We are inching towards the point at which I blow away this entire
package, or at least all tests that interpret the standard "testing"
package.
Change-Id: I06d99aac6d7baab14ee6c6a61afe0af34b814767
Reviewed-on: https://go-review.googlesource.com/80356
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
This change implements an AST walker, Apply, with the capability
to perform various actions (including rewrites) during the AST
walk, in pre- and post-iteration order.
Ideally we would like to have this functionality in the std lib
(go/ast), but as this change contains significant new API, it
is better to first gain some experience with it in x/tools.
Credits: This is joint work of Josh Bleecher Snyder, Roger
Peppe, and myself (author of this CL). The code is based on
proposal golang/go#17108 which I had started, together with an
initial implementation. It was then reworked significantly by
Josh in golang.org/cl/55790, further refined by suggestions
from Roger (allocation reduction), and finally cleaned up a bit
by me (AST simplifications).
Fixesgolang/go#17108.
Change-Id: I56d56b7f2bc5be2acfeb927f76aea7f264bb7b94
Reviewed-on: https://go-review.googlesource.com/77811
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
Manually apply same change as CL 68110 did for cmd/go/internal/get,
but for golang.org/x/tools/go/vcs, to help keep them in sync.
Updates golang/go#22125.
Helps golang/go#11490.
Change-Id: I255f7a494d9572389fc8dc8ce96891b6fcc214a0
Reviewed-on: https://go-review.googlesource.com/68352
Reviewed-by: Russ Cox <rsc@golang.org>
Before this change, astutil would only do a prefix match of a new import
with all the existing ones, to try to place it in the correct group. If
none was found, the new import would be placed at the beginning of the
first import group.
This works well for new std imports, but it doesn't work well for new
third-party packages that don't share any prefix with any of the
existing imports.
Example:
import (
"time"
"github.com/golang/snappy"
)
When adding "golang.org/x/sys/unix" with astutil.AddImport, the import
is inserted as follows:
import (
"golang.org/x/sys/unix"
"time"
"github.com/golang/snappy"
)
And goimports reorganizes the imports to separate std and third-party
packages:
import (
"time"
"golang.org/x/sys/unix"
"github.com/golang/snappy"
)
We usually don't want to introduce a new import group; in most cases,
the desired behavior is separating std from third-party packages.
With this CL, new imports that don't share prefix with any existing ones
will be placed with the first group of third-party imports, if any
exist. If no third-party import group exists, a new one will be added.
In the case of our example above, this will be the new outcome:
import (
"time"
"github.com/golang/snappy"
"golang.org/x/sys/unix"
)
Fixesgolang/go#19190.
Change-Id: Id4630015c029bd815234a6c8726cb97f4af16f1c
Reviewed-on: https://go-review.googlesource.com/37552
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
(*build.Context).Import has signature of
(ctx, importPath, fromDir, mode)
not
(ctx, fromDir, importPath, mode)
and the loader actually uses the first version.
The thinko was introduced in d6e83e53 (go/loader: changes for vendor
support; CL 18053) and was possible because importPath and fromDir both
have the same type string.
Fix it.
Change-Id: I52deaec6d141846b8a495835b121c1bdc2864215
Reviewed-on: https://go-review.googlesource.com/47343
Reviewed-by: Alan Donovan <adonovan@google.com>
The loader was calling (*types.Checker).Files on the "unsafe" package,
a global variable. Even with zero files, this operation is not a no-op
because it sets the package's "complete" flag, leading to a data race.
(Because Unsafe.complete is already set at construction, the
race is benign, but is reported by -race nonetheless.)
Fixesgolang/go#20718
Change-Id: I5a4f95be5ab4c60ea3b6c2a7fb6f1b67acbf42bc
Reviewed-on: https://go-review.googlesource.com/46071
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
https://go-review.googlesource.com/37157 introduced a bug that caused
some live φ-nodes to be removed from the CFG. The cause was that
reachability traversal considered edges only among "new" φ-nodes
(those introduced during SSA renaming) but not existing φ-nodes from
&& and || expressions. The fix is to mark existing phis, and thus
other phis reachable from them, as live. We also clear the Phi.block
field when eliminating a φ-node.
Also, during reachability, we treat DebugRef instructions as roots
like any other non-Phi instruction. This eliminates a related known
bug whereby the operand of a DebugRef may be a dead φ.
This change also adds a sanity check that all operands of an SSA value
that are themselves instructions must belong to a block. The sanity
check would fail 7 times on the standard library without the fix.
Fixesgolang/go#19622
Change-Id: If3a897a6a593a17bc3f0f8228d1edf483be7a3d0
Reviewed-on: https://go-review.googlesource.com/45832
Run-TryBot: Dominik Honnef <dominik@honnef.co>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Dominik Honnef <dominik@honnef.co>
In the dom tree traversal, the final child of each node now
inherits the parent's renaming map, reducing garbage.
This reduces allocations by 1.4% and bytes allocated by 2.0% when
building SSA for the entire standard library.
Change-Id: Id19b6d6766b3e0bf32d1db1238eff8a42d11b242
Reviewed-on: https://go-review.googlesource.com/45833
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
This time, chose a package that's not changed across several builds.
Follow-up on https://go-review.googlesource.com/45151.
Change-Id: I0b487a45f8b12179b3aa495d852c0b7caa44c921
Reviewed-on: https://go-review.googlesource.com/45154
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
This is now allowed in go1.8.
Fixesgolang/go#19646.
Change-Id: Iece4fd2a881144bdbe841e0a26ba4348d6b8828e
Reviewed-on: https://go-review.googlesource.com/38452
Reviewed-by: Alan Donovan <adonovan@google.com>
Run-TryBot: Alan Donovan <adonovan@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
DeleteNamedImport assumes the import declaration is in the form of:
import (
"foo"
)
If an import path is deleted there might be a blank line-sized hole:
import (
)
It'll merge the black hole with the last line to change it to:
import (
)
However the import declaration might be in the following form as well:
import (
"foo")
Whic means after deleting the import path, it changes to:
import (
)
In this case it still tries to merge the line with a non existing line,
causing token.File.MergeLine to panic.
We fix the issue by checking that the import path line is not the last
line to avoid panicing.
Fixesgolang/go#20229
Change-Id: I37537a4eaa83d14db59a2926d7bb14c27167a2e4
Reviewed-on: https://go-review.googlesource.com/44372
Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
This mostly reverts commit 1c59bc354d
(CL 43350) which had fixed Go tip but broken Go 1.7 and Go 1.8
builders.
Change-Id: I4e7bdfafde74c9e730bb870b9db05b663b4f56a3
Reviewed-on: https://go-review.googlesource.com/43391
Reviewed-by: Alan Donovan <adonovan@google.com>
http://golang.org/cl/41476 added the internal/cpu package
to unify cpu feature flag detection in the standard library.
Add a replacement for the assembler function internal/cpu.cpuid
that simulates a x86 cpu with no feature detection capabilities.
Remove bytes.init and strings.init from the external function list
because they do not depend on assembler functions anymore.
Remove hash/crc32.haveSSE42 and math.hasSSE4 because they
have been removed from the go standard library.
Change-Id: Icab6ed3cb13eb14b28d23f2b9c5ae94688f2dc95
Reviewed-on: https://go-review.googlesource.com/43350
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
importPath and srcDir are both of string type.
Change-Id: Ia5230bd19ea83bc210cb0b1a50046e4e0ef2accb
Reviewed-on: https://go-review.googlesource.com/42890
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Manually apply same change as CL 41822 did for cmd/go/internal/get,
but for golang.org/x/tools/go/vcs, to help keep them in sync.
Updates golang/go#18660.
Helps golang/go#11490.
Change-Id: I6c7759c073583dea771bc438b70f8c2eb7b5ebfb
Reviewed-on: https://go-review.googlesource.com/42017
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Backport of CL 41619.
Generated by copying bimport.go and reverting
the chunk containing the "This file is a copy"
comment near the top.
Fixesgolang/go#20121
Change-Id: If24c97d01a550318ab919c37cd0c4a8f34d776c7
Reviewed-on: https://go-review.googlesource.com/41756
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
(specifically: strings.Count, testing.callerEntry)
The interpreter tests were very useful for finding bugs during
development of go/ssa but now seem to be all cost and no benefit.
It may be time to delete this package.
Change-Id: I22348be9fb37bb0fd0c572c3e6f57e70fc069e02
Reviewed-on: https://go-review.googlesource.com/40871
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
This is already tested in the std library. Ok to take the easy way
out here.
Change-Id: Ie17e16fddba827bfe279e1f790006b3874f830ad
Reviewed-on: https://go-review.googlesource.com/38455
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
The existing API for querying values only allows directly querying
pointer-like values, or pointers to such values. The new, extended
queries make it possible to destructure values and access arbitrarily
nested pointers, such as fields in structs, or specific values in
tuples.
We use single Go expressions as the queries, which we further restrict
to a subset of the language (indexing, field access, pointer
dereferences and channel receives).
Instead of adding another map to Result, we return a *Pointer and
populate it during analysis. This does mean that a Config cannot be
reused for multiple analyses, as it now holds analysis-specific state.
That is, however, not a big problem as reusing one Config had little
use.
Since the new API can do everything the old could, it deprecates
AddQuery and AddIndirectQuery.
Change-Id: I4b2ae7d85c462bc7faa6bab76456106e76be5a65
Reviewed-on: https://go-review.googlesource.com/37635
Reviewed-by: Alan Donovan <adonovan@google.com>
Package "unsafe" has no files, and is treated like a regular package since
https://go-review.googlesource.com/37710, and the interpreter needs an
init function with a body for every package.
Change-Id: I27bd7cdb1ecaf01d0b5ed91db57145d1b4f551e1
Reviewed-on: https://go-review.googlesource.com/37711
Reviewed-by: Robert Griesemer <gri@golang.org>
https://golang.org/cl/37694 is a backwards-incompatible change to the
go/types API that causes "unsafe" to be included among the results of
(*types.Package).Imports(). Client packages such as go/loader and
go/ssa now need to recognize the special unsafe package and its
unusual *Bultin members and not create a types.Package for it.
Change-Id: I902725d3053cb887246978efdb5ec5dcdc0e4818
Reviewed-on: https://go-review.googlesource.com/37710
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Found with honnef.co/go/tools/cmd/unused.
Change-Id: Ib8acac744a171b70f3f842f8f5f8c34064869383
Reviewed-on: https://go-review.googlesource.com/37604
Reviewed-by: Alan Donovan <adonovan@google.com>
Found with honnef.co/go/tools/cmd/unused.
