2019-05-16 22:23:05 +00:00
// Copyright (c) 2013 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 or at
// https://developers.google.com/open-source/licenses/bsd.
// Package lint contains a linter for Go source code.
package lint // import "golang.org/x/lint"
import (
"bufio"
"bytes"
"fmt"
"go/ast"
"go/parser"
"go/printer"
"go/token"
"go/types"
"regexp"
"sort"
"strconv"
"strings"
"unicode"
"unicode/utf8"
"golang.org/x/tools/go/ast/astutil"
"golang.org/x/tools/go/gcexportdata"
)
const styleGuideBase = "https://golang.org/wiki/CodeReviewComments"
// A Linter lints Go source code.
type Linter struct {
}
// Problem represents a problem in some source code.
type Problem struct {
Position token . Position // position in source file
Text string // the prose that describes the problem
Link string // (optional) the link to the style guide for the problem
Confidence float64 // a value in (0,1] estimating the confidence in this problem's correctness
LineText string // the source line
Category string // a short name for the general category of the problem
// If the problem has a suggested fix (the minority case),
// ReplacementLine is a full replacement for the relevant line of the source file.
ReplacementLine string
}
func ( p * Problem ) String ( ) string {
if p . Link != "" {
return p . Text + "\n\n" + p . Link
}
return p . Text
}
type byPosition [ ] Problem
func ( p byPosition ) Len ( ) int { return len ( p ) }
func ( p byPosition ) Swap ( i , j int ) { p [ i ] , p [ j ] = p [ j ] , p [ i ] }
func ( p byPosition ) Less ( i , j int ) bool {
pi , pj := p [ i ] . Position , p [ j ] . Position
if pi . Filename != pj . Filename {
return pi . Filename < pj . Filename
}
if pi . Line != pj . Line {
return pi . Line < pj . Line
}
if pi . Column != pj . Column {
return pi . Column < pj . Column
}
return p [ i ] . Text < p [ j ] . Text
}
// Lint lints src.
func ( l * Linter ) Lint ( filename string , src [ ] byte ) ( [ ] Problem , error ) {
return l . LintFiles ( map [ string ] [ ] byte { filename : src } )
}
// LintFiles lints a set of files of a single package.
// The argument is a map of filename to source.
func ( l * Linter ) LintFiles ( files map [ string ] [ ] byte ) ( [ ] Problem , error ) {
pkg := & pkg {
fset : token . NewFileSet ( ) ,
files : make ( map [ string ] * file ) ,
}
var pkgName string
for filename , src := range files {
if isGenerated ( src ) {
continue // See issue #239
}
f , err := parser . ParseFile ( pkg . fset , filename , src , parser . ParseComments )
if err != nil {
return nil , err
}
if pkgName == "" {
pkgName = f . Name . Name
} else if f . Name . Name != pkgName {
return nil , fmt . Errorf ( "%s is in package %s, not %s" , filename , f . Name . Name , pkgName )
}
pkg . files [ filename ] = & file {
pkg : pkg ,
f : f ,
fset : pkg . fset ,
src : src ,
filename : filename ,
}
}
if len ( pkg . files ) == 0 {
return nil , nil
}
return pkg . lint ( ) , nil
}
var (
genHdr = [ ] byte ( "// Code generated " )
genFtr = [ ] byte ( " DO NOT EDIT." )
)
// isGenerated reports whether the source file is generated code
// according the rules from https://golang.org/s/generatedcode.
func isGenerated ( src [ ] byte ) bool {
sc := bufio . NewScanner ( bytes . NewReader ( src ) )
for sc . Scan ( ) {
b := sc . Bytes ( )
if bytes . HasPrefix ( b , genHdr ) && bytes . HasSuffix ( b , genFtr ) && len ( b ) >= len ( genHdr ) + len ( genFtr ) {
return true
}
}
return false
}
// pkg represents a package being linted.
type pkg struct {
fset * token . FileSet
files map [ string ] * file
typesPkg * types . Package
typesInfo * types . Info
// sortable is the set of types in the package that implement sort.Interface.
sortable map [ string ] bool
// main is whether this is a "main" package.
