// Copyright 2009 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. // Parse input AST and prepare Prog structure. package main import ( "fmt" "go/ast" "go/parser" "go/scanner" "go/token" "os" "path/filepath" "strings" ) func parse(name string, flags parser.Mode) *ast.File { ast1, err := parser.ParseFile(fset, name, nil, flags) if err != nil { if list, ok := err.(scanner.ErrorList); ok { // If err is a scanner.ErrorList, its String will print just // the first error and then (+n more errors). // Instead, turn it into a new Error that will return // details for all the errors. for _, e := range list { fmt.Fprintln(os.Stderr, e) } os.Exit(2) } fatalf("parsing %s: %s", name, err) } return ast1 } func sourceLine(n ast.Node) int { return fset.Position(n.Pos()).Line } // ReadGo populates f with information learned from reading the // Go source file with the given file name. It gathers the C preamble // attached to the import "C" comment, a list of references to C.xxx, // a list of exported functions, and the actual AST, to be rewritten and // printed. func (f *File) ReadGo(name string) { // Create absolute path for file, so that it will be used in error // messages and recorded in debug line number information. // This matches the rest of the toolchain. See golang.org/issue/5122. if aname, err := filepath.Abs(name); err == nil { name = aname } // Two different parses: once with comments, once without. // The printer is not good enough at printing comments in the // right place when we start editing the AST behind its back, // so we use ast1 to look for the doc comments on import "C" // and on exported functions, and we use ast2 for translating // and reprinting. ast1 := parse(name, parser.ParseComments) ast2 := parse(name, 0) f.Package = ast1.Name.Name f.Name = make(map[string]*Name) // In ast1, find the import "C" line and get any extra C preamble. sawC := false for _, decl := range ast1.Decls { d, ok := decl.(*ast.GenDecl) if !ok { continue } for _, spec := range d.Specs { s, ok := spec.(*ast.ImportSpec) if !ok || string(s.Path.Value) != `"C"` { continue } sawC = true if s.Name != nil { error_(s.Path.Pos(), `cannot rename import "C"`) } cg := s.Doc if cg == nil && len(d.Specs) == 1 { cg = d.Doc } if cg != nil { f.Preamble += fmt.Sprintf("#line %d %q\n", sourceLine(cg), name) f.Preamble += commentText(cg) + "\n" } } } if !sawC { error_(token.NoPos, `cannot find import "C"`) } // In ast2, strip the import "C" line. w := 0 for _, decl := range ast2.Decls { d, ok := decl.(*ast.GenDecl) if !ok { ast2.Decls[w] = decl w++ continue } ws := 0 for _, spec := range d.Specs { s, ok := spec.(*ast.ImportSpec) if !ok || string(s.Path.Value) != `"C"` { d.Specs[ws] = spec ws++ } } if ws == 0 { continue } d.Specs = d.Specs[0:ws] ast2.Decls[w] = d w++ } ast2.Decls = ast2.Decls[0:w] // Accumulate pointers to uses of C.x. if f.Ref == nil { f.Ref = make([]*Ref, 0, 8) } f.walk(ast2, "prog", (*File).saveRef) // Accumulate exported functions. // The comments are only on ast1 but we need to // save the function bodies from ast2. // The first walk fills in ExpFunc, and the // second walk changes the entries to // refer to ast2 instead. f.walk(ast1, "prog", (*File).saveExport) f.walk(ast2, "prog", (*File).saveExport2) f.Comments = ast1.Comments f.AST = ast2 } // Like ast.CommentGroup's Text method but preserves // leading blank lines, so that line numbers line up. func commentText(g *ast.CommentGroup) string { if g == nil { return "" } var pieces []string for _, com := range g.List { c := string(com.Text) // Remove comment markers. // The parser has given us exactly the comment text. switch c[1] { case '/': //-style comment (no newline at the end) c = c[2:] + "\n" case '*': /*-style comment */ c = c[2 : len(c)-2] } pieces = append(pieces, c) } return strings.Join(pieces, "") } // Save references to C.xxx for later processing. func (f *File) saveRef(x interface{}, context string) { n, ok := x.