summaryrefslogtreecommitdiffstats
path: root/src/cmd/go/internal/load/pkg.go
diff options
context:
space:
mode:
Diffstat (limited to 'src/cmd/go/internal/load/pkg.go')
-rw-r--r--src/cmd/go/internal/load/pkg.go2594
1 files changed, 2594 insertions, 0 deletions
diff --git a/src/cmd/go/internal/load/pkg.go b/src/cmd/go/internal/load/pkg.go
new file mode 100644
index 0000000..61fde89
--- /dev/null
+++ b/src/cmd/go/internal/load/pkg.go
@@ -0,0 +1,2594 @@
+// Copyright 2011 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+// Package load loads packages.
+package load
+
+import (
+ "bytes"
+ "context"
+ "encoding/json"
+ "errors"
+ "fmt"
+ "go/build"
+ "go/scanner"
+ "go/token"
+ "io/fs"
+ "os"
+ "path"
+ pathpkg "path"
+ "path/filepath"
+ "runtime"
+ "sort"
+ "strconv"
+ "strings"
+ "unicode"
+ "unicode/utf8"
+
+ "cmd/go/internal/base"
+ "cmd/go/internal/cfg"
+ "cmd/go/internal/fsys"
+ "cmd/go/internal/modinfo"
+ "cmd/go/internal/modload"
+ "cmd/go/internal/par"
+ "cmd/go/internal/search"
+ "cmd/go/internal/str"
+ "cmd/go/internal/trace"
+ "cmd/internal/sys"
+
+ "golang.org/x/mod/module"
+)
+
+var IgnoreImports bool // control whether we ignore imports in packages
+
+// A Package describes a single package found in a directory.
+type Package struct {
+ PackagePublic // visible in 'go list'
+ Internal PackageInternal // for use inside go command only
+}
+
+type PackagePublic struct {
+ // Note: These fields are part of the go command's public API.
+ // See list.go. It is okay to add fields, but not to change or
+ // remove existing ones. Keep in sync with list.go
+ Dir string `json:",omitempty"` // directory containing package sources
+ ImportPath string `json:",omitempty"` // import path of package in dir
+ ImportComment string `json:",omitempty"` // path in import comment on package statement
+ Name string `json:",omitempty"` // package name
+ Doc string `json:",omitempty"` // package documentation string
+ Target string `json:",omitempty"` // installed target for this package (may be executable)
+ Shlib string `json:",omitempty"` // the shared library that contains this package (only set when -linkshared)
+ Root string `json:",omitempty"` // Go root, Go path dir, or module root dir containing this package
+ ConflictDir string `json:",omitempty"` // Dir is hidden by this other directory
+ ForTest string `json:",omitempty"` // package is only for use in named test
+ Export string `json:",omitempty"` // file containing export data (set by go list -export)
+ BuildID string `json:",omitempty"` // build ID of the compiled package (set by go list -export)
+ Module *modinfo.ModulePublic `json:",omitempty"` // info about package's module, if any
+ Match []string `json:",omitempty"` // command-line patterns matching this package
+ Goroot bool `json:",omitempty"` // is this package found in the Go root?
+ Standard bool `json:",omitempty"` // is this package part of the standard Go library?
+ DepOnly bool `json:",omitempty"` // package is only as a dependency, not explicitly listed
+ BinaryOnly bool `json:",omitempty"` // package cannot be recompiled
+ Incomplete bool `json:",omitempty"` // was there an error loading this package or dependencies?
+
+ // Stale and StaleReason remain here *only* for the list command.
+ // They are only initialized in preparation for list execution.
+ // The regular build determines staleness on the fly during action execution.
+ Stale bool `json:",omitempty"` // would 'go install' do anything for this package?
+ StaleReason string `json:",omitempty"` // why is Stale true?
+
+ // Source files
+ // If you add to this list you MUST add to p.AllFiles (below) too.
+ // Otherwise file name security lists will not apply to any new additions.
+ GoFiles []string `json:",omitempty"` // .go source files (excluding CgoFiles, TestGoFiles, XTestGoFiles)
+ CgoFiles []string `json:",omitempty"` // .go source files that import "C"
+ CompiledGoFiles []string `json:",omitempty"` // .go output from running cgo on CgoFiles
+ IgnoredGoFiles []string `json:",omitempty"` // .go source files ignored due to build constraints
+ IgnoredOtherFiles []string `json:",omitempty"` // non-.go source files ignored due to build constraints
+ CFiles []string `json:",omitempty"` // .c source files
+ CXXFiles []string `json:",omitempty"` // .cc, .cpp and .cxx source files
+ MFiles []string `json:",omitempty"` // .m source files
+ HFiles []string `json:",omitempty"` // .h, .hh, .hpp and .hxx source files
+ FFiles []string `json:",omitempty"` // .f, .F, .for and .f90 Fortran source files
+ SFiles []string `json:",omitempty"` // .s source files
+ SwigFiles []string `json:",omitempty"` // .swig files
+ SwigCXXFiles []string `json:",omitempty"` // .swigcxx files
+ SysoFiles []string `json:",omitempty"` // .syso system object files added to package
+
+ // Embedded files
+ EmbedPatterns []string `json:",omitempty"` // //go:embed patterns
+ EmbedFiles []string `json:",omitempty"` // files matched by EmbedPatterns
+
+ // Cgo directives
+ CgoCFLAGS []string `json:",omitempty"` // cgo: flags for C compiler
+ CgoCPPFLAGS []string `json:",omitempty"` // cgo: flags for C preprocessor
+ CgoCXXFLAGS []string `json:",omitempty"` // cgo: flags for C++ compiler
+ CgoFFLAGS []string `json:",omitempty"` // cgo: flags for Fortran compiler
+ CgoLDFLAGS []string `json:",omitempty"` // cgo: flags for linker
+ CgoPkgConfig []string `json:",omitempty"` // cgo: pkg-config names
+
+ // Dependency information
+ Imports []string `json:",omitempty"` // import paths used by this package
+ ImportMap map[string]string `json:",omitempty"` // map from source import to ImportPath (identity entries omitted)
+ Deps []string `json:",omitempty"` // all (recursively) imported dependencies
+
+ // Error information
+ // Incomplete is above, packed into the other bools
+ Error *PackageError `json:",omitempty"` // error loading this package (not dependencies)
+ DepsErrors []*PackageError `json:",omitempty"` // errors loading dependencies
+
+ // Test information
+ // If you add to this list you MUST add to p.AllFiles (below) too.
+ // Otherwise file name security lists will not apply to any new additions.
+ TestGoFiles []string `json:",omitempty"` // _test.go files in package
+ TestImports []string `json:",omitempty"` // imports from TestGoFiles
+ TestEmbedPatterns []string `json:",omitempty"` // //go:embed patterns
+ TestEmbedFiles []string `json:",omitempty"` // files matched by TestEmbedPatterns
+ XTestGoFiles []string `json:",omitempty"` // _test.go files outside package
+ XTestImports []string `json:",omitempty"` // imports from XTestGoFiles
+ XTestEmbedPatterns []string `json:",omitempty"` // //go:embed patterns
+ XTestEmbedFiles []string `json:",omitempty"` // files matched by XTestEmbedPatterns
+}
+
+// AllFiles returns the names of all the files considered for the package.
+// This is used for sanity and security checks, so we include all files,
+// even IgnoredGoFiles, because some subcommands consider them.
+// The go/build package filtered others out (like foo_wrongGOARCH.s)
+// and that's OK.
+func (p *Package) AllFiles() []string {
+ files := str.StringList(
+ p.GoFiles,
+ p.CgoFiles,
+ // no p.CompiledGoFiles, because they are from GoFiles or generated by us
+ p.IgnoredGoFiles,
+ p.IgnoredOtherFiles,
+ p.CFiles,
+ p.CXXFiles,
+ p.MFiles,
+ p.HFiles,
+ p.FFiles,
+ p.SFiles,
+ p.SwigFiles,
+ p.SwigCXXFiles,
+ p.SysoFiles,
+ p.TestGoFiles,
+ p.XTestGoFiles,
+ )
+
+ // EmbedFiles may overlap with the other files.
+ // Dedup, but delay building the map as long as possible.
+ // Only files in the current directory (no slash in name)
+ // need to be checked against the files variable above.
+ var have map[string]bool
+ for _, file := range p.EmbedFiles {
+ if !strings.Contains(file, "/") {
+ if have == nil {
+ have = make(map[string]bool)
+ for _, file := range files {
+ have[file] = true
+ }
+ }
+ if have[file] {
+ continue
+ }
+ }
+ files = append(files, file)
+ }
+ return files
+}
+
+// Desc returns the package "description", for use in b.showOutput.
+func (p *Package) Desc() string {
+ if p.ForTest != "" {
+ return p.ImportPath + " [" + p.ForTest + ".test]"
+ }
+ return p.ImportPath
+}
+
+type PackageInternal struct {
+ // Unexported fields are not part of the public API.
+ Build *build.Package
+ Imports []*Package // this package's direct imports
+ CompiledImports []string // additional Imports necessary when using CompiledGoFiles (all from standard library)
+ RawImports []string // this package's original imports as they appear in the text of the program
+ ForceLibrary bool // this package is a library (even if named "main")
+ CmdlineFiles bool // package built from files listed on command line
+ CmdlinePkg bool // package listed on command line
+ CmdlinePkgLiteral bool // package listed as literal on command line (not via wildcard)
+ Local bool // imported via local path (./ or ../)
+ LocalPrefix string // interpret ./ and ../ imports relative to this prefix
+ ExeName string // desired name for temporary executable
+ CoverMode string // preprocess Go source files with the coverage tool in this mode
+ CoverVars map[string]*CoverVar // variables created by coverage analysis
+ OmitDebug bool // tell linker not to write debug information
+ GobinSubdir bool // install target would be subdir of GOBIN
+ BuildInfo string // add this info to package main
+ TestmainGo *[]byte // content for _testmain.go
+ Embed map[string][]string // //go:embed comment mapping
+
+ Asmflags []string // -asmflags for this package
+ Gcflags []string // -gcflags for this package
+ Ldflags []string // -ldflags for this package
+ Gccgoflags []string // -gccgoflags for this package
+}
+
+// A NoGoError indicates that no Go files for the package were applicable to the
+// build for that package.
+//
+// That may be because there were no files whatsoever, or because all files were
+// excluded, or because all non-excluded files were test sources.
+type NoGoError struct {
+ Package *Package
+}
+
+func (e *NoGoError) Error() string {
+ if len(e.Package.IgnoredGoFiles) > 0 {
+ // Go files exist, but they were ignored due to build constraints.
+ return "build constraints exclude all Go files in " + e.Package.Dir
+ }
+ if len(e.Package.TestGoFiles)+len(e.Package.XTestGoFiles) > 0 {
+ // Test Go files exist, but we're not interested in them.
+ // The double-negative is unfortunate but we want e.Package.Dir
+ // to appear at the end of error message.
+ return "no non-test Go files in " + e.Package.Dir
+ }
+ return "no Go files in " + e.Package.Dir
+}
+
+// setLoadPackageDataError presents an error found when loading package data
+// as a *PackageError. It has special cases for some common errors to improve
+// messages shown to users and reduce redundancy.
+//
+// setLoadPackageDataError returns true if it's safe to load information about
+// imported packages, for example, if there was a parse error loading imports
+// in one file, but other files are okay.
+func (p *Package) setLoadPackageDataError(err error, path string, stk *ImportStack, importPos []token.Position) {
+ matchErr, isMatchErr := err.(*search.MatchError)
+ if isMatchErr && matchErr.Match.Pattern() == path {
+ if matchErr.Match.IsLiteral() {
+ // The error has a pattern has a pattern similar to the import path.
+ // It may be slightly different (./foo matching example.com/foo),
+ // but close enough to seem redundant.
+ // Unwrap the error so we don't show the pattern.
+ err = matchErr.Err
+ }
+ }
+
+ // Replace (possibly wrapped) *build.NoGoError with *load.NoGoError.
+ // The latter is more specific about the cause.
+ var nogoErr *build.NoGoError
+ if errors.As(err, &nogoErr) {
+ if p.Dir == "" && nogoErr.Dir != "" {
+ p.Dir = nogoErr.Dir
+ }
+ err = &NoGoError{Package: p}
+ }
+
+ // Take only the first error from a scanner.ErrorList. PackageError only
+ // has room for one position, so we report the first error with a position
+ // instead of all of the errors without a position.