main bool
problems [ ] Problem
}
func ( p * pkg ) lint ( ) [ ] Problem {
if err := p . typeCheck ( ) ; err != nil {
/ * TODO ( dsymonds ) : Consider reporting these errors when golint operates on entire packages .
if e , ok := err . ( types . Error ) ; ok {
pos := p . fset . Position ( e . Pos )
conf := 1.0
if strings . Contains ( e . Msg , "can't find import: " ) {
// Golint is probably being run in a context that doesn't support
// typechecking (e.g. package files aren't found), so don't warn about it.
conf = 0
}
if conf > 0 {
p . errorfAt ( pos , conf , category ( "typechecking" ) , e . Msg )
}
// TODO(dsymonds): Abort if !e.Soft?
}
* /
}
p . scanSortable ( )
p . main = p . isMain ( )
for _ , f := range p . files {
f . lint ( )
}
sort . Sort ( byPosition ( p . problems ) )
return p . problems
}
// file represents a file being linted.
type file struct {
pkg * pkg
f * ast . File
fset * token . FileSet
src [ ] byte
filename string
}
func ( f * file ) isTest ( ) bool { return strings . HasSuffix ( f . filename , "_test.go" ) }
func ( f * file ) lint ( ) {
f . lintPackageComment ( )
f . lintImports ( )
f . lintBlankImports ( )
f . lintExported ( )
f . lintNames ( )
f . lintElses ( )
f . lintRanges ( )
f . lintErrorf ( )
f . lintErrors ( )
f . lintErrorStrings ( )
f . lintReceiverNames ( )
f . lintIncDec ( )
f . lintErrorReturn ( )
f . lintUnexportedReturn ( )
f . lintTimeNames ( )
f . lintContextKeyTypes ( )
f . lintContextArgs ( )
}
type link string
type category string
// The variadic arguments may start with link and category types,
// and must end with a format string and any arguments.
// It returns the new Problem.
func ( f * file ) errorf ( n ast . Node , confidence float64 , args ... interface { } ) * Problem {
pos := f . fset . Position ( n . Pos ( ) )
if pos . Filename == "" {
pos . Filename = f . filename
}
return f . pkg . errorfAt ( pos , confidence , args ... )
}
func ( p * pkg ) errorfAt ( pos token . Position , confidence float64 , args ... interface { } ) * Problem {
problem := Problem {
Position : pos ,
Confidence : confidence ,
}
if pos . Filename != "" {
// The file might not exist in our mapping if a //line directive was encountered.
if f , ok := p . files [ pos . Filename ] ; ok {
problem . LineText = srcLine ( f . src , pos )
}
}
argLoop :
for len ( args ) > 1 { // always leave at least the format string in args
switch v := args [ 0 ] . ( type ) {
case link :
problem . Link = string ( v )
case category :
problem . Category = string ( v )
default :
break argLoop
}
args = args [ 1 : ]
}
problem . Text = fmt . Sprintf ( args [ 0 ] . ( string ) , args [ 1 : ] ... )
p . problems = append ( p . problems , problem )
return & p . problems [ len ( p . problems ) - 1 ]
}
var newImporter = func ( fset * token . FileSet ) types . ImporterFrom {
return gcexportdata . NewImporter ( fset , make ( map [ string ] * types . Package ) )
}
func ( p * pkg ) typeCheck ( ) error {
config := & types . Config {
// By setting a no-op error reporter, the type checker does as much work as possible.
Error : func ( error ) { } ,
Importer : newImporter ( p . fset ) ,
}
info := & types . Info {
Types : make ( map [ ast . Expr ] types . TypeAndValue ) ,
Defs : make ( map [ * ast . Ident ] types . Object ) ,
Uses : make ( map [ * ast . Ident ] types . Object ) ,
Scopes : make ( map [ ast . Node ] * types . Scope ) ,
}
var anyFile * file
var astFiles [ ] * ast . File
for _ , f := range p . files {
anyFile = f
astFiles = append ( astFiles , f . f )
}
pkg , err := config . Check ( anyFile . f . Name . Name , p . fset , astFiles , info )
// Remember the typechecking info, even if config.Check failed,
// since we will get partial information.