(*ast.Expr) if !ok { return } if sel, ok := (*n).(*ast.SelectorExpr); ok { // For now, assume that the only instance of capital C is // when used as the imported package identifier. // The parser should take care of scoping in the future, // so that we will be able to distinguish a "top-level C" // from a local C. if l, ok := sel.X.(*ast.Ident); ok && l.Name == "C" { if context == "as2" { context = "expr" } if context == "embed-type" { error_(sel.Pos(), "cannot embed C type") } goname := sel.Sel.Name if goname == "errno" { error_(sel.Pos(), "cannot refer to errno directly; see documentation") return } if goname == "_CMalloc" { error_(sel.Pos(), "cannot refer to C._CMalloc; use C.malloc") return } if goname == "malloc" { goname = "_CMalloc" } name := f.Name[goname] if name == nil { name = &Name{ Go: goname, } f.Name[goname] = name } f.Ref = append(f.Ref, &Ref{ Name: name, Expr: n, Context: context, }) return } } } // If a function should be exported add it to ExpFunc. func (f *File) saveExport(x interface{}, context string) { n, ok := x.(*ast.FuncDecl) if !ok { return } if n.Doc == nil { return } for _, c := range n.Doc.List { if !strings.HasPrefix(string(c.Text), "//export ") { continue } name := strings.TrimSpace(string(c.Text[9:])) if name == "" { error_(c.Pos(), "export missing name") } if name != n.Name.Name { error_(c.Pos(), "export comment has wrong name %q, want %q", name, n.Name.Name) } doc := "" for _, c1 := range n.Doc.List { if c1 != c { doc += c1.Text + "\n" } } f.ExpFunc = append(f.ExpFunc, &ExpFunc{ Func: n, ExpName: name, Doc: doc, }) break } } // Make f.ExpFunc[i] point at the Func from this AST instead of the other one. func (f *File) saveExport2(x interface{}, context string) { n, ok := x.(*ast.FuncDecl) if !ok { return } for _, exp := range f.ExpFunc { if exp.Func.Name.Name == n.Name.Name { exp.Func = n break } } } // walk walks the AST x, calling visit(f, x, context) for each node. func (f *File) walk(x interface{}, context string, visit func(*File, interface{}, string)) { visit(f, x, context) switch n := x.(type) { case *ast.Expr: f.walk(*n, context, visit) // everything else just recurs default: error_(token.NoPos, "unexpected type %T in walk", x, visit) panic("unexpected type") case nil: // These are ordered and grouped to match ../../go/ast/ast.go case *ast.Field: if len(n.Names) == 0 && context == "field" { f.walk(&n.Type, "embed-type", visit) } else { f.walk(&n.Type, "type", visit) } case *ast.FieldList: for _, field := range n.List { f.walk(field, context, visit) } case *ast.BadExpr: case *ast.Ident: case *ast.Ellipsis: case *ast.BasicLit: case *ast.FuncLit: f.walk(n.Type, "type", visit) f.walk(n.Body, "stmt", visit) case *ast.CompositeLit: f.walk(&n.Type, "type", visit) f.walk(n.Elts, "expr", visit) case *ast.ParenExpr: f.walk(&n.X, context, visit) case *ast.SelectorExpr: f.walk(&n.X, "selector", visit) case *ast.IndexExpr: f.walk(&n.X, "expr", visit) f.walk(&n.Index, "expr", visit) case *ast.SliceExpr: f.walk(&n.X, "expr", visit) if n.Low != nil { f.walk(&n.Low, "expr", visit) } if n.High != nil { f.walk(&n.High, "expr", visit) } if n.Max != nil { f.walk(&n.Max, "expr", visit) } case *ast.TypeAssertExpr: f.walk(&n.X, "expr", visit) f.walk(&n.Type, "type", visit) case *ast.CallExpr: if context == "as2" { f.walk(&n.Fun, "call2", visit) } else { f.walk(&n.Fun, "call", visit) } f.walk(n.Args, "expr", visit) case *ast.StarExpr: f.walk(&n.X, context, visit) case *ast.UnaryExpr: f.walk(&n.X, "expr", visit) case *ast.BinaryExpr: f.walk(&n.X, "expr", visit) f.walk(&n.Y, "expr", visit) case *ast.KeyValueExpr: f.walk(&n.Key, "expr", visit) f.walk(&n.Value, "expr", visit) case *ast.ArrayType: f.walk(&n.Len, "expr", visit) f.walk(&n.Elt, "type", visit) case *ast.StructType: f.walk(n.Fields, "field", visit) case *ast.FuncType: f.walk(n.Params, "param", visit) if n.Results != nil { f.walk(n.Results, "param", visit) } case *ast.InterfaceType: f.walk(n.Methods, "field", visit) case *ast.MapType: f.walk(&n.Key, "type", visit) f.walk(&n.Value, "type", visit) case *ast.ChanType: f.walk(&n.Value, "type", visit) case *ast.BadStmt: case *ast.DeclStmt: f.walk(n.Decl, "decl", visit) case *ast.EmptyStmt: case *ast.LabeledStmt: f.walk(n.Stmt, "stmt", visit) case *ast.ExprStmt: f.walk(&n.X, "expr", visit) case *ast.SendStmt: f.walk(&n.Chan, "expr", visit) f.walk(&n.Value, "expr", visit) case *ast.IncDecStmt: f.walk(&n.X, "expr", visit) case *ast.AssignStmt: f.walk(n.Lhs, "expr", visit) if len(n.Lhs) == 2 && len(n.Rhs) == 1 { f.walk(n.Rhs, "as2", visit) } else { f.walk(n.Rhs, "expr", visit) } case *ast.GoStmt: f.walk(n.Call, "expr", visit) case *ast.DeferStmt: f.walk(n.Call, "expr", visit) case *ast.ReturnStmt: f.walk(n.Results, "expr", visit) case *ast.BranchStmt: case *ast.BlockStmt: f.walk(n.List, context, visit) case *ast.IfStmt: f.walk(n.Init, "stmt", visit) f.walk(&n.Cond, "expr", visit) f.walk(n.Body, "stmt", visit) f.walk(n.Else, "stmt", visit) case *ast.CaseClause: if context == "typeswitch" { context = "type" } else { context = "expr" } f.walk(n.List, context, visit) f.walk(n.Body, "stmt", visit) case *ast.SwitchStmt: f.walk(n.Init, "stmt", visit) f.walk(&n.Tag, "expr", visit) f.walk(n.Body, "switch", visit) case *ast.TypeSwitchStmt: f.walk(n.Init, "stmt", visit) f.walk(n.Assign, "stmt", visit) f.walk(n.Body, "typeswitch", visit) case *ast.CommClause: f.walk(n.Comm, "stmt", visit) f.walk(n.Body, "stmt", visit) case *ast.SelectStmt: f.walk(n.Body, "stmt", visit) case *ast.ForStmt: f.walk(n.Init, "stmt", visit) f.walk(&n.Cond, "expr", visit) f.walk(n.Post, "stmt", visit) f.walk(n.Body, "stmt", visit) case *ast.RangeStmt: f.walk(&n.Key, "expr", visit) f.walk(&n.Value, "expr", visit) f.walk(&n.X, "expr", visit) f.walk(n.Body, "stmt", visit) case *ast.ImportSpec: case *ast.ValueSpec: f.walk(&n.Type, "type", visit) f.walk(n.Values, "expr", visit) case *ast.TypeSpec: f.walk(&n.Type, "type", visit) case *ast.BadDecl: case *ast.GenDecl: f.walk(n.Specs, "spec", visit) case *ast.FuncDecl: if n.Recv != nil { f.walk(n.Recv, "param", visit) } f.walk(n.Type, "type", visit) if n.Body != nil { f.walk(n.Body, "stmt", visit) } case *ast.File: f.walk(n.Decls, "decl", visit) case *ast.Package: for _, file := range n.Files { f.walk(file, "file", visit) } case []ast.Decl: for _, d := range n { f.walk(d, context, visit) } case []ast.Expr: for i := range n { f.walk(&n[i], context, visit) } case []ast.Stmt: for _, s := range n { f.walk(s, context, visit) } case []ast.Spec: for _, s := range n { f.walk(s, context, visit) } } }