NewStream implies the default stream type, which it is not.
NewHeaderStream is actually the default choice.
Change-Id: I1744d7e902d27c13393f3b367fe2d29e5d7dc283
Reviewed-on: https://go-review.googlesource.com/c/tools/+/231618
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Jonathan Amsterdam <jba@google.com>
It was not very useful, it basically renamed io.EOF under some limited
circumstances, and was only there for the befit of tests.
Instead, the test now checks io.EOF directly, but also checks for
io.ErrClosedPipe which was also happening.
We also make sure we wrap errors rather than replacing them.
This prevents some weird random test failures due to races in the way they were
closed.
Change-Id: I236b03ac5ba16bf763299b95d882cf58b1f74776
Reviewed-on: https://go-review.googlesource.com/c/tools/+/230303
Run-TryBot: Ian Cottrell <iancottrell@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
When jsonrpc2.Serve times out or is cancelled, we leak the goroutine
that is accepting connections, because it is stuck trying to write its
error back to the doneListening channel.
Fix this by adding a context cancellation for the serve func, and
selecting on this context when writing the error.
Change-Id: I3383535f58b44616983816e8b257a975e3c337a7
Reviewed-on: https://go-review.googlesource.com/c/tools/+/229978
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Cottrell <iancottrell@google.com>
Also moves core.Key to label.Key, but leaves the implementations
behind for now.
After using for a while, the word Tag conveys slightly the wrong
concept, tagging implies the entire set of information, label maps
better to a single named piece of information.
A label is just a named key/value pair, it is not really tied to the
event package, separating it makes it much easier to understand the
public symbols of the event and core packages, and allows us to also
move the key implementations somewhere else, which otherwise dominate
the API.
Change-Id: I46275d531cec91e28af6ab1e74a2713505d52533
Reviewed-on: https://go-review.googlesource.com/c/tools/+/229239
Run-TryBot: Ian Cottrell <iancottrell@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
event.Log removed
event.Print -> event.Log
event.Record -> event.Metric
event.StartSpan -> event.Start
In order to support this core now exposes the MakeEvent and Export functions.
Change-Id: Ic7550d88dbf400e32c419adbb61d1546c471841e
Reviewed-on: https://go-review.googlesource.com/c/tools/+/229238
Reviewed-by: Robert Findley <rfindley@google.com>
internal/telemetry/event was renamed to internal/event/core
Some things were partly moved from internal/telemetry/event straight to
internal/event to minimize churn in the following restructuring.
Change-Id: I8511241c68d2d05f64c52dbe04748086dd325158
Reviewed-on: https://go-review.googlesource.com/c/tools/+/229237
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
ID's are now by value not pointer, which caused it to not use the Format
method, resulting in broken id strings
The id maps need to be crossover (set and get go to different maps for a given direction of message)
Change-Id: Ide2b63ec1b009ae3587ee10e8bce018732ea342c
Reviewed-on: https://go-review.googlesource.com/c/tools/+/229244
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Peter Weinberger <pjw@google.com>
messages are the atomic unit of communication, changing streams
to read and write whole messages makes the code clearer.
It also avoids the confusion about what should be an atomic
operation or when a stream should flush.
Change-Id: I4b731c9518ad7c2be92fc92211c33f32d809f38b
Reviewed-on: https://go-review.googlesource.com/c/tools/+/228722
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
The previous implementation was exposing the details of the wire format
and resulted in non idomatic go, detecting the presence of absence of
values in fields to deterimine the message type.
Now the messages are distinct types and we use type switches instead.
Request still exists as an interface to expose the shared behaviour of
Call and Notification, as this is the type accepted by handlers.
The set of messages is deliberately closed by using a private methods on the
interfaces.
Change-Id: I2cf15ee3923ef4688670c62896f81f760c77fe04
Reviewed-on: https://go-review.googlesource.com/c/tools/+/228719
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
This runs the tests as sub tests, and makes sure they clean up
correctly.
It also installs the telemetry debug handlers.
Change-Id: I75b4f7a19be378603d97875fc31091be316949e0
Reviewed-on: https://go-review.googlesource.com/c/tools/+/228718
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
It does not need to expose wire level fields, especially not by embedding.