p . typesPkg = pkg
p . typesInfo = info
return err
}
func ( p * pkg ) typeOf ( expr ast . Expr ) types . Type {
if p . typesInfo == nil {
return nil
}
return p . typesInfo . TypeOf ( expr )
}
func ( p * pkg ) isNamedType ( typ types . Type , importPath , name string ) bool {
n , ok := typ . ( * types . Named )
if ! ok {
return false
}
tn := n . Obj ( )
return tn != nil && tn . Pkg ( ) != nil && tn . Pkg ( ) . Path ( ) == importPath && tn . Name ( ) == name
}
// scopeOf returns the tightest scope encompassing id.
func ( p * pkg ) scopeOf ( id * ast . Ident ) * types . Scope {
var scope * types . Scope
if obj := p . typesInfo . ObjectOf ( id ) ; obj != nil {
scope = obj . Parent ( )
}
if scope == p . typesPkg . Scope ( ) {
// We were given a top-level identifier.
// Use the file-level scope instead of the package-level scope.
pos := id . Pos ( )
for _ , f := range p . files {
if f . f . Pos ( ) <= pos && pos < f . f . End ( ) {
scope = p . typesInfo . Scopes [ f . f ]
break
}
}
}
return scope
}
func ( p * pkg ) scanSortable ( ) {
p . sortable = make ( map [ string ] bool )
// bitfield for which methods exist on each type.
const (
Len = 1 << iota
Less
Swap
)
nmap := map [ string ] int { "Len" : Len , "Less" : Less , "Swap" : Swap }
has := make ( map [ string ] int )
for _ , f := range p . files {
f . walk ( func ( n ast . Node ) bool {
fn , ok := n . ( * ast . FuncDecl )
if ! ok || fn . Recv == nil || len ( fn . Recv . List ) == 0 {
return true
}
// TODO(dsymonds): We could check the signature to be more precise.
recv := receiverType ( fn )
if i , ok := nmap [ fn . Name . Name ] ; ok {
has [ recv ] |= i
}
return false
} )
}
for typ , ms := range has {
if ms == Len | Less | Swap {
p . sortable [ typ ] = true
}
}
}
func ( p * pkg ) isMain ( ) bool {
for _ , f := range p . files {
if f . isMain ( ) {
return true
}
}
return false
}
func ( f * file ) isMain ( ) bool {
if f . f . Name . Name == "main" {
return true
}
return false
}
// lintPackageComment checks package comments. It complains if
// there is no package comment, or if it is not of the right form.
// This has a notable false positive in that a package comment
// could rightfully appear in a different file of the same package,
// but that's not easy to fix since this linter is file-oriented.
func ( f * file ) lintPackageComment ( ) {
if f . isTest ( ) {
return
}
const ref = styleGuideBase + "#package-comments"
prefix := "Package " + f . f . Name . Name + " "
// Look for a detached package comment.
// First, scan for the last comment that occurs before the "package" keyword.
var lastCG * ast . CommentGroup
for _ , cg := range f . f . Comments {
if cg . Pos ( ) > f . f . Package {
// Gone past "package" keyword.
break
}
lastCG = cg
}
if lastCG != nil && strings . HasPrefix ( lastCG . Text ( ) , prefix ) {
endPos := f . fset . Position ( lastCG . End ( ) )
pkgPos := f . fset . Position ( f . f . Package )
if endPos . Line + 1 < pkgPos . Line {
// There isn't a great place to anchor this error;
// the start of the blank lines between the doc and the package statement
// is at least pointing at the location of the problem.
pos := token . Position {
Filename : endPos . Filename ,
// Offset not set; it is non-trivial, and doesn't appear to be needed.