Change-Id: I1dc8e4dbfd784bb22b2f0dabb7c1a08cf1ef44ee
Reviewed-on: https://go-review.googlesource.com/37603
Reviewed-by: Alan Donovan <adonovan@google.com>
Found with honnef.co/go/tools/cmd/unused.
Change-Id: I68b4af07cc64a46f5794a0e339859f4d6bc9dad1
Reviewed-on: https://go-review.googlesource.com/37605
Reviewed-by: Alan Donovan <adonovan@google.com>
The previous "dead φ" check was simple and naive but left cycles of
dead φ-nodes. This confused some downstream static analysis tools.
This change makes the φ-nodes liveness check transitive.
+ Test.
Also, number phi nodes so they're not all called t0 during debugging.
Reduces memory consumption by 1%.
Increases execution time by <1%.
Change-Id: I2908662c1478d455fdf4a179f4a12d6184a456c0
Reviewed-on: https://go-review.googlesource.com/37157
Reviewed-by: Robert Griesemer <gri@golang.org>
Tip introduces a new internal/poll package that breaks the expected
output.
Fixesgolang/go#19150
Updates golang/go#19152
Change-Id: I5ff7e8a92afe4d25feb6365933062e931c9b435f
Reviewed-on: https://go-review.googlesource.com/37148
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
Details:
- remove go1.5 "default version" labels on most files
- remove go1.6 labels on a few files
- go/loader: use conf.Cwd not "." in a couple places; update faulty
test expectations. (Not sure why this wasn't caught sooner.)
- go/ssa/interp: add 'mono' result to time.now intrinsic
- go/gcimporter15/bimport.go: make consistent with the version in gc
- go/ssa/interp: update test error message
- go/ssa: update a comment
The go/gcimporter15/bexport.go logic is stale and needs to be brought
up to date. Needs a separate CL since it's tricky.
Tested on go1.6, go1.7, go1.8.
Change-Id: I841189d30e131b7c49a4e8690ea7c40b55041bae
Reviewed-on: https://go-review.googlesource.com/36540
Reviewed-by: Robert Griesemer <gri@golang.org>
This is essentially a copy of the changes in https://golang.org/cl/35268/
but with Go1.9 specific tests factored out into a separate file with build
tag.
For golang/go#18130.
Change-Id: I235693e1bc6145bb53bfd3b343647704c8e82e61
Reviewed-on: https://go-review.googlesource.com/35269
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
- bimport.go is a 1:1 copy of the respective version in the std
library: $GOROOT/src/go/internal/gcimporter/bimport.go .
- bexport.go is mimicking the respective code in the cmd/compile.
- isAlias18/19.go are needed because types.TypeName.IsAlias does
not exist before Go 1.9.
Tested against Go 1.6, 1.7, 1.8, 1.9 (dev.typealias branch).
For golang/go#18130.
Change-Id: Ic46c5850923fab2a35d4dc33850f2b0667c30398
Reviewed-on: https://go-review.googlesource.com/35104
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
Resolve symlinks in source directories (GOPATH, GOROOT, etc.) and source
files in order to find correct package. All I/O performed through
build.Context. Also add minor fix to guru unit tests in order to pass on
Windows.
Change-Id: Ie6134b9cd74eb7386e1d93603eb37c8e44b083b8
Reviewed-on: https://go-review.googlesource.com/33924
Reviewed-by: Alan Donovan <adonovan@google.com>
Reason: ContainingPackage should do all I/O through build.Context.
This reverts commit 3a9a2cbbc4.
Change-Id: I3897d8da5026ddc470989e3f239540286e89df4d
Reviewed-on: https://go-review.googlesource.com/33922
Reviewed-by: Alan Donovan <adonovan@google.com>
Reason: ContainingPackage must do all I/O through build.Context.
This reverts commit c945ee3be4.
Change-Id: I625410bc754ea2d150be097bf424de2be42acde4
Reviewed-on: https://go-review.googlesource.com/33921
Reviewed-by: Alan Donovan <adonovan@google.com>
Restructure tests to account for possibility of being run on Windows
(which doesn't handle symlinks).
Change-Id: I428db26c9a1aad337d8972baa2b71468be3a2e58
Reviewed-on: https://go-review.googlesource.com/33920
Reviewed-by: Alan Donovan <adonovan@google.com>
Resolve symlinks in source directories (GOPATH, GOROOT, etc.) and source
files in order to find correct package.
Fixesgolang/go#16219
Change-Id: I6fae14908827d5ebac55dfe455eaf616f71f8767
Reviewed-on: https://go-review.googlesource.com/33919
Reviewed-by: Alan Donovan <adonovan@google.com>
Fix reference to OverlayContext in the doc for
ParseOverlayArchive.
Change-Id: I0b1db478489764fa309cb2e2f788865ac7ff71bb
Reviewed-on: https://go-review.googlesource.com/33916
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Also: drop all pretense of support for platforms other than darwin and linux.
This package is just a test of go/ssa, not a portable interpreter, and these
are the only platforms to which I have easy access.
Builds on: freebsd darwin plan9 linux windows
Change-Id: I965abc67b1280d33e933b83607a4372d65e070cf
Reviewed-on: https://go-review.googlesource.com/33163
Reviewed-by: Robert Griesemer <gri@golang.org>
The export data formats (text and binary) emitted by gc in Go 1.7 did
not record the package that "owns" each blank field, even though blank
is an unexported identifier. Before, gcimporter would assume the
package of the export data file owns blank fields within it, even
blank fields reexported from another package. As a result, identical
types would become nonidentical during reexporting.
For bug compatibility with gc, gcimporter now treats blank fields as
if they all belong to the same dummy package, avoiding spurious "can't
assign A to B" errors in tools based on go/types.
Change-Id: I0dbf71491a0ec0f376e9dc8a91efe0376c855a28
Reviewed-on: https://go-review.googlesource.com/33146
Reviewed-by: Robert Griesemer <gri@golang.org>
We already are compatible with the latest format. Update version.
Tested against 1.6, 1.7, and tip.
Fixesgolang/go#17734.
Change-Id: Ia167adf0ed1de371c348329513819fb9e8ca2628
Reviewed-on: https://go-review.googlesource.com/33113
Reviewed-by: Alan Donovan <adonovan@google.com>
We don't support Go1.5 anymore.
Tested with 1.6, 1.7, and tip.
For golang/go#17734.
Change-Id: I718c7996b99f9b98744346e119ee209b4709a070
Reviewed-on: https://go-review.googlesource.com/33111
Reviewed-by: Alan Donovan <adonovan@google.com>
If the name of an archive member is longer than 16 bytes, an ELF archive
file contains a string table of file names under the name "//", and this
member has no mode. Skip such members.
+ Test.
Change-Id: Ib10db1cc42816c9002433be6176240e490678560
Reviewed-on: https://go-review.googlesource.com/32973
Reviewed-by: Ian Lance Taylor <iant@golang.org>
...to match the new interface between "go test" and the standard testing
package.
Rather than generate SSA code directly, which was tricky and fragile, we
now generate source just as "go test" does, type-check it, and build an
SSA package from it. crawshaw suggested I do this in the very first
version of testmain.go, but at the time I believed it to be infeasible.
The testMainStartBodyHook mechanism has gone away; installations that
needed it can now achieve the same results more easily by overriding the
templates.
Tested with Go 1.6, 1.7 and 1.8.
Fixesgolang/go#17722
Change-Id: I3ffd25f01157f6fb7a39acd18af46f17e9c07b99
Reviewed-on: https://go-review.googlesource.com/32888
Reviewed-by: Robert Griesemer <gri@golang.org>
Allow extra arguments to be passed to the specified gccgo compiler
when it is called in GccgoInstallation.InitFromDriver. This allows
the call to gccgo to be adjusted depending on the build environment.
For example, some build environments require the
'no-canonical-prefixes' option to prevent gccgo from resolving
symlinks when generating relative prefixes.
Change-Id: I0ecf338ee7a3780f1f65b30e214e69c1698041bb
Reviewed-on: https://go-review.googlesource.com/32874
Reviewed-by: Alan Donovan <adonovan@google.com>
Reason: Decision to back out current alias implementation.
For golang/go#16339 (comment).
Change-Id: Id2a394d78a8661c767bcc05370b81f79d9bfb714
Reviewed-on: https://go-review.googlesource.com/32756
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Chris Manghane <cmang@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
This reverts commit 50193ec1c5.
Reason for revert: aliases will not be part of Go 1.8
Change-Id: Idb3b74243eeb8dbeb2a2f4bd69a248c1dafa5348
Reviewed-on: https://go-review.googlesource.com/32835
Reviewed-by: Robert Griesemer <gri@golang.org>
+ test
Tested on 1.6, 1.7, and 1.8.
Fixesgolang/go#17731.
Change-Id: I06dff4a72ff08ed5e8ae1d23a1e65fe719c03180
Reviewed-on: https://go-review.googlesource.com/32581
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Alan Donovan <adonovan@google.com>
The -newexport flag is on by default for 1.7,
no need to special-case it anymore.
Tested against 1.6, 1.7, and 1.8.
Change-Id: I9c4a31f80d1309564e2a01514fca4b17e4378b9c
Reviewed-on: https://go-review.googlesource.com/32582
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
This CL only copies files and updates build tags.
Substantive changes will come in follow-ups.
This is a workaround for git's lack of rename/copy tracking.
Tested with go1.6, go1.7, and tip (go1.8).
Change-Id: Id88a05273fb963586b228d5e5dfacab32133a960
Reviewed-on: https://go-review.googlesource.com/32630
Reviewed-by: Robert Griesemer <gri@golang.org>
because its export data doesn't contain file position information.
Change-Id: Ia3ab54ae8a493666b9ad63577aa8fad513385052
Reviewed-on: https://go-review.googlesource.com/32299
Reviewed-by: Robert Griesemer <gri@golang.org>
This is a short-term stop-gap to keep the builders happy.
Change-Id: I87171c20bf44f36fd2c4d7213211217de6110fdd
Reviewed-on: https://go-review.googlesource.com/32293
Reviewed-by: Robert Griesemer <gri@golang.org>
(The first was go-review.googlesource.com/c/31813)
Fixes issue golang/go#17565
Change-Id: I61c124e041af689913aedebdde654197c5526228
Reviewed-on: https://go-review.googlesource.com/32034
Reviewed-by: Robert Griesemer <gri@golang.org>
This package is provided as a stop-gap until gccgo uses the same export
data format as gc. Once that occurs, this package will be deprecated
and eventually deleted. The API is similar to (a subset of) gcexportdata.
