1
0
mirror of https://github.com/golang/go synced 2024-11-18 23:34:45 -07:00
go/internal/jsonrpc2/serve.go

140 lines
3.8 KiB
Go
Raw Normal View History

internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
// Copyright 2020 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package jsonrpc2
import (
"context"
"fmt"
"io"
internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
"net"
"os"
"time"
"golang.org/x/tools/internal/event"
errors "golang.org/x/xerrors"
internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
)
// NOTE: This file provides an experimental API for serving multiple remote
// jsonrpc2 clients over the network. For now, it is intentionally similar to
// net/http, but that may change in the future as we figure out the correct
// semantics.
// A StreamServer is used to serve incoming jsonrpc2 clients communicating over
// a newly created connection.
internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
type StreamServer interface {
ServeStream(context.Context, Conn) error
internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
}
// The ServerFunc type is an adapter that implements the StreamServer interface
// using an ordinary function.
type ServerFunc func(context.Context, Conn) error
internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
// ServeStream calls f(ctx, s).
func (f ServerFunc) ServeStream(ctx context.Context, c Conn) error {
return f(ctx, c)
internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
}
// HandlerServer returns a StreamServer that handles incoming streams using the
// provided handler.
func HandlerServer(h Handler) StreamServer {
return ServerFunc(func(ctx context.Context, conn Conn) error {
conn.Go(ctx, h)
<-conn.Done()
return conn.Err()
internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
})
}
// ListenAndServe starts an jsonrpc2 server on the given address. If
// idleTimeout is non-zero, ListenAndServe exits after there are no clients for
// this duration, otherwise it exits only on error.
func ListenAndServe(ctx context.Context, network, addr string, server StreamServer, idleTimeout time.Duration) error {
ln, err := net.Listen(network, addr)
internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
if err != nil {
return err
}
defer ln.Close()
if network == "unix" {
defer os.Remove(addr)
}
return Serve(ctx, ln, server, idleTimeout)
internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
}
// Serve accepts incoming connections from the network, and handles them using
// the provided server. If idleTimeout is non-zero, ListenAndServe exits after
// there are no clients for this duration, otherwise it exits only on error.
func Serve(ctx context.Context, ln net.Listener, server StreamServer, idleTimeout time.Duration) error {
ctx, cancel := context.WithCancel(ctx)
defer cancel()
// Max duration: ~290 years; surely that's long enough.
const forever = 1<<63 - 1
if idleTimeout <= 0 {
idleTimeout = forever
}
connTimer := time.NewTimer(idleTimeout)
newConns := make(chan net.Conn)
doneListening := make(chan error)
closedConns := make(chan error)
go func() {
for {
nc, err := ln.Accept()
if err != nil {
select {
case doneListening <- fmt.Errorf("Accept(): %w", err):
case <-ctx.Done():
}
return
}
newConns <- nc
}
}()
activeConns := 0
internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
for {
select {
case netConn := <-newConns:
activeConns++
connTimer.Stop()
stream := NewHeaderStream(netConn)
go func() {
conn := NewConn(stream)
closedConns <- server.ServeStream(ctx, conn)
stream.Close()
}()
case err := <-doneListening:
internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
return err
case err := <-closedConns:
if !isClosingError(err) {
event.Error(ctx, "closed a connection", err)
}
activeConns--
if activeConns == 0 {
connTimer.Reset(idleTimeout)
}
case <-connTimer.C:
return ErrIdleTimeout
case <-ctx.Done():
return ctx.Err()
}
internal/lsp: refactor LSP server instantiation Previously, the process of instantiating and running the LSP server was sharded across the lsp, protocol, and cmd packages, and this resulted in some APIs that are hard to work with. For example, it's hard to guess the difference between lsp.NewClientServer, lsp.NewServer, protocol.NewServer (which returns a client), and protocol.NewClient (which returns a server). This change reorganizes Server instantiation as follows: + The lsp.Server is now purely an implementation of the protocol.Server interface. It is no longer responsible for installing itself into the jsonrpc2 Stream, nor for running itself. + A new package 'lsprpc' is added, to implement the logic of binding an incoming connection to an LSP server session. This is put in a separate package for lack of a clear home: it didn't really philosophically belong in any of the lsp, cmd, or protocol packages. We can perhaps move it to cmd in the future, but I'd like to keep it as a separate package while I develop request forwarding. simplified import graph: jsonrpc2 ⭠ lsprpc ⭠ cmd ⭩ ⭦ lsp (t.b.d. client tests) ⭩ ⭨ protocol source + The jsonrpc2 package is extended to have a minimal API for running a 'StreamServer': something analogous to an HTTP server that listens for new connections and delegates to a handler (but we couldn't use the word 'Handler' for this delegate as it was already taken). After these changes, I hope that the concerns of "serving the LSP", "serving jsonrpc2", and "installing the LSP on jsonrpc2" are more logically organized, though one legitimate criticism is that the word 'Server' is still heavily overloaded. This change prepares a subsequent change which hijacks the jsonrpc2 connection when forwarding messages to a shared gopls instance. To test this change, the following improvements are made: + A servertest package is added to make it easier to run a test against an in-process jsonrpc2 server. For now, this uses TCP but it could easily be modified to use io.Pipe. + cmd tests are updated to use the servertest package. Unfortunately it wasn't yet possible to eliminate the concept of `remote=internal` in favor of just using multiple sessions, because view initialization involves calling both `go env` and `packages.Load`, which slow down session startup significantly. See also golang.org/issue/35968. Instead, the syntax for `-remote=internal` is modified to be `-remote=internal@127.0.0.1:12345`. + An additional test for request cancellation is added for the sessionserver package. This test uncovered a bug: when calling Canceller.Cancel, we were using id rather than &id, which resulted in incorrect json serialization (as only the pointer receiver implements the json.Marshaller interface). Updates golang/go#34111 Change-Id: I75c219df634348cdf53a9e57839b98588311a9ef Reviewed-on: https://go-review.googlesource.com/c/tools/+/215742 Run-TryBot: Robert Findley <rfindley@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rebecca Stambler <rstambler@golang.org> Reviewed-by: Heschi Kreinick <heschi@google.com>
2020-01-21 17:34:50 -07:00
}
}
// isClosingError reports if the error occurs normally during the process of
// closing a network connection. It uses imperfect heuristics that err on the
// side of false negatives, and should not be used for anything critical.
func isClosingError(err error) bool {
if errors.Is(err, io.EOF) {
return true
}
// Per https://github.com/golang/go/issues/4373, this error string should not
// change. This is not ideal, but since the worst that could happen here is
// some superfluous logging, it is acceptable.
if err.Error() == "use of closed network connection" {
return true
}
return false
}