The Check for upgrades codelens was only looking for upgrades for
the current module, but was applying diagnostics to all go.mod
files in the workspace. This change makes sure to only apply the
diagnostics in the same selected go.mod.
Fixesgolang/go#54556
Change-Id: I1eacbc8af2e9dcfe1e1a67516f047bcb94099872
Reviewed-on: https://go-review.googlesource.com/c/tools/+/425195
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Suzy Mueller <suzmue@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
CL 417576 externalized shouldLoad tracking into a map, which was used to
trigger a reload and cleared once reload completes. Unfortunately, it
overlooked the fact that we may also reload the entire workspace (via
reinitialization). In this case, we should clear newly loaded IDs from
the shouldLoad map, so that they are not subsequently loaded again.
Fixesgolang/go#54473
Change-Id: I26f49552cae502644142dc4a4e946294db37f6f7
Reviewed-on: https://go-review.googlesource.com/c/tools/+/424074
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
Reloading has been significantly refactored recently. Unskip this test
to see if it flakes:
- If it does not flake, that is a useful signal.
- If it does flake, that is also a useful signal.
Notably, following CL 419500 we allow network when reloading the
workspace, and so do not need to apply quick-fixes in order to download
the new module from the proxy.
For golang/go#46375
For golang/go#53878
Change-Id: Idde7195730c32bdb434a26b28aac82649dd1b5ac
Reviewed-on: https://go-review.googlesource.com/c/tools/+/422910
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Peter Weinberger <pjw@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
The go command can't always resolve arbitrary commits as versions, and
some commits have an x/tools replace directive that must be honored, so
use 'git clone' with an arbitrary commit ref, instead of 'go install'
with a Go module version, to install gopls.
Also rename BenchmarkIWL to BenchmarkInitialWorkspaceLoad.
For golang/go#53538
Change-Id: Ic3a08e4c023e0292f6595cc5b2ab59954d073546
Reviewed-on: https://go-review.googlesource.com/c/tools/+/422908
Run-TryBot: Robert Findley <rfindley@google.com>
Reviewed-by: Alan Donovan <adonovan@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Metadata reloading has been significantly refactored recently. Unskip
this test to see if it still flakes.
For golang/go#51352
For golang/go#53878
Change-Id: I9f2ae1835bbace1b5095c2d45db082c4e709437b
Reviewed-on: https://go-review.googlesource.com/c/tools/+/423974
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Reviewed-by: Peter Weinberger <pjw@google.com>
This query parameter is not needed.
Change-Id: Id45d7be0b1cbe5d383bcc6768ef20df26de3e7b3
Reviewed-on: https://go-review.googlesource.com/c/tools/+/422901
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
Auto-Submit: Robert Findley <rfindley@google.com>
A debugging print statement was left in the analyzer, which breaks
gopls' communication over stdin/stdout.
Fix this, and add tests.
Fixesgolang/vscode-go#2406
Change-Id: I1b785fa09e66eae2f1b1e03806e5b59d2015e75e
Reviewed-on: https://go-review.googlesource.com/c/tools/+/422902
TryBot-Result: Gopher Robot <gobot@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
Reviewed-by: Tim King <taking@google.com>
To facilitate testing specific gopls versions, add a -gopls_version flag
that installs and uses a specified gopls version.
Also fix a bug in the DidChange benchmark that causes it to be
inaccurate when the benchmark function is invoked multiple times (by the
benchmark runner or via -count): because the server environment is
shared across invocations, we can't use the benchmark iteration to
derive the expected number of changes: we must use our count of actual
DidChange notifications from the server.
Change-Id: I1ad733cb3e695d382e81dfb7e31346162d9b7635
Reviewed-on: https://go-review.googlesource.com/c/tools/+/422368
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
Reviewed-by: Nooras Saba <saba@golang.org>
This test was asserting immediately on the non-presence of published
diagnostics, and therefore was racing with the diagnostics calculation
pass. Following https://go.dev/cl/420539 this became a flake, because
once gopls has calculated diagnostics for the open event, it will
actually publish empty diagnostics for the opened file.
Update the test to use a OnceMet so that we only evaluate the
expectation once we've finished the diagnostics pass. As a result, the
test deterministically failed, which was fixed by using the
EmptyOrNoDiagnostics expectation.
Fixesgolang/go#54271
Change-Id: Id3f220ce44c7996132699a724b6c627f034e367f
Reviewed-on: https://go-review.googlesource.com/c/tools/+/422136
gopls-CI: kokoro <noreply+kokoro@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
Reviewed-by: Peter Weinberger <pjw@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Follow Lua Style Guide for function names.
Query the client's offset encoding, with a fallback to utf-16.
Omit handling of r.command since the source.organizeImports code action
doesn't return any.
Omit passing a value for wait_ms. It is optional and defaults to 1000
(see NeoVim help for vim.lsp.buf_request_sync).
Change-Id: I5e5eb6cee3a46fee1edc1e6d15b40ad88498a26c
Reviewed-on: https://go-review.googlesource.com/c/tools/+/421295
Reviewed-by: Dylan Le <dungtuanle@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Jamal Carvalho <jamal@golang.org>
When the go.work file is set by the GOWORK environment variable, we must
create an additional file watching pattern.
Fixesgolang/go#53631
Change-Id: I2d78c5a9ee8a71551d5274db7eb4e6c623d8db74
Reviewed-on: https://go-review.googlesource.com/c/tools/+/421501
gopls-CI: kokoro <noreply+kokoro@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
In order to correctly process changes to the go.work file, the workspace
must know about GOWORK settings configured in the users environment.
Compute this when initializing the view, and thread this through to the
workspace.
At this point, workspace information is spread around in a few places.
Add some TODOs to clean this up.
Also remove some module data that was not used in
TestBrokenWorkspace_DuplicateModules.
Updates golang/go#53631
Change-Id: Ie0577d702c8a229304387bc7fe53a8befb544acb
Reviewed-on: https://go-review.googlesource.com/c/tools/+/421500
Reviewed-by: Suzy Mueller <suzmue@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
For users of Go >= 1.18, update the error message presented to users
with broken workspaces to suggest using a go.work. For older Go
versions, suggest updating to 1.18+.
Also:
- add support for changing workspace folders from the fake editor
- update the test to confirm that changing workspace folders resolves
the error message
- inline validBuildConfiguration, and make a few TODOs for how the code
could be further cleaned up
Fixesgolang/go#53882
Change-Id: Ia03dcfec59569b1a3ac941dc40d079b9c2593825
Reviewed-on: https://go-review.googlesource.com/c/tools/+/421499
TryBot-Result: Gopher Robot <gobot@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
Reviewed-by: Dylan Le <dungtuanle@google.com>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
This test is really about workspace setup, not diagnostics. Move it
in advance of changes to this feature.
Pure code move, no other changes.
Change-Id: Ib78f1fe5ce701673f5aa071f399da11f208874df
Reviewed-on: https://go-review.googlesource.com/c/tools/+/421498
Reviewed-by: Suzy Mueller <suzmue@golang.org>
gopls.run_vulncheck_exp runs `gopls vulncheck`
(fork of govulncheck) and pipes its stderr (logs)
as progress messages. The default log format
includes timestamp and that is too long for progress
message. Tell gopls vulncheck to omit timestamp
in the log message.
Use "govulncheck" as the progress message prefix,
instead of the long "Checking vulnerabilities".
Change-Id: I92fe9958b20d0260711a42af9b5f9f399e267587
Reviewed-on: https://go-review.googlesource.com/c/tools/+/420998
Run-TryBot: Hyang-Ah Hana Kim <hyangah@gmail.com>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
Reviewed-by: Jamal Carvalho <jamal@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
And, make gopls.run_vulncheck_exp show an information/error
message popup after a successful run. This is temporary.
We plan to publish the results as diagnostics and quick-fix.
Finally, changed the stdlib vulnerability info id in
testdata to GO-0000-0001 which looks more like a vulnerability
ID than STD.
Changed TestRunVulncheckExp to include tests on codelens
and use the command included in the codelens, instead of
directly calling the gopls.run_vulncheck_exp command.
Change-Id: Iaf91e4e61b2dfc1e050b887946a69efd3e3785b0
Reviewed-on: https://go-review.googlesource.com/c/tools/+/420995
Run-TryBot: Hyang-Ah Hana Kim <hyangah@gmail.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
In order to avoid shutdown races we must always close the second editor
created for simultaneous edit tests.
Rather than introduce additional indirection, just merge the two tests
into one.
For golang/go#54241
Change-Id: I6604141baa77d07f6281d3a90efa13c02b94079a
Reviewed-on: https://go-review.googlesource.com/c/tools/+/421258
gopls-CI: kokoro <noreply+kokoro@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
commandHandler.run expects a valid file for forURI and
forURI is necessary to fully populate commandDeps.
Our use of directory URI does not work.
We plan to use this custom command triggered through
codelenses on documents (e.g. go.mod), so we use
the document's URI.
Change-Id: I4de6d488dec5a4b71716499e7fc5e8328ecf3e49
Reviewed-on: https://go-review.googlesource.com/c/tools/+/420994
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
Run-TryBot: Hyang-Ah Hana Kim <hyangah@gmail.com>
To support the experimentalUseInvalidMetadata mode, we keep around
invalid metadata. This can help gopls features work when, for example,
the go.mod file is broken. It is debatable whether this feature is worth
supporting (see golang/go#54180), but in the meantime there is a very
negative side-effect when module paths are changed in the go.mod file:
we keep around a bunch of workspace packages with a stale module path.
As a result we can be left with a lots of extra type-checked packages
in memory, and many inaccurate diagnostics.
Fix this by skipping packages with invalid metadata when computing
workspace packages. While we may want to use invalid metadata when
finding the best package for a file, it does not make sense to re-
type-check and diagnose all those stale packages.
Fixesgolang/go#43186
Change-Id: Id73b47ea138ec80a9de63b03dae41d4e509b8d5a
Reviewed-on: https://go-review.googlesource.com/c/tools/+/420956
Run-TryBot: Robert Findley <rfindley@google.com>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Semantic tokens, as defined by the LSP, have no way of marking parts of
strings or numbers, for instance, to emphasize escape character. But
if gopls returns no semantic tokens for strings, then the editor
will use its coloring for strings, which may be more useful (and
similarly for components of numbers).
This change introduces boolean flags noSemanticString and
noSemanticNumber that can be set to true to suppress the semantic
token and let the editor's formatting shine through.
Fixes: Fixesgolang/go#45753
Change-Id: Ibae880a08fb9a67daa73aa172375a1c949431e11
Reviewed-on: https://go-review.googlesource.com/c/tools/+/421256
Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Run-TryBot: Peter Weinberger <pjw@google.com>
By making gopls.run_vulncheck_exp (RunVulncheckExp implements)
call `gopls vulncheck`, we achieve
- gopls.run_vulncheck_exp can run asynchronously and be cancellable
- log information can be forwarded as progress messages
- isolate any failures during vulncheck execution
In this CL, we also changed not to include test files in the analysis
(match the default of govulncheck). We will add an option in the future.
TODO:
- prevent concurrent gopls.run_vulncheck_exp
- convert the gopls vulncheck output to diagnostics and publish it
- remove timestamps from the `gopls vulncheck` log messages
for simplify progress messages
- add test to check vulnerability in third-party dependencies
Change-Id: I21592e03794cd9e9d96ed3989973a2ab7d75c538
Reviewed-on: https://go-review.googlesource.com/c/tools/+/420717
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
Run-TryBot: Hyang-Ah Hana Kim <hyangah@gmail.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
Based on dle8's CL 414454, wire in directoryFilters into the goimports
ModuleResolver scan. A few changes/fixes/additions from that CL:
- Fix a bug in filter validation that was causing -**/a not to validate.
- Fix a bug in regex building where -a was not treated as an excluded
prefix (it was matching 'a' anywhere).
- Use absolute paths in the SkipPathInScan, so that we can evaluate
directory filters relative to the workspace folder.
- Add several regression tests.
- Consolidate directoryFilters regression tests under a new
directoryfilters_test.go file.
- Add several TODOs.
Fixesgolang/go#46438
Change-Id: I55cd3d6410905216cc8cfbdf528f301d67c2bbac
Reviewed-on: https://go-review.googlesource.com/c/tools/+/420959
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Reviewed-by: Dylan Le <dungtuanle@google.com>
Significantly refactor the gopls benchmarks to turn them into proper
benchmarks, eliminate the need for passing flags, and allow running them
on external gopls processes so that they may be used to test older gopls
versions.
Doing this required decoupling the benchmarks themselves from the
regtest.Runner. Instead, they just create their own regtest.Env to use
for scripting operations. In order to facilitate this, I tried to
redefine Env as a convenience wrapper around other primitives.
By using a separate environment setup for benchmarks, I was able to
eliminate a lot of regtest.Options that existed only to prevent the
regtest runner from adding instrumentation that would affect
benchmarking. This also helped clean up Runner.Run somewhat, though it
is still too complicated.
Also eliminate the unused AnyDiagnosticAtCurrentVersion, and make a few
other TODOs about future cleanup.
For golang/go#53992
For golang/go#53538
Change-Id: Idbf923178d4256900c3c05bc8999c0c9839a3c07
Reviewed-on: https://go-review.googlesource.com/c/tools/+/419988
gopls-CI: kokoro <noreply+kokoro@google.com>
Reviewed-by: Peter Weinberger <pjw@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
I believe the races described in the issue have been fixed: we should
invalidate mod tidy results on any metadata change. If this invalidation
doesn't work due to a race, we want to know about it.
Update the test to wait for file-related events to complete before
removing files, in an attempt to avoid windows file-locking issues.
For golang/go#40269
For golang/go#53878
Change-Id: I91f0cb4969851010b34904a0b78ab9bd2808f92e
Reviewed-on: https://go-review.googlesource.com/c/tools/+/420718
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
Fixgolang/go#50267 by ensuring that diagnostics are re-sent following
didOpen or didClose events. Additionally, introduce a new hidden
'chattyDiagnostics' option that causes diagnostics to be resent on
*every* file change event. This latter option is for LSP clients that
get confused when diagnostics are not re-sent for later file versions.
For now, be conservative and only force diagnostic publication on
didOpen and didClose.
Update tests whose 'NoDiagnostics' assertions were broken by the new
behavior.
Fixesgolang/go#50267
Change-Id: I6332d66a1851e0d8261599d37020a03b4c598f7d
Reviewed-on: https://go-review.googlesource.com/c/tools/+/420539
Run-TryBot: Robert Findley <rfindley@google.com>
Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Requesting gc_details for an empty file never worked, but in the past it
would inadvertently get handled by our forgiving position helpers. With
https://go.dev/cl/409935, it became a crasher.
Fix the crash by adding handling when the package name position is
invalid.
Also move gc_details related codelens to their own file.
Fixesgolang/go#54199
Change-Id: I7339b3903abd1315f1fea3dd9929d81855a8264a
Reviewed-on: https://go-review.googlesource.com/c/tools/+/420715
gopls-CI: kokoro <noreply+kokoro@google.com>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Add a test that when a user adds a "//go:build ignore" to a file, gopls
correctly invalidates diagnostics.
This used to be broken, but was fixed by CL 417576.
Fixesgolang/go#54147
Change-Id: I554fcfc0a56b72f657e19b3c0ae53a66d1a99a76
Reviewed-on: https://go-review.googlesource.com/c/tools/+/420537
TryBot-Result: Gopher Robot <gobot@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
Run-TryBot: Robert Findley <rfindley@google.com>
This CL contains a partial implementation of package renaming. Currently gopls is only able to rename references to the renaming package within the renaming package itself.
prepareRename is still expected to return an error for renaming a package.
For golang/go#41567
Change-Id: I3683a0a7128bba7620ef30db528f45b753e6c08f
Reviewed-on: https://go-review.googlesource.com/c/tools/+/420219
Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Dylan Le <dungtuanle@google.com>
Document that modules matching GOPRIVATE will not be linked.
Updates golang/vscode-go#2362
Change-Id: I7e2447bb50a2cd0d7d394f8589ccd4498f889048
Reviewed-on: https://go-review.googlesource.com/c/tools/+/419979
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
Auto-Submit: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>
Per the explanation at golang/go#54069, allow network access whenever
loading the workspace. Enable one test that exercises this behavior.
More tests will be added in subsequent CLs.
Updates golang/go#47540
Updates golang/go#53676
Updates golang/go#54069
Change-Id: I9c3bb19d36702bc6b8051bee6b7cddaec5b97c0c
Reviewed-on: https://go-review.googlesource.com/c/tools/+/419500
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
When a go.work file fails to validate, the workspace is left in an
invalid state: we will detect that the workspace is defined by the
go.work, but will not actually parse any active modules. This should be
a critical error.
Fix this by adding allowing the workspace to surface critical errors via
a new cache.workspace.criticalError method.
Additionally:
- only build the workspace mod file in workspace.build if the mode is
fileSystemWorkspace (in all other modes the modfile is already
determined)
- rename workspace.invalidate to workspace.Clone, to be consistent with
other data structures
- rename CriticalError.DiagList to CriticalError.Diagnostics
- add several TODOs for observations while reading the code
- create a new file for regtests related to broken workspaces
- make the regtest sandbox panic when duplicate paths are present in
the sandbox file set (an error I made while writing the test)
Updates golang/go#53933
Change-Id: If8625ab190129bc9c57e784314bc9cc92644c955
Reviewed-on: https://go-review.googlesource.com/c/tools/+/417593
Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Robert Findley <rfindley@google.com>
added
Add logic to invalidate any packages with missing dependencies when a
file is added.
Also fix a latent bug overwriting 'anyFileOpenenedOrClosed' for each
loop iteration.
Fixesgolang/go#54073
Change-Id: I000ceb354885bd4863a1dfdda09e4d5f0e5481f3
Reviewed-on: https://go-review.googlesource.com/c/tools/+/419501
Run-TryBot: Robert Findley <rfindley@google.com>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Due to shared caching, the "default" tests can run faster than other
execution modes and still retain most of the test coverage. Update the
test runner to only run the singleton tests if testing.Short() is true,
independent of GOOS.
On the other hand, we lost noticeable coverage when disabling the
Forwarded testing mode. Now that the regtests are lighter weight in
general, reenable it on the longtests builders.
While at it, clean up tests that used the server-side Options hook to
instead use Settings, since clients would configure their server via
Settings and Options can't work with a shared daemon.
Updates golang/go#39384
Change-Id: I33e8b746188d795e88841727e6b7116cd4851dc2
Reviewed-on: https://go-review.googlesource.com/c/tools/+/418774
Reviewed-by: Bryan Mills <bcmills@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
Each regtest does a significant amount of extra work re-doing things
like parsing and type-checking the runtime package. We can share this
work across regtests by using a shared cache, significantly speeding
them up at the cost of potentially hiding bugs related to timing.
Sharing this work still retains most of the benefit of the regtests, so
implement this in the default mode (formerly called "singleton" and now
renamed to "default"). In a subsequent CL, modes will be cleaned up so
that "default" is the only mode that runs with -short.
Making this change actually revealed a caching bug: our cached package
stores error messages extracted from go/packages errors, but does not
include these errors in the cache key. Fix this by hashing all metadata
errors into the package cache key.
Updates golang/go#39384
Change-Id: I37ab9604149d34c9a79fc02b0e1bc23fcb17c454
Reviewed-on: https://go-review.googlesource.com/c/tools/+/417587
TryBot-Result: Gopher Robot <gobot@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
Reviewed-by: Bryan Mills <bcmills@google.com>
Update the markdown documentation for inlay hints and fix a couple
of typos in the examples.
Change-Id: I114502a81999bc5e4f25384ab619888f3e31a731
Reviewed-on: https://go-review.googlesource.com/c/tools/+/419496
Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
Run-TryBot: Suzy Mueller <suzmue@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Update all dependencies, except sergi/go-diff.
Also tidy x/tools with -compat=1.16, as it had recently been broken at
1.16.
Change-Id: I2e6c9bf48c6bedb2dff0fa418bf588dd07918866
Reviewed-on: https://go-review.googlesource.com/c/tools/+/419494
gopls-CI: kokoro <noreply+kokoro@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>
This test was originally skipped due to deadline exceeded errors. In the
time since, we've made performance improvements, fixed races, and
altered the regtests to remove arbitrary deadlines. Unskip it to see if
it still flakes.
For golang/go#44098
For golang/go#53878
Change-Id: I06530f2bc9c6883f415dc9147cfcbf260abb2a00
Reviewed-on: https://go-review.googlesource.com/c/tools/+/418898
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Bryan Mills <bcmills@google.com>
This test ran slowly sometimes. Now that we have improved performance,
elimininated arbitrary timeouts, and improved cacheability of
computed results when running with -short, I suspect this test should no
longer flake.
If it does, we can reduce its cost in other ways rather than turning it
off entirely.
Updates golang/go#48773
Updates golang/go#53878
Change-Id: I878e78117df5a1a25f4ac5f72e02f28fc078ec73
Reviewed-on: https://go-review.googlesource.com/c/tools/+/419106
Run-TryBot: Robert Findley <rfindley@google.com>
Reviewed-by: Bryan Mills <bcmills@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
The new implementation is based on Myers' paper, and is in the
package diff/lcs.
There is a new option newDiff, that can be set to 'old', 'new',
or 'both'. The default is 'both', although that may not be
the right choice for a release. See gopls/hooks/diff.go.
'both' runs both the old and new diff algorithm and saves some
statistics in a file in os.Tempdir(),
When (or if) the new code becomes the default, this logging (and
some internal checking) will be removed.
The new implementation has internal checking, which currently
panics. The code in gopls/hooks/diff.go tries to save an encrypted
(for privacy) version of the failing input.
The package diff/myers has not been replaced, but it could be.
Fixesgolang/go#52966
Change-Id: Id38d76ed383c4330d9373580561765b5a2412587
Reviewed-on: https://go-review.googlesource.com/c/tools/+/396855
Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
Run-TryBot: Peter Weinberger <pjw@google.com>
The fundamental bug causing TestChangePackageName to fail has been
fixed, yet unskipping it revealed a new bug: tracking whether or not a
package should be loaded requires that we actually store that package in
s.meta. In cases where we drop metadata, we also lose the information
that a package path needs to be reloaded.
Fix this by significantly reworking the tracking of pending loads, to
simplify the code and separate the reloading logic from the logic of
tracking metadata. As a nice side-effect, this eliminates the needless
work necessary to mark/unmark packages as needing loading, since this is
no longer tracked by the immutable metadata graph.
Additionally, eliminate the "shouldLoad" guard inside of snapshot.load.
We should never ask for loads that we do not want, and the shouldLoad
guard either masks bugs or leads to bugs. For example, we would
repeatedly call load from reloadOrphanedFiles for files that are part of
a package that needs loading, because we skip loading the file scope.
Lift the responsibility for determining if we should load to the callers
of load.
Along the way, make a few additional minor improvements:
- simplify the code where possible
- leave TODOs for likely bugs or things that should be simplified in
the future
- reduce the overly granular locking in getOrLoadIDsForURI, which could
lead to strange races
- remove a stale comment for a test that is no longer flaky.
Updates golang/go#53878
Change-Id: I6d9084806f1fdebc43002c7cc75dc1b94f8514b9
Reviewed-on: https://go-review.googlesource.com/c/tools/+/417576
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
Reviewed-by: Suzy Mueller <suzmue@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Suzy Mueller
Robert Findley
Leonardo Henrique
Dylan Le
Antoine Cotten
Hana (Hyang-Ah) Kim
pjw
Erik Dubbelboer