Change-Id: I3398dbb8eab508a24d12036bbadaa60c2c1e77b3
Reviewed-on: https://go-review.googlesource.com/31822
Reviewed-by: Robert Griesemer <gri@golang.org>
The standard go/internal/gccgoimporter package is essentially unusable
behind the go/importer.For API, so this change makes a copy of it in
x/tools. A follow-up change will expose a small but usable API to it.
Change-Id: Ica5092267ecafb78e1d983c86aa46e4e0bef02d5
Reviewed-on: https://go-review.googlesource.com/31854
Reviewed-by: Robert Griesemer <gri@golang.org>
+ test
Change-Id: I0842795ece634e4441b17a10009d26bfc2a8481a
Reviewed-on: https://go-review.googlesource.com/31818
Reviewed-by: Robert Griesemer <gri@golang.org>
Supporting user-defined TestMain functions requires that we generate a
"testmain" package for each testable package, rather than a single one
for the entire program. This entails these API changes:
1. (*ssa.Program).{CreateTestMainPackage,FindTests} both now
accept only a single package. Existing clients that pass them
multiple packages must call them from a loop.
2. (*ssa.Program).FindTests returns an additional result, the the
optional TestMain *ssa.Function. Existing clients may discard it.
Also:
- Test the new logic using the SSA interpreter
- add ssautil.MainPackages helper
- callgraph: allow multiple main packages, and analyze them all
- ssadump -run: allow multiple main/test packages, and run each in a new interpreter
- minor simplifications to some callers (e.g. guru)
Fixesgolang/go#9553
Change-Id: Ia7de9bd27448fb08b8d172ba5cdbcf37a762b7a0
Reviewed-on: https://go-review.googlesource.com/25102
Reviewed-by: Robert Griesemer <gri@golang.org>
When adding imports to an existing import declaration, ensure that
the target declaration will be printed with parentheses. This allows
all of the imported packages to be printed, not just the first one.
Fixesgolang/go#17212
Change-Id: Ie5de5ec9bca6169650336ee2ea98334817e64e48
Reviewed-on: https://go-review.googlesource.com/29688
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
The "sort" package is no longer quite as low-level as it once was.
Change-Id: Ic62d780841ef1172f65d4c00ec500994f94cb4b7
Reviewed-on: https://go-review.googlesource.com/30931
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
If the parens of an ast.GenDecl are dropped, move the documentation
of the ImportSpec above the import statement, otherwise the the
code is invalid.
Fixesgolang/go#15432.
Change-Id: I715750b8f528380b96a6bc8b5f216043937976c2
Reviewed-on: https://go-review.googlesource.com/22415
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Previously, a trailing slash would cause "foo/" to be treated as the
name of a package, with comical results.
Fixedgolang/go#14584
Change-Id: I660f8a079bbd63d3645812e516a9264c8e080e61
Reviewed-on: https://go-review.googlesource.com/30452
Reviewed-by: Robert Griesemer <gri@golang.org>
When merging import declarations into a single block, AddNamedImport
modifies the list of declarations in the provided file while
iterating over the list. Take care to adjust the index into the list
so as to not skip entries or fall off the end.
Fixesgolang/go#17213
Change-Id: I807246f762c965ea1fc51eb57759f6088336db86
Reviewed-on: https://go-review.googlesource.com/29681
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
...to avoid executing the supportAVX2 assembly function.
Also, combine all no-op intrinsics into a single function.
Change-Id: Ic65a80d3a6df52c3850c34406f034781057a0991
Reviewed-on: https://go-review.googlesource.com/28711
Reviewed-by: Robert Griesemer <gri@golang.org>
Backport of https://go-review.googlesource.com/27814 changes to
bexport.go.
With this, the gcimporter15 is again up-to-date and in sync with both
the compiler's export code, and the compiler's and std lib gcimporter's
import code.
Change-Id: I8e229660eb78ddc1506b7b96a89c81ff083e0412
Reviewed-on: https://go-review.googlesource.com/27914
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Both functions had documentation referring to BuildAll when the
function is now called Build.
Change-Id: I59cce397a0a72bf7fa36f9798e2b07bb6b1da726
Reviewed-on: https://go-review.googlesource.com/25084
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Only unary CallExprs with no ellipsis, f(x), are ambiguous.
Change-Id: If4f17445ab0725dee916992db133eac5536133a7
Reviewed-on: https://go-review.googlesource.com/24552
Reviewed-by: Robert Griesemer <gri@golang.org>
Go is being proposed as an officially supported language for elements of
OpenStack:
https://review.openstack.org/#/c/312267/
As such, repos that exist in OpenStack's git infrastructure
are likely to become places from which people might want to go get
things. Allow optional .git suffixes to allow writing code that
depends on git.openstack.org repos that will work with older go versions
while we wait for this support to roll out.
Change-Id: I1f7be6b7aae63f9c554dbcdbfa46855bcff321df
Reviewed-on: https://go-review.googlesource.com/23362
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
BImportData now requires a token.FileSet.
Column numbers are not exported, and are always reported as 1.
Line numbers greater than the limit (currently 64K) are reported as 1.
+ Test that file/line info is correctly preserved for the entire
standard library.
Change-Id: I80cf370685320240dfb262d36fafd6cdf8569bfb
Reviewed-on: https://go-review.googlesource.com/22788
Reviewed-by: Robert Griesemer <gri@golang.org>
...now that gc's export data records fewer dependencies.
Change-Id: Ie0f68f1f7b0825f6d2c100dfe189be017383db60
Reviewed-on: https://go-review.googlesource.com/22584
Reviewed-by: Robert Griesemer <gri@golang.org>
...which is used by $GOROOT/test/goprint.go.
Change-Id: I4626b8fae3f87d9c8dd8cdcd8c05036955a36262
Reviewed-on: https://go-review.googlesource.com/22560
Reviewed-by: Robert Griesemer <gri@golang.org>
Don't export position info for now. Enabled/disabled with a flag.
The importer recognizes the chosen format automatically.
Change-Id: I055818eb9dba50cc97f40eb92220258a1ddbfbec
Reviewed-on: https://go-review.googlesource.com/22427
Reviewed-by: Alan Donovan <adonovan@google.com>
Read and write position info.
Change-Id: Ibe4a914ff51911bbda656b08f1397132e495ab8a
Reviewed-on: https://go-review.googlesource.com/22098
Reviewed-by: Alan Donovan <adonovan@google.com>
This extracts the overlay context and archive parsing from guru into
buildutil.
At least one tool (gogetdoc) has a vendored copy of this code already,
and more tools implementing the same functionality will follow.
The new code in buildutil is an almost identical copy of the code in
guru (names aside), except for the following changes:
- Instead of reading into a bytes.Buffer, we read directly into a []byte
of appropriate size
- sameFile first attempts a simple comparison of path.Clean'ed paths.
Change-Id: I97cd978ccc10722e3648e5e10625fa7f1407f202
Reviewed-on: https://go-review.googlesource.com/21805
Reviewed-by: Alan Donovan <adonovan@google.com>
Apply style improvements to TestFromDir from golang/go@b6cd6d7d32,
in order to keep them in sync.
Check for error when creating a directory, its successful existence is a
precondition for the test to run.
Helps golang/go#11490.
Change-Id: I87054114c84aead96977f603ca3bd9eccfcfbd5e
Reviewed-on: https://go-review.googlesource.com/21795
Reviewed-by: Alex Brainman <alex.brainman@gmail.com>
Per https://groups.google.com/forum/#!topic/golang-announce/qu_rAphYdxY
this change deletes the packages
go/exact
go/gccgoimporter
go/gcimporter
go/importer
go/types
cmd/vet
from the x/tools repo and any files depending on those packages
building against Go 1.4.
x/tools packages depending on any of these libraries must use the
respective versions from the std lib or use vendored versions if
building against 1.4.
Remaining packages may or may not build against Go 1.4 anymore
and will not be supported against 1.4.
Change-Id: I1c655fc30aee49b6c7326ebd4eb1bb0836ac97e0
Reviewed-on: https://go-review.googlesource.com/20810
Reviewed-by: Alan Donovan <adonovan@google.com>
This is simply a copy of std lib's go/internal/gcimporter/bimport.go
with updated header comment and build tag. No semantic changes.
This will fix part of the x/tools build break at tip. The other part
is to adjust bexport.go (next CL).
Change-Id: Ibc37fae7e0d0447fdea9e3a733aa38589735c59a
Reviewed-on: https://go-review.googlesource.com/21543
Reviewed-by: Alan Donovan <adonovan@google.com>
Although these flags are harmless on linux, they
cause clang on darwin to issue an error about
unused flags (-lpthread). We only care about compilation
so we don't need them.
Change-Id: I0fc756e2f4d7a829d43b5aa912c4e4b24a802a1c
Reviewed-on: https://go-review.googlesource.com/21283
Reviewed-by: Robert Griesemer <gri@golang.org>
The os/user.lookup function no longer exists so instead we use
os/user.current. (This function also existed in go1.5.)
Change-Id: Ic66d80acc186289331f1023e0145370feab6a001
Reviewed-on: https://go-review.googlesource.com/21281
Reviewed-by: Robert Griesemer <gri@golang.org>
It has few interesting effects on aliasing and it contains a lot of
unsafe code that we can't analyze.
Change-Id: I66d3592ea7797802c6be36eb30fd17e2ee307e50
Reviewed-on: https://go-review.googlesource.com/21260
Reviewed-by: Robert Griesemer <gri@golang.org>
This fixes some print calls with wrong format directives. Additionally,
struct initialisers were changed to use keyed fields, purely to reduce
the amount of noise generated by go vet.
Change-Id: Ib9f6fd8f2dff7ce84826478de0ba83dda9746270
Reviewed-on: https://go-review.googlesource.com/21180
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Fix the following issue detected by golint:
import.go:9:1: package comment should be of the form "Package importer ..."
It makes the package comment more consistent with other packages.
Change-Id: I1415066c136ce249af19d70848a38203caf7c9c7
Reviewed-on: https://go-review.googlesource.com/21177
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
The previous CL added a duplicate InvalidType to the predecl table so
that it tracked the one used by gc, but this caused it to fail an
assertion of uniqueness, and I ran the wrong tests.