+ var pos string
+ var isScanErr bool
+ if scanErr, ok := err.(scanner.ErrorList); ok && len(scanErr) > 0 {
+ isScanErr = true // For stack push/pop below.
+
+ scanPos := scanErr[0].Pos
+ scanPos.Filename = base.ShortPath(scanPos.Filename)
+ pos = scanPos.String()
+ err = errors.New(scanErr[0].Msg)
+ }
+
+ // Report the error on the importing package if the problem is with the import declaration
+ // for example, if the package doesn't exist or if the import path is malformed.
+ // On the other hand, don't include a position if the problem is with the imported package,
+ // for example there are no Go files (NoGoError), or there's a problem in the imported
+ // package's source files themselves (scanner errors).
+ //
+ // TODO(matloob): Perhaps make each of those the errors in the first group
+ // (including modload.ImportMissingError, ImportMissingSumError, and the
+ // corresponding "cannot find package %q in any of" GOPATH-mode error
+ // produced in build.(*Context).Import; modload.AmbiguousImportError,
+ // and modload.PackageNotInModuleError; and the malformed module path errors
+ // produced in golang.org/x/mod/module.CheckMod) implement an interface
+ // to make it easier to check for them? That would save us from having to
+ // move the modload errors into this package to avoid a package import cycle,
+ // and from having to export an error type for the errors produced in build.
+ if !isMatchErr && (nogoErr != nil || isScanErr) {
+ stk.Push(path)
+ defer stk.Pop()
+ }
+
+ p.Error = &PackageError{
+ ImportStack: stk.Copy(),
+ Pos: pos,
+ Err: err,
+ }
+
+ if path != stk.Top() {
+ p.Error.setPos(importPos)
+ }
+}
+
+// Resolve returns the resolved version of imports,
+// which should be p.TestImports or p.XTestImports, NOT p.Imports.
+// The imports in p.TestImports and p.XTestImports are not recursively
+// loaded during the initial load of p, so they list the imports found in
+// the source file, but most processing should be over the vendor-resolved
+// import paths. We do this resolution lazily both to avoid file system work
+// and because the eventual real load of the test imports (during 'go test')
+// can produce better error messages if it starts with the original paths.
+// The initial load of p loads all the non-test imports and rewrites
+// the vendored paths, so nothing should ever call p.vendored(p.Imports).
+func (p *Package) Resolve(imports []string) []string {
+ if len(imports) > 0 && len(p.Imports) > 0 && &imports[0] == &p.Imports[0] {
+ panic("internal error: p.Resolve(p.Imports) called")
+ }
+ seen := make(map[string]bool)
+ var all []string
+ for _, path := range imports {
+ path = ResolveImportPath(p, path)
+ if !seen[path] {
+ seen[path] = true
+ all = append(all, path)
+ }
+ }
+ sort.Strings(all)
+ return all
+}
+
+// CoverVar holds the name of the generated coverage variables targeting the named file.
+type CoverVar struct {
+ File string // local file name
+ Var string // name of count struct
+}
+
+func (p *Package) copyBuild(pp *build.Package) {
+ p.Internal.Build = pp
+
+ if pp.PkgTargetRoot != "" && cfg.BuildPkgdir != "" {
+ old := pp.PkgTargetRoot
+ pp.PkgRoot = cfg.BuildPkgdir
+ pp.PkgTargetRoot = cfg.BuildPkgdir
+ pp.PkgObj = filepath.Join(cfg.BuildPkgdir, strings.TrimPrefix(pp.PkgObj, old))
+ }
+
+ p.Dir = pp.Dir
+ p.ImportPath = pp.ImportPath
+ p.ImportComment = pp.ImportComment
+ p.Name = pp.Name
+ p.Doc = pp.Doc
+ p.Root = pp.Root
+ p.ConflictDir = pp.ConflictDir
+ p.BinaryOnly = pp.BinaryOnly
+
+ // TODO? Target
+ p.Goroot = pp.Goroot
+ p.Standard = p.Goroot && p.ImportPath != "" && search.IsStandardImportPath(p.ImportPath)
+ p.GoFiles = pp.GoFiles
+ p.CgoFiles = pp.CgoFiles
+ p.IgnoredGoFiles = pp.IgnoredGoFiles
+ p.IgnoredOtherFiles = pp.IgnoredOtherFiles
+ p.CFiles = pp.CFiles
+ p.CXXFiles = pp.CXXFiles
+ p.MFiles = pp.MFiles
+ p.HFiles = pp.HFiles
+ p.FFiles = pp.FFiles
+ p.SFiles = pp.SFiles
+ p.SwigFiles = pp.SwigFiles
+ p.SwigCXXFiles = pp.SwigCXXFiles
+ p.SysoFiles = pp.SysoFiles
+ p.CgoCFLAGS = pp.CgoCFLAGS
+ p.CgoCPPFLAGS = pp.CgoCPPFLAGS
+ p.CgoCXXFLAGS = pp.CgoCXXFLAGS
+ p.CgoFFLAGS = pp.CgoFFLAGS
+ p.CgoLDFLAGS = pp.CgoLDFLAGS
+ p.CgoPkgConfig = pp.CgoPkgConfig
+ // We modify p.Imports in place, so make copy now.
+ p.Imports = make([]string, len(pp.Imports))
+ copy(p.Imports, pp.Imports)
+ p.Internal.RawImports = pp.Imports
+ p.TestGoFiles = pp.TestGoFiles
+ p.TestImports = pp.TestImports
+ p.XTestGoFiles = pp.XTestGoFiles
+ p.XTestImports = pp.XTestImports
+ if IgnoreImports {
+ p.Imports = nil
+ p.Internal.RawImports = nil
+ p.TestImports = nil
+ p.XTestImports = nil
+ }
+ p.EmbedPatterns = pp.EmbedPatterns
+ p.TestEmbedPatterns = pp.TestEmbedPatterns
+ p.XTestEmbedPatterns = pp.XTestEmbedPatterns
+}
+
+// A PackageError describes an error loading information about a package.
+type PackageError struct {
+ ImportStack []string // shortest path from package named on command line to this one
+ Pos string // position of error
+ Err error // the error itself
+ IsImportCycle bool // the error is an import cycle
+ Hard bool // whether the error is soft or hard; soft errors are ignored in some places
+ alwaysPrintStack bool // whether to always print the ImportStack
+}
+
+func (p *PackageError) Error() string {
+ // TODO(#43696): decide when to print the stack or the position based on
+ // the error type and whether the package is in the main module.
+ // Document the rationale.
+ if p.Pos != "" && (len(p.ImportStack) == 0 || !p.alwaysPrintStack) {
+ // Omit import stack. The full path to the file where the error
+ // is the most important thing.
+ return p.Pos + ": " + p.Err.Error()
+ }
+
+ // If the error is an ImportPathError, and the last path on the stack appears
+ // in the error message, omit that path from the stack to avoid repetition.
+ // If an ImportPathError wraps another ImportPathError that matches the
+ // last path on the stack, we don't omit the path. An error like
+ // "package A imports B: error loading C caused by B" would not be clearer
+ // if "imports B" were omitted.
+ if len(p.ImportStack) == 0 {
+ return p.Err.Error()
+ }
+ var optpos string
+ if p.Pos != "" {
+ optpos = "\n\t" + p.Pos
+ }
+ return "package " + strings.Join(p.ImportStack, "\n\timports ") + optpos + ": " + p.Err.Error()
+}
+
+func (p *PackageError) Unwrap() error { return p.Err }
+
+// PackageError implements MarshalJSON so that Err is marshaled as a string
+// and non-essential fields are omitted.
+func (p *PackageError) MarshalJSON() ([]byte, error) {
+ perr := struct {
+ ImportStack []string
+ Pos string
+ Err string
+ }{p.ImportStack, p.Pos, p.Err.Error()}
+ return json.Marshal(perr)
+}
+
+func (p *PackageError) setPos(posList []token.Position) {
+ if len(posList) == 0 {
+ return
+ }
+ pos := posList[0]
+ pos.Filename = base.ShortPath(pos.Filename)
+ p.Pos = pos.String()
+}
+
+// ImportPathError is a type of error that prevents a package from being loaded
+// for a given import path. When such a package is loaded, a *Package is
+// returned with Err wrapping an ImportPathError: the error is attached to
+// the imported package, not the importing package.
+//
+// The string returned by ImportPath must appear in the string returned by
+// Error. Errors that wrap ImportPathError (such as PackageError) may omit
+// the import path.
+type ImportPathError interface {
+ error
+ ImportPath() string
+}
+
+var (
+ _ ImportPathError = (*importError)(nil)
+ _ ImportPathError = (*modload.ImportMissingError)(nil)
+ _ ImportPathError = (*modload.ImportMissingSumError)(nil)
+)
+
+type importError struct {
+ importPath string
+ err error // created with fmt.Errorf
+}
+
+func ImportErrorf(path, format string, args ...interface{}) ImportPathError {
+ err := &importError{importPath: path, err: fmt.Errorf(format, args...)}
+ if errStr := err.Error(); !strings.Contains(errStr, path) {
+ panic(fmt.Sprintf("path %q not in error %q", path, errStr))
+ }
+ return err
+}
+
+func (e *importError) Error() string {
+ return e.err.Error()
+}
+
+func (e *importError) Unwrap() error {
+ // Don't return e.err directly, since we're only wrapping an error if %w
+ // was passed to ImportErrorf.
+ return errors.Unwrap(e.err)
+}
+
+func (e *importError) ImportPath() string {
+ return e.importPath
+}
+
+// An ImportStack is a stack of import paths, possibly with the suffix " (test)" appended.
+// The import path of a test package is the import path of the corresponding
+// non-test package with the suffix "_test" added.
+type ImportStack []string
+
+func (s *ImportStack) Push(p string) {
+ *s = append(*s, p)
+}
+
+func (s *ImportStack) Pop() {
+ *s = (*s)[0 : len(*s)-1]
+}
+
+func (s *ImportStack) Copy() []string {
+ return append([]string{}, *s...)
+}
+
+func (s *ImportStack) Top() string {
+ if len(*s) == 0 {
+ return ""
+ }
+ return (*s)[len(*s)-1]
+}
+
+// shorterThan reports whether sp is shorter than t.
+// We use this to record the shortest import sequence
+// that leads to a particular package.
+func (sp *ImportStack) shorterThan(t []string) bool {
+ s := *sp
+ if len(s) != len(t) {
+ return len(s) < len(t)
+ }
+ // If they are the same length, settle ties using string ordering.
+ for i := range s {
+ if s[i] != t[i] {
+ return s[i] < t[i]
+ }
+ }
+ return false // they are equal
+}
+
+// packageCache is a lookup cache for LoadImport,
+// so that if we look up a package multiple times
+// we return the same pointer each time.
+var packageCache = map[string]*Package{}
+
+// ClearPackageCache clears the in-memory package cache and the preload caches.
+// It is only for use by GOPATH-based "go get".
+// TODO(jayconrod): When GOPATH-based "go get" is removed, delete this function.
+func ClearPackageCache() {
+ for name := range packageCache {
+ delete(packageCache, name)
+ }
+ resolvedImportCache.Clear()
+ packageDataCache.Clear()
+}
+
+// ClearPackageCachePartial clears packages with the given import paths from the
+// in-memory package cache and the preload caches. It is only for use by
+// GOPATH-based "go get".
+// TODO(jayconrod): When GOPATH-based "go get" is removed, delete this function.
+func ClearPackageCachePartial(args []string) {
+ shouldDelete := make(map[string]bool)
+ for _, arg := range args {
+ shouldDelete[arg] = true
+ if p := packageCache[arg]; p != nil {
+ delete(packageCache, arg)
+ }
+ }
+ resolvedImportCache.DeleteIf(func(key interface{}) bool {
+ return shouldDelete[key.(importSpec).path]
+ })
+ packageDataCache.DeleteIf(func(key interface{}) bool {
+ return shouldDelete[key.(string)]
+ })
+}
+
+// ReloadPackageNoFlags is like LoadImport but makes sure
+// not to use the package cache.
+// It is only for use by GOPATH-based "go get".
+// TODO(rsc): When GOPATH-based "go get" is removed, delete this function.
+func ReloadPackageNoFlags(arg string, stk *ImportStack) *Package {
+ p := packageCache[arg]
+ if p != nil {
+ delete(packageCache, arg)
+ resolvedImportCache.DeleteIf(func(key interface{}) bool {
+ return key.(importSpec).path == p.ImportPath
+ })
+ packageDataCache.Delete(p.ImportPath)
+ }
+ return LoadImport(context.TODO(), arg, base.Cwd, nil, stk, nil, 0)
+}
+
+// dirToImportPath returns the pseudo-import path we use for a package
+// outside the Go path. It begins with _/ and then contains the full path
+// to the directory. If the package lives in c:\home\gopher\my\pkg then
+// the pseudo-import path is _/c_/home/gopher/my/pkg.
+// Using a pseudo-import path like this makes the ./ imports no longer
+// a special case, so that all the code to deal with ordinary imports works
+// automatically.
+func dirToImportPath(dir string) string {
+ return pathpkg.Join("_", strings.Map(makeImportValid, filepath.ToSlash(dir)))
+}
+
+func makeImportValid(r rune) rune {
+ // Should match Go spec, compilers, and ../../go/parser/parser.go:/isValidImport.
+ const illegalChars = `!"#$%&'()*,:;<=>?[\]^{|}` + "`\uFFFD"
+ if !unicode.IsGraphic(r) || unicode.IsSpace(r) || strings.ContainsRune(illegalChars, r) {
+ return '_'
+ }
+ return r
+}
+
+// Mode flags for loadImport and download (in get.go).
+const (
+ // ResolveImport means that loadImport should do import path expansion.
+ // That is, ResolveImport means that the import path came from
+ // a source file and has not been expanded yet to account for
+ // vendoring or possible module adjustment.
+ // Every import path should be loaded initially with ResolveImport,
+ // and then the expanded version (for example with the /vendor/ in it)
+ // gets recorded as the canonical import path. At that point, future loads
+ // of that package must not pass ResolveImport, because
+ // disallowVendor will reject direct use of paths containing /vendor/.
+ ResolveImport = 1 << iota
+
+ // ResolveModule is for download (part of "go get") and indicates
+ // that the module adjustment should be done, but not vendor adjustment.
+ ResolveModule
+
+ // GetTestDeps is for download (part of "go get") and indicates
+ // that test dependencies should be fetched too.
+ GetTestDeps
+)
+
+// LoadImport scans the directory named by path, which must be an import path,
+// but possibly a local import path (an absolute file system path or one beginning
+// with ./ or ../). A local relative path is interpreted relative to srcDir.
+// It returns a *Package describing the package found in that directory.
+// LoadImport does not set tool flags and should only be used by
+// this package, as part of a bigger load operation, and by GOPATH-based "go get".
+// TODO(rsc): When GOPATH-based "go get" is removed, unexport this function.
+func LoadImport(ctx context.Context, path, srcDir string, parent *Package, stk *ImportStack, importPos []token.Position, mode int) *Package {
+ return loadImport(ctx, nil, path, srcDir, parent, stk, importPos, mode)
+}
+
+func loadImport(ctx context.Context, pre *preload, path, srcDir string, parent *Package, stk *ImportStack, importPos []token.Position, mode int) *Package {
+ if path == "" {
+ panic("LoadImport called with empty package path")
+ }
+
+ var parentPath, parentRoot string
+ parentIsStd := false
+ if parent != nil {
+ parentPath = parent.ImportPath
+ parentRoot = parent.Root
+ parentIsStd = parent.Standard
+ }
+ bp, loaded, err := loadPackageData(path, parentPath, srcDir, parentRoot, parentIsStd, mode)
+ if loaded && pre != nil && !IgnoreImports {
+ pre.preloadImports(bp.Imports, bp)
+ }
+ if bp == nil {
+ if importErr, ok := err.(ImportPathError); !ok || importErr.ImportPath() != path {
+ // Only add path to the error's import stack if it's not already present on the error.
+ stk.Push(path)
+ defer stk.Pop()
+ }
+ return &Package{
+ PackagePublic: PackagePublic{
+ ImportPath: path,
+ Error: &PackageError{
+ ImportStack: stk.Copy(),
+ Err: err,
+ },
+ },
+ }
+ }
+
+ importPath := bp.ImportPath
+ p := packageCache[importPath]
+ if p != nil {
+ stk.Push(path)
+ p = reusePackage(p, stk)
+ stk.Pop()
+ } else {
+ p = new(Package)
+ p.Internal.Local = build.IsLocalImport(path)
+ p.ImportPath = importPath
+ packageCache[importPath] = p
+
+ // Load package.
+ // loadPackageData may return bp != nil even if an error occurs,
+ // in order to return partial information.
+ p.load(ctx, path, stk, importPos, bp, err)
+
+ if !cfg.ModulesEnabled && path != cleanImport(path) {
+ p.Error = &PackageError{
+ ImportStack: stk.Copy(),
+ Err: ImportErrorf(path, "non-canonical import path %q: should be %q", path, pathpkg.Clean(path)),
+ }
+ p.Incomplete = true
+ p.Error.setPos(importPos)
+ }
+ }
+
+ // Checked on every import because the rules depend on the code doing the importing.
+ if perr := disallowInternal(srcDir, parent, parentPath, p, stk); perr != p {
+ perr.Error.setPos(importPos)
+ return perr
+ }
+ if mode&ResolveImport != 0 {
+ if perr := disallowVendor(srcDir, path, parentPath, p, stk); perr != p {
+ perr.Error.setPos(importPos)
+ return perr
+ }
+ }
+
+ if p.Name == "main" && parent != nil && parent.Dir != p.Dir {
+ perr := *p
+ perr.Error = &PackageError{
+ ImportStack: stk.Copy(),
+ Err: ImportErrorf(path, "import %q is a program, not an importable package", path),
+ }
+ perr.Error.setPos(importPos)
+ return &perr
+ }
+
+ if p.Internal.Local && parent != nil && !parent.Internal.Local {
+ perr := *p
+ var err error
+ if path == "." {
+ err = ImportErrorf(path, "%s: cannot import current directory", path)
+ } else {
+ err = ImportErrorf(path, "local import %q in non-local package", path)
+ }
+ perr.Error = &PackageError{
+ ImportStack: stk.Copy(),
+ Err: err,
+ }
+ perr.Error.setPos(importPos)
+ return &perr
+ }
+
+ return p
+}
+
+// loadPackageData loads information needed to construct a *Package. The result
+// is cached, and later calls to loadPackageData for the same package will return
+// the same data.
+//
+// loadPackageData returns a non-nil package even if err is non-nil unless
+// the package path is malformed (for example, the path contains "mod/" or "@").
+//
+// loadPackageData returns a boolean, loaded, which is true if this is the
+// first time the package was loaded. Callers may preload imports in this case.
+func loadPackageData(path, parentPath, parentDir, parentRoot string, parentIsStd bool, mode int) (bp *build.Package, loaded bool, err error) {
+ if path == "" {
+ panic("loadPackageData called with empty package path")
+ }
+
+ if strings.HasPrefix(path, "mod/") {
+ // Paths beginning with "mod/" might accidentally
+ // look in the module cache directory tree in $GOPATH/pkg/mod/.
+ // This prefix is owned by the Go core for possible use in the
+ // standard library (since it does not begin with a domain name),
+ // so it's OK to disallow entirely.
+ return nil, false, fmt.Errorf("disallowed import path %q", path)
+ }
+
+ if strings.Contains(path, "@") {
+ return nil, false, errors.New("can only use path@version syntax with 'go get' and 'go install' in module-aware mode")
+ }
+
+ // Determine canonical package path and directory.
+ // For a local import the identifier is the pseudo-import path
+ // we create from the full directory to the package.
+ // Otherwise it is the usual import path.
+ // For vendored imports, it is the expanded form.
+ //
+ // Note that when modules are enabled, local import paths are normally
+ // canonicalized by modload.LoadPackages before now. However, if there's an
+ // error resolving a local path, it will be returned untransformed
+ // so that 'go list -e' reports something useful.
+ importKey := importSpec{
+ path: path,
+ parentPath: parentPath,
+ parentDir: parentDir,
+ parentRoot: parentRoot,
+ parentIsStd: parentIsStd,
+ mode: mode,
+ }
+ r := resolvedImportCache.Do(importKey, func() interface{} {
+ var r resolvedImport
+ if build.IsLocalImport(path) {
+ r.dir = filepath.Join(parentDir, path)
+ r.path = dirToImportPath(r.dir)
+ } else if cfg.ModulesEnabled {
+ r.dir, r.path, r.err = modload.Lookup(parentPath, parentIsStd, path)
+ } else if mode&ResolveImport != 0 {
+ // We do our own path resolution, because we want to
+ // find out the key to use in packageCache without the
+ // overhead of repeated calls to buildContext.Import.
+ // The code is also needed in a few other places anyway.
+ r.path = resolveImportPath(path, parentPath, parentDir, parentRoot, parentIsStd)
+ } else if mode&ResolveModule != 0 {
+ r.path = moduleImportPath(path, parentPath, parentDir, parentRoot)
+ }
+ if r.path == "" {
+ r.path = path
+ }
+ return r
+ }).(resolvedImport)
+ // Invariant: r.path is set to the resolved import path. If the path cannot
+ // be resolved, r.path is set to path, the source import path.
+ // r.path is never empty.
+
+ // Load the package from its directory. If we already found the package's
+ // directory when resolving its import path, use that.
+ data := packageDataCache.Do(r.path, func() interface{} {
+ loaded = true
+ var data packageData
+ if r.dir != "" {
+ var buildMode build.ImportMode
+ if !cfg.ModulesEnabled {
+ buildMode = build.ImportComment
+ }
+ data.p, data.err = cfg.BuildContext.ImportDir(r.dir, buildMode)
+ if data.p.Root == "" && cfg.ModulesEnabled {
+ if info := modload.PackageModuleInfo(path); info != nil {
+ data.p.Root = info.Dir
+ }
+ }
+ } else if r.err != nil {
+ data.p = new(build.Package)
+ data.err = r.err
+ } else if cfg.ModulesEnabled && path != "unsafe" {
+ data.p = new(build.Package)
+ data.err = fmt.Errorf("unknown import path %q: internal error: module loader did not resolve import", r.path)
+ } else {
+ buildMode := build.ImportComment
+ if mode&ResolveImport == 0 || r.path != path {
+ // Not vendoring, or we already found the vendored path.
+ buildMode |= build.IgnoreVendor
+ }
+ data.p, data.err = cfg.BuildContext.Import(r.path, parentDir, buildMode)
+ }
+ data.p.ImportPath = r.path
+
+ // Set data.p.BinDir in cases where go/build.Context.Import
+ // may give us a path we don't want.
+ if !data.p.Goroot {
+ if cfg.GOBIN != "" {
+ data.p.BinDir = cfg.GOBIN
+ } else if cfg.ModulesEnabled {
+ data.p.BinDir = modload.BinDir()
+ }
+ }
+
+ if !cfg.ModulesEnabled && data.err == nil &&
+ data.p.ImportComment != "" && data.p.ImportComment != path &&
+ !strings.Contains(path, "/vendor/") && !strings.HasPrefix(path, "vendor/") {
+ data.err = fmt.Errorf("code in directory %s expects import %q", data.p.Dir, data.p.ImportComment)
+ }
+ return data
+ }).(packageData)
+
+ return data.p, loaded, data.err
+}
+
+// importSpec describes an import declaration in source code. It is used as a
+// cache key for resolvedImportCache.
+type importSpec struct {
+ path string
+ parentPath, parentDir, parentRoot string
+ parentIsStd bool
+ mode int
+}
+
+// resolvedImport holds a canonical identifier for a package. It may also contain
+// a path to the package's directory and an error if one occurred. resolvedImport
+// is the value type in resolvedImportCache.
+type resolvedImport struct {
+ path, dir string
+ err error
+}
+
+// packageData holds information loaded from a package. It is the value type
+// in packageDataCache.
+type packageData struct {
+ p *build.Package
+ err error
+}
+
+// resolvedImportCache maps import strings (importSpec) to canonical package names
+// (resolvedImport).
+var resolvedImportCache par.Cache
+
+// packageDataCache maps canonical package names (string) to package metadata
+// (packageData).
+var packageDataCache par.Cache
+
+// preloadWorkerCount is the number of concurrent goroutines that can load
+// packages. Experimentally, there are diminishing returns with more than
+// 4 workers. This was measured on the following machines.
+//
+// * MacBookPro with a 4-core Intel Core i7 CPU
+// * Linux workstation with 6-core Intel Xeon CPU
+// * Linux workstation with 24-core Intel Xeon CPU
+//
+// It is very likely (though not confirmed) that this workload is limited
+// by memory bandwidth. We don't have a good way to determine the number of
+// workers that would saturate the bus though, so runtime.GOMAXPROCS
+// seems like a reasonable default.
+var preloadWorkerCount = runtime.GOMAXPROCS(0)
+
+// preload holds state for managing concurrent preloading of package data.
+//
+// A preload should be created with newPreload before loading a large
+// package graph. flush must be called when package loading is complete
+// to ensure preload goroutines are no longer active. This is necessary
+// because of global mutable state that cannot safely be read and written
+// concurrently. In particular, packageDataCache may be cleared by "go get"
+// in GOPATH mode, and modload.loaded (accessed via modload.Lookup) may be
+// modified by modload.LoadPackages.
+type preload struct {
+ cancel chan struct{}
+ sema chan struct{}
+}
+
+// newPreload creates a new preloader. flush must be called later to avoid
+// accessing global state while it is being modified.
+func newPreload() *preload {
+ pre := &preload{
+ cancel: make(chan struct{}),
+ sema: make(chan struct{}, preloadWorkerCount),
+ }
+ return pre
+}
+
+// preloadMatches loads data for package paths matched by patterns.
+// When preloadMatches returns, some packages may not be loaded yet, but
+// loadPackageData and loadImport are always safe to call.
+func (pre *preload) preloadMatches(matches []*search.Match) {
+ for _, m := range matches {
+ for _, pkg := range m.Pkgs {
+ select {
+ case <-pre.cancel:
+ return
+ case pre.sema <- struct{}{}:
+ go func(pkg string) {
+ mode := 0 // don't use vendoring or module import resolution
+ bp, loaded, err := loadPackageData(pkg, "", base.Cwd, "", false, mode)
+ <-pre.sema
+ if bp != nil && loaded && err == nil && !IgnoreImports {
+ pre.preloadImports(bp.Imports, bp)
+ }
+ }(pkg)
+ }
+ }
+ }
+}
+
+// preloadImports queues a list of imports for preloading.
+// When preloadImports returns, some packages may not be loaded yet,
+// but loadPackageData and loadImport are always safe to call.
+func (pre *preload) preloadImports(imports []string, parent *build.Package) {
+ parentIsStd := parent.Goroot && parent.ImportPath != "" && search.IsStandardImportPath(parent.ImportPath)
+ for _, path := range imports {
+ if path == "C" || path == "unsafe" {
+ continue
+ }
+ select {
+ case <-pre.cancel:
+ return
+ case pre.sema <- struct{}{}:
+ go func(path string) {
+ bp, loaded, err := loadPackageData(path, parent.ImportPath, parent.Dir, parent.Root, parentIsStd, ResolveImport)
+ <-pre.sema
+ if bp != nil && loaded && err == nil && !IgnoreImports {
+ pre.preloadImports(bp.Imports, bp)
+ }
+ }(path)
+ }
+ }
+}
+
+// flush stops pending preload operations. flush blocks until preload calls to
+// loadPackageData have completed. The preloader will not make any new calls
+// to loadPackageData.
+func (pre *preload) flush() {
+ // flush is usually deferred.
+ // Don't hang program waiting for workers on panic.
+ if v := recover(); v != nil {
+ panic(v)
+ }
+
+ close(pre.cancel)
+ for i := 0; i < preloadWorkerCount; i++ {
+ pre.sema <- struct{}{}
+ }
+}
+
+func cleanImport(path string) string {
+ orig := path
+ path = pathpkg.Clean(path)
+ if strings.HasPrefix(orig, "./") && path != ".." && !strings.HasPrefix(path, "../") {
+ path = "./" + path
+ }
+ return path
+}
+
+var isDirCache par.Cache
+
+func isDir(path string) bool {
+ return isDirCache.Do(path, func() interface{} {
+ fi, err := fsys.Stat(path)
+ return err == nil && fi.IsDir()
+ }).(bool)
+}
+
+// ResolveImportPath returns the true meaning of path when it appears in parent.
+// There are two different resolutions applied.
+// First, there is Go 1.5 vendoring (golang.org/s/go15vendor).
+// If vendor expansion doesn't trigger, then the path is also subject to
+// Go 1.11 module legacy conversion (golang.org/issue/25069).
+func ResolveImportPath(parent *Package, path string) (found string) {
+ var parentPath, parentDir, parentRoot string
+ parentIsStd := false
+ if parent != nil {
+ parentPath = parent.ImportPath
+ parentDir = parent.Dir
+ parentRoot = parent.Root
+ parentIsStd = parent.Standard
+ }
+ return resolveImportPath(path, parentPath, parentDir, parentRoot, parentIsStd)
+}
+
+func resolveImportPath(path, parentPath, parentDir, parentRoot string, parentIsStd bool) (found string) {
+ if cfg.ModulesEnabled {
+ if _, p, e := modload.Lookup(parentPath, parentIsStd, path); e == nil {
+ return p
+ }
+ return path
+ }
+ found = vendoredImportPath(path, parentPath, parentDir, parentRoot)
+ if found != path {
+ return found
+ }
+ return moduleImportPath(path, parentPath, parentDir, parentRoot)
+}
+
+// dirAndRoot returns the source directory and workspace root
+// for the package p, guaranteeing that root is a path prefix of dir.
+func dirAndRoot(path string, dir, root string) (string, string) {
+ origDir, origRoot := dir, root
+ dir = filepath.Clean(dir)
+ root = filepath.Join(root, "src")
+ if !str.HasFilePathPrefix(dir, root) || path != "command-line-arguments" && filepath.Join(root, path) != dir {
+ // Look for symlinks before reporting error.
+ dir = expandPath(dir)
+ root = expandPath(root)
+ }
+
+ if !str.HasFilePathPrefix(dir, root) || len(dir) <= len(root) || dir[len(root)] != filepath.Separator || path != "command-line-arguments" && !build.IsLocalImport(path) && filepath.Join(root, path) != dir {
+ base.Fatalf("unexpected directory layout:\n"+
+ " import path: %s\n"+
+ " root: %s\n"+
+ " dir: %s\n"+
+ " expand root: %s\n"+
+ " expand dir: %s\n"+
+ " separator: %s",
+ path,
+ filepath.Join(origRoot, "src"),
+ filepath.Clean(origDir),
+ origRoot,
+ origDir,
+ string(filepath.Separator))
+ }
+
+ return dir, root
+}
+
+// vendoredImportPath returns the vendor-expansion of path when it appears in parent.
+// If parent is x/y/z, then path might expand to x/y/z/vendor/path, x/y/vendor/path,
+// x/vendor/path, vendor/path, or else stay path if none of those exist.
+// vendoredImportPath returns the expanded path or, if no expansion is found, the original.
+func vendoredImportPath(path, parentPath, parentDir, parentRoot string) (found string) {
+ if parentRoot == "" {
+ return path
+ }
+
+ dir, root := dirAndRoot(parentPath, parentDir, parentRoot)
+
+ vpath := "vendor/" + path
+ for i := len(dir); i >= len(root); i-- {
+ if i < len(dir) && dir[i] != filepath.Separator {
+ continue
+ }
+ // Note: checking for the vendor directory before checking
+ // for the vendor/path directory helps us hit the
+ // isDir cache more often. It also helps us prepare a more useful
+ // list of places we looked, to report when an import is not found.
+ if !isDir(filepath.Join(dir[:i], "vendor")) {
+ continue
+ }
+ targ := filepath.Join(dir[:i], vpath)
+ if isDir(targ) && hasGoFiles(targ) {
+ importPath := parentPath
+ if importPath == "command-line-arguments" {
+ // If parent.ImportPath is 'command-line-arguments'.
+ // set to relative directory to root (also chopped root directory)
+ importPath = dir[len(root)+1:]
+ }
+ // We started with parent's dir c:\gopath\src\foo\bar\baz\quux\xyzzy.
+ // We know the import path for parent's dir.
+ // We chopped off some number of path elements and
+ // added vendor\path to produce c:\gopath\src\foo\bar\baz\vendor\path.
+ // Now we want to know the import path for that directory.
+ // Construct it by chopping the same number of path elements
+ // (actually the same number of bytes) from parent's import path
+ // and then append /vendor/path.
+ chopped := len(dir) - i
+ if chopped == len(importPath)+1 {
+ // We walked up from c:\gopath\src\foo\bar
+ // and found c:\gopath\src\vendor\path.
+ // We chopped \foo\bar (length 8) but the import path is "foo/bar" (length 7).
+ // Use "vendor/path" without any prefix.
+ return vpath
+ }
+ return importPath[:len(importPath)-chopped] + "/" + vpath
+ }
+ }
+ return path
+}
+
+var (
+ modulePrefix = []byte("\nmodule ")
+ goModPathCache par.Cache
+)
+
+// goModPath returns the module path in the go.mod in dir, if any.
+func goModPath(dir string) (path string) {
+ return goModPathCache.Do(dir, func() interface{} {
+ data, err := os.ReadFile(filepath.Join(dir, "go.mod"))
+ if err != nil {
+ return ""
+ }
+ var i int
+ if bytes.HasPrefix(data, modulePrefix[1:]) {
+ i = 0
+ } else {
+ i = bytes.Index(data, modulePrefix)
+ if i < 0 {
+ return ""
+ }
+ i++
+ }
+ line := data[i:]
+
+ // Cut line at \n, drop trailing \r if present.
+ if j := bytes.IndexByte(line, '\n'); j >= 0 {
+ line = line[:j]
+ }
+ if line[len(line)-1] == '\r' {
+ line = line[:len(line)-1]
+ }
+ line = line[len("module "):]
+
+ // If quoted, unquote.
+ path = strings.TrimSpace(string(line))
+ if path != "" && path[0] == '"' {
+ s, err := strconv.Unquote(path)
+ if err != nil {
+ return ""
+ }
+ path = s
+ }
+ return path
+ }).(string)
+}
+
+// findVersionElement returns the slice indices of the final version element /vN in path.
+// If there is no such element, it returns -1, -1.
+func findVersionElement(path string) (i, j int) {
+ j = len(path)
+ for i = len(path) - 1; i >= 0; i-- {
+ if path[i] == '/' {
+ if isVersionElement(path[i+1 : j]) {
+ return i, j
+ }
+ j = i
+ }
+ }
+ return -1, -1
+}
+
+// isVersionElement reports whether s is a well-formed path version element:
+// v2, v3, v10, etc, but not v0, v05, v1.
+func isVersionElement(s string) bool {
+ if len(s) < 2 || s[0] != 'v' || s[1] == '0' || s[1] == '1' && len(s) == 2 {
+ return false
+ }
+ for i := 1; i < len(s); i++ {
+ if s[i] < '0' || '9' < s[i] {
+ return false
+ }
+ }
+ return true
+}
+
+// moduleImportPath translates import paths found in go modules
+// back down to paths that can be resolved in ordinary builds.
+//
+// Define “new” code as code with a go.mod file in the same directory
+// or a parent directory. If an import in new code says x/y/v2/z but
+// x/y/v2/z does not exist and x/y/go.mod says “module x/y/v2”,
+// then go build will read the import as x/y/z instead.
+// See golang.org/issue/25069.
+func moduleImportPath(path, parentPath, parentDir, parentRoot string) (found string) {
+ if parentRoot == "" {
+ return path
+ }
+
+ // If there are no vN elements in path, leave it alone.
+ // (The code below would do the same, but only after
+ // some other file system accesses that we can avoid
+ // here by returning early.)
+ if i, _ := findVersionElement(path); i < 0 {
+ return path
+ }
+
+ dir, root := dirAndRoot(parentPath, parentDir, parentRoot)
+
+ // Consider dir and parents, up to and including root.
+ for i := len(dir); i >= len(root); i-- {
+ if i < len(dir) && dir[i] != filepath.Separator {
+ continue
+ }
+ if goModPath(dir[:i]) != "" {
+ goto HaveGoMod
+ }
+ }
+ // This code is not in a tree with a go.mod,
+ // so apply no changes to the path.
+ return path
+
+HaveGoMod:
+ // This import is in a tree with a go.mod.
+ // Allow it to refer to code in GOPATH/src/x/y/z as x/y/v2/z
+ // if GOPATH/src/x/y/go.mod says module "x/y/v2",
+
+ // If x/y/v2/z exists, use it unmodified.
+ if bp, _ := cfg.BuildContext.Import(path, "", build.IgnoreVendor); bp.Dir != "" {
+ return path
+ }
+
+ // Otherwise look for a go.mod supplying a version element.
+ // Some version-like elements may appear in paths but not
+ // be module versions; we skip over those to look for module
+ // versions. For example the module m/v2 might have a
+ // package m/v2/api/v1/foo.
+ limit := len(path)
+ for limit > 0 {
+ i, j := findVersionElement(path[:limit])
+ if i < 0 {
+ return path
+ }
+ if bp, _ := cfg.BuildContext.Import(path[:i], "", build.IgnoreVendor); bp.Dir != "" {
+ if mpath := goModPath(bp.Dir); mpath != "" {
+ // Found a valid go.mod file, so we're stopping the search.
+ // If the path is m/v2/p and we found m/go.mod that says
+ // "module m/v2", then we return "m/p".
+ if mpath == path[:j] {
+ return path[:i] + path[j:]
+ }
+ // Otherwise just return the original path.
+ // We didn't find anything worth rewriting,
+ // and the go.mod indicates that we should
+ // not consider parent directories.
+ return path
+ }
+ }
+ limit = i
+ }
+ return path
+}
+
+// hasGoFiles reports whether dir contains any files with names ending in .go.
+// For a vendor check we must exclude directories that contain no .go files.
+// Otherwise it is not possible to vendor just a/b/c and still import the
+// non-vendored a/b. See golang.org/issue/13832.
+func hasGoFiles(dir string) bool {
+ files, _ := os.ReadDir(dir)
+ for _, f := range files {
+ if !f.IsDir() && strings.HasSuffix(f.Name(), ".go") {
+ return true
+ }
+ }
+ return false
+}
+
+// reusePackage reuses package p to satisfy the import at the top
+// of the import stack stk. If this use causes an import loop,
+// reusePackage updates p's error information to record the loop.
+func reusePackage(p *Package, stk *ImportStack) *Package {
+ // We use p.Internal.Imports==nil to detect a package that
+ // is in the midst of its own loadPackage call
+ // (all the recursion below happens before p.Internal.Imports gets set).
+ if p.Internal.Imports == nil {
+ if p.Error == nil {
+ p.Error = &PackageError{
+ ImportStack: stk.Copy(),
+ Err: errors.New("import cycle not allowed"),
+ IsImportCycle: true,
+ }
+ } else if !p.Error.IsImportCycle {
+ // If the error is already set, but it does not indicate that
+ // we are in an import cycle, set IsImportCycle so that we don't
+ // end up stuck in a loop down the road.
+ p.Error.IsImportCycle = true
+ }
+ p.Incomplete = true
+ }
+ // Don't rewrite the import stack in the error if we have an import cycle.
+ // If we do, we'll lose the path that describes the cycle.
+ if p.Error != nil && !p.Error.IsImportCycle && stk.shorterThan(p.Error.ImportStack) {
+ p.Error.ImportStack = stk.Copy()
+ }
+ return p
+}
+
+// disallowInternal checks that srcDir (containing package importerPath, if non-empty)
+// is allowed to import p.
+// If the import is allowed, disallowInternal returns the original package p.
+// If not, it returns a new package containing just an appropriate error.
+func disallowInternal(srcDir string, importer *Package, importerPath string, p *Package, stk *ImportStack) *Package {
+ // golang.org/s/go14internal:
+ // An import of a path containing the element “internal”
+ // is disallowed if the importing code is outside the tree
+ // rooted at the parent of the “internal” directory.
+
+ // There was an error loading the package; stop here.
+ if p.Error != nil {
+ return p
+ }
+
+ // The generated 'testmain' package is allowed to access testing/internal/...,
+ // as if it were generated into the testing directory tree
+ // (it's actually in a temporary directory outside any Go tree).
+ // This cleans up a former kludge in passing functionality to the testing package.
+ if str.HasPathPrefix(p.ImportPath, "testing/internal") && importerPath == "testmain" {
+ return p
+ }
+
+ // We can't check standard packages with gccgo.
+ if cfg.BuildContext.Compiler == "gccgo" && p.Standard {
+ return p
+ }
+
+ // The sort package depends on internal/reflectlite, but during bootstrap
+ // the path rewriting causes the normal internal checks to fail.
+ // Instead, just ignore the internal rules during bootstrap.
+ if p.Standard && strings.HasPrefix(importerPath, "bootstrap/") {
+ return p
+ }
+
+ // importerPath is empty: we started
+ // with a name given on the command line, not an
+ // import. Anything listed on the command line is fine.
+ if importerPath == "" {
+ return p
+ }
+
+ // Check for "internal" element: three cases depending on begin of string and/or end of string.
+ i, ok := findInternal(p.ImportPath)
+ if !ok {
+ return p
+ }
+
+ // Internal is present.
+ // Map import path back to directory corresponding to parent of internal.
+ if i > 0 {
+ i-- // rewind over slash in ".../internal"
+ }
+
+ if p.Module == nil {
+ parent := p.Dir[:i+len(p.Dir)-len(p.ImportPath)]
+
+ if str.HasFilePathPrefix(filepath.Clean(srcDir), filepath.Clean(parent)) {
+ return p
+ }
+
+ // Look for symlinks before reporting error.
+ srcDir = expandPath(srcDir)
+ parent = expandPath(parent)
+ if str.HasFilePathPrefix(filepath.Clean(srcDir), filepath.Clean(parent)) {
+ return p
+ }
+ } else {
+ // p is in a module, so make it available based on the importer's import path instead
+ // of the file path (https://golang.org/issue/23970).
+ if importer.Internal.CmdlineFiles {
+ // The importer is a list of command-line files.
+ // Pretend that the import path is the import path of the
+ // directory containing them.
+ // If the directory is outside the main module, this will resolve to ".",
+ // which is not a prefix of any valid module.
+ importerPath = modload.DirImportPath(importer.Dir)
+ }
+ parentOfInternal := p.ImportPath[:i]
+ if str.HasPathPrefix(importerPath, parentOfInternal) {
+ return p
+ }
+ }
+
+ // Internal is present, and srcDir is outside parent's tree. Not allowed.
+ perr := *p
+ perr.Error = &PackageError{
+ alwaysPrintStack: true,
+ ImportStack: stk.Copy(),
+ Err: ImportErrorf(p.ImportPath, "use of internal package "+p.ImportPath+" not allowed"),
+ }
+ perr.Incomplete = true
+ return &perr
+}
+
+// findInternal looks for the final "internal" path element in the given import path.
+// If there isn't one, findInternal returns ok=false.
+// Otherwise, findInternal returns ok=true and the index of the "internal".
+func findInternal(path string) (index int, ok bool) {
+ // Three cases, depending on internal at start/end of string or not.
+ // The order matters: we must return the index of the final element,
+ // because the final one produces the most restrictive requirement
+ // on the importer.
+ switch {
+ case strings.HasSuffix(path, "/internal"):
+ return len(path) - len("internal"), true
+ case strings.Contains(path, "/internal/"):
+ return strings.LastIndex(path, "/internal/") + 1, true
+ case path == "internal", strings.HasPrefix(path, "internal/"):
+ return 0, true
+ }
+ return 0, false
+}
+
+// disallowVendor checks that srcDir is allowed to import p as path.
+// If the import is allowed, disallowVendor returns the original package p.
+// If not, it returns a new package containing just an appropriate error.
+func disallowVendor(srcDir string, path string, importerPath string, p *Package, stk *ImportStack) *Package {
+ // If the importerPath is empty, we started
+ // with a name given on the command line, not an
+ // import. Anything listed on the command line is fine.
+ if importerPath == "" {
+ return p
+ }
+
+ if perr := disallowVendorVisibility(srcDir, p, importerPath, stk); perr != p {
+ return perr
+ }
+
+ // Paths like x/vendor/y must be imported as y, never as x/vendor/y.
+ if i, ok := FindVendor(path); ok {
+ perr := *p
+ perr.Error = &PackageError{
+ ImportStack: stk.Copy(),
+ Err: ImportErrorf(path, "%s must be imported as %s", path, path[i+len("vendor/"):]),
+ }
+ perr.Incomplete = true
+ return &perr
+ }
+
+ return p
+}
+
+// disallowVendorVisibility checks that srcDir is allowed to import p.
+// The rules are the same as for /internal/ except that a path ending in /vendor
+// is not subject to the rules, only subdirectories of vendor.
+// This allows people to have packages and commands named vendor,
+// for maximal compatibility with existing source trees.
+func disallowVendorVisibility(srcDir string, p *Package, importerPath string, stk *ImportStack) *Package {
+ // The stack does not include p.ImportPath.
+ // If there's nothing on the stack, we started
+ // with a name given on the command line, not an
+ // import. Anything listed on the command line is fine.
+ if importerPath == "" {
+ return p
+ }
+
+ // Check for "vendor" element.
+ i, ok := FindVendor(p.ImportPath)
+ if !ok {
+ return p
+ }
+
+ // Vendor is present.
+ // Map import path back to directory corresponding to parent of vendor.
+ if i > 0 {
+ i-- // rewind over slash in ".../vendor"
+ }
+ truncateTo := i + len(p.Dir) - len(p.ImportPath)
+ if truncateTo < 0 || len(p.Dir) < truncateTo {
+ return p
+ }
+ parent := p.Dir[:truncateTo]
+ if str.HasFilePathPrefix(filepath.Clean(srcDir), filepath.Clean(parent)) {
+ return p
+ }
+
+ // Look for symlinks before reporting error.
+ srcDir = expandPath(srcDir)
+ parent = expandPath(parent)
+ if str.HasFilePathPrefix(filepath.Clean(srcDir), filepath.Clean(parent)) {
+ return p
+ }
+
+ // Vendor is present, and srcDir is outside parent's tree. Not allowed.
+ perr := *p
+ perr.Error = &PackageError{
+ ImportStack: stk.Copy(),
+ Err: errors.New("use of vendored package not allowed"),
+ }
+ perr.Incomplete = true
+ return &perr
+}
+
+// FindVendor looks for the last non-terminating "vendor" path element in the given import path.
+// If there isn't one, FindVendor returns ok=false.
+// Otherwise, FindVendor returns ok=true and the index of the "vendor".
+//
+// Note that terminating "vendor" elements don't count: "x/vendor" is its own package,
+// not the vendored copy of an import "" (the empty import path).
+// This will allow people to have packages or commands named vendor.
+// This may help reduce breakage, or it may just be confusing. We'll see.
+func FindVendor(path string) (index int, ok bool) {
+ // Two cases, depending on internal at start of string or not.
+ // The order matters: we must return the index of the final element,
+ // because the final one is where the effective import path starts.
+ switch {
+ case strings.Contains(path, "/vendor/"):
+ return strings.LastIndex(path, "/vendor/") + 1, true
+ case strings.HasPrefix(path, "vendor/"):
+ return 0, true
+ }
+ return 0, false
+}
+
+type TargetDir int
+
+const (
+ ToTool TargetDir = iota // to GOROOT/pkg/tool (default for cmd/*)
+ ToBin // to bin dir inside package root (default for non-cmd/*)
+ StalePath // an old import path; fail to build
+)
+
+// InstallTargetDir reports the target directory for installing the command p.
+func InstallTargetDir(p *Package) TargetDir {
+ if strings.HasPrefix(p.ImportPath, "code.google.com/p/go.tools/cmd/") {
+ return StalePath
+ }
+ if p.Goroot && strings.HasPrefix(p.ImportPath, "cmd/") && p.Name == "main" {
+ switch p.ImportPath {
+ case "cmd/go", "cmd/gofmt":
+ return ToBin
+ }
+ return ToTool
+ }
+ return ToBin
+}
+
+var cgoExclude = map[string]bool{
+ "runtime/cgo": true,
+}
+
+var cgoSyscallExclude = map[string]bool{
+ "runtime/cgo": true,
+ "runtime/race": true,
+ "runtime/msan": true,
+}
+
+var foldPath = make(map[string]string)
+
+// exeFromImportPath returns an executable name
+// for a package using the import path.
+//
+// The executable name is the last element of the import path.
+// In module-aware mode, an additional rule is used on import paths
+// consisting of two or more path elements. If the last element is
+// a vN path element specifying the major version, then the
+// second last element of the import path is used instead.
+func (p *Package) exeFromImportPath() string {
+ _, elem := pathpkg.Split(p.ImportPath)
+ if cfg.ModulesEnabled {
+ // If this is example.com/mycmd/v2, it's more useful to
+ // install it as mycmd than as v2. See golang.org/issue/24667.
+ if elem != p.ImportPath && isVersionElement(elem) {
+ _, elem = pathpkg.Split(pathpkg.Dir(p.ImportPath))
+ }
+ }
+ return elem
+}
+
+// exeFromFiles returns an executable name for a package
+// using the first element in GoFiles or CgoFiles collections without the prefix.
+//
+// Returns empty string in case of empty collection.
+func (p *Package) exeFromFiles() string {
+ var src string
+ if len(p.GoFiles) > 0 {
+ src = p.GoFiles[0]
+ } else if len(p.CgoFiles) > 0 {
+ src = p.CgoFiles[0]
+ } else {
+ return ""
+ }
+ _, elem := filepath.Split(src)
+ return elem[:len(elem)-len(".go")]
+}
+
+// DefaultExecName returns the default executable name for a package
+func (p *Package) DefaultExecName() string {
+ if p.Internal.CmdlineFiles {
+ return p.exeFromFiles()
+ }
+ return p.exeFromImportPath()
+}
+
+// load populates p using information from bp, err, which should
+// be the result of calling build.Context.Import.
+// stk contains the import stack, not including path itself.
+func (p *Package) load(ctx context.Context, path string, stk *ImportStack, importPos []token.Position, bp *build.Package, err error) {
+ p.copyBuild(bp)
+
+ // The localPrefix is the path we interpret ./ imports relative to.
+ // Synthesized main packages sometimes override this.
+ if p.Internal.Local {
+ p.Internal.LocalPrefix = dirToImportPath(p.Dir)
+ }
+
+ // setError sets p.Error if it hasn't already been set. We may proceed
+ // after encountering some errors so that 'go list -e' has more complete
+ // output. If there's more than one error, we should report the first.
+ setError := func(err error) {
+ if p.Error == nil {
+ p.Error = &PackageError{
+ ImportStack: stk.Copy(),
+ Err: err,
+ }
+
+ // Add the importer's position information if the import position exists, and
+ // the current package being examined is the importer.
+ // If we have not yet accepted package p onto the import stack,
+ // then the cause of the error is not within p itself: the error
+ // must be either in an explicit command-line argument,
+ // or on the importer side (indicated by a non-empty importPos).
+ if path != stk.Top() && len(importPos) > 0 {
+ p.Error.setPos(importPos)
+ }
+ }
+ }
+
+ if err != nil {
+ p.Incomplete = true
+ p.setLoadPackageDataError(err, path, stk, importPos)
+ }
+
+ useBindir := p.Name == "main"
+ if !p.Standard {
+ switch cfg.BuildBuildmode {
+ case "c-archive", "c-shared", "plugin":
+ useBindir = false
+ }
+ }
+
+ if useBindir {
+ // Report an error when the old code.google.com/p/go.tools paths are used.
+ if InstallTargetDir(p) == StalePath {
+ // TODO(matloob): remove this branch, and StalePath itself. code.google.com/p/go is so
+ // old, even this code checking for it is stale now!
+ newPath := strings.Replace(p.ImportPath, "code.google.com/p/go.", "golang.org/x/", 1)
+ e := ImportErrorf(p.ImportPath, "the %v command has moved; use %v instead.", p.ImportPath, newPath)
+ setError(e)
+ return
+ }
+ elem := p.DefaultExecName()
+ full := cfg.BuildContext.GOOS + "_" + cfg.BuildContext.GOARCH + "/" + elem
+ if cfg.BuildContext.GOOS != base.ToolGOOS || cfg.BuildContext.GOARCH != base.ToolGOARCH {
+ // Install cross-compiled binaries to subdirectories of bin.
+ elem = full
+ }
+ if p.Internal.Build.BinDir == "" && cfg.ModulesEnabled {
+ p.Internal.Build.BinDir = modload.BinDir()
+ }
+ if p.Internal.Build.BinDir != "" {
+ // Install to GOBIN or bin of GOPATH entry.
+ p.Target = filepath.Join(p.Internal.Build.BinDir, elem)
+ if !p.Goroot && strings.Contains(elem, "/") && cfg.GOBIN != "" {
+ // Do not create $GOBIN/goos_goarch/elem.
+ p.Target = ""
+ p.Internal.GobinSubdir = true
+ }
+ }
+ if InstallTargetDir(p) == ToTool {
+ // This is for 'go tool'.
+ // Override all the usual logic and force it into the tool directory.
+ if cfg.BuildToolchainName == "gccgo" {
+ p.Target = filepath.Join(base.ToolDir, elem)
+ } else {
+ p.Target = filepath.Join(cfg.GOROOTpkg, "tool", full)
+ }
+ }
+ if p.Target != "" && cfg.BuildContext.GOOS == "windows" {
+ p.Target += ".exe"
+ }
+ } else if p.Internal.Local {
+ // Local import turned into absolute path.
+ // No permanent install target.
+ p.Target = ""
+ } else {
+ p.Target = p.Internal.Build.PkgObj
+ if cfg.BuildLinkshared && p.Target != "" {
+ // TODO(bcmills): The reliance on p.Target implies that -linkshared does
+ // not work for any package that lacks a Target — such as a non-main
+ // package in module mode. We should probably fix that.
+ shlibnamefile := p.Target[:len(p.Target)-2] + ".shlibname"
+ shlib, err := os.ReadFile(shlibnamefile)
+ if err != nil && !os.IsNotExist(err) {
+ base.Fatalf("reading shlibname: %v", err)
+ }
+ if err == nil {
+ libname := strings.TrimSpace(string(shlib))
+ if cfg.BuildContext.Compiler == "gccgo" {
+ p.Shlib = filepath.Join(p.Internal.Build.PkgTargetRoot, "shlibs", libname)
+ } else {
+ p.Shlib = filepath.Join(p.Internal.Build.PkgTargetRoot, libname)
+ }
+ }
+ }
+ }
+
+ // Build augmented import list to add implicit dependencies.
+ // Be careful not to add imports twice, just to avoid confusion.
+ importPaths := p.Imports
+ addImport := func(path string, forCompiler bool) {
+ for _, p := range importPaths {
+ if path == p {
+ return
+ }
+ }
+ importPaths = append(importPaths, path)
+ if forCompiler {
+ p.Internal.CompiledImports = append(p.Internal.CompiledImports, path)
+ }
+ }
+
+ // Cgo translation adds imports of "unsafe", "runtime/cgo" and "syscall",
+ // except for certain packages, to avoid circular dependencies.
+ if p.UsesCgo() {
+ addImport("unsafe", true)
+ }
+ if p.UsesCgo() && (!p.Standard || !cgoExclude[p.ImportPath]) && cfg.BuildContext.Compiler != "gccgo" {
+ addImport("runtime/cgo", true)
+ }
+ if p.UsesCgo() && (!p.Standard || !cgoSyscallExclude[p.ImportPath]) {
+ addImport("syscall", true)
+ }
+
+ // SWIG adds imports of some standard packages.
+ if p.UsesSwig() {
+ addImport("unsafe", true)
+ if cfg.BuildContext.Compiler != "gccgo" {
+ addImport("runtime/cgo", true)
+ }
+ addImport("syscall", true)
+ addImport("sync", true)
+
+ // TODO: The .swig and .swigcxx files can use
+ // %go_import directives to import other packages.
+ }
+
+ // The linker loads implicit dependencies.
+ if p.Name == "main" && !p.Internal.ForceLibrary {
+ for _, dep := range LinkerDeps(p) {
+ addImport(dep, false)
+ }
+ }
+
+ // Check for case-insensitive collisions of import paths.
+ fold := str.ToFold(p.ImportPath)
+ if other := foldPath[fold]; other == "" {
+ foldPath[fold] = p.ImportPath
+ } else if other != p.ImportPath {
+ setError(ImportErrorf(p.ImportPath, "case-insensitive import collision: %q and %q", p.ImportPath, other))
+ return
+ }
+
+ if !SafeArg(p.ImportPath) {
+ setError(ImportErrorf(p.ImportPath, "invalid import path %q", p.ImportPath))
+ return
+ }
+
+ // Errors after this point are caused by this package, not the importing
+ // package. Pushing the path here prevents us from reporting the error
+ // with the position of the import declaration.
+ stk.Push(path)
+ defer stk.Pop()
+
+ p.EmbedFiles, p.Internal.Embed, err = resolveEmbed(p.Dir, p.EmbedPatterns)
+ if err != nil {
+ p.Incomplete = true
+ setError(err)
+ embedErr := err.(*EmbedError)
+ p.Error.setPos(p.Internal.Build.EmbedPatternPos[embedErr.Pattern])
+ }
+
+ // Check for case-insensitive collision of input files.
+ // To avoid problems on case-insensitive files, we reject any package
+ // where two different input files have equal names under a case-insensitive
+ // comparison.
+ inputs := p.AllFiles()
+ f1, f2 := str.FoldDup(inputs)
+ if f1 != "" {
+ setError(fmt.Errorf("case-insensitive file name collision: %q and %q", f1, f2))
+ return
+ }
+
+ // If first letter of input file is ASCII, it must be alphanumeric.
+ // This avoids files turning into flags when invoking commands,
+ // and other problems we haven't thought of yet.
+ // Also, _cgo_ files must be generated by us, not supplied.
+ // They are allowed to have //go:cgo_ldflag directives.
+ // The directory scan ignores files beginning with _,
+ // so we shouldn't see any _cgo_ files anyway, but just be safe.
+ for _, file := range inputs {
+ if !SafeArg(file) || strings.HasPrefix(file, "_cgo_") {
+ setError(fmt.Errorf("invalid input file name %q", file))
+ return
+ }
+ }
+ if name := pathpkg.Base(p.ImportPath); !SafeArg(name) {
+ setError(fmt.Errorf("invalid input directory name %q", name))
+ return
+ }
+
+ // Build list of imported packages and full dependency list.
+ imports := make([]*Package, 0, len(p.Imports))
+ for i, path := range importPaths {
+ if path == "C" {
+ continue
+ }
+ p1 := LoadImport(ctx, path, p.Dir, p, stk, p.Internal.Build.ImportPos[path], ResolveImport)
+
+ path = p1.ImportPath
+ importPaths[i] = path
+ if i < len(p.Imports) {
+ p.Imports[i] = path
+ }
+
+ imports = append(imports, p1)
+ if p1.Incomplete {
+ p.Incomplete = true
+ }
+ }
+ p.Internal.Imports = imports
+ p.collectDeps()
+
+ // unsafe is a fake package.
+ if p.Standard && (p.ImportPath == "unsafe" || cfg.BuildContext.Compiler == "gccgo") {
+ p.Target = ""
+ }
+
+ // If cgo is not enabled, ignore cgo supporting sources
+ // just as we ignore go files containing import "C".
+ if !cfg.BuildContext.CgoEnabled {
+ p.CFiles = nil
+ p.CXXFiles = nil
+ p.MFiles = nil
+ p.SwigFiles = nil
+ p.SwigCXXFiles = nil
+ // Note that SFiles are okay (they go to the Go assembler)
+ // and HFiles are okay (they might be used by the SFiles).
+ // Also Sysofiles are okay (they might not contain object
+ // code; see issue #16050).
+ }
+
+ // The gc toolchain only permits C source files with cgo or SWIG.
+ if len(p.CFiles) > 0 && !p.UsesCgo() && !p.UsesSwig() && cfg.BuildContext.Compiler == "gc" {
+ setError(fmt.Errorf("C source files not allowed when not using cgo or SWIG: %s", strings.Join(p.CFiles, " ")))
+ return
+ }
+
+ // C++, Objective-C, and Fortran source files are permitted only with cgo or SWIG,
+ // regardless of toolchain.
+ if len(p.CXXFiles) > 0 && !p.UsesCgo() && !p.UsesSwig() {
+ setError(fmt.Errorf("C++ source files not allowed when not using cgo or SWIG: %s", strings.Join(p.CXXFiles, " ")))
+ return
+ }
+ if len(p.MFiles) > 0 && !p.UsesCgo() && !p.UsesSwig() {
+ setError(fmt.Errorf("Objective-C source files not allowed when not using cgo or SWIG: %s", strings.Join(p.MFiles, " ")))
+ return
+ }
+ if len(p.FFiles) > 0 && !p.UsesCgo() && !p.UsesSwig() {
+ setError(fmt.Errorf("Fortran source files not allowed when not using cgo or SWIG: %s", strings.Join(p.FFiles, " ")))
+ return
+ }
+
+ if cfg.ModulesEnabled && p.Error == nil {
+ mainPath := p.ImportPath
+ if p.Internal.CmdlineFiles {
+ mainPath = "command-line-arguments"
+ }
+ p.Module = modload.PackageModuleInfo(mainPath)
+ if p.Name == "main" && len(p.DepsErrors) == 0 {
+ p.Internal.BuildInfo = modload.PackageBuildInfo(mainPath, p.Deps)
+ }
+ }
+}
+
+// An EmbedError indicates a problem with a go:embed directive.
+type EmbedError struct {
+ Pattern string
+ Err error
+}
+
+func (e *EmbedError) Error() string {
+ return fmt.Sprintf("pattern %s: %v", e.Pattern, e.Err)
+}
+
+func (e *EmbedError) Unwrap() error {
+ return e.Err
+}
+
+// ResolveEmbed resolves //go:embed patterns and returns only the file list.
+// For use by go mod vendor to find embedded files it should copy into the
+// vendor directory.
+// TODO(#42504): Once go mod vendor uses load.PackagesAndErrors, just
+// call (*Package).ResolveEmbed
+func ResolveEmbed(dir string, patterns []string) ([]string, error) {
+ files, _, err := resolveEmbed(dir, patterns)
+ return files, err
+}
+
+// resolveEmbed resolves //go:embed patterns to precise file lists.
+// It sets files to the list of unique files matched (for go list),
+// and it sets pmap to the more precise mapping from
+// patterns to files.
+func resolveEmbed(pkgdir string, patterns []string) (files []string, pmap map[string][]string, err error) {
+ var pattern string
+ defer func() {
+ if err != nil {
+ err = &EmbedError{
+ Pattern: pattern,
+ Err: err,
+ }
+ }
+ }()
+
+ // TODO(rsc): All these messages need position information for better error reports.
+ pmap = make(map[string][]string)
+ have := make(map[string]int)
+ dirOK := make(map[string]bool)
+ pid := 0 // pattern ID, to allow reuse of have map
+ for _, pattern = range patterns {
+ pid++
+
+ // Check pattern is valid for //go:embed.
+ if _, err := path.Match(pattern, ""); err != nil || !validEmbedPattern(pattern) {
+ return nil, nil, fmt.Errorf("invalid pattern syntax")
+ }
+
+ // Glob to find matches.
+ match, err := fsys.Glob(pkgdir + string(filepath.Separator) + filepath.FromSlash(pattern))
+ if err != nil {
+ return nil, nil, err
+ }
+
+ // Filter list of matches down to the ones that will still exist when
+ // the directory is packaged up as a module. (If p.Dir is in the module cache,
+ // only those files exist already, but if p.Dir is in the current module,
+ // then there may be other things lying around, like symbolic links or .git directories.)
+ var list []string
+ for _, file := range match {
+ rel := filepath.ToSlash(file[len(pkgdir)+1:]) // file, relative to p.Dir
+
+ what := "file"
+ info, err := fsys.Lstat(file)
+ if err != nil {
+ return nil, nil, err
+ }
+ if info.IsDir() {
+ what = "directory"
+ }
+
+ // Check that directories along path do not begin a new module
+ // (do not contain a go.mod).
+ for dir := file; len(dir) > len(pkgdir)+1 && !dirOK[dir]; dir = filepath.Dir(dir) {
+ if _, err := fsys.Stat(filepath.Join(dir, "go.mod")); err == nil {
+ return nil, nil, fmt.Errorf("cannot embed %s %s: in different module", what, rel)
+ }
+ if dir != file {
+ if info, err := fsys.Lstat(dir); err == nil && !info.IsDir() {
+ return nil, nil, fmt.Errorf("cannot embed %s %s: in non-directory %s", what, rel, dir[len(pkgdir)+1:])
+ }
+ }
+ dirOK[dir] = true
+ if elem := filepath.Base(dir); isBadEmbedName(elem) {
+ if dir == file {
+ return nil, nil, fmt.Errorf("cannot embed %s %s: invalid name %s", what, rel, elem)
+ } else {
+ return nil, nil, fmt.Errorf("cannot embed %s %s: in invalid directory %s", what, rel, elem)
+ }
+ }
+ }
+
+ switch {
+ default:
+ return nil, nil, fmt.Errorf("cannot embed irregular file %s", rel)
+
+ case info.Mode().IsRegular():
+ if have[rel] != pid {
+ have[rel] = pid
+ list = append(list, rel)
+ }
+
+ case info.IsDir():
+ // Gather all files in the named directory, stopping at module boundaries
+ // and ignoring files that wouldn't be packaged into a module.
+ count := 0
+ err := fsys.Walk(file, func(path string, info os.FileInfo, err error) error {
+ if err != nil {
+ return err
+ }
+ rel := filepath.ToSlash(path[len(pkgdir)+1:])
+ name := info.Name()
+ if path != file && (isBadEmbedName(name) || name[0] == '.' || name[0] == '_') {
+ // Ignore bad names, assuming they won't go into modules.
+ // Also avoid hidden files that user may not know about.
+ // See golang.org/issue/42328.
+ if info.IsDir() {
+ return fs.SkipDir
+ }
+ return nil
+ }
+ if info.IsDir() {
+ if _, err := fsys.Stat(filepath.Join(path, "go.mod")); err == nil {
+ return filepath.SkipDir
+ }
+ return nil
+ }
+ if !info.Mode().IsRegular() {
+ return nil
+ }
+ count++
+ if have[rel] != pid {
+ have[rel] = pid
+ list = append(list, rel)
+ }
+ return nil
+ })
+ if err != nil {
+ return nil, nil, err
+ }
+ if count == 0 {
+ return nil, nil, fmt.Errorf("cannot embed directory %s: contains no embeddable files", rel)
+ }
+ }
+ }
+
+ if len(list) == 0 {
+ return nil, nil, fmt.Errorf("no matching files found")
+ }
+ sort.Strings(list)
+ pmap[pattern] = list
+ }
+
+ for file := range have {
+ files = append(files, file)
+ }
+ sort.Strings(files)
+ return files, pmap, nil
+}
+
+func validEmbedPattern(pattern string) bool {
+ return pattern != "." && fs.ValidPath(pattern)
+}
+
+// isBadEmbedName reports whether name is the base name of a file that
+// can't or won't be included in modules and therefore shouldn't be treated
+// as existing for embedding.
+func isBadEmbedName(name string) bool {
+ if err := module.CheckFilePath(name); err != nil {
+ return true
+ }
+ switch name {
+ // Empty string should be impossible but make it bad.
+ case "":
+ return true
+ // Version control directories won't be present in module.
+ case ".bzr", ".hg", ".git", ".svn":
+ return true
+ }
+ return false
+}
+
+// collectDeps populates p.Deps and p.DepsErrors by iterating over
+// p.Internal.Imports.
+//
+// TODO(jayconrod): collectDeps iterates over transitive imports for every
+// package. We should only need to visit direct imports.
+func (p *Package) collectDeps() {
+ deps := make(map[string]*Package)
+ var q []*Package
+ q = append(q, p.Internal.Imports...)
+ for i := 0; i < len(q); i++ {
+ p1 := q[i]
+ path := p1.ImportPath
+ // The same import path could produce an error or not,
+ // depending on what tries to import it.
+ // Prefer to record entries with errors, so we can report them.
+ p0 := deps[path]
+ if p0 == nil || p1.Error != nil && (p0.Error == nil || len(p0.Error.ImportStack) > len(p1.Error.ImportStack)) {
+ deps[path] = p1
+ for _, p2 := range p1.Internal.Imports {
+ if deps[p2.ImportPath] != p2 {
+ q = append(q, p2)
+ }
+ }
+ }
+ }
+
+ p.Deps = make([]string, 0, len(deps))
+ for dep := range deps {
+ p.Deps = append(p.Deps, dep)
+ }
+ sort.Strings(p.Deps)
+ for _, dep := range p.Deps {
+ p1 := deps[dep]
+ if p1 == nil {
+ panic("impossible: missing entry in package cache for " + dep + " imported by " + p.ImportPath)
+ }
+ if p1.Error != nil {
+ p.DepsErrors = append(p.DepsErrors, p1.Error)
+ }
+ }
+}
+
+// SafeArg reports whether arg is a "safe" command-line argument,
+// meaning that when it appears in a command-line, it probably
+// doesn't have some special meaning other than its own name.
+// Obviously args beginning with - are not safe (they look like flags).
+// Less obviously, args beginning with @ are not safe (they look like
+// GNU binutils flagfile specifiers, sometimes called "response files").
+// To be conservative, we reject almost any arg beginning with non-alphanumeric ASCII.
+// We accept leading . _ and / as likely in file system paths.
+// There is a copy of this function in cmd/compile/internal/gc/noder.go.
+func SafeArg(name string) bool {
+ if name == "" {
+ return false
+ }
+ c := name[0]
+ return '0' <= c && c <= '9' || 'A' <= c && c <= 'Z' || 'a' <= c && c <= 'z' || c == '.' || c == '_' || c == '/' || c >= utf8.RuneSelf
+}
+
+// LinkerDeps returns the list of linker-induced dependencies for main package p.
+func LinkerDeps(p *Package) []string {
+ // Everything links runtime.
+ deps := []string{"runtime"}
+
+ // External linking mode forces an import of runtime/cgo.
+ if externalLinkingForced(p) && cfg.BuildContext.Compiler != "gccgo" {
+ deps = append(deps, "runtime/cgo")
+ }
+ // On ARM with GOARM=5, it forces an import of math, for soft floating point.
+ if cfg.Goarch == "arm" {
+ deps = append(deps, "math")
+ }
+ // Using the race detector forces an import of runtime/race.
+ if cfg.BuildRace {
+ deps = append(deps, "runtime/race")
+ }
+ // Using memory sanitizer forces an import of runtime/msan.
+ if cfg.BuildMSan {
+ deps = append(deps, "runtime/msan")
+ }
+
+ return deps
+}
+
+// externalLinkingForced reports whether external linking is being
+// forced even for programs that do not use cgo.
+func externalLinkingForced(p *Package) bool {
+ if !cfg.BuildContext.CgoEnabled {
+ return false
+ }
+
+ // Some targets must use external linking even inside GOROOT.
+ switch cfg.BuildContext.GOOS {
+ case "android":
+ if cfg.BuildContext.GOARCH != "arm64" {
+ return true
+ }
+ case "ios":
+ return true
+ }
+
+ // Currently build modes c-shared, pie (on systems that do not
+ // support PIE with internal linking mode (currently all
+ // systems: issue #18968)), plugin, and -linkshared force
+ // external linking mode, as of course does
+ // -ldflags=-linkmode=external. External linking mode forces
+ // an import of runtime/cgo.
+ // If there are multiple -linkmode options, the last one wins.
+ pieCgo := cfg.BuildBuildmode == "pie" && !sys.InternalLinkPIESupported(cfg.BuildContext.GOOS, cfg.BuildContext.GOARCH)
+ linkmodeExternal := false
+ if p != nil {
+ ldflags := BuildLdflags.For(p)
+ for i := len(ldflags) - 1; i >= 0; i-- {
+ a := ldflags[i]
+ if a == "-linkmode=external" ||
+ a == "-linkmode" && i+1 < len(ldflags) && ldflags[i+1] == "external" {
+ linkmodeExternal = true
+ break
+ } else if a == "-linkmode=internal" ||
+ a == "-linkmode" && i+1 < len(ldflags) && ldflags[i+1] == "internal" {
+ break
+ }
+ }
+ }
+
+ return cfg.BuildBuildmode == "c-shared" || cfg.BuildBuildmode == "plugin" || pieCgo || cfg.BuildLinkshared || linkmodeExternal
+}
+
+// mkAbs rewrites list, which must be paths relative to p.Dir,
+// into a sorted list of absolute paths. It edits list in place but for
+// convenience also returns list back to its caller.
+func (p *Package) mkAbs(list []string) []string {
+ for i, f := range list {
+ list[i] = filepath.Join(p.Dir, f)
+ }
+ sort.Strings(list)
+ return list
+}
+
+// InternalGoFiles returns the list of Go files being built for the package,
+// using absolute paths.
+func (p *Package) InternalGoFiles() []string {
+ return p.mkAbs(str.StringList(p.GoFiles, p.CgoFiles, p.TestGoFiles))
+}
+
+// InternalXGoFiles returns the list of Go files being built for the XTest package,
+// using absolute paths.
+func (p *Package) InternalXGoFiles() []string {
+ return p.mkAbs(p.XTestGoFiles)
+}
+
+// InternalGoFiles returns the list of all Go files possibly relevant for the package,
+// using absolute paths. "Possibly relevant" means that files are not excluded
+// due to build tags, but files with names beginning with . or _ are still excluded.
+func (p *Package) InternalAllGoFiles() []string {
+ return p.mkAbs(str.StringList(p.IgnoredGoFiles, p.GoFiles, p.CgoFiles, p.TestGoFiles, p.XTestGoFiles))
+}
+
+// usesSwig reports whether the package needs to run SWIG.
+func (p *Package) UsesSwig() bool {
+ return len(p.SwigFiles) > 0 || len(p.SwigCXXFiles) > 0
+}
+
+// usesCgo reports whether the package needs to run cgo
+func (p *Package) UsesCgo() bool {
+ return len(p.CgoFiles) > 0
+}
+
+// PackageList returns the list of packages in the dag rooted at roots
+// as visited in a depth-first post-order traversal.
+func PackageList(roots []*Package) []*Package {
+ seen := map[*Package]bool{}
+ all := []*Package{}
+ var walk func(*Package)
+ walk = func(p *Package) {
+ if seen[p] {
+ return
+ }
+ seen[p] = true
+ for _, p1 := range p.Internal.Imports {
+ walk(p1)
+ }
+ all = append(all, p)
+ }
+ for _, root := range roots {
+ walk(root)
+ }
+ return all
+}
+
+// TestPackageList returns the list of packages in the dag rooted at roots
+// as visited in a depth-first post-order traversal, including the test
+// imports of the roots. This ignores errors in test packages.
+func TestPackageList(ctx context.Context, roots []*Package) []*Package {
+ seen := map[*Package]bool{}
+ all := []*Package{}
+ var walk func(*Package)
+ walk = func(p *Package) {
+ if seen[p] {
+ return
+ }
+ seen[p] = true
+ for _, p1 := range p.Internal.Imports {
+ walk(p1)
+ }
+ all = append(all, p)
+ }
+ walkTest := func(root *Package, path string) {
+ var stk ImportStack
+ p1 := LoadImport(ctx, path, root.Dir, root, &stk, root.Internal.Build.TestImportPos[path], ResolveImport)
+ if p1.Error == nil {
+ walk(p1)
+ }
+ }
+ for _, root := range roots {
+ walk(root)
+ for _, path := range root.TestImports {
+ walkTest(root, path)
+ }
+ for _, path := range root.XTestImports {
+ walkTest(root, path)
+ }
+ }
+ return all
+}
+
+// LoadImportWithFlags loads the package with the given import path and
+// sets tool flags on that package. This function is useful loading implicit
+// dependencies (like sync/atomic for coverage).
+// TODO(jayconrod): delete this function and set flags automatically
+// in LoadImport instead.
+func LoadImportWithFlags(path, srcDir string, parent *Package, stk *ImportStack, importPos []token.Position, mode int) *Package {
+ p := LoadImport(context.TODO(), path, srcDir, parent, stk, importPos, mode)
+ setToolFlags(p)
+ return p
+}
+
+// ModResolveTests indicates whether calls to the module loader should also
+// resolve test dependencies of the requested packages.
+//
+// If ModResolveTests is true, then the module loader needs to resolve test
+// dependencies at the same time as packages; otherwise, the test dependencies
+// of those packages could be missing, and resolving those missing dependencies
+// could change the selected versions of modules that provide other packages.
+//
+// TODO(#40775): Change this from a global variable to an explicit function
+// argument where needed.
+var ModResolveTests bool
+
+// PackagesAndErrors returns the packages named by the command line arguments
+// 'patterns'. If a named package cannot be loaded, PackagesAndErrors returns
+// a *Package with the Error field describing the failure. If errors are found
+// loading imported packages, the DepsErrors field is set. The Incomplete field
+// may be set as well.
+//
+// To obtain a flat list of packages, use PackageList.
+// To report errors loading packages, use ReportPackageErrors.
+func PackagesAndErrors(ctx context.Context, patterns []string) []*Package {
+ ctx, span := trace.StartSpan(ctx, "load.PackagesAndErrors")
+ defer span.Done()
+
+ for _, p := range patterns {
+ // Listing is only supported with all patterns referring to either:
+ // - Files that are part of the same directory.
+ // - Explicit package paths or patterns.
+ if strings.HasSuffix(p, ".go") {
+ // We need to test whether the path is an actual Go file and not a
+ // package path or pattern ending in '.go' (see golang.org/issue/34653).
+ if fi, err := fsys.Stat(p); err == nil && !fi.IsDir() {
+ return []*Package{GoFilesPackage(ctx, patterns)}
+ }
+ }
+ }
+
+ var matches []*search.Match
+ if modload.Init(); cfg.ModulesEnabled {
+ loadOpts := modload.PackageOpts{
+ ResolveMissingImports: true,
+ LoadTests: ModResolveTests,
+ SilenceErrors: true,
+ }
+ matches, _ = modload.LoadPackages(ctx, loadOpts, patterns...)
+ } else {
+ matches = search.ImportPaths(patterns)
+ }
+
+ var (
+ pkgs []*Package
+ stk ImportStack
+ seenPkg = make(map[*Package]bool)
+ )
+
+ pre := newPreload()
+ defer pre.flush()
+ pre.preloadMatches(matches)
+
+ for _, m := range matches {
+ for _, pkg := range m.Pkgs {
+ if pkg == "" {
+ panic(fmt.Sprintf("ImportPaths returned empty package for pattern %s", m.Pattern()))
+ }
+ p := loadImport(ctx, pre, pkg, base.Cwd, nil, &stk, nil, 0)
+ p.Match = append(p.Match, m.Pattern())
+ p.Internal.CmdlinePkg = true
+ if m.IsLiteral() {
+ // Note: do not set = m.IsLiteral unconditionally
+ // because maybe we'll see p matching both
+ // a literal and also a non-literal pattern.
+ p.Internal.CmdlinePkgLiteral = true
+ }
+ if seenPkg[p] {
+ continue
+ }
+ seenPkg[p] = true
+ pkgs = append(pkgs, p)
+ }
+
+ if len(m.Errs) > 0 {
+ // In addition to any packages that were actually resolved from the
+ // pattern, there was some error in resolving the pattern itself.
+ // Report it as a synthetic package.
+ p := new(Package)
+ p.ImportPath = m.Pattern()
+ // Pass an empty ImportStack and nil importPos: the error arose from a pattern, not an import.
+ var stk ImportStack
+ var importPos []token.Position
+ p.setLoadPackageDataError(m.Errs[0], m.Pattern(), &stk, importPos)
+ p.Incomplete = true
+ p.Match = append(p.Match, m.Pattern())
+ p.Internal.CmdlinePkg = true
+ if m.IsLiteral() {
+ p.Internal.CmdlinePkgLiteral = true
+ }
+ pkgs = append(pkgs, p)
+ }
+ }
+
+ // Now that CmdlinePkg is set correctly,
+ // compute the effective flags for all loaded packages
+ // (not just the ones matching the patterns but also
+ // their dependencies).
+ setToolFlags(pkgs...)
+
+ return pkgs
+}
+
+// CheckPackageErrors prints errors encountered loading pkgs and their
+// dependencies, then exits with a non-zero status if any errors were found.
+func CheckPackageErrors(pkgs []*Package) {
+ printed := map[*PackageError]bool{}
+ for _, pkg := range pkgs {
+ if pkg.Error != nil {
+ base.Errorf("%v", pkg.Error)
+ printed[pkg.Error] = true
+ }
+ for _, err := range pkg.DepsErrors {
+ // Since these are errors in dependencies,
+ // the same error might show up multiple times,
+ // once in each package that depends on it.
+ // Only print each once.
+ if !printed[err] {
+ printed[err] = true
+ base.Errorf("%v", err)
+ }
+ }
+ }
+ base.ExitIfErrors()
+
+ // Check for duplicate loads of the same package.
+ // That should be impossible, but if it does happen then
+ // we end up trying to build the same package twice,
+ // usually in parallel overwriting the same files,
+ // which doesn't work very well.
+ seen := map[string]bool{}
+ reported := map[string]bool{}
+ for _, pkg := range PackageList(pkgs) {
+ if seen[pkg.ImportPath] && !reported[pkg.ImportPath] {
+ reported[pkg.ImportPath] = true
+ base.Errorf("internal error: duplicate loads of %s", pkg.ImportPath)
+ }
+ seen[pkg.ImportPath] = true
+ }
+ base.ExitIfErrors()
+}
+
+func setToolFlags(pkgs ...*Package) {
+ for _, p := range PackageList(pkgs) {
+ p.Internal.Asmflags = BuildAsmflags.For(p)
+ p.Internal.Gcflags = BuildGcflags.For(p)
+ p.Internal.Ldflags = BuildLdflags.For(p)
+ p.Internal.Gccgoflags = BuildGccgoflags.For(p)
+ }
+}
+
+// GoFilesPackage creates a package for building a collection of Go files
+// (typically named on the command line). The target is named p.a for
+// package p or named after the first Go file for package main.
+func GoFilesPackage(ctx context.Context, gofiles []string) *Package {
+ modload.Init()
+
+ for _, f := range gofiles {
+ if !strings.HasSuffix(f, ".go") {
+ pkg := new(Package)
+ pkg.Internal.Local = true
+ pkg.Internal.CmdlineFiles = true
+ pkg.Name = f
+ pkg.Error = &PackageError{
+ Err: fmt.Errorf("named files must be .go files: %s", pkg.Name),
+ }
+ return pkg
+ }
+ }
+
+ var stk ImportStack
+ ctxt := cfg.BuildContext
+ ctxt.UseAllFiles = true
+
+ // Synthesize fake "directory" that only shows the named files,
+ // to make it look like this is a standard package or
+ // command directory. So that local imports resolve
+ // consistently, the files must all be in the same directory.
+ var dirent []fs.FileInfo
+ var dir string
+ for _, file := range gofiles {
+ fi, err := fsys.Stat(file)
+ if err != nil {
+ base.Fatalf("%s", err)
+ }
+ if fi.IsDir() {
+ base.Fatalf("%s is a directory, should be a Go file", file)
+ }
+ dir1, _ := filepath.Split(file)
+ if dir1 == "" {
+ dir1 = "./"
+ }
+ if dir == "" {
+ dir = dir1
+ } else if dir != dir1 {
+ base.Fatalf("named files must all be in one directory; have %s and %s", dir, dir1)
+ }
+ dirent = append(dirent, fi)
+ }
+ ctxt.ReadDir = func(string) ([]fs.FileInfo, error) { return dirent, nil }
+
+ if cfg.ModulesEnabled {
+ modload.ImportFromFiles(ctx, gofiles)
+ }
+
+ var err error
+ if dir == "" {
+ dir = base.Cwd
+ }
+ dir, err = filepath.Abs(dir)
+ if err != nil {
+ base.Fatalf("%s", err)
+ }
+
+ bp, err := ctxt.ImportDir(dir, 0)
+ pkg := new(Package)
+ pkg.Internal.Local = true
+ pkg.Internal.CmdlineFiles = true
+ pkg.load(ctx, "command-line-arguments", &stk, nil, bp, err)
+ pkg.Internal.LocalPrefix = dirToImportPath(dir)
+ pkg.ImportPath = "command-line-arguments"
+ pkg.Target = ""
+ pkg.Match = gofiles
+
+ if pkg.Name == "main" {
+ exe := pkg.DefaultExecName() + cfg.ExeSuffix
+
+ if cfg.GOBIN != "" {
+ pkg.Target = filepath.Join(cfg.GOBIN, exe)
+ } else if cfg.ModulesEnabled {
+ pkg.Target = filepath.Join(modload.BinDir(), exe)
+ }
+ }
+
+ setToolFlags(pkg)
+
+ return pkg
+}