Line : endPos . Line + 1 ,
Column : 1 ,
}
f . pkg . errorfAt ( pos , 0.9 , link ( ref ) , category ( "comments" ) , "package comment is detached; there should be no blank lines between it and the package statement" )
return
}
}
if f . f . Doc == nil {
f . errorf ( f . f , 0.2 , link ( ref ) , category ( "comments" ) , "should have a package comment, unless it's in another file for this package" )
return
}
s := f . f . Doc . Text ( )
if ts := strings . TrimLeft ( s , " \t" ) ; ts != s {
f . errorf ( f . f . Doc , 1 , link ( ref ) , category ( "comments" ) , "package comment should not have leading space" )
s = ts
}
// Only non-main packages need to keep to this form.
if ! f . pkg . main && ! strings . HasPrefix ( s , prefix ) {
f . errorf ( f . f . Doc , 1 , link ( ref ) , category ( "comments" ) , ` package comment should be of the form "%s..." ` , prefix )
}
}
// lintBlankImports complains if a non-main package has blank imports that are
// not documented.
func ( f * file ) lintBlankImports ( ) {
// In package main and in tests, we don't complain about blank imports.
if f . pkg . main || f . isTest ( ) {
return
}
// The first element of each contiguous group of blank imports should have
// an explanatory comment of some kind.
for i , imp := range f . f . Imports {
pos := f . fset . Position ( imp . Pos ( ) )
if ! isBlank ( imp . Name ) {
continue // Ignore non-blank imports.
}
if i > 0 {
prev := f . f . Imports [ i - 1 ]
prevPos := f . fset . Position ( prev . Pos ( ) )
if isBlank ( prev . Name ) && prevPos . Line + 1 == pos . Line {
continue // A subsequent blank in a group.
}
}
// This is the first blank import of a group.
if imp . Doc == nil && imp . Comment == nil {
ref := ""
f . errorf ( imp , 1 , link ( ref ) , category ( "imports" ) , "a blank import should be only in a main or test package, or have a comment justifying it" )
}
}
}
// lintImports examines import blocks.
func ( f * file ) lintImports ( ) {
for i , is := range f . f . Imports {
_ = i
if is . Name != nil && is . Name . Name == "." && ! f . isTest ( ) {
f . errorf ( is , 1 , link ( styleGuideBase + "#import-dot" ) , category ( "imports" ) , "should not use dot imports" )
}
}
}
const docCommentsLink = styleGuideBase + "#doc-comments"
// lintExported examines the exported names.
// It complains if any required doc comments are missing,
// or if they are not of the right form. The exact rules are in
// lintFuncDoc, lintTypeDoc and lintValueSpecDoc; this function
// also tracks the GenDecl structure being traversed to permit
// doc comments for constants to be on top of the const block.
// It also complains if the names stutter when combined with
// the package name.
func ( f * file ) lintExported ( ) {
if f . isTest ( ) {
return
}
var lastGen * ast . GenDecl // last GenDecl entered.
// Set of GenDecls that have already had missing comments flagged.
genDeclMissingComments := make ( map [ * ast . GenDecl ] bool )
f . walk ( func ( node ast . Node ) bool {
switch v := node . ( type ) {
case * ast . GenDecl :
if v . Tok == token . IMPORT {
return false
}
// token.CONST, token.TYPE or token.VAR
lastGen = v
return true
case * ast . FuncDecl :
f . lintFuncDoc ( v )
if v . Recv == nil {
// Only check for stutter on functions, not methods.
// Method names are not used package-qualified.
f . checkStutter ( v . Name , "func" )
}
// Don't proceed inside funcs.
return false
case * ast . TypeSpec :
// inside a GenDecl, which usually has the doc
doc := v . Doc
if doc == nil {
doc = lastGen . Doc
}
f . lintTypeDoc ( v , doc )
f . checkStutter ( v . Name , "type" )
// Don't proceed inside types.
return false
case * ast . ValueSpec :
f . lintValueSpecDoc ( v , lastGen , genDeclMissingComments )
return false
}
return true
} )
}
var (
allCapsRE = regexp . MustCompile ( ` ^[A-Z0-9_]+$ ` )
anyCapsRE = regexp . MustCompile ( ` [A-Z] ` )
)
// knownNameExceptions is a set of names that are known to be exempt from naming checks.
// This is usually because they are constrained by having to match names in the
// standard library.
var knownNameExceptions = map [ string ] bool {
"LastInsertId" : true , // must match database/sql
"kWh" : true ,
}
func isInTopLevel ( f * ast . File , ident * ast . Ident ) bool {
path , _ := astutil . PathEnclosingInterval ( f , ident . Pos ( ) , ident . End ( ) )
for _ , f := range path {
switch f . ( type ) {
case * ast . File , * ast . GenDecl , * ast . ValueSpec , * ast . Ident :
continue
}
return false
}
return true
}
// lintNames examines all names in the file.