Change-Id: I56342046cea328e503d917127f5b12205df7999a
Reviewed-on: https://go-review.googlesource.com/20870
Reviewed-by: Robert Griesemer <gri@golang.org>
Although invalid types and unknown constant values should never appear
in .a files (since gc will stop with an error before writing export
data), they can now be faithfully encoded and decoded. This makes the
protocol robust for IDE-like applications that must deal with incomplete
or incorrect programs.
(Corresponding std lib CL: https://go-review.googlesource.com/20828)
Change-Id: I539ffd951b90f01705a7f23ec778c623c729d9a0
Reviewed-on: https://go-review.googlesource.com/20827
Reviewed-by: Robert Griesemer <gri@golang.org>
The binary import/export functionality depends on go/constant from
go1.6 - make it only available if we build against go1.6. Eventually
this packet will be replaced by std lib go/importer functionality.
Change-Id: If96cf2e889daf1250152d7719afa64ad1ba8fb0e
Reviewed-on: https://go-review.googlesource.com/20716
Reviewed-by: Alan Donovan <adonovan@google.com>
Also: renamed file to match 1.7 tag.
Fixesgolang/go#14824.
Change-Id: Iea92292d93c1140e3396678bc37f50f0348e616a
Reviewed-on: https://go-review.googlesource.com/20694
Reviewed-by: Michael Hudson-Doyle <michael.hudson@canonical.com>
Reviewed-by: Alan Donovan <adonovan@google.com>
This change copies the respective changes from https://golang.org/cl/20605.
There is a format conflict here - we are going to track tip, not 1.6.
This change should fix the issue when testing against tip.
Fixesgolang/go#14824.
Change-Id: I58e79cc65748e7a3e5c8486c6cee339884110a07
Reviewed-on: https://go-review.googlesource.com/20693
Reviewed-by: Alan Donovan <adonovan@google.com>
The BuilderModeFlag function does not work with flag packages (such as
Google's) that define their own FlagSet. This change exposes a type
that satisifies flag.Value insted of a function. That type
is *BuilderMode; there's no need for a separate wrapper type.
Change-Id: I8095b80de499e3c52a29a5c1996d1b1fe3799358
Reviewed-on: https://go-review.googlesource.com/20330
Reviewed-by: Michael Matloob <matloob@golang.org>
This CL makes little observable difference to the behavior but paves the
way for streaming 'referrers' and (later) 'implements' queries which
scan the entire workspace, but print each result as soon as it is found.
The go/loader package now exposes a hook, AfterTypeCheck, that lets
clients inspect each package as soon as it is type-checked, and also
modify it, for instance to release unneeded data structures.
A 'referrers' query applied to an exported object must scan the entire
workspace. It uses this hook so to gather uses of the query object in
streaming fashion. However, for now, it still accumulates the results
and prints them all at the end, though I propose to change that in a
follow-up.
Code details:
- The referrers logic had a 2-iteration loop to load first the query
package and then if necessary the enlarged program. The second
iteration has now been unrolled and split into globalReferrers.
- Queries for package names (whether in a package declaration or
a qualified identifier) have been split off into packageReferrers.
It now loads all direct importers of the query package,
which catches some references that were missing before.
(It used to inspect only the forward dependencies of the query
package.)
Also:
- Referrers.Pos (the position of query identifier) was removed from the
JSON output. It's a nuisance to compute now, and it's already
absent from the plain output.
(In a follow-up, I plan to simplify the information content of the
JSON output exactly what is currently printed in the plain output.)
Change-Id: Ia5677636dc7b0fe4461a5d393107665757fb9a97
Reviewed-on: https://go-review.googlesource.com/19794
Reviewed-by: Daniel Morsing <daniel.morsing@gmail.com>
This fixes a bug in guru describe.
Also, add a test of IntuitiveMethodSet.
Change-Id: Ied3780807afd88e664fdb186619499670529fe33
Reviewed-on: https://go-review.googlesource.com/20166
Reviewed-by: Daniel Morsing <daniel.morsing@gmail.com>
+ test
Change-Id: I8db03e19ebb4cf3888f259457aaa3c931da23f24
Reviewed-on: https://go-review.googlesource.com/20109
Reviewed-by: Robert Griesemer <gri@golang.org>
DeleteNamedImport deletes the import with a given
name and path from a parsed file, if present.
imports uses this function to delete named imports.
Fixesgolang/go#8149.
Change-Id: I84539d5627191c45f06db2d008507aee4d3b1eb2
Reviewed-on: https://go-review.googlesource.com/19951
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
This will be used by the guru command's -scope argument.
+ test
Change-Id: I5bf38b544809e4518e2c22a73ec3349a5d2c09fc
Reviewed-on: https://go-review.googlesource.com/19746
Reviewed-by: Michael Matloob <matloob@golang.org>
These nodes are common in incomplete programs.
Change-Id: Iff9750050c78762f0cb8bebc7739584c197d661e
Reviewed-on: https://go-review.googlesource.com/19509
Reviewed-by: Michael Matloob <matloob@golang.org>
This is a backport of the respective changes in golang.org/cl/19393.
For golang/go#14215.
Change-Id: I8d60dd6daa827a60597f3af925e6732914537319
Reviewed-on: https://go-review.googlesource.com/19394
Reviewed-by: Alan Donovan <adonovan@google.com>
For the default build.Context, the Package.Goroot flag indicates when a
package was loaded from the standard library. Until now, the loader
used it to enable the typechecker's intrinsics for the "unsafe" package.
This seemed like a good check, but it is troublesome for clients that
use a nonstandard build.Context. For example, if a client defines
nonstandard Context hooks that load all packages, whether standard or
user-defined, from a flat sstable, there is no way for those hooks to
indicate which packages should have this flag set and which not. As a
result the contents of the "unsafe" package directory are treated as Go
source code when they are merely documentation.
Change-Id: Iea0a7cc9877507d73606391293971a28279c4e4b
Reviewed-on: https://go-review.googlesource.com/19188
Reviewed-by: Robert Griesemer <gri@golang.org>
This change prevents test code to slip into package gcimporter15 and
fixes a weird situation that the test passes but the package doesn't
exist.
For example, we can see the situation with Go 1.6 like the follwoing:
go build
./gcimporter_test16.go:20: undefined: MustHaveGoBuild
./gcimporter_test16.go:45: undefined: testPath
go test
PASS
ok golang.org/x/tools/go/gcimporter15 0.896s
Change-Id: I95550574ccd1b2273072c700c28a82c791c16c63
Reviewed-on: https://go-review.googlesource.com/18950
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
The hosting service was shut down on January 25, 2016 and a few tests
started to fail like the following:
=== RUN TestRepoRootForImportPath
--- FAIL: TestRepoRootForImportPath (2.47s)
vcs_test.go:129: RepoRootForImport("code.google.com/p/go"):
unable to detect version control system for code.google.com/ path
FAIL
This change drops support for code.google.com and fixes test. See
http://google-opensource.blogspot.jp/2015/03/farewell-to-google-code.html
for further information.
Fixesgolang/go#14105
Change-Id: I87cb9a3d666c2a1af0f8bf39a66ba0f669c99fd5
Reviewed-on: https://go-review.googlesource.com/18951
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Also: rename setname15.go to setname.go for consistency
Change-Id: I3f7e37ef4ade716ce64674edb4d53a543a1150c5
Reviewed-on: https://go-review.googlesource.com/18772
Reviewed-by: Alan Donovan <adonovan@google.com>
This makes gcimporter15 build against Go 1.5.
Change-Id: I14e7ff80b28d99f996abc19a0a74b08e5e1bbd75
Reviewed-on: https://go-review.googlesource.com/18771
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
Import path is a '/'-separated path. FromDir documentation says
on return, root is the import path corresponding to the root of
the repository. On Windows and other OSes where os.PathSeparator
is not '/', that wasn't true since root would contain characters
other than '/', and therefore it wasn't a valid import path
corresponding to the root of the repository. Fix that by using
filepath.ToSlash.
Add test coverage for root value returned from FromDir, it was
previously not tested.
Additionally, remove a dubious statement from the documentation
"(thus root is a prefix of importPath)". There is no variable
importPath that is being referred to. It's also redundant and
confusing. Without it, the description of root value matches
the documentation of RepoRoot.Root struct field:
// Root is the import path corresponding to the root of the
// repository.
Root string
Fixesgolang/go#7723.
Change-Id: If9f5f55b5751e01a7f88b79d9b039402af3e9312
Reviewed-on: https://go-review.googlesource.com/18461
Reviewed-by: Chris Manghane <cmang@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
...by making the cache non-blocking and duplicate-suppressing.
(The bottleneck was introduced while adding vendoring support.)
This reduces the 95th percentile map shard time for type-checking 120K
packages to 8min from 20min.
Also: move the I/O-limiting counting semaphore and the NoGoError check
from FindPackage to findPackage so that all implementations benefit.
Change-Id: I43527122262cf80475dd3212d78c340e1c71e36c
Reviewed-on: https://go-review.googlesource.com/18580
Reviewed-by: Robert Griesemer <gri@golang.org>
The loader treats GOROOT's "unsafe" package specially,
with no source files and a Package of types.Unsafe.
Tested on Go 1.4.1, 1.5, and ~1.6 (tip).
Fixes issue #13882
Change-Id: I86c4e394665d86a50ec3852d6d702f0e9c5d2276
Reviewed-on: https://go-review.googlesource.com/18457
Reviewed-by: Robert Griesemer <gri@golang.org>
(late response to comment in review of CL 18452)
Change-Id: I1fa80fd1c7fd22cec752c839d20a483e0732337c
Reviewed-on: https://go-review.googlesource.com/18454
Reviewed-by: Alan Donovan <adonovan@google.com>
(See https://go-review.googlesource.com/#/c/18450)
Details:
- delete buildutil.AllowVendor flag.
Instead use unexported 'go16' var in each package that needs it.
- delete buildutil.StripVendor; was never needed.
- remove inapplicable vendoring comments from loader14.go
- importgraph: go1.5: don't bother checking for 1.6 before calling
absolutize (minor simplifiication and deoptimization).
Tested on 1.4.1, 1.5, and ~1.6 (tip).
Change-Id: If692b0df2eb6c120a9c09d7b1ed99f1b4c6b0826
Reviewed-on: https://go-review.googlesource.com/18452
Reviewed-by: Russ Cox <rsc@golang.org>
Because the std go/importer's Lookup functionality is incomplete,
clients have no way to create a Go 1.5 types.Package from an io.Reader
reading from gc export data. This package provides a stopgap until
the standard library is complete.
