When resolving a position to a package we must consider all packages,
including intermediate test variants. This manifests, for example, when
jumping to definition in a package that is imported as a test variant
(see golang/go#47825).
For now, fix this by threading through an 'includeTestVariants' flag to
PackagesForFile. This isn't pretty, but should be a trivially safe
change: the only effect will be to increase the number of packages
considered in FindPackageFromPos. Since we are discussing future changes
to the API for querying packages from the snapshot, now did not seem
like a good time to undertake significant refactoring.
A regtest based on the original issue is included.
This CL is joint with rstambler@golang.org.
Fixesgolang/go#47825
Co-authored-by: Rebecca Stambler <rstambler@golang.org>
Change-Id: I4693ec69b50ed4acd569cff87883769c1edf332b
Reviewed-on: https://go-review.googlesource.com/c/tools/+/347563
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
In preparation for moving metadata related functionality to a separate
package, move around some types and export some symbols. This is purely
to reduce diffs in subsequent CLs, and contains no functional changes.
Change-Id: I24d4fbd71df78e4d7a84f6598cdf820b41d542a2
Reviewed-on: https://go-review.googlesource.com/c/tools/+/340729
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Coupling workspace symbols to package checking means that they do not
function when the workspace is contracted, and also forces us to do
duplicate work traversing file declarations.
This CL changes the workspace symbol implementation to precompute
symbols based only on syntactic information, allowing them to function
in degraded workspace mode, improving their performance, and laying the
groundwork for more significant performance improvement later on.
There is some loss of precision where we can't determine the kind of a
symbol from syntactic information alone, but this is minor: we fall back
on 'Class' if we can't determine whether a type definition is a basic
type, struct, or interface.
Benchmark ("test" in x/tools): 56ms->40ms
Benchmark ("test" in kuberneted): 874ms->799ms
Change-Id: Ic48df29b387bf029dd374d7d09720746bc27ae5e
Reviewed-on: https://go-review.googlesource.com/c/tools/+/338692
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
In my testing, the gopls degraded memory mode (currently set via
"memoryMode": "DegradeClosed") did not save as much memory as expected
due to still type checking all packages in the workspace (even if in
ParseExported mode). It is also annoying to get incomplete results from
references and renaming.
I think we can (and should) fix both problems: don't even consider
packages that aren't 'reachable' via open files, but fully type check
the reverse transitive closure of the packages you're working on. This
CL does exactly that, by swapping out the concept of 'workspace
packages' with 'active packages'.
In testing, this decreased my memory footprint while working on std by
3-4x when compared to normal mode, and 2x when compared to the previous
implementation of DegradeClosed.
It still needs more testing before we move this option out of
experimental.
For golang/go#46902
Change-Id: I1e319d0b1607d344d27e797ce32de057d7a583f9
Reviewed-on: https://go-review.googlesource.com/c/tools/+/336410
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
When we open a file in a package, independent of whether it is in the
workspace, we type check in ParseFull mode. However, several other
code paths don't find this better parse mode.
We need a better abstraction, but for now improve a couple code paths
specifically for the purpose of fixing Hover content.
Updates golang/go#46158
Updates golang/go#46902
Change-Id: I34c0432fdba406d569ea963ab4366336068767a2
Reviewed-on: https://go-review.googlesource.com/c/tools/+/333689
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Sometimes, we may want to report progress from functions inside of the
cache package, so move the progress tracker to the session to allow for
that.
Change-Id: I15409577a7a5080e7f0224a95d159de42856ffa7
Reviewed-on: https://go-review.googlesource.com/c/tools/+/319330
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
This change treats the GOMODCACHE like a directory filter, and it
excludes any modules under the module cache from being considered part
of the workspace. This can happen when users open their entire GOPATH.
To do this, I had to propagate the view's root and gomodcache through
the exclusion functions, and I also had to add BuildGoplsMod to the
snapshot's interface.
Change-Id: Id80b359d73d09a525b380389917451e85357b784
Reviewed-on: https://go-review.googlesource.com/c/tools/+/326816
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Each analyzer in Staticcheck is annotated with the appropriate
severity to use for its diagnostics. For example, most checks in SA*
produce warnings, but some produce errors (e.g. when passing an
invalid regular expression to regexp.Compile).
This will be especially important for a follow-up CL that enables
Staticcheck's new quickfix category, which contains optional
refactorings that shouldn't be flagged as warnings.
Change-Id: I6235303a3bb188ef79f52952c01e9585301a3270
Reviewed-on: https://go-review.googlesource.com/c/tools/+/322491
Trust: Dominik Honnef <dominik@honnef.co>
Run-TryBot: Dominik Honnef <dominik@honnef.co>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Despite the name, ParseExported only hollowed out declarations -- it
didn't actually drop any from the AST. This leaves a fair amount of
unexported crud behind. Unfortunately, there are a *lot* of ways to
expose an unexported declaration from an exported one, and it can be
done across files. Because of that, discarding unexported declarations
requires a lot of work.
This CL implements a decent attempt at pruning as much as possible from
the AST in ParseExported mode.
First, we analyze the AST of all the files in the package for exported
uses of unexported identifiers, iterating to a fixed point. Then, we
type check those ASTs. If there are missing identifiers (probably due to
a bug in the dependency analysis) we use those errors to re-parse. After
that we give up and fall back to the older, less effective trimming. The
pkg type changes slightly to accomodate the new control flow.
We have to analyze all the files at once because an unexported type
might be exposed in another file. Unfortunately, that means we can't
parse a single file at a time any more -- the result of parsing a file
depends on the result of parsing its siblings. To avoid cache
corruption, we have to do the parsing directly in type checking,
uncached.
This, in turn, required changes to the PosTo* functions. Previously,
they operated just on files, but a file name is no longer sufficient to
get a ParseExported AST. Change them to work on Packages instead. I
squeezed in a bit of refactoring while I was touching them.
Change-Id: I61249144ffa43ad645ed38d79e873e3998b0f38d
Reviewed-on: https://go-review.googlesource.com/c/tools/+/312471
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Provide some support for template files, implementing most of
https://docs.google.com/document/d/1clKAywucZVBXvL_v4mMhLQXso59lmQPMk1gtSpkV-Xw
Template support is controlled by the option 'experimentalTemplateSupport'
which defaults to false.
Most of the code is in a new 'template' package. Implemented are
semantic tokens, diagnostics, definitions, hover, and references,
and there is a stub for completions.
This code treats all the template files of a package together, so as
to follow cross-references.
Change-Id: I793606d8a0c9e96a0c015162d68f56b5d8599294
Reviewed-on: https://go-review.googlesource.com/c/tools/+/297871
Run-TryBot: Peter Weinberger <pjw@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Trust: Peter Weinberger <pjw@google.com>
ast.NewPackage mutates the input files, making it difficult to avoid
races with our caching model. I had avoided a race resulting from cache
handle cancellation, just to run into another race in multi-session
servers.
But there's no reason to use ast.NewPackage when we only have a single
file. We can just interrogate the file scope wherever needed.
Fixesgolang/go#45868
Change-Id: I521475b51ee3b1c3e408916affecafbc629b0191
Reviewed-on: https://go-review.googlesource.com/c/tools/+/315629
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
Calling PackagesForFile on builtin.go loads is at a
command-line-arguments package, which has many type checking errors.
Add a new snapshot method IsBuiltin, which is used to avoid calling
PackagesForFile on builtin.go when diagnosing changed files or checking
for orphaned files. There may be other places where this should be used,
but this functionality can't reasonably be pushed down, as
PackagesForFile should always return something.
This exacerbated an existing race to building the builtin, because
ast.NewPackage unfortunately mutates the ast.File. Fix this by just
building the builtin package directly when building the handle. It
should be very fast.
Fixesgolang/go#44866
Change-Id: Ie6c07478493fa011e92e6966289c2fa822d87b35
Reviewed-on: https://go-review.googlesource.com/c/tools/+/314290
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
Some of the refactoring changed the way that we label code action kinds,
and we need to add quickfix and fixall kinds for each diagnostic type.
Support a per-kind suggested fix, and fix a small issue in setting the
analyzer for a fixall code action.
Fixesgolang/go#45111
Change-Id: I6bb32c9aa7427b690f42910672d3139579e84478
Reviewed-on: https://go-review.googlesource.com/c/tools/+/303209
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
Locating the workspace module by convention has multiple problems:
+ gopls's view of $TMPDIR might be different from the client
+ there might be multiple views
+ there might be multiple gopls sessions per pid
Instead, assign a temp workspace directory for each workspace folder,
and provide a command to access this information.
Cleaning up all these temp directories was overcomplicated. Instead,
create a temp directory for the gopls server to nest them under, that
can be removed up on server shutdown.
Also fix a bug where the snapshot was not acquired before copying its
workspace.
Updates golang/go#42252
Change-Id: I0641cebe09cd376dfa27373cac30397711c64a8f
Reviewed-on: https://go-review.googlesource.com/c/tools/+/300409
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
As much as possible, try to unify the codeAction code paths. We always
run analysis now. And rather than assuming certain categories of
analyzers will generate certain kinds of code actions, mark them
explicitly and use that information to filter the actions afterward.
Change-Id: I8154cd67aa8b59b2a6c8aa9c3ea811de2e190db4
Reviewed-on: https://go-review.googlesource.com/c/tools/+/300170
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Now that we're generating quick fixes at analysis time, we can use those
in code action requests and delete a fair amount of redundancy. The
codeAction function is a little cluttered, but I want to get it all in
one place before I decide how to split it up.
Change-Id: Icd91e2547542cce0a05c18c02a088833f71232a4
Reviewed-on: https://go-review.googlesource.com/c/tools/+/297532
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Type error analyzers can be viewed as enhancing type errors, rather
than analyzers in their own right. Create a source.DiagnosePackage
function that combines the list/parse/typecheck diagnostics with type
error analyzers. This allows us to remove some special cases from the
analysis path, and is a first step in removing all the special
handling for analysis quick fixes.
Along the way:
Pass pointers to source.Analyzer after I spent half an hour chasing a
loop capture bug. Spend a further 2-3 hours chasing slowdown in the
command tests as a result.
Move Unnecessary tag generation into diagnostic creation rather than
as a mutating post-processing step that required cloning diagnostics.
Change-Id: Id246667a9dcf484dc79516f92d5524261c435794
Reviewed-on: https://go-review.googlesource.com/c/tools/+/297879
Trust: Heschi Kreinick <heschi@google.com>
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Most callers of source.Package.GetDiagnostics do it via
GetTypeCheckDiagnostics. Push its logic up or down as appropriate and
delete it.
Rather than requiring fully populated maps of diagnostics, which was
rather subtle, call storeDiagnostics for every Go file in the package.
Change-Id: If43b0cc922af1013e80f969362246538df14985b
Reviewed-on: https://go-review.googlesource.com/c/tools/+/297878
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Hopefully improve some of the details around parsing that have always
confused me.
- parser.ParseFile will never return an error other than
scanner.ErrorList. Encode that in ParsedGoFile.
- parser.ParseFile will never return a nil file. Eliminate the code
path that handled that.
- Explain why we might fail to find a token.File.
- Trying to fix errors appears quite expensive even if there aren't any
to fix. Don't waste the time.
Change-Id: I87e082ed52c98a438bc7fd3b29e1a486f32fb347
Reviewed-on: https://go-review.googlesource.com/c/tools/+/297069
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
In typeCheckDiagnostics, we have logic to suppress go list and type
checking errors based on the presence of other errors. Nobody seems to
know why the logic is exactly what it is, and suppressing list errors
means that bad //go:embed patterns don't show up.
Intuitively, it makes sense to me that we don't want to type check if
listing or parsing fails: list errors mean that whole files may be
missing, and parsing errors may wipe out arbitrary chunks of files.
There's little point reporting errors from type checking that. However,
list errors and parse errors should be mostly orthogonal: go list
parses very little of the file and in practice only reports errors
parsing the package statement. So, at least for now, we report both
parse and list errors, and stop there if there are any.
Finally, move the suppression logic to the actual typeCheck function
that generates the diagnostics. typeCheckDiagnostics is the primary
consumer, but I think it's better to do the suppression at the source.
Because we are now showing list errors, and they are prone to getting
stuck due to bad overlay support, a couple of tests now require 1.16.
Change-Id: I7801e44c4d3da30bda61e0c1710a2f52de6215f5
Reviewed-on: https://go-review.googlesource.com/c/tools/+/295414
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
We currently write directly to go.mod/sum via the go command, expecting
that editors will pick up the changes. While that's true for VS Code,
vim doesn't necessarily reload unchanged buffers. Change to send
explicit edits instead, but only if the file is open. Behavior when
using Go versions that don't support -modfile is unchanged.
Fixesgolang/go#44035.
Change-Id: Ie4e5cf60673b860f5dfcbdb726bee0efe6aae405
Reviewed-on: https://go-review.googlesource.com/c/tools/+/290189
Trust: Heschi Kreinick <heschi@google.com>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Fully switch to the new generated command API, and remove the old
dynamic command configuration.
This involved several steps:
+ Switch the command dispatch in internal/lsp/command.go to go through
the command package. This means that all commands must now use the new
signature.
+ Update commandHandler to use the new command signatures.
+ Fix some errors discovered in the command interface now that we're
actually using it.
+ Regenerate bindings.
+ Update all code lens and suggested fixes to new the new command
constructors.
+ Generate values in the command package to hold command names and the
full set of commands, so that they may be referenced by name.
+ Update any references to command names to use the command package.
+ Delete command metadata from the source package. Rename command.go to
fix.go.
+ Update lsp tests to execute commands directly rather than use an
internal API. This involved a bit of hackery to collect the edits.
+ Update document generation to use command metadata. Documenting the
arguments is left to a later CL.
+ Various small fixes related to the above.
This change is intended to be invisible to users. We have changed the
command signatures, but have not (previously) committed to backwards
compatibility for commands. Notably, the gopls.test and gopls.gc_details
signatures are preserved, as these are the two cases where we are aware
of LSP clients calling them directly, not from a code lens or
diagnostic.
For golang/go#40438
Change-Id: Ie1b92c95d6ce7e2fc25fc029d1f85b942f40e851
Reviewed-on: https://go-review.googlesource.com/c/tools/+/290111
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
Collapse Diagnostic.Kind, Source, and Category into just Source. Remove
code that converted from Diagnostic to Diagnostic. Notes on the changes
I had to make along the way:
- We used to use Kind to determine Severity. Set Severity when the
Diagnostic is created instead.
- Use constants for Source as much as possible -- we still need to use
Analyzer.Name for analysis diagnostics. It would be nice to break that
dependency so that Source was totally opaque, but that's a separate
issue.
- Introduce a new Source for gc_details, "optimizer details". It was "go
compiler" previously.
- Some of the assignments are a little arbitrary. Is inconsistent
vendoring really a "go list" error?
- GetTypeCheckDiagnostics had code to cope with diagnostics that had no
URI associated with them. We now spread such diagnostics to all files
when they are generated.
- Analyze modifies Diagnostics by adding a Tag to them. That means it
has to own them, so I had it clone them. I would like to push that logic
down to the diagnostics, per the TODO, but that's another CL.
And some observations:
- It's obviously tempting to combine DiagnosticSource and
diagnosticSource, but they mean very different things. I'm open to a
better name for one or the other.
Change-Id: If2e861d6fe16bfd2e5ba216cf7e29cf338d0fd25
Reviewed-on: https://go-review.googlesource.com/c/tools/+/288215
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
source.Error and source.Diagnostic are almost identical types, used
arbitrarily in different parts of the code. This CL is the first step in
cleaning up that redundancy: it deletes the source.Error type.
To do that, I added the fields from source.Error to source.Diagnostic,
and made absolutely no other semantic code changes -- I just renamed
things that were named Error to Diagnostic. With only aesthetic concerns
in play, I hope this CL will be easy to review. The next CL will clean
up all the stupid-looking code that converts a Diagnostic to a
Diagnostic, etc.
Change-Id: I1298cc8144c686b8a37fc2cc106930105e511353
Reviewed-on: https://go-review.googlesource.com/c/tools/+/288214
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
The Typescript source is still at version 3.16, but there are new
requests, more detailed client capabilities, and an attempt to be
more specific about ranges of number in the Typescript code.
Vscode defines types integer and uinteger (32-bit signed and unsigned),
so the Go code now uses int32 and uint32.
They've changed the use of TextDocument, so version information is sometimes
missing. cache/session.go:625 was changed correspondingly.
This CL also make CodeAction.Disabled into a pointer.
New requests or notifications:
DidCreateFiles, DidRenameFiles, DidDeleteFiles (notifications)
ShowDocument, WillCreateFiles,WillRenameFiles, WillDeleteFiles (request)
It's a lot of code; I've probably missed something.
Change-Id: I8449ad8473ac00947d0344c5f6133f9bd73b9e10
Reviewed-on: https://go-review.googlesource.com/c/tools/+/286192
Run-TryBot: Peter Weinberger <pjw@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Trust: Peter Weinberger <pjw@google.com>
Using structured errors in gopls has proven to be difficult to manage:
it's hard to know whether a given error return is expected to be
structured without any type information. We have mostly eliminated
them; finish the job.
I don't intend any semantic changes here.
I considered eliminating CriticalError altogether, but it does seem
useful to have a convenient bundle for return values. So I left it
alone for now.
Change-Id: I4b5f85a8de9712babffbc1383088151b596bd815
Reviewed-on: https://go-review.googlesource.com/c/tools/+/287792
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Now that workspace module mode generates a combined go.sum there are
relatively few blockers to enabling -mod=readonly. Fix them and do it.
This CL is a bit of a grab bag, but the fixes are relatively separate. I
can split it into multiple CLs if desired.
- If module A depends on module B at v1.0.0, the go command will want to
upgrade the workspace module from v0.0.0-goplsworkspace to v1.0.0. To
prevent that, use vN.999999.0 as the base pseudoversion, adjusting v0 to
v1 where appropriate. A few test cases needed updating as a result.
- For old Go versions, sort the generated workspace module and
synthesize a go statement from the maximum go version declared in the
workspace.
- Some regtests need go.sum files created.
- matchErrorToModule created incorrect quick fixes: it would try to
download the top-level module mentioned in the error message, not the
one that actually caused the problem. Now it issues quick fixes for the
lowest-level module.
- TestMultiModuleModDiagnostics accidentally included the same module
in the workspace twice. Fix it, and make that an error.
Fixesgolang/go#43346.
Change-Id: I605f762a4d23bedd914241525e64c1b3ecc42150
Reviewed-on: https://go-review.googlesource.com/c/tools/+/287032
Trust: Heschi Kreinick <heschi@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
In CL 271297, I disabled the constantly-running upgrade check, which
removed the upgrade commands for individual dependencies. This seems to
have been a relatively popular feature. Re-introduce it, but requiring
explicit user interaction.
We now run an upgrade check when the user clicks "Check for upgrades".
Those results are stored on the View and used to show diagnostics on
any requires they apply to. Right now we only check the go.mod the user
has open; in multi-module workspaces it might make sense to check all of
them, but I'm not sure.
Fixesgolang/go#42969.
Change-Id: I65205dc99a4fa9daafdb83145b0294b6f3be5336
Reviewed-on: https://go-review.googlesource.com/c/tools/+/286474
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
The fragmentation of the critical error reporting is getting in the way
of small fixes. This change adds a GetCriticalError function that reports
critical errors, while ModTidy and WorkspacePackages no longer report
critical errors. Any function that wants to process critical errors
should call AwaitLoaded.
Some other smaller changes are made to account for these changes. One
change is that we may report multiple *source.Errors, with the
assumption that duplicate diagnostics will be caught by the diagnostic
caching. Also, any `go list` error message that ends with "to add it" is
now considered an error that gets the "add dependency" suggested fix.
Fixesgolang/go#43338Fixesgolang/go#43307
Change-Id: I056b32e0a0495d9a1dcc64f9f5103649a6102645
Reviewed-on: https://go-review.googlesource.com/c/tools/+/280093
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Rebecca Stambler <rstambler@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
This change adds a diagnostic and error message that appears when a user
edits a nested module in legacy mode. At first, I thought a diagnostic
would be enough, but it's actually quite difficult to spot it when you
have a bunch of "undeclared name" diagnostics caused by the nested
module, so I figured a progress bar error message would also be useful.
This error message just indicates to the user that they should open the
nested module as its own workspace folder.
Also, while debugging this, I noticed that command-line-arguments
packages can have test variants, which we were never handling.
So I addressed that in this change.
Fixesgolang/go#42109
Change-Id: Ifa6f6af401a3725835c09b76e35f889ec5cb8901
Reviewed-on: https://go-review.googlesource.com/c/tools/+/275554
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
For every file change, we want to diagnose the resulting snapshots.
However, we also want to diagnose the file in its "best possible" view
(for example, for a nested module, we should use the inner module rather
than the outer module).
Add a bestView function that operates on a set of views, which allows
us to a pick a view out of the *views affected* by a file modification,
not out of *all* of the views in the session. This means that we can
pick the best view, if any, for each file change and then only return
the snapshots that we should diagnose, as well as the changed URIs for
those snapshots.
Fixesgolang/go#42890
Change-Id: Iad23d5ed4832dfd857f1424a7244cf3bd8900e3b
Reviewed-on: https://go-review.googlesource.com/c/tools/+/274235
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
The "you are neither in a module nor your GOPATH" warning doesn't fit
our current notification style, so adjust it to appear as a progress
report and clean up the associated tests.
Change-Id: I32b96f17f3b9715403e465e3e0f9705f5d859147
Reviewed-on: https://go-review.googlesource.com/c/tools/+/275537
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
When GO111MODULE=auto and the workspace is outside of $GOPATH/src, we
should still show diagnostics about how to correctly configure
multi-module workspaces. We should not show these warnings for
workspaces outside of a module within GOPATH.
This adds an extra piece to the WorkspacePackages error logic--we may
still need to show the errors even if WorkspacePackages returned
results. We should eventually consolidate all of this logic to be more
cohesive, but for now I think it's more important to cover all of the
different cases and add tests.
Updates golang/go#42109
Change-Id: I673a03c9840cdaaf7f058de1cda3bf36b96fa7d3
Reviewed-on: https://go-review.googlesource.com/c/tools/+/275553
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
In Go 1.16 error codes as well as start and end positions are added for
go/types errors. This information is temporarily stored in unexported
fields, until we're more confident in the correctness of both the API
and the underlying data.
Read this information using reflection and, if available, use it to set
the corresponding field in compiler diagnostics. This establishes a
positive feedback loop: in most cases this should improve the gopls
diagnostic, and wherever it doesn't we can make a note and fall back on
the old heuristics for that error code.
For golang/go#42290
Change-Id: I37475189cbd14a0a5bcfde163f599c9a7b957372
Reviewed-on: https://go-review.googlesource.com/c/tools/+/268539
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Trust: Robert Findley <rfindley@google.com>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
There was a race to the background context that allowed work for earlier
snapshots to run on the background context of later snapshots. For
example, the following sequence is possible and has been observed in
tests (admittedly at unrealistic time scales):
- snapshot N is created
- view.cancelBackground() is called
- snapshot N+1 is created
- snapshot N+1 is diagnosed
- snapshot N is diagnosed
- snapshot N is destroyed
Fix this by pinning the background context to the snapshot. This also
makes more sense; I'm guessing the background context being on view is a
vestige of when more work was done on the view.
For golang/go#42837
Change-Id: If9cbad7361de823da9bfd2f55cb2f8b3aed95075
Reviewed-on: https://go-review.googlesource.com/c/tools/+/275239
Run-TryBot: Robert Findley <rfindley@google.com>
Trust: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
For users with poor network connections, go command invocations that
access the network may hang up for long periods of time. Even for
users with good network connections, accessing proxy.golang.org or
github can take a few seconds, which is a long time to lock up an
editor.
Disable network access via GOPROXY=off when running most go commands.
There are two notable exceptions. First, the initial workspace load is
allowed, though subsequent loads are not. My reasoning is that if the
user is opening a project they've just downloaded, they probably expect
to fetch its dependencies. (Also, it's convenient to keep the regtests
going.) The second is the go mod tidy diagnostics, which I hope to
address in a later change. All the other commands that access the
network are at the user's request.
When the workspace load fails due to a dependency that hasn't been
downloaded, we present a quick fix on the go.mod line to do so. The only
way that happens is if there's a manual modification to the go.mod file,
since all of the other quick fixes will do the download. So I don't
think there's a need to present the fix anywhere else.
Updates golang/go#38462.
Change-Id: I470bc1ba790db7c1afee54d0b28fa0c6fd203fb9
Reviewed-on: https://go-review.googlesource.com/c/tools/+/274120
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Go 1.16 will set -mod=readonly to be the default behavior of the go
command. This CL changes gopls' behavior to be more in line with that.
The model we end up with is more explicit about managing the go.mod.
Previously, adding an import for a package in the module cache but not
in scope would succeed, but add a warning with a quick fix. Now, the
import will fail, again with a quick fix, and the project will not
compile until the require is added. It also means that the go.sum needs
to be in sync. Previously, the go command would update the temporary
go.sum if the user's was out of date, and then we would discard those
changes. Now, loads will fail until the go.sum is brought into sync,
either by `go mod tidy` or the new "update go.sum" command.
The go.sum requirements affect many regtests. I added a DumpGoSum
function to the environment, which prints the txtar format go.sum file
that needs to be added to the workspace. We'll need to keep them in
sync, unfortunately. We could update them on the fly, but that would
slow down the regression tests somewhat and potentially mess up the test
setup. So I'm not sure what to do exactly. Perhaps maintain them
out-of-line in a separate file that could be auto-updated? Dunno.
I also had to add go directives to go.mod files to keep old Go versions
happy.
Multi-module workspace mode currently doesn't create a go.sum file at
all and is held back to the old behavior until it does.
Change-Id: Ib31afc17614afac2f5fbdf31c7fc03a90bd13e3a
Reviewed-on: https://go-review.googlesource.com/c/tools/+/268597
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Currently, when a user opens a workspace with no top-level module but
multiple modules in subdirectories, gopls treats that as an invalid
build configuration and reports an error message that may be difficult
for the user to understand (a go list error message about creating a
main module in the top-level directory). Instead, show a more useful
error message about the gopls workspace layout in both the progress bar
and as a diagnostic on every open file.
This fix only works for GO111MODULE=on (for now) because it's a lot
easier to interpret user intent, and the go command will return no
packages. The next step will be to improve error messaging for
GO111MODULE=auto and for users with nested modules.
Updates golang/go#42109
Change-Id: I702ca6745f7e080ff6704ade7843972ab469ccf3
Reviewed-on: https://go-review.googlesource.com/c/tools/+/272346
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Expanding the set of modifications in the cache package leads to a
mismatch when the didModifyFiles code in text_synchronization.go
processes the list of modifications.
Change-Id: Id0b7fea445103b34838fb9b03aee95a01e454306
Reviewed-on: https://go-review.googlesource.com/c/tools/+/274192
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
VS Code's file watching API doesn't send notifications about directory
deletions unless you register for them explicitly (see
https://github.com/microsoft/vscode/issues/109754). Rather than watch
every file in the workspace, keep track of every relevant directory and
register file watchers for it.
This CL moves the snapshot's WorkspaceDirectories function to the
session and changes it to a KnownDirectories function. It returns all of
the directories and subdirectories known the session at a given moment.
Top-level directories are marked as such so that their *.{go,mod,sum}
contents can be watched, while subdirectories are just watched by path
so we can be notified of deletions.
Fixesgolang/go#42348
Change-Id: Ic6d02dba55b5de89370522ea5d3cf1d198927997
Reviewed-on: https://go-review.googlesource.com/c/tools/+/271438
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
Trust: Rebecca Stambler <rstambler@golang.org>
Our handling of diagnostics has gotten complicated and has recently been
a source of bugs. Heschi and I took a step back and refactored the
diagnostic pass, starting with a new data structure for tracking
diagnostics.
The LSP server now holds a map of URI to collection of diagnostic
reports, which is used to track diagnostics as they are computed by
multiple sources. Additionally, this collection holds a hash of the
last published diagnostic set.
This new information allows us to implement an algorithm for
incrementally updating diagnostics on the server: diagnostics reports
are stored as they are computed for a snapshot, and then published in
(possibly multiple) passes, with the last pass for a snapshot being
marked as 'final'. In non-final passes, 'publishReports' publishes any
diagnostics that have already been computed for the snapshot, but does
nothing with URIs for which no diagnostics have been computed. In final
passes all diagnostics are reported, with empty diagnostics being
published for any URIs with no diagnostic reports. Any URIs for which no
diagnostic reports were computed are pruned from the diagnostic set. In
all cases, tracking the hash of published diagnostics prevents us from
duplicate publication.
This enables some simplifications of the existing diagnostic logic.
Computing new diagnostics and tracking send diagnostics is now handled
by the same algorithm, avoiding some bookkeeping. It is also no longer
necessary to explicitly clear diagnostics for deletions. This fixes some
previously broken edge cases, for example when packages go out of scope
due to go.mod or gopls.mod changes.
Fixesgolang/go#42198
Change-Id: Id0d8d0f7a60f6ffa8c33f0ae41763461f61dab7b
Reviewed-on: https://go-review.googlesource.com/c/tools/+/269677
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Trust: Robert Findley <rfindley@google.com>
We should be consistent about whether we pass around Error or *Error.
*Error was somewhat more common, and its Error method has a pointer
receiver, so I picked *Error.
ErrorList, on the other hand, is usually dereferenced when used, and
slice headers are small enough to pass around casually. So switch that
over to a non-pointer and change its Error method to a value receiver.
Change-Id: I19e70201ed1d6ef36d217db21c9101bb810bcf0c
Reviewed-on: https://go-review.googlesource.com/c/tools/+/272092
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Throwing away all of the metadata in the package means it will be
expensive to reload, so it makes sense to keep it around until a go.mod
file has been saved, indicating that the user is more likely to be done
typing. Otherwise, we will be triggering full package metadata reloads
on every keystroke.
Fixesgolang/go#42529
Change-Id: Ibc4f4bc8937f8f7176da79e9fefe166468578993
Reviewed-on: https://go-review.googlesource.com/c/tools/+/271299
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
Reviewed-by: Heschi Kreinick <heschi@google.com>
This allows us to show parsing errors as part of module diagnostics.
Change-Id: I558b95c145135482fdfceef8a5c68c62a6d32721
Reviewed-on: https://go-review.googlesource.com/c/tools/+/271630
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
Refactor to use goCommandInvocation to get the evironment for
goimports. That takes care of most of the work; all that's left is to
track the go.mod hash so we know when to clear the cache.
Fixesgolang/go#41836.
Change-Id: I614dacad04adfb879ead55e3b8640e57816534cb
Reviewed-on: https://go-review.googlesource.com/c/tools/+/270877
Trust: Heschi Kreinick <heschi@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
VS Code's file watching API doesn't send notifications about directory
deletions unless you register for them explicitly (see
https://github.com/microsoft/vscode/issues/109754). Rather than watch
every file in the workspace, keep track of every relevant directory and
register file watchers for it.
This CL moves the snapshot's WorkspaceDirectories function to the
session and changes it to a KnownDirectories function. It returns all of
the directories and subdirectories known the session at a given moment.
Top-level directories are marked as such so that their *.{go,mod,sum}
contents can be watched, while subdirectories are just watched by path
so we can be notified of deletions.
Fixesgolang/go#42348
Change-Id: If80a26cf6af5a2c2f7d87bb57b9ef5cb3dcf04ac
Reviewed-on: https://go-review.googlesource.com/c/tools/+/267120
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
Editors need a way to run commands in the same workspace that gopls
sees. Longer term, we need a good solution for this that supports
multiple workspace folders, but for now just write the first folder's
workspace to a deterministic location:
$TMPDIR/gopls-<client PID>.workspace.
Using the client-provided PID allows this mechanism to work even for
multi-session daemons.
Along the way, simplify the snapshot reinitialization logic a bit.
Fixesgolang/go#42126
Change-Id: I5b9f454fcf1a1a8fa49a4b0a122e55e762d398b4
Reviewed-on: https://go-review.googlesource.com/c/tools/+/264618
Run-TryBot: Robert Findley <rfindley@google.com>
gopls-CI: kokoro <noreply+kokoro@google.com>
Reviewed-by: Heschi Kreinick <heschi@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Robert Findley <rfindley@google.com>
This CL uses the approach of a source.ErrorList for the `go mod vendor`
error--rather than a ShowMessageRequest. The suggested fix is a command
that runs `go mod vendor`, so I added a CommandFix to the source.Error
type.
I'm not sure if a diagnostic is better than a ShowMessageRequest--
perhaps it should be both?
Fixesgolang/go#41819Fixesgolang/go#42152
Change-Id: I4ba2d7af0233f13583395e871166caed83454d6e
Reviewed-on: https://go-review.googlesource.com/c/tools/+/261737
Trust: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
gopls-CI: kokoro <noreply+kokoro@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Findley <rfindley@google.com>
Robert Findley
Rebecca Stambler
Dominik Honnef
Heschi Kreinick
Peter Weinbergr