// It complains if any use underscores or incorrect known initialisms.
func ( f * file ) lintNames ( ) {
// Package names need slightly different handling than other names.
if strings . Contains ( f . f . Name . Name , "_" ) && ! strings . HasSuffix ( f . f . Name . Name , "_test" ) {
f . errorf ( f . f , 1 , link ( "http://golang.org/doc/effective_go.html#package-names" ) , category ( "naming" ) , "don't use an underscore in package name" )
}
if anyCapsRE . MatchString ( f . f . Name . Name ) {
f . errorf ( f . f , 1 , link ( "http://golang.org/doc/effective_go.html#package-names" ) , category ( "mixed-caps" ) , "don't use MixedCaps in package name; %s should be %s" , f . f . Name . Name , strings . ToLower ( f . f . Name . Name ) )
}
check := func ( id * ast . Ident , thing string ) {
if id . Name == "_" {
return
}
if knownNameExceptions [ id . Name ] {
return
}
// Handle two common styles from other languages that don't belong in Go.
if len ( id . Name ) >= 5 && allCapsRE . MatchString ( id . Name ) && strings . Contains ( id . Name , "_" ) {
capCount := 0
for _ , c := range id . Name {
if 'A' <= c && c <= 'Z' {
capCount ++
}
}
if capCount >= 2 {
f . errorf ( id , 0.8 , link ( styleGuideBase + "#mixed-caps" ) , category ( "naming" ) , "don't use ALL_CAPS in Go names; use CamelCase" )
return
}
}
if thing == "const" || ( thing == "var" && isInTopLevel ( f . f , id ) ) {
if len ( id . Name ) > 2 && id . Name [ 0 ] == 'k' && id . Name [ 1 ] >= 'A' && id . Name [ 1 ] <= 'Z' {
should := string ( id . Name [ 1 ] + 'a' - 'A' ) + id . Name [ 2 : ]
f . errorf ( id , 0.8 , link ( styleGuideBase + "#mixed-caps" ) , category ( "naming" ) , "don't use leading k in Go names; %s %s should be %s" , thing , id . Name , should )
}
}
should := lintName ( id . Name )
if id . Name == should {
return
}
if len ( id . Name ) > 2 && strings . Contains ( id . Name [ 1 : ] , "_" ) {
f . errorf ( id , 0.9 , link ( "http://golang.org/doc/effective_go.html#mixed-caps" ) , category ( "naming" ) , "don't use underscores in Go names; %s %s should be %s" , thing , id . Name , should )
return
}
f . errorf ( id , 0.8 , link ( styleGuideBase + "#initialisms" ) , category ( "naming" ) , "%s %s should be %s" , thing , id . Name , should )
}
checkList := func ( fl * ast . FieldList , thing string ) {
if fl == nil {
return
}
for _ , f := range fl . List {
for _ , id := range f . Names {
check ( id , thing )
}
}
}
f . walk ( func ( node ast . Node ) bool {
switch v := node . ( type ) {
case * ast . AssignStmt :
if v . Tok == token . ASSIGN {
return true
}
for _ , exp := range v . Lhs {
if id , ok := exp . ( * ast . Ident ) ; ok {
check ( id , "var" )
}
}
case * ast . FuncDecl :
if f . isTest ( ) && ( strings . HasPrefix ( v . Name . Name , "Example" ) || strings . HasPrefix ( v . Name . Name , "Test" ) || strings . HasPrefix ( v . Name . Name , "Benchmark" ) ) {
return true
}
thing := "func"
if v . Recv != nil {
thing = "method"
}
// Exclude naming warnings for functions that are exported to C but
// not exported in the Go API.