The go/gcimporter package remains unchanged, and uses only the
golang.org/x/tools/go/types package.
Change-Id: I47a817f4b6a52ddab26c6b01de6e28099301faf5
Reviewed-on: https://go-review.googlesource.com/18382
Reviewed-by: Robert Griesemer <gri@golang.org>
A few files have been forked and tagged "go1.5,!go1.6" to work around
minor API changes between the two types packages:
- constant.Value.String() in oracle/describe.go and its tests;
- constant.ToInt must now be called before constant.Int64Val.
- types.Config{Importer: importer.Default()} in a number of places
- go/types/typeutil/import_test.go uses lowercase names to avoid 'import "C"'.
Files in go/types/typesutil, missing from my previous CL, have been
tagged !go1.5; these files will be deleted in February.
All affected packages were tested using 1.4.1, 1.5, and ~1.6 (tip).
Change-Id: Iec7fd370e1434508149b378438fb37f65b8d2ba8
Reviewed-on: https://go-review.googlesource.com/18207
Reviewed-by: Robert Griesemer <gri@golang.org>
This change will ensure that the tree continues to work with go1.4.1.
All files continue to depend on golang.org/x/tools/go/types, but in a
follow-up change, I will switch the primary files to depend on the
standard go/types package. Another (smaller) set of files will be
forked and tagged, this time !1.6, due to API differences between the
two packages.
All tests pass using 1.4.1, 1.5, and ~1.6 (tip).
Change-Id: Ifd75a6330e120957d646be91693daaba1ce0e8c9
Reviewed-on: https://go-review.googlesource.com/18333
Reviewed-by: Robert Griesemer <gri@golang.org>
Packages specified on the command line should be interpreted relative to
cwd iff they are local (e.g. ./http within $GOROOT/src/net), otherwise a
request for, say, "golang.org/x/net/http2/hpack" might return the vendored
package depending on the working directory.
The FindPackage hook function now takes a build.ImportMode parameter, so
it matches the signature of (*build.Context).Import. The AllowVendor
flag is enabled only for imports within source files, but not for the
initial packages.
+ test.
Change-Id: I756dc46b70928d2fd9f824e6670092d8169e0d64
Reviewed-on: https://go-review.googlesource.com/18318
Reviewed-by: Robert Griesemer <gri@golang.org>
Use an concurrency-limiting semaphore to reduce I/O parallelism in Import.
Also, start the producer in a new goroutine so that it runs in parallel
with the consumer. Paradoxically, this reduces the peak number of
goroutines.
Also, in buildutil.ForEachPackage, make the concurrency limiting
semaphore global, since I/O parallelism is a process-wide resource.
Change-Id: I282b717c50603361826e5675077c9f464c874132
Reviewed-on: https://go-review.googlesource.com/18215
Reviewed-by: Michael Matloob <matloob@golang.org>
...using same logic as in $GOROOT/go/src/cmd/go/discovery.go
See issue #13757
Change-Id: I2ace0abed0743f4a8675fa8780e190d640b57e6c
Reviewed-on: https://go-review.googlesource.com/18196
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
We need a better approach for tests such as these.
Change-Id: I6fb608acc096088fe377f6f2c40ff6124cf3d09c
Reviewed-on: https://go-review.googlesource.com/18194
Reviewed-by: Alan Donovan <adonovan@google.com>
1. Added intrinsic for os.Pipe to avoid use of "unsafe". This makes
Example tests work in the interpreter, fixing the broken "unicode"
package (which recently sprouted some Examples) and allowing us to
enable tests for:
encoding/csv
flag
path
text/scanner
unicode/utf8
The implementation of os.Pipe may require tweaking for non-Linux platforms.
2. Added intrinsic for strings.Index to avoid assembly dependency.
3. Added dumb intrinsic for testing۰runExample to work around a bug in
go/ssa/testmain.go that requires an invasive fix
(Until then, the output of Example functions will not be checked.)
Change-Id: I6374c9c47aa802275b7cdc98525e057f5db0615a
Reviewed-on: https://go-review.googlesource.com/18191
Reviewed-by: Alan Donovan <adonovan@google.com>
(gcimporter does not support vendoring---see issue #13756)
Change-Id: Ief4101e4f33007d497ed1519845febfcffdded38
Reviewed-on: https://go-review.googlesource.com/18193
Reviewed-by: Alan Donovan <adonovan@google.com>
Prior to this change, the loader made the simplifying assumption
that an import path is equal to a package path, that is, a
subdirectory of src/. (This assumption was already false because
relative imports "../foo" are possible, though discouraged.)
Now, an import "foo" may denote "a/vendor/foo" or "b/vendor/foo"
depending on whether it appears beneath a/ or b/. Thus import
paths and package paths are no longer the same, and the directory
containing an import declaration is a necessary input to the
import resolution function.
This change makes the loader hygienic w.r.t. the directory of
each import declaration, and cleans up the terminology.
The only API change is to the FindPackage hook, which most
clients never use.
Details:
- Add a 'fromDir string' parameter to the FindPackage hook function.
- Add a dir field to each PackageInfo.
- doImport (called by go/types) now consists of two steps:
use FindPackage(dir, importPath) to locate the package,
then consult the import map using the canonical package path.
Only the first step can fail.
- Memoize FindPackage.
- Simplify importInfo now that it no longer has to deal with errors.
Replace a condition variable with a channel.
- Use a separate type to report importErrors.
- Rename loadAll to importAll
- Delete the importMode constant.
- Test.
Change-Id: I3defab51bfa12b48b1511a2172fb48dc8e9150e6
Reviewed-on: https://go-review.googlesource.com/18053
Reviewed-by: Robert Griesemer <gri@golang.org>
Makes programs like ssadump work on packages using vendored code,
for example net/http.
For golang/go#12278.
Depends on CL 17726 in main repository.
Change-Id: Ibabf564e397044a0f449087124dd96161081baaf
Reviewed-on: https://go-review.googlesource.com/17727
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>
Since Go 1.5, if a key expression in a map literal is itself a composite
literal, the type may be omitted. An & operation may be implied,
but was not generated, resulting in a type mismatch crash.
Added regression test.
Also, added missing intrinsic math.hasSSE4 to interpreter.
Fixes issue 13341
Change-Id: I6a5e560460cdcac295a500a3fc3738cbe8aa5853
Reviewed-on: https://go-review.googlesource.com/17123
Reviewed-by: Robert Griesemer <gri@golang.org>
The example incorrectly used buildutil.TagsDoc identifier which does not
exist. It should be buildutil.TagsFlagDoc instead.
The issue was introduced in https://golang.org/cl/9172 when this code was
first written.
Also use more idiomatic style for Go comments, '//' followed by a space,
then tab (instead of skipping the space). Although both render correctly
in godoc, this form is more correct and consistent.
Change-Id: I3b3c9767fe313106a8ff81e7887f3241c5806b59
Reviewed-on: https://go-review.googlesource.com/16743
Reviewed-by: Alan Donovan <adonovan@google.com>
This would cause oracle callee lookup failures when the RHS is an
invoke-style call.
Fixesgolang/go#12999
Change-Id: Ifd561c4e7bf26f57ace5f62afac746b926c70993
Reviewed-on: https://go-review.googlesource.com/16210
Reviewed-by: Alan Donovan <adonovan@google.com>
This adds a missing memory barrier at the end of Package.Build.
Change-Id: Ife35d5ad5a48ba121f35656fef682863d4f2aef6
Reviewed-on: https://go-review.googlesource.com/14761
Reviewed-by: Robert Griesemer <gri@golang.org>
until vendoring support is added (soon).
Change-Id: I954514b5c6b82622c0c831a289ccdcc16dbd37f8
Reviewed-on: https://go-review.googlesource.com/14130
Reviewed-by: Alan Donovan <adonovan@google.com>
This makes 'ulimit -n 128; go test -short -v' pass.
It did not before, and that was breaking the openbsd/386 builder.
For golang/go#11811.
Change-Id: Idfdb2f4007ed06c6084486c0e58a561add552d2c
Reviewed-on: https://go-review.googlesource.com/13695
Reviewed-by: Robert Griesemer <gri@golang.org>
The output of ExampleConfig_CreateFromFiles and ExampleConfig_Import
are different for Windows that for other platforms: They contain
internal/syscall/windows packages and unicode/utf16 not present in
the output for other platforms.
For golang/go#11811.
Change-Id: Id391fbeec8123616da86cb68fc3cefcd513b2493
Reviewed-on: https://go-review.googlesource.com/13032
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Blank line was inserted by mistake in CL 12830.
Change-Id: I9e4b3562a4efcab43c5a85c2960a0d22c0aab752
Reviewed-on: https://go-review.googlesource.com/12883
Reviewed-by: Dmitri Shuralyov <shurcool@gmail.com>
Reviewed-by: Robert Griesemer <gri@golang.org>
Some standard library dependencies have changed (packages and files).
Both ExampleConfig_CreateFromFiles and ExampleConfig_Import Output
needs to be adjusted. Do that.
Update golang/go#11811
Change-Id: I523f2adc1aa46f0932a71ccb23dd7c5a6b07fb27
Reviewed-on: https://go-review.googlesource.com/12832
Reviewed-by: Alan Donovan <adonovan@google.com>
Backport of the go/types-specific changes of:
https://go-review.googlesource.com/#/c/11673/
(from std repo go/types). This will allow running
the go/types tests again in the x/tools repo.
Change-Id: I97d4664d5aed6309e74b571f86f36f8bb4df4fca
Reviewed-on: https://go-review.googlesource.com/12042
Reviewed-by: Ian Lance Taylor <iant@golang.org>
The optional Qualifier function determines what prefix to attach to
package-level names, enabling clients to qualify packages in different
ways, for example, using only the package name instead of its complete
path, or using the locally appropriate name for package given a set of
(possibly renaming) imports.
Prior to this change, clients wanting this behavior had to copy
hundreds of lines of complex printing logic.
Fun fact: (*types.Package).Path and (*types.Package).Name are valid
Qualifier functions.
We provide the RelativeTo helper function to create Qualifiers so that
the old behavior remains a one-liner.
Fixesgolang/go#11133
Change-Id: Ibd63f639c7b3aa1738826d6165f2d810efeb8293
Reviewed-on: https://go-review.googlesource.com/11692
Reviewed-by: Robert Griesemer <gri@golang.org>
Also: make (*Scope).Innermost work for Package scopes.
Change-Id: I9836676e94f95df897101606bed6f29ba46e0f9d
Reviewed-on: https://go-review.googlesource.com/11691
Reviewed-by: Robert Griesemer <gri@golang.org>
since we can't interpret os.Pipe yet.
Change-Id: If27dadc18532274ce97ad7e7557e8614dd15279e
Reviewed-on: https://go-review.googlesource.com/11712
Reviewed-by: Robert Griesemer <gri@golang.org>
Revert "go/types: fix Eval to use correct file set when evaluating an expression"
This reverts commit d241a1448b.
Change-Id: Ie16c57da2732d8b079108efef100fc956b71f737
Reviewed-on: https://go-review.googlesource.com/11010
Reviewed-by: David Symonds <dsymonds@golang.org>
Scopes now have "extent" information; that is they provide a
range [scope.Pos(), scope.End()) which describes the source
text range covered by the scope. It requires that the incoming
AST has correct position information; also the extent for the
Universe and for package scopes is not set (positions are invalid).
Objects have a new (currently unexported) scopePos position,
which is the position at which the object becomes visible inside
its *Scope.
Scope.LookupParent takes an addition parameter pos. If valid, an
identifier is looked up as if found at position pos. This can be
used to find the object corresponding to an identifier at position
pos after scopes have been completely populated (and thus may
contain the same identifier which may be defined only later in the
source text).
Fixes#9980.
Change-Id: Icb49c44c5c3d4b93c0718ce2a769ec468877709d
Reviewed-on: https://go-review.googlesource.com/10800
Reviewed-by: Alan Donovan <adonovan@google.com>
since they use quick.Check, which requires the unimplemented
(reflect.Value).SetString.
Also:
- Add reflect.Type.{In,NumIn} methods, whose absence
was only the proximate cause of the failed test.
- Delete bodies of reflect.Value methods so that it's obvious a
function that should be intrinsic is missing.
Change-Id: Ib64b8f4953a913f4ead90e376bda70419adb87cb
Reviewed-on: https://go-review.googlesource.com/10796
Reviewed-by: Robert Griesemer <gri@golang.org>
And add a TODO.
Change-Id: I51c63b32e9ac4309cdfb8228348a5d528f36a919
Reviewed-on: https://go-review.googlesource.com/10489
Reviewed-by: Robert Griesemer <gri@golang.org>
This backports an anlogous change made to std repo go/types and
enables running the tests again in a reasonable amount of time.
Change-Id: Ied59f6788b7b180f34c918a3c94d50c892b15f32
Reviewed-on: https://go-review.googlesource.com/10467
Reviewed-by: Alan Donovan <adonovan@google.com>
This change sets ParserMode=AllErrors so that the parser is never
allowed to discard the AST and use a dummy one just because it saw too
many errors.
Also, change (*loader.Program).PathEnclosingInterval so that other
clients that forget to set this flag don't panic while calling
fset.File(f.Pos()).Base() on an ast.File f with no position info.
Change-Id: Ie544f169d367d2aa85426212b27063dc72e36fb1
Reviewed-on: https://go-review.googlesource.com/10290
Reviewed-by: Robert Griesemer <gri@golang.org>
Failing to set this when adding a named import
to an existing block at the 0th position
caused the Lparen position to be set to zero.
As a result, the specs were printed as if
they were a single spec, not a group.
This made it appear as if imports had
been swallowed.
See CL 8663 for more context
and the original bug report.
CL 2050 fixed most similar cases
but missed this one.
Change-Id: Ic578fbb8040fa3d3d41db5bde2b839e394801608
Reviewed-on: https://go-review.googlesource.com/10252
Reviewed-by: Robert Griesemer <gri@golang.org>
Now:
% gorename -h
Usage of ./gorename:
...
-tags build tags
a list of build tags to consider...
Change-Id: I46d6906f683407bad6f3dee25c63b139f47e4588
Reviewed-on: https://go-review.googlesource.com/9655
Reviewed-by: Dominik Honnef <dominik@honnef.co>
Reviewed-by: David Crawshaw <crawshaw@golang.org>
Not sure how it ended up there...
Change-Id: I0d48025fd6595714b0c52cadf305b3e06e2b9b84
Reviewed-on: https://go-review.googlesource.com/9171
Reviewed-by: David Crawshaw <crawshaw@golang.org>
And other minor cosmetic tweaks.
Change-Id: Ic75d405e6eca8f29b7e97de66fb86f1f39bcae1e
Reviewed-on: https://go-review.googlesource.com/9035
Reviewed-by: Robert Griesemer <gri@golang.org>
The loader package now loads, parses, and type-checks a whole program
from source, and that is all.
Also:
- simplified loader logic
- ssa.Create is gone; use ssautil.CreateProgram.
- ssautil.LoadPackage renamed to BuildPackage.
It is now independent of go/types' Import hook and the Packages map.
- ssadump: -importbin flag removed.
The value of this flag was that it caused the tool to print IR
for only a single package; this is now the normal behaviour.
Fixes#9955
Change-Id: I4571118258ab1a46dccece3241b7dc51401a3acc
Reviewed-on: https://go-review.googlesource.com/8953
Reviewed-by: Robert Griesemer <gri@golang.org>
Also: make it easier to find packages. Hide the importMap field
(never used) and expose a (*Program).Package method which searches
importMap and then Created.
Also: move huge comments into doc.go.
Change-Id: Iad96a12524b7c41ad9acd1e806af23171e71fa7c
Reviewed-on: https://go-review.googlesource.com/9030
Reviewed-by: Robert Griesemer <gri@golang.org>
All importers should populate the set of imported packages. In Go 1.5
importer API, there will be no package map from which to compute the
transitive closure of dependencies, and SSA package creation needs
this information.
+ test.
Change-Id: I1c2823b07bf7316aa62c80e2ef2a0755cf6f5384
Reviewed-on: https://go-review.googlesource.com/8924
Reviewed-by: Robert Griesemer <gri@golang.org>
Remove all dependencies from non-test code in go/ssa to go/loader,
except the deprecated Create function which will be eliminated in
favor of ssautil.CreateProgram in a mechnanical followup.
Add Examples of two main use cases of SSA construction:
loading a complete program from source; and
building a single package, loading its dependencies from import data.
Add tests to ssautil of the two load functions.
Suggestions welcome for better names.
Planned follow-ups:
- replace all references to ssa.Create with ssautil.CreateProgram and eliminate it.
- eliminate support in go/loader for the ImportBinary flag, and the
PackageCreated hook which is no longer needed since clients can
create the package themselves (see Example).
Step 1 to fixing issue 9955.
Change-Id: I4e64df67fcd5b7f0c0388047e06cea247fddfec5
Reviewed-on: https://go-review.googlesource.com/8669
Reviewed-by: Robert Griesemer <gri@golang.org>
...otherwise the interpreted program can make the whole test suite
slow. Just ignore the argument and return the current GOMAXPROCS
value.
Change-Id: Ife2ad6c53e6fdf9feea1d1b231d8d796b3db3a24
Also: add missing intrinsic for os.runtime_beforeExit.
Reviewed-on: https://go-review.googlesource.com/8591
Reviewed-by: Robert Griesemer <gri@golang.org>
This package was only imported for the trivial Unparen function.
Change-Id: I14f8d91bc0afaa6ab3aa797a53e42e56b59ffcbe
Reviewed-on: https://go-review.googlesource.com/8499
Reviewed-by: Alan Donovan <adonovan@google.com>
This is a putative fix for the file descriptor exhaustion problem
described in https://github.com/golang/go/issues/10306.
Change-Id: If603fb9bbaec1b53f6b44d15b2c202e4670035ce
Reviewed-on: https://go-review.googlesource.com/8421
Reviewed-by: Matt Joiner <anacrolix@gmail.com>
Reviewed-by: David Crawshaw <crawshaw@golang.org>
AddImport and AddNamedImport attempt to place new
imports in roughly the correct place--and thus the
correct group--by matching prefixes. Matching prefixes
byte-by-byte led to "regexp" being grouped with "rsc.io/p".
Instead, match prefixes by segments.
Fixesgolang/go#9961.
Change-Id: I52b7c58a9a2fbe85c2b5297e50c87d409364bda3
Reviewed-on: https://go-review.googlesource.com/8090
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Composite literals are initialized in place where possible, but in
cases the initializer expression refers to the variable that
is being updated
x = T{a: x.a}
we must ensure that the RHS is fully evaluated before we execute any
stores to x. This means we need to record the sequence of stores in a
"store buffer" and execute it only once the entire composite literal
has been evaluated.
Fixes issue #10127
Change-Id: If94e3b179beb25feea5b298ed43de6a199aaf347
Reviewed-on: https://go-review.googlesource.com/7533
Reviewed-by: Robert Griesemer <gri@golang.org>
Be more tolerant in the presence of incorrect range clauses
if for loops and type-check loop body with minimal assumptions
about iteration variables. (Before, in some cases we would simply
ignore the loop body in such cases).
Fixes#10148.
Change-Id: I0b66f81875348088c1a7fa04ccdcbfe768f2eb6c
Reviewed-on: https://go-review.googlesource.com/7525
Reviewed-by: Alan Donovan <adonovan@google.com>
Since all SSA values are immutable, no value copying is required for
any operations except those that load from or store to a variable;
those operations must do an aggregate copy, i.e., descend into struct
and array elements. All other calls to copyVal have been removed;
they were pieces of duct tape, as I had long suspected.
The descent must be based on the static type information, not the
"shape" of the dynamic value, since two reflect.Value structs may have
different internal shapes. We clobber the true definition of
reflect.Value's underlying type, replacing it with struct{interface{},
interface{}}, which is close enough to make the load/store functions
work.
+ Test
Change-Id: I5e239d91ed0cb2a669a9f75766024fe1f9a5c347
Reviewed-on: https://go-review.googlesource.com/7532
Reviewed-by: Robert Griesemer <gri@golang.org>
Config.Cwd sets the base directory; os.Getwd is its default.
+Test.
Change-Id: I213abfb30085cd1306719ed6f94aeae6a3170bc0
Reviewed-on: https://go-review.googlesource.com/7502
Reviewed-by: Robert Griesemer <gri@golang.org>
The callbacks are intentionally concurrent, making this function very
easy to misuse (most clients so far have got it wrong, even my own).
Using a channel in the API makes the concurrency obvious, the
correct usage easy, and the client control flow simpler.
Change-Id: Ied38c3ed5c98b40eb1b322a984ed9dc092ac0918
Reviewed-on: https://go-review.googlesource.com/3250
Reviewed-by: Sameer Ajmani <sameer@golang.org>
Before this change, declared init functions were not package members;
this choice dates from when go/types did not create Func objects for them.
Now, they have an Object. They appear in Members, keyed by "init#%d"
(sequence number) for uniqueness. They can be enumerated. They can
be looked up from a *types.Func via (*Program).FuncValue.
Caveat: fn.Object.Name() no longer equals fn.Name() in all cases.
NB: incompatible API change! (Your build will not break though.)
Change-Id: I2de873079fd57329e6c2f55a282940f6699a77a1
Reviewed-on: https://go-review.googlesource.com/6950
Reviewed-by: Robert Griesemer <gri@golang.org>
Reviewed-by: Peter Collingbourne <pcc@google.com>
Add tests for:
- this naming
- missing file in created package
Change-Id: I07c66b66e845c52d4685509c362b34f1f0c92648
Reviewed-on: https://go-review.googlesource.com/6310
Reviewed-by: Robert Griesemer <gri@golang.org>
...since the zero value is more useful by far.
This is a breaking API change, obviously. (One or two tests in this
CL have intentional been left using the zero value, i.e., they now
load source.)
Change-Id: I42287bfcdb1afef8ee84e5eac12534dd0a1fd5d2
Reviewed-on: https://go-review.googlesource.com/5653
Reviewed-by: Robert Griesemer <gri@golang.org>
Google's proprietary build system, for example, does not use the
_test.go suffix to distinguish test from non-test files; this
information is stated explicitly in another form.
Change-Id: I3a8e919dbc556b6d5cfea1d2123da2616bd934d4
Reviewed-on: https://go-review.googlesource.com/5450
Reviewed-by: David Crawshaw <crawshaw@golang.org>
ssadump's -build=G option is now spelled -import; it was never related to ssa.
Change-Id: Ic21cd8b6990c0ffd25651c17a842a63bfa5019cf
Reviewed-on: https://go-review.googlesource.com/5172
Reviewed-by: David Crawshaw <crawshaw@golang.org>
Before this change, many kinds of error would cause the loader to stop.
making it brittle when analyzing large codebases, as in "godoc -analysis".
This change moves operations that used to occur during
configuration---(*build.Context).Import, loading, and parsing of
initial packages---into the Load call, and ensures that all failures
during Loading are reported at the end so that the maximum amount of
progress is made.
Also: redesign the tests and add many new cases.
Change-Id: Ia8cd99416af7c5d4a5fe133908adfa83676d401f
Reviewed-on: https://go-review.googlesource.com/3626
Reviewed-by: Robert Griesemer <gri@golang.org>
- use import path not file path in go/buildutil.FakeContext OpenFile;
- use regexp to compare error messages in TestErrors, because
they contain windows file paths;
- use OS file path (not unix path), when checking move results
in TestMoves.
Change-Id: Ib62d344acb551fb612d8a0773ae1ab5f18341294
Reviewed-on: https://go-review.googlesource.com/3171
Reviewed-by: Alan Donovan <adonovan@google.com>
And log its value in godoc -analysis.
Related to issue 8968
Change-Id: I96a96922a3fa5c434c69e0faff1cc8ec4686b6f2
Reviewed-on: https://go-review.googlesource.com/3154
Reviewed-by: Robert Griesemer <gri@golang.org>
Details:
- rename (*Program).TypesWithMethodSets() to RuntimeTypes()
- delete (*Package).TypesWithMethodSets() method and simplify
- move code to methods.go
- update test to use
1-2% improvement in space and time (though I barely trust this data
because the GC at tip is in such terrible state).
Change-Id: I38eab78b11e0ad0ff16e0530e775b6ff6a2ab246
Reviewed-on: https://go-review.googlesource.com/3148
Reviewed-by: Robert Griesemer <gri@golang.org>
...to avoid namespace conflicts.
Also make its name "main", since it defines func main().
And fix 2 typos.
Change-Id: I7cf7894d6bed134907b3d2742255e5a82426071b
Reviewed-on: https://go-review.googlesource.com/3150
Reviewed-by: Robert Griesemer <gri@golang.org>
This refactoring of the tests of go/loader and refactor/rename made it
possible to write some loader tests I wanted, but the new tests reveal
bugs so they're commented out for now. I will fix them in a follow-up.
Change-Id: Iae3a20681a0a0791076debd4b82bb5ed74b0c577
Reviewed-on: https://go-review.googlesource.com/2825
Reviewed-by: Robert Griesemer <gri@golang.org>
See loader.go comments for orientation.
This is what happens when you use start using Dmitry's new trace tool. :)
Tests:
- added test of cycles
- load/parse/typecheck of 'godoc' is ~2.5x faster
- race detector finds no errors.
Change-Id: Icb5712c7825002342baf471b216252cecd9149d6
Reviewed-on: https://go-review.googlesource.com/1582
Reviewed-by: Robert Griesemer <gri@golang.org>
PackageCreated is a hook called when a types.Package
is created but before it has been populated.
The package's import Path() and Scope() are defined,
but not its Name() since no package declaration has
been seen yet.
Clients may use this to insert synthetic items into
the package scope, for example.
Change-Id: I210a0c4c766f03f715f03f26d5cd765f15f56e04
Reviewed-on: https://go-review.googlesource.com/2138
Reviewed-by: Alan Donovan <adonovan@google.com>
Classic Go pitfall: "defer" in a loop does not do what you might expect.
+ test case
Fixes issue 9570
Fixes issue 9569
Change-Id: Iec05420872ef71190083a7192f76c92f54f4a2a1
Reviewed-on: https://go-review.googlesource.com/2655
Reviewed-by: Robert Griesemer <gri@golang.org>
The parser was assuming it would find <body> or </head>.
If the entire response is just <meta> tags, it finds EOF and
treats that as an error. It's not.
This is the same change as in https://golang.org/cl/68520044.
Fixes#9556.
Change-Id: If51ed36e7364c15788311039caf8323eb5fe9a6c
Reviewed-on: https://go-review.googlesource.com/2650
Reviewed-by: Minux Ma <minux@golang.org>
Also: Better position for error messages related to wrong use of ... .
Fixes#9473.
Change-Id: I90565f51a42897b7292f651a84a23611a5d8f359
Reviewed-on: https://go-review.googlesource.com/2390
Reviewed-by: Alan Donovan <adonovan@google.com>
It is sometimes useful for API clients to be able to disable the unused
import check, although for good reason this should not be exposed as part of
a user-facing interface. There are at least two use cases that I am aware of:
1) It allows for automated test case reduction tools such as delta or C-Reduce
to be more easily applied to type checker input. Disabling the check
makes it possible for the tool to identify and remove code that depends
on imported packages without any specific knowledge of Go, as the import
need not be removed simultaneously with the code.
2) Interactive tools (such as REPLs) that may have previously received a
list of imports and subsequently receive a line of code that may use any
number of these imports. It is simpler for such tools to import all the
packages in its list than to try to identify the correct set of imports.
Change-Id: I00091a4e5c8e1bd664efd82a636f255eaaa5a2db
Reviewed-on: https://go-review.googlesource.com/2136
Reviewed-by: Robert Griesemer <gri@golang.org>
The new interface makes the functions more useful by allowing clients to
check the various properties that TypeAndValue provides.
Change-Id: I8b41a27316081bea24a18ffe6fa1812e809d6f67
Reviewed-on: https://go-review.googlesource.com/2134
Reviewed-by: Robert Griesemer <gri@golang.org>
+ Tests:
Extend function name uniqueness check to exported wrappers.
Check that expected synthetic functions are created, reachable, and not duplicated.
Change-Id: I0e87ebb2712e33e1f49da3fa9a9dde0085bf3850
Reviewed-on: https://go-review.googlesource.com/2013
Reviewed-by: Robert Griesemer <gri@golang.org>
+ test
Change-Id: Ie37835577ffcdd764cf6a0b611e02f04386755cf
Reviewed-on: https://go-review.googlesource.com/1580
Reviewed-by: Robert Griesemer <gri@golang.org>
...like we do for "runtime" functions, so that they fail informatively
if executed. They all need intrinsics, but only some are yet defined.
Also:
- added test for issue 9462
- "BUG" in test output is now a failure in all tests (not just $GOROOT tests)
- added intrinsic for reflect.SliceOf
- show dynamic type of panic value
Fixes issue 9462
Change-Id: I3a504c7faeed81e922fedc7dd59222717f3a7e95
Reviewed-on: https://go-review.googlesource.com/2145
Reviewed-by: Robert Griesemer <gri@golang.org>
Fixes various problems reported by go vet.
Change-Id: I12a6fdba8f911b21805d8e42903f8f6a5033790a
Reviewed-on: https://go-review.googlesource.com/2163
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
"static" ignores dynamic calls altogether.
"cha" uses Class Hierarchy Analysis, which assumes that a
dynamic call may dispatch to any func or method that satisfies
the type.
Both these algorithms can work on partial programs,
e.g. libraries without a main function or tests.
(This feature was requested after my talk last night.)
+ Tests.
LGTM=sameer
R=sameer, minux
CC=golang-codereviews, gri
https://golang.org/cl/176780043
Also use type assertions in a more defensive
way (check for != nil rather than ok).
LGTM=dsymonds, adonovan
R=adonovan, dsymonds
CC=golang-codereviews
https://golang.org/cl/169480043
This is an algorithm for callgraph construction that is faster
but much less precise than pointer analysis.
(I evaluated this for the Go Oracle last year but shelved it,
but it's a natural fit for the work Brian is doing on
automatic program minimization.)
LGTM=sameer
R=gri, crawshaw, sameer
CC=bwkster, golang-codereviews
https://golang.org/cl/124690043
Rewrite performed with this command:
sed -i '' 's_code.google.com/p/go\._golang.org/x/_g' \
$(grep -lr 'code.google.com/p/go.' *)
LGTM=rsc
R=rsc
CC=golang-codereviews
https://golang.org/cl/170920043
See regression test for explanation.
I audited the code for similar issues and found none.
Many thanks to Daniel Morsing for providing a small
reproducible test case, a rarity for PTA bugs!
Fixesgolang/go#9002
LGTM=crawshaw
R=crawshaw
CC=daniel.morsing, golang-codereviews
https://golang.org/cl/163350043
Also removes a potential race condition regarding the
used flag of Var objects when type-checking packages
concurrently.