It also no longer needs to know what Conn it came from.
Change-Id: I1aede2baaa2daa40da4e7a1278b2eaada25b2310
Reviewed-on: https://go-review.googlesource.com/c/tools/+/227919
Reviewed-by: Jonathan Amsterdam <jba@google.com>
The same behavior can now be achieved by wrapping the Replier as we traverse the
handler stack instead.
Request is no longer mutated during handler invocation.
Change-Id: I2213de500f39e048f3f80161e234f3ae30464d70
Reviewed-on: https://go-review.googlesource.com/c/tools/+/227918
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Jonathan Amsterdam <jba@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
The single "l" form is used throughout the adjacent tools package
and the Go standard library.
Change-Id: I88c3530ef9d3b1354895d342e39403fa20ccd4d1
Reviewed-on: https://go-review.googlesource.com/c/tools/+/228237
Reviewed-by: Ian Cottrell <iancottrell@google.com>
reply is now passed to handlers separately from request, which allows it to be
substituted by handlers.
This also makes the handler signature much closer to http (which has
ResponseWriter)
Change-Id: I12be2e8e8b9bd508982ba43c9092709429284eaf
Reviewed-on: https://go-review.googlesource.com/c/tools/+/227839
Reviewed-by: Robert Findley <rfindley@google.com>
The wire structures do not need to be public, and making them so might make it
harder to keep the package correct without breaking changes if the protocol
changes in the future.
Change-Id: I03a5618c63c9f7691183d4285f88a177ccdd3b35
Reviewed-on: https://go-review.googlesource.com/c/tools/+/227838
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
We utilize error wrapping to recover the error codes when needed.
The code constants are also replaced by fully declared errors with
human readable messages.
Change-Id: I8edeb05f5028e99966e4ca28151f644f008d4e7d
Reviewed-on: https://go-review.googlesource.com/c/tools/+/227837
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
This required changing the jsonrpc.Conn.Call signature to also return the
request ID so it can be cancelled.
The protocol package now declares the Call function which wrapps up
Conn.Call and then sends a cancel message if the context was
cancelled during the call.
There is a small chance that a context can be cancelled on a
request that has already completed, but it is safe to do so.
Change-Id: Ic8040c193e1dd4ef376ad21194b1d0ea82145976
Reviewed-on: https://go-review.googlesource.com/c/tools/+/227558
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Replace the String method with a Format method so we can use it for extra
formats.
Add some tests to make sure it is all correct
Change-Id: I39f361ffba036fad99c93f8c0944164f7cf199ec
Reviewed-on: https://go-review.googlesource.com/c/tools/+/227486
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
It is no longer used after the changes to the logging system.
Change-Id: I7b96fb8297eb66f2ebad67c74c82fa7ed96c3139
Reviewed-on: https://go-review.googlesource.com/c/tools/+/227485
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Now it only uses the telemetry messages directly rather than the metric system.
This is much faster and more direct, and removes a blocker on improving the
metrics support.
Also fixes the fact that recieced and sent were the wrong way round before,
probably as an artifact of the old protocol logging code, and also removes
the bytes histogram which was a lie (it was a histogram of write sizes that
was presented as a histogram of message sizes)
fixesgolang/go#38168
Change-Id: Ib1c3459c0ff1cf0c6087a828981e80c1ce1c5c1b
Reviewed-on: https://go-review.googlesource.com/c/tools/+/227139
Run-TryBot: Ian Cottrell <iancottrell@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
This pushes the handler construction out to the user, allowing flexability of
use, and is the final stage of the switch to the new handler API.
Change-Id: Id2e61813a817df0d6e4d20dd47ce8c92b0ae87db
Reviewed-on: https://go-review.googlesource.com/c/tools/+/227024
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
We can do cancelling at the top level handler now, it can drop the cancel
messages themselves before they enter the queue stage, and also track
all the events as they flow through it.
The ugly part is the OnCancelled interface, which is a bit clunky.
Change-Id: I3fa972198625fb3517fdecc740d1a3fdb19a188a
Reviewed-on: https://go-review.googlesource.com/c/tools/+/226959
Run-TryBot: Ian Cottrell <iancottrell@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
This removes the state machine from the request.
