https://segmentfault.com/a/
前言
parser package 包含了 golang 语法分析相关的数据结构和方法,源代码位于 <go-src>/src/go/parser
之前大概看了点 PHP 和 Ruby 的源代码,感叹 go 确实如宣传的一样,简洁如 C,parser.go 代码总共 几千行(Ruby 语法规则定义文件有 1w 多行),使用递归下降语法分析方法(感觉 go 语言的语法规则很适合递归下降)
example_test.go
parser package 里面也有一个示例 example_test.go,如何使用 parser
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
| func ExampleParseFile() { fset := token.NewFileSet() f, err := parser.ParseFile(fset, "example_test.go", nil, parser.ImportsOnly) if err != nil { fmt.Println(err) return } for _, s := range f.Imports { fmt.Println(s.Path.Value) } }
|
parser struct
The parser structure holds the parser’s internal state.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
| type parser struct { file *token.File errors scanner.ErrorList scanner scanner.Scanner ... pos token.Pos tok token.Token lit string ... pkgScope *ast.Scope topScope *ast.Scope unresolved []*ast.Ident imports []*ast.ImportSpec ... }
|
parser 结构体以小写字母开头,意味着它是一个仅供内部使用的数据结构,里面字段比较多,一时不明白用途关系不大,有个大概印象
总结