Implementation: Rather than marking all used dot-imported
objects and then deduce which corresponding package was used,
now we consider all dot-imported packages as unused and remove
each package from the unused packages map as objects are used.
Now only objects that can be marked as used have a used field
(variables, labels, and packages).
As a result, the code became cleaner and simpler.
Fixesgolang/go#8969.
LGTM=adonovan
R=adonovan
CC=golang-codereviews
https://golang.org/cl/163740043
The previous logic would descend into (e.g.) .git repositories
and vendored packages with "_"-prefixed names.
Fixesgolang/go#8907
LGTM=gri
R=gri
CC=golang-codereviews, shurcool
https://golang.org/cl/157800043
The type-checker depended on (token.Pos) position information of
the presented files to determine source order. That information
is determined by the parse order of the files rather than the
order in which the files are presented to the type-checker.
Introduced an order number strictly determined by the file
order as presented to the type-checker and the AST structure
of each file; thus providing source order information even in
the absence of (token.Pos) position information.
Added test case (provided by adonovan).
LGTM=adonovan
R=adonovan
CC=golang-codereviews
https://golang.org/cl/151160043
- ParseFile (core of go/loader's parseFiles(); also used by refactor/rename)
- ContainingPackage (core of Oracle's guessImportPath; also used by refactor/rename)
- Accessors for effective "methods" of build.Context:
FileExists, OpenFile, IsAbsPath, JoinPath.
LGTM=sameer
R=dave, sameer
CC=golang-codereviews, gri
https://golang.org/cl/146120043
The needMethods cache logic was wrong: it would treat any
previous call as a cache hit, even if 'skip' was true for that
call. As a result it could fail to generate methods for some
'skip' types, i.e. anonymous structs.
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/144750043
This CL is the first refactoring automated by "gorename". :)
Though I had to update the comments and run 'hg gofmt'. :(
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/142930043
It now has a main() function, which was confusing the logic to find the entry point.
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/142860044
This function has been copied at least 6 times throughout
go.tools. This implementation is superior since it does
all I/O through the virtualized go/build file system, and it
is highly parallel (and much faster).
We expose two flavours, simple (for existing tests) and
parallel (for high-performance tools such as gorename).
This CL creates the go/buildutil package, which is intended for
utilities related to go/build.
+ test.
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/137430043
- for unused packages where base(package path) != package name
- for conflicts between imported packages or dot-imported objects
and local declarations
Per suggestions from adonovan, inspired by the gc error messages.
LGTM=adonovan
R=adonovan, bradfitz
CC=golang-codereviews
https://golang.org/cl/135550043
(godoc is excluded from this CL since it will continue to use
/src/pkg in its URL namespace, making the necessary cleanup
more subtle.)
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/141770043
Recent changes in the runtime caused the interpreter to call 'getg',
and it should never have gotten that far.
Also, delete bodies of "runtime" functions, since they're too magical.
This makes missing intrinsics cause very obvious failures.
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/135330043
It returns the value formerly returned by Pkg(), i.e. the imported package.
Pkg() now returns the package enclosing the import statement,
which is consistent with all other Objects.
Fixesgolang/go#8628.
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/136090043
Also: Added Config.Strict flag to enable type checker tests
beyond the Go 1 specification.
Fixesgolang/go#8561.
LGTM=adonovan
R=adonovan
CC=golang-codereviews
https://golang.org/cl/134050043
LookupFieldOrMethod now also decides whether a found
method is actually in the method set. Simplifies call
sites. Added corresponding API tests.
TODO (separate CL): Decide what the correct value for
the indirect result should be (as required for code
generation). For now, the result value for indirect
is unchanged from before if a field/method is found.
Fixesgolang/go#8584.
LGTM=adonovan
R=adonovan
CC=golang-codereviews
https://golang.org/cl/132260043
(1) support Example functions defined in programs that don't
import "testing". We emit code to testmain.main() to call
them directly, since we can't call testing.Main.
(2) expose a FindTests function which reports the set of
Test, Example and Benchmark functions it finds.
Certain clients need this.
Added test for logic in FindTests.
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/115290048
Previously these were recorded in the PackageInfo, but not
reported to the user's error handler (types.Config.Error),
which is typically what prints them.
Minor subtlety: that function must now be able to handle error
values that are not of type types.Error.
+ Test (and renamed it).
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/121290043
- Break out parts of coverage.go into more specific files.
- Re-enable test of nil interface-to-interface conversion.
- Update initorder test to reflect spec ambiguity and gc vs go/types variance.
- Re-enable test dependent on now-fixed bug 8189 ("value,ok" yields an untyped bool)
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/119530043
And serialize the printing of each item with a mutex.
It is the formatted output of this tool, after all.
Also: minor doc tweaks.
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/114620044
Examples:
- "foo$1" becomes "pkg.foo$1"
- "init$1" (meaning the first declared "init" function) becomes "init#1",
to distinguish it from "init$1" (meaning the first anonymous function
within the synthetic "init" function that initializes package-level vars).
It is now an invariant that all source-level (non-synthetic)
functions have distinct names, and that all names include the
enclosing package. Added test for this.
+ updated various clients.
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/122750043
It was missing from the four conversions and the Make* instructions.
(Experiments with pure bytes.Buffer-based printing were not faster; various TODOs removed.)
LGTM=crawshaw
R=gri, crawshaw
CC=golang-codereviews
https://golang.org/cl/58040043
This transforms the virtualized directory (build.Context).Dir to a physical one,
for proprietary build systems that distinguish them.
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/116230043
Users need only add an extra file to the package to specify
additional imports and initialization steps in testmain, to
match their build system.
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/120090043
Without it, no value appears to be sent on NewTicker/NewTimer channels.
+ test
Also:
- add (callgraph.Edge).{Description,Pos} convenience methods
to simplify client code when Site==nil.
LGTM=gri
R=gri, friestein68503
CC=golang-codereviews
https://golang.org/cl/112610043
If an expression is addressable, we compute its address then load, rather than
extracting the value of subelements. For aggregates this avoids large copies.
Example:
var x [2]struct{y [3]int}
print(x[1].y[2])
Was:
t0 = local [3]struct{x [5]int} (x) *[3]struct{x [5]int}
t1 = *t0 [3]struct{x [5]int}
t2 = t1[1:int] struct{x [5]int}
t3 = t2.x [#0] [5]int
t4 = t3[2:int] int
Now:
t1 = &t0[1:int] *struct{x [5]int}
t2 = &t1.x [#0] *[5]int
t3 = &t2[2:int] *int
t4 = *t3 int
Also:
- make emitFieldSelections responsible for calling emitDebugRef, as
one of its two calls was forgetting to do it.
- relax the specification of (*Program).VarValue because not all
subexpressions are materalized as values now.
- fix up the objlookup.go test expectations to match.
go/ssa/interp test runs 10% faster.
Thanks to Peter Collingbourne for pointing this out.
LGTM=pcc
R=pcc, gri
CC=golang-codereviews
https://golang.org/cl/109710043
PackageInfo:
- deleted IsType
- inlined + deleted: ValueOf, TypeCaseVar, ImportSpecPkg
- on failure, TypeOf accessor now returns nil (was: panic)
go/ssa: avoid extra map lookups by using Uses or Defs directly when safe to do so,
and keeping the TypeAndValue around in expr0().
LGTM=gri
R=gri, pcc
CC=golang-codereviews
https://golang.org/cl/107650043
The InitOrder needs to be reset.
+ Test.
This bug manifested itself in duplicate HTML in the godoc -analysis view,
e.g. "f((x)" or "funcfunc f()"
LGTM=gri
R=gri
CC=golang-codereviews
https://golang.org/cl/107660044
This extends the sanity checker to identify and report referrers
which do not appear in the function's instruction lists, and fixes two
bugs in the lifting algorithm which were caught by the sanity check.
LGTM=adonovan
R=adonovan
CC=axwalk, golang-codereviews
https://golang.org/cl/110210045
We introduce a method (*Interface).Complete(), which is intended
to be called from clients after all embedded interfaces have been
fully defined. For importers, this will definitely be the case
after the import has finished, so each importer have been updated
to do so, with the exception of the gcimporter, which does not use
embedded interfaces, therefore Complete() can be called immediately
after construction.
Building the method set separately from the constructor type caused
some problems with go/importer, which copies the types.Interface
object, leading to there existing two almost-identical interface
types referenced from interface method receivers, only one of which
has been completed. To avoid this situation, the importer has been
modified to construct the interface object only once.
Fixesgolang/go#8177.
LGTM=gri
R=gri, dave, gordon.klaus, adonovan
CC=golang-codereviews
https://golang.org/cl/105060044
This method was deemed unfit for the API. See the original CL for discussion.
««« original CL description
go.tools/go/loader: Add Program.FilePath convenience method for getting the full path of a source file.
LGTM=gri
R=gri, adonovan
CC=golang-codereviews
https://golang.org/cl/107160049
»»»
LGTM=gri
R=gri, adonovan
CC=golang-codereviews
https://golang.org/cl/107570043
This builtin is a little weird in this form as it is (to my knowledge)
the only function that takes a variadic argument of non-slice
type. The language provides no syntax to express this, so we pick
a stringification for such arguments that does not appear in the
language. Specifically, use T... instead of ...T to distinguish it
from the normal case where the type is a slice.
This change lets the go/ssa package produce more efficient IR by
avoiding an extra conversion of the second argument.
LGTM=gri
R=gri
CC=adonovan, golang-codereviews
https://golang.org/cl/108230044
+ regression test.
Fixesgolang/go#8172
Also: return error (not panic) when called with empty input.
LGTM=gri
R=crawshaw, gri
CC=golang-codereviews, jon
https://golang.org/cl/104270043
Alan Donovan
Robert Griesemer
Dmitri Shuralyov
haya14busa
Kirill Smelkov
Spencer Nelson
Fatih Arslan
Brad Fitzpatrick
Martin Möhrmann
Koichi Shiraishi
Josh Bleecher Snyder
Ross Light
Matthew Dempsky
Dominik Honnef
David R. Jenni
Ian Lance Taylor
Samuel Tan
Rebecca Stambler
Zac Bergquist
Alan Donovan
Rhys Hiltner
Russ Cox
Tim Henderson
Monty Taylor
Akihiro Suda
Idora Shinatose
Alberto Donizetti
Mikio Hara
Kenji Kaneda
David Symonds
Alex Brainman
Daniel Morsing
Michal Cierniak
Peter Collingbourne
Emil Hessman
Andrew Gerrand
Gordon Klaus
Aram Hăvărneanu