It adds a done handler and uses that to manage the unlock channel instead
This also allows us to remove the nextRequest channel from request.
This is the last major piece that allows us to split up the run method into
composable handlers.
Change-Id: I5517ed5a51e30534754522a58453c27b5178ffa8
Reviewed-on: https://go-review.googlesource.com/c/tools/+/226839
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
It is now a programatic error to have a handler registered to a connection that
does not call reply for all messages, including notifications.
This normalizes the flow making the code easier to understand and fixes a
couple of long standing hard to find bugs.
Change-Id: If41c39ece70e3bc64420abefac75ec647a8f8b37
Reviewed-on: https://go-review.googlesource.com/c/tools/+/226838
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Instead of having a Parallel method, we only have Reply, which must also be used
for Notify messages (with a nil response).
This has no real functional impact but makes it easier to refactor in the next
cl.
Change-Id: Ifd4316dd71706de7913c69e6be539b966800e9dd
Reviewed-on: https://go-review.googlesource.com/c/tools/+/226837
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Handler is now a function type that mapps to what used to be the Deliver method.
The only handler that used other methods was Canceller, for now that still
exists as LegacyHooks. Once the handlers are fully cleaned up we should be able
to re-implement canceller as handler middleware.
Each connection is now only allowed one handler, and it is passed to the Run
method, but handlers are composable.
Change-Id: I370e0459df851bb9c9c2a679b99cff073b94489e
Reviewed-on: https://go-review.googlesource.com/c/tools/+/226479
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
This allows us to reduce the handler interface and delete the telemetry handler.
It is also safer and faster, and can be easily disabled along with the rest of
the telemetry system.
Change-Id: Ia4961d7f2e374f7dc22360d6a4020a065bfeae6f
Reviewed-on: https://go-review.googlesource.com/c/tools/+/225957
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
While experimenting with different static analysis on x/tools, I noticed
that there are many actionable diagnostics found by staticcheck. Fix the
ones that were not false positives.
Change-Id: I0b68cf1f636b57b557db879fad84fff9b7237a89
Reviewed-on: https://go-review.googlesource.com/c/tools/+/222248
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Gopls behavior on disconnection is currently somewhat undefined, because
it hasn't mattered when there was a single gopls session per binary
invocation. With golang/go#34111, this changes.
Checks are added to ensure clients and sessions are cleaned up when an LSP
connection closes. Also, normal client disconnection is differentiated
with the jsonrpc2.ErrDisconnected value.
Updates golang/go#34111
Change-Id: I74d48ad6dcfc30d11f7f751dcffb20c18a4cbaa3
Reviewed-on: https://go-review.googlesource.com/c/tools/+/220519
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
When running gopls against an automatically started remote instance, we
want the lifecycle of the remote to be detached from that of its
clients, so that it doesn't shut down while clients are still connected.
On the other hand, a gopls process can consume significant resources, so
we don't want it to remain when there are no more connected clients.
The jsonrpc2 package is updated to support the concept of idle timeout:
a duration after which the server is shut down when there are no
connected clients. This is exposed in the gopls serve command via the
-listen.timeout flag.
Update golang/go#34111
Change-Id: Id62b3d4a2fa66de2c9306d130ca431717f01d1e5
Reviewed-on: https://go-review.googlesource.com/c/tools/+/220281
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
Not all regtests resulted in LSP shutdown, which caused temp modfiles to
be leaked. After this fix I have confirmed that /tmp is clean after a
successful run of the regtests.
Also proactively clean up the unix socket file when serving jsonrpc2
over UDS.
Change-Id: I745fbd3d2adeeb165cadf7c54fd815d8df81d4e4
Reviewed-on: https://go-review.googlesource.com/c/tools/+/220061
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rohan Challa <rohan@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
For tests (and perhaps later, for daemon discovery), unix domain sockets
offer advantages over TCP: we can know the exact socket address that will be
used when starting a server subprocess. They also offer performance and
security advantages over TCP, and were specifically requested on
golang.org/issues/34111.
This CL adds support for listening on UDS, and uses this to implement an
additional regtest environment mode that starts up an external process.
This mode is disabled by default, but may be enabled by the
-enable_gopls_subprocess_tests.
The regtest TestMain may be hijacked to instead run as gopls, if a
special environment variable is set. This allows the the test runner to
start a separate process by using os.Argv[0]. The -gopls_test_binary
flag may be used to point tests at a separate gopls binary.
Updates golang/go#36879
Updates golang/go#34111
Change-Id: I1cfdf55040e81ffa69a6726878a96529e5522e82
Reviewed-on: https://go-review.googlesource.com/c/tools/+/218839
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
Update the servertest package to support connecting to a jsonrpc2 server
using either TCP or io.Pipes. The latter is provided so that regtests
can more accurately mimic the current gopls execution mode, where gopls
is run as a sidecar and communicated with via a pipe.
Updates golang/go#36879
Change-Id: I0e14ed0e628333ba2cc7b088009f1887fcaa82a5
Reviewed-on: https://go-review.googlesource.com/c/tools/+/218777
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
We had a deadlock in cases where a request was cancelled (1) after being
written to the stream, but (2) before a response was received. This
resulted in the request ID being removed from the pending map while the
server has the request, after which point the server response would hang
in Conn.Run trying to send to a nil channel.
After fixing this nil send there was still a race: it was possible that
Conn.Run could get the pending request, and Conn.Call would select
ctx.Done before Conn.Run could send to the response channel, again
resulting in a blocking send. Fix this by adding a buffer to the
response channel.
The response channel management is also made less forgiving, because we
should be able to reason precisely about how many sends and receives
will occur:
+ Don't close the response channel after sending a response: there
should only be one recipient.
+ Don't delete the ID from pending map twice: it should only be cleaned
up by Conn.Call.
Cancellation tests in the lsprpc package are updated to exercise the
race conditions.
Fixesgolang/go#37159
Change-Id: Ie3207442ea910f79247b18d8647fd52f39fb15db
Reviewed-on: https://go-review.googlesource.com/c/tools/+/219126
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
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>
Found by running the go vet pass 'testinggoroutine'.
Change-Id: I38f17877e2a97ffb823bb97850d21107743271d7
Reviewed-on: https://go-review.googlesource.com/c/tools/+/217179
Run-TryBot: Heschi Kreinick <heschi@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
CL 215738 didn't work because the canceller was embedded in the
serverHandler, which already had a Deliver method. Add it as an actual
handler instead.
Change-Id: I0c79f1bee67aa3b4da53d92547804de859f1938c
Reviewed-on: https://go-review.googlesource.com/c/tools/+/216303
Run-TryBot: Heschi Kreinick <heschi@google.com>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
This change allows us to hanel cancel messages as they go into the queue, and
cancel messages that are ahead of them in the queue but not being processed yet.
This should reduce the amount of redundant work that we do when we are handling
a cancel storm.
Change-Id: Id1a58991407d75b68d65bacf96350a4dd69d4d2b
Reviewed-on: https://go-review.googlesource.com/c/tools/+/200766
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
A detatched context ends up attributing all background work to the initialize
function.
Change-Id: I81206462752228d5ac81408fb1e3fb86ab36796e
Reviewed-on: https://go-review.googlesource.com/c/tools/+/186457
Run-TryBot: Ian Cottrell <iancottrell@google.com>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
We merge them into a single interface and support multiple of them rather than
just one.
This will allow us to stack handlers with different responsabilities and extract
some core logic (like tracing) out to a handler where it belongs.
Change-Id: I6aab92138550c5062fcb1bed86171e0850d1eb38
Reviewed-on: https://go-review.googlesource.com/c/tools/+/185879
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
also change the return type to be and end function and not an incomplete span
Change-Id: Icd99d93ac98a0f8088f33e905cf1ee3fe410c024
Reviewed-on: https://go-review.googlesource.com/c/tools/+/185349
Run-TryBot: Ian Cottrell <iancottrell@google.com>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
This is the basic library that allows for recording of stats about the program
operation.
Change-Id: I09f7e3de5fc37aaf29bc0db46f15b15056fc0eb2
Reviewed-on: https://go-review.googlesource.com/c/tools/+/185338
Run-TryBot: Ian Cottrell <iancottrell@google.com>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
