After some discussion about how to handle insert and filter text
(https://github.com/microsoft/vscode-languageserver-node/issues/488), it
seems that it is better practice to overwrite the prefix in completion
items, rather than trimming the prefix from the insert text.
Change-Id: I7c794b4b1d4518af31e7318a283aa3681a0cf66a
Reviewed-on: https://go-review.googlesource.com/c/tools/+/176958
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Cottrell <iancottrell@google.com>
Fix the following issues:
- We were trying to complete struct literal field names for
selector expressions (e.g. "Foo{a.B<>}"). Now we only complete field
names in this case if the expression is an *ast.Ident.
- We weren't including lexical completions in cases where you might be
completing a field name or a variable name (e.g. "Foo{A<>}").
I refactored composite literal logic to live mostly in one place. Now
enclosingCompositeLiteral computes all the bits of information related
to composite literals. The expected type, completion, and snippet code
make use of those precalculated facts instead of redoing the work.
Change-Id: I29fc808544382c3c77f0bba1843520e04f38e79b
GitHub-Last-Rev: 3489062be342ab0f00325d3b3ae9ce681df7cf2e
GitHub-Pull-Request: golang/tools#96
Reviewed-on: https://go-review.googlesource.com/c/tools/+/176601
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
(Forked off Yury Smolsky's CL 176618)
With this change, the playground.js client now asks the server to do a
vet check in the same HTTP request as the /compile (and run) step. If
the server replies that it understands the request (VetErrors or VetOK
in the repsonse), then the client can avoid the latency of a second
HTTP roundtrip. We'll remove the /vet handler after we see it fall out
of use from older clients.
Updates golang/go#31970
Change-Id: I5b123883e19cbc6a8ec30c50705e6b945a4d322d
Reviewed-on: https://go-review.googlesource.com/c/tools/+/176939
Reviewed-by: Andrew Bonventre <andybons@golang.org>
If you have x == 1 || x == 2 || x == 3 || x == 4, the pass considered the set
{x==1, x==2, x==3, x==4}
and then also
{x==2, x==3, x==4}
{x==3, x==4}
Since the comparison is itself linear in the size of the set, this was overall
taking time quadratic in the length of the || or && sequence.
Worse, if it found duplicates, they'd be reported a quadratic number of times.
This CL cuts the time and output to linear by avoiding already-checked
subexpressions. This cuts the time spent analyzing cmd/compile/internal/ssa
(with all passes enabled, not just this one) by 20%.
Fixesgolang/go#28086.
Change-Id: I812f64bd5a44fea995c9ab0c4fa2fbefb44037ce
Reviewed-on: https://go-review.googlesource.com/c/tools/+/176457
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Rob Pike <r@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
Partly fixesgolang/go#31333 (responds to shutdown request, but does
not reject later requests)
Fixes: golang/go#31375 by adding DO NOT EDIT comments to tsclient.go,
tsserver.go, and further details in the package doc in tsprotocol.go,
following the rule in golang/go#13560
Change-Id: I412e3dc6661373b8b70cdd15a385ad7bbd55a897
Reviewed-on: https://go-review.googlesource.com/c/tools/+/176921
Run-TryBot: Peter Weinberger <pjw@google.com>
Reviewed-by: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Also log gopls version information on startup in server mode.
Change-Id: If7bf85d19f993430709b1fae83083e6fdfe1b2ca
Reviewed-on: https://go-review.googlesource.com/c/tools/+/175199
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
This also required us to change the way we map files to a view, as it may change
over time.
Fixesgolang/go#31635
Change-Id: Ic82467a1185717081487389f4c25ad69df1af290
Reviewed-on: https://go-review.googlesource.com/c/tools/+/175477
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
specifically it uses them for the guru compatability tests
This change radically increases the test coverage of the godef tests as it now
works for all the jump to definition tests not just the specialized ones.
Change-Id: I63547138566ac3de56344dcfddb758ed5f362a06
Reviewed-on: https://go-review.googlesource.com/c/tools/+/174937
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
This required extensive changes to the "internal" server handling and also to
the way we wait for diagnostics in the "check" verb.