// See https://github.com/golang/lint/issues/144.
if ast . IsExported ( v . Name . Name ) || ! isCgoExported ( v ) {
check ( v . Name , thing )
}
checkList ( v . Type . Params , thing + " parameter" )
checkList ( v . Type . Results , thing + " result" )
case * ast . GenDecl :
if v . Tok == token . IMPORT {
return true
}
var thing string
switch v . Tok {
case token . CONST :
thing = "const"
case token . TYPE :
thing = "type"
case token . VAR :
thing = "var"
}
for _ , spec := range v . Specs {
switch s := spec . ( type ) {
case * ast . TypeSpec :
check ( s . Name , thing )
case * ast . ValueSpec :
for _ , id := range s . Names {
check ( id , thing )
}
}
}
case * ast . InterfaceType :
// Do not check interface method names.
// They are often constrainted by the method names of concrete types.
for _ , x := range v . Methods . List {
ft , ok := x . Type . ( * ast . FuncType )
if ! ok { // might be an embedded interface name
continue
}
checkList ( ft . Params , "interface method parameter" )
checkList ( ft . Results , "interface method result" )
}
case * ast . RangeStmt :
if v . Tok == token . ASSIGN {
return true
}
if id , ok := v . Key . ( * ast . Ident ) ; ok {
check ( id , "range var" )
}
if id , ok := v . Value . ( * ast . Ident ) ; ok {
check ( id , "range var" )
}
case * ast . StructType :
for _ , f := range v . Fields . List {
for _ , id := range f . Names {
check ( id , "struct field" )
}
}
}
return true
} )
}
// lintName returns a different name if it should be different.
func lintName ( name string ) ( should string ) {
// Fast path for simple cases: "_" and all lowercase.
if name == "_" {
return name
}
allLower := true
for _ , r := range name {
if ! unicode . IsLower ( r ) {
allLower = false
break
}
}
if allLower {
return name
}
// Split camelCase at any lower->upper transition, and split on underscores.
// Check each word for common initialisms.
runes := [ ] rune ( name )
w , i := 0 , 0 // index of start of word, scan
for i + 1 <= len ( runes ) {
eow := false // whether we hit the end of a word
if i + 1 == len ( runes ) {
eow = true
} else if runes [ i + 1 ] == '_' {
// underscore; shift the remainder forward over any run of underscores
eow = true
n := 1
for i + n + 1 < len ( runes ) && runes [ i + n + 1 ] == '_' {
n ++
}
// Leave at most one underscore if the underscore is between two digits
if i + n + 1 < len ( runes ) && unicode . IsDigit ( runes [ i ] ) && unicode . IsDigit ( runes [ i + n + 1 ] ) {
n --
}
copy ( runes [ i + 1 : ] , runes [ i + n + 1 : ] )
runes = runes [ : len ( runes ) - n ]
} else if unicode . IsLower ( runes [ i ] ) && ! unicode . IsLower ( runes [ i + 1 ] ) {
// lower->non-lower
eow = true
}
i ++
if ! eow {
continue
}
// [w,i) is a word.
word := string ( runes [ w : i ] )
if u := strings . ToUpper ( word ) ; commonInitialisms [ u ] {
// Keep consistent case, which is lowercase only at the start.
if w == 0 && unicode . IsLower ( runes [ w ] ) {
u = strings . ToLower ( u )
}
// All the common initialisms are ASCII,
// so we can replace the bytes exactly.
copy ( runes [ w : ] , [ ] rune ( u ) )
} else if w > 0 && strings . ToLower ( word ) == word {
// already all lowercase, and not the first word, so uppercase the first character.
runes [ w ] = unicode . ToUpper ( runes [ w ] )
}
w = i
}
return string ( runes )
}
// commonInitialisms is a set of common initialisms.
// Only add entries that are highly unlikely to be non-initialisms.
// For instance, "ID" is fine (Freudian code is rare), but "AND" is not.
var commonInitialisms = map [ string ] bool {
"ACL" : true ,
"API" : true ,
"ASCII" : true ,
"CPU" : true ,
"CSS" : true ,
"DNS" : true ,
"EOF" : true ,
"GUID" : true ,
"HTML" : true ,
"HTTP" : true ,
"HTTPS" : true ,
"ID" : true ,
"IP" : true ,
"JSON" : true ,
"LHS" : true ,
"QPS" : true ,
"RAM" : true ,
"RHS" : true ,
"RPC" : true ,
"SