR rebases onto its trait shape. Either order works.

Test plan

  • cargo xtask check fmt -v clean
  • cargo xtask check lints -v clean (workspace, all-targets, with helper + __bench features)
  • cargo xtask check tests -v passes
  • cargo build --workspace --all-targets clean

Tracking: #1209.

Introduces a typed error story for ironrdp-server, mirroring the shape
already used by ironrdp-connector and the rest of the connection-management
layer:

  pub type ServerError = ironrdp_error::Error<ServerErrorKind>;
  pub type ServerResult<T> = Result<T, ServerError>;

ServerErrorKind is a non-exhaustive enum with concretely typed variants
for the failure modes the crate surfaces (Encode, Decode, Io, Channel,
Unsupported, Reason, General, Custom). Sources are attached through
ironrdp_error::Error::with_source rather than embedded as Box<dyn Error>
in variant data, matching ConnectorErrorKind exactly.

This first PR converts public boundaries only:

  - RdpServer::run                -> ServerResult<()>
  - RdpServer::run_connection<S>  -> ServerResult<()>
  - TlsIdentityCtx::init_from_paths -> ServerResult<Self>
  - TlsIdentityCtx::make_acceptor   -> ServerResult<TlsAcceptor>
  - EchoServerHandle::send_request  -> ServerResult<()>

Internal call sites continue to use anyhow::Result during the staged
migration. A private  helper bridges at the public boundary.
A follow-up PR converts internal sites to use the typed kinds directly
and drops the anyhow dependency. A second follow-up changes the
ConnectionHandler::on_disconnected error parameter from
Option<&anyhow::Error> to Option<&ServerError>.

This is a breaking change to consumers of the listed public functions:
return types change from anyhow::Result<T> to ServerResult<T>.

Tracking: Devolutions#1209
…rverError

Second step of the staged migration started in the previous commit.
Replaces anyhow construction sites in modules whose internal flow does
not pass through the ConnectionHandler::on_disconnected callback (which
still takes &anyhow::Error and is the subject of a separate follow-up
PR per Devolutions#1209):

  - encoder/mod.rs: ~15 anyhow! / .context() / bail! sites converted
    to typed ServerError variants (Encode, Reason, Custom).
    EncodeError sources go through ServerError::encode (matching
    ConnectorErrorExt::encode). spawn_blocking JoinError, qoi codec
    errors, and zstd error codes go through ServerError::custom or
    ServerError::reason as appropriate.
  - helper.rs: TLS cert/key loading paths construct ServerError::io
    for std::io::Error sources, ServerError::reason for missing-key
    cases, ServerError::custom for x509-cert and PEM parsing errors.
    Removes the from_anyhow bridge and the inner-fn split introduced
    in the previous commit.
  - echo.rs: build_echo_request returns ServerResult, builds errors
    via ServerError::custom directly. send_request keeps its already-
    typed Channel and Reason variants.

server.rs internals stay on anyhow because they propagate into
ConnectionHandler::on_disconnected which still takes &anyhow::Error;
that conversion plus the remaining server.rs internal sites land in
PR Devolutions#3. The anyhow dep stays in Cargo.toml for now and will be dropped
when the public traits (ConnectionHandler, RdpServerDisplay, and
RdpServerDisplayUpdates) finish their typed migration.
…rError

Third step of the staged migration started in Devolutions#1242 and continued in
the previous commit. Combines the on_disconnected signature change with
the server.rs internal site conversion since both touch anyhow-flowing
code; doing them together avoids a temporary anyhow-to-ServerError
two-way bridge during the intermediate state.

Public API change:

  ConnectionHandler::on_disconnected
      error: Option<&anyhow::Error>  ->  Option<&ServerError>

This is a breaking change for handler implementations.

Internal changes:

  - The two run_inner / run_connection_inner wrapper functions are
    folded back into run / run_connection. The accept loop calls the
    public method directly; result.as_ref().err() now feeds the new
    ServerError-typed parameter naturally.
  - ~25 .context() / bail! sites in run, run_connection, accept_finalize,
    handle_io_channel_data, handle_x224, handle_input_backlog, and the
    encode_share_data_pdu / deactivate_all helpers replaced with typed
    ServerError variants. Pattern alignment with ConnectorErrorExt:
    EncodeError sources -> ServerError::encode, DecodeError sources ->
    ServerError::decode, std::io::Error sources -> ServerError::io,
    Option<channel> with .ok_or_else -> ServerError::channel,
    bail!("Fastpath output not supported!") ->
    ServerError::unsupported, anything else -> ServerError::custom with
    a static context.
  - The from_anyhow bridge is retained only at one boundary: the
    RdpServerDisplay::updates() trait method still returns
    anyhow::Result, so the call site in run_connection wraps the anyhow
    result via from_anyhow. PR Devolutions#4 of this migration converts the
    display traits and drops the bridge entirely.

The anyhow dependency stays in Cargo.toml because RdpServerDisplay,
RdpServerDisplayUpdates, and a small handful of consumer-facing trait
returns still use anyhow::Result. PR Devolutions#4 finishes that conversion and
drops the dep.

Tracking: Devolutions#1209
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Development

Successfully merging this pull request may close these issues.

2 participants