It improves both memory and time by over an order of magnitude, hopefully
allowing us to renable the tests on the builders
Change-Id: I84e84ca4c449e9970ebf1d922a0a2ce0a8a49c72
Reviewed-on: https://go-review.googlesource.com/c/tools/+/175878
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Improve expected type determination for the following cases:
- search back further through ast path to handle cases where the
position's node is more than two nodes from the ancestor node with
type information
- generate expected type for return statements
- wrap and unwrap pointerness from expected type when position is
preceded by "*" (dereference) or "&" (reference) operators,
respectively
- fix some false positive expected types when completing the "Fun"
(left) side of a CallExpr
Change-Id: I907ee3e405bd8420031a7b03329de5df1c3493b9
GitHub-Last-Rev: 20a0ac9bf2b5350494c6738f5960676cc50fb454
GitHub-Pull-Request: golang/tools#93
Reviewed-on: https://go-review.googlesource.com/c/tools/+/174477
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Now the "type" of a *ast.PkgName is the package it points to. Of
course, a package is not a real types.Type, but we can still jump you
there. We have to pick one of the package's files, so we choose the
longest one, hoping it is the most interesting.
Similarly, the "definition" of an *ast.ImportSpec is the package being
imported.
I also added a nil check for the package in SignatureHelp. This panics
for me occasionally.
Change-Id: Ide4640530a28bcec9da6de36723eb7f0e4cc941c
GitHub-Last-Rev: 8190baa0b908065db5b53f236de03d2f3bff39b5
GitHub-Pull-Request: golang/tools#92
Reviewed-on: https://go-review.googlesource.com/c/tools/+/174081
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Previously help for flags was printed to stderr.
This CL makes all the output to be printed to stdout.
Updates golang/go#31885
Change-Id: If95edeccd79581326502dd5c7fc2b49d8f160be7
Reviewed-on: https://go-review.googlesource.com/c/tools/+/175900
Reviewed-by: Ian Cottrell <iancottrell@google.com>
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Prior to this change, if a package was rendered invalid by a change in
one of its dependencies, diagnostics would not be propagated until the
user typed in one of the package's files. Now, these updated diagnostics
are sent along with the diagnostics for the dependency.
Fixesgolang/go#29817
Change-Id: I4761de31c4bdee820e024005f6112b3b3d2e1da6
Reviewed-on: https://go-review.googlesource.com/c/tools/+/174977
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Cottrell <iancottrell@google.com>
asmdecl:
- MOVW $x+0(FP) is OK if x is big, because $x is an address
(happens in internal/cpu, golang.org/x/sys/cpu, runtime)
- ignore TEXT lines in comments
(happens in runtime/internal/atomic)
- wasm's CallImport instruction writes return results
(happens in syscall)
- allow write out-of-bounds (SP) references in NOFRAME functions
(happens in runtime)
- recognize "NOP SP" as an SP "write" to disable SP bounds checking
- 'go test' in passes/asmdecl was not testing all architectures; fix that
stdmethods:
- ignore WriteTo if obviously not io.WriterTo (as in go/types and runtime/pprof)
errorsas:
- don't complain about package errors testing invalid calls
structtag:
- don't complain about encoding/json and encoding/xml testing invalid tags
unmarshal:
- don't complain about encoding/gob, encoding/json, encoding/xml testing invalid calls
For golang/go#31916.
Fixesgolang/go#25822.
Change-Id: I322c08b5991ffc4995112b8ea945161a4c5193ce
Reviewed-on: https://go-review.googlesource.com/c/tools/+/176097
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Michael Matloob <matloob@golang.org>
Reporting MemStats requires the legacy profile format (see golang/go#18641).
This CL detects when it couldn't get the MemStats from the profile and
reports this, rather than reporting 0 allocs.
Change-Id: Ib621ad975290cf05835fafa81e8e47762d82a519
Reviewed-on: https://go-review.googlesource.com/c/tools/+/175802
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
This links cmd/compile, which is the largest binary in GOROOT.
Unfortunately, this can't measure the linker's allocation metrics
prior to CL 176057 because the fix for golang/go#18641 was applied to
the compiler but not the linker.
Change-Id: Ia44d80f1d727c1608cedaa4b8ad2a05903774875
Reviewed-on: https://go-review.googlesource.com/c/tools/+/175801
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
We do a lot of sophisticated things around running the compile command
to gather benchmark metrics from it. We're about to add a benchmark
for the linker and will want all of the same mechanism. This CL
factors this out into a function.
Change-Id: I499d15d30417618f372982404ceb3e0f9c762e5c
Reviewed-on: https://go-review.googlesource.com/c/tools/+/175800
Run-TryBot: Austin Clements <austin@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
compilebench contains three different types of benchmarks and we're
about to add a fourth. Currently we dispatch based on name. This CL
cleans this up so the list of benchmarks points to how to run each
benchmark. It also joins together common error-reporting paths.
Change-Id: Icd4f77bdda9776863792dcd85f27b17829a833e6
Reviewed-on: https://go-review.googlesource.com/c/tools/+/175799
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Updates that only add text have a start and end that are the same, we were
erroring on those and failing to apply the changes.
Fixesgolang/go#31800
Change-Id: Ia31b90f108742e5532d2da35137c347c26090a6a
Reviewed-on: https://go-review.googlesource.com/c/tools/+/174949
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
I was about to add some more tests and it caused a huge number of golden files,
which was hard to deal with. Now all the golden files are packed into a single
.golden archive in the txtar format.
I also changed the tagging key for hover results to use the marker name rather
than the line and column, as it makes it more stable against test data changes.
Change-Id: Iaa1f54ab55a41d380db67b9f6f928fa7a52d9a5e
Reviewed-on: https://go-review.googlesource.com/c/tools/+/174877
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
RangeFormatting was broken with the introduction of the diff library,
but not noticed until golang/go#31150. Temporarily disable this behavior
until we fix it.
Change-Id: Ie3e50a763e0c0e8672ad20c3a2e37d901325d356
Reviewed-on: https://go-review.googlesource.com/c/tools/+/175938
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Cottrell <iancottrell@google.com>
golang.org/x/tools/go/vcs is significantly behind the upstream
cmd/go/internal/get code, and has no support for modules. It continues
to implement mechanics for GOPATH mode only.
This change is a minimal fix to get it to continue to work
in the presence of the module mode-only "mod" VCS type
(documented at https://golang.org/cmd/go/#hdr-Remote_import_paths)
by effectively implementing IgnoreMod ModuleMode behavior.
It is similar to issue golang/go#24751 and a small subset of CL 109340
that fixed it.
This helps with module adoption by reducing the harm of adding the
"mod" VCS type for vanity import paths, something that was meant to
be backwards compatible.
While here, also backport CL 14482 (the Token to RawToken change).
Fixesgolang/go#31845
Updates golang/go#24751
Change-Id: I0852f52cb9bda56879f923337c7f361df8412845
Reviewed-on: https://go-review.googlesource.com/c/tools/+/175219
Run-TryBot: Dmitri Shuralyov <dmitshur@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
It used to be that when the start offset was valid, it was presumed the end was
as well.
This was not true in the case where the start offset was not supplied but could
be inferred (at the very start of the file).
Fixesgolang/go#31797
Change-Id: Ie5a079796fa0f77cef5571a4e5b309c798e1e06b
Reviewed-on: https://go-review.googlesource.com/c/tools/+/174943
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
This prints the base of the path rather than the whole path when printing a
span.
This is useful for places where you are printing for the user rather than for
machines.
The format is %f
Change-Id: I6d1a52e4583099ff298c1fb645272578a49472eb
Reviewed-on: https://go-review.googlesource.com/c/tools/+/174942
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
This change uses an *ast.Package built from the file
go/src/builtin/builtin.go. Completion (and ultimately other features)
will be resolved using this AST instead of being hardcoded.
Change-Id: I3e34030b3236994faa484cf17cf75da511165133
Reviewed-on: https://go-review.googlesource.com/c/tools/+/174381
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Cottrell <iancottrell@google.com>
This uses golden files to hold the hover text as they are no more fragile than
hard coding the text in the tests, and much easier to update.
We need a lot more tests, and ones with actual comments, but this is a start and
at least adds the machienery it would take.
Change-Id: Ia2f79257642759e4c2f972d4037f258134e0fb33
Reviewed-on: https://go-review.googlesource.com/c/tools/+/174380
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
The generated code adds some server methods that are unsupported, but were
prviously unimplemented, and makes small adjusments to some parameter types.
Change-Id: I3dd32cafa3e457193f0771e0f8e150c88b381c82
Reviewed-on: https://go-review.googlesource.com/c/tools/+/173420
Run-TryBot: Peter Weinberger <pjw@google.com>
Reviewed-by: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
This is mostly to set things up to use golden files the same way for other commands.
Change-Id: I7fcc7165706763e655b0e46f0790b367fe5d3d59
Reviewed-on: https://go-review.googlesource.com/c/tools/+/174018
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
The current tests contain a bug in the priming of funnyString; the
subslicing leaves the resulting content slice with a capacity greater
than its length. This allowed a bug ToUTF16Column to sneak through where
we were not using 0-based column as the offset within the line.
Fix the priming of funnyString, and fix the implementation of
ToUTF16Column.
Change-Id: I2618878d85bba26f52f99a3fc136ad21fe198dfc
Reviewed-on: https://go-review.googlesource.com/c/tools/+/174357
Reviewed-by: Ian Cottrell <iancottrell@google.com>
Run-TryBot: Ian Cottrell <iancottrell@google.com>
Now when you accept a struct literal field name completion, you will
get a snippet that includes the colon, a tab stop, and a comma if
the literal is multi-line. If you have "gopls.usePlaceholders"
enabled, you will get a placeholder with the field's type as well.
I pushed snippet generation into the "source" package so ast and type
info is available. This allows for smarter, more context aware snippet
generation. For example, this let me fix an issue with the function
snippets where "foo<>()" was completing to "foo(<>)()". Now we don't
add the function call snippet if the position is already in a CallExpr.
I also added a new "Insert" field to CompletionItem to store the plain
object name. This way, we don't have to undo label decorations when
generating the insert text for the completion response. I also changed
"filterText" to use this "Insert" field since you don't want the
filter text to include the extra label decorations.
Fixesgolang/go#31556
Change-Id: I75266b2a4c0fe4036c44b315582f51738e464a39
GitHub-Last-Rev: 1ec28b2395c7bbe748940befe8c38579f5d75f61
GitHub-Pull-Request: golang/tools#89
Reviewed-on: https://go-review.googlesource.com/c/tools/+/173577
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
This change separates the completion formatting functions from the
completion logic. It also simplifies the completion logic by necessary
values per-request into a struct that is used throughout.
Change-Id: Ieb6b09b7076ecf89c8b76ec12c1f1c9b10618cfe
Reviewed-on: https://go-review.googlesource.com/c/tools/+/173779
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Cottrell <iancottrell@google.com>
Also remove error case that can no longer happen and the related test.
Fixesgolang/go#31341
Change-Id: I534956f6e835dea4334b68d0ad0297cf93920af2
Reviewed-on: https://go-review.googlesource.com/c/tools/+/173960
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Based on the work Paul Jolly did in https://go-review.googlesource.com/c/tools/+/173797
but not as internal tests and with a mildly obsessive attention to coverage.
Also has a failing test for golang/go#31341 that you can enable with -b31341
Change-Id: I528eee5304cd7191eafd3bcddb2f636c8722846f
Reviewed-on: https://go-review.googlesource.com/c/tools/+/173978
Run-TryBot: Ian Cottrell <iancottrell@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Amends prematurely submitted CL 173918. We now use file:/// URLs for
go1.13 and later and file:// URLs for go1.12 and earlier.
Fixesgolang/go#31675
Change-Id: I009c63a900bdfd091bf46def5cea5a0843639b47
Reviewed-on: https://go-review.googlesource.com/c/tools/+/173919
Run-TryBot: Jay Conrod <jayconrod@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Cottrell <iancottrell@google.com>
The "analyses" cache in lsp/cache.(*Package).GetActionGraph was not
getting cleared on errors. This could result in future calls to
GetActionGraph waiting on the "ready" channel indefinitely. This in
turn caused the goroutine in cacheAndDiagnose to block indefinitely
and never send the diagnostic results back.
Now we use a defer statement to always close the channel. If we did
not succeed, we also clear out the cache entry and set a "succeeded =
false" flag to signal waiters that they need to retry. If in the
future errors other than context.Canceled/Timeout are possible, this
retry behavior may need to be revisited.
Fixesgolang/go#30786
Change-Id: Icacc9188f1500b00f2178521ce373a2c1363f932
GitHub-Last-Rev: 7c43afd4286a69b0d35a625716e6934c72c4cef5
GitHub-Pull-Request: golang/tools#91
Reviewed-on: https://go-review.googlesource.com/c/tools/+/173977
Reviewed-by: Rebecca Stambler <rstambler@golang.org>
Run-TryBot: Rebecca Stambler <rstambler@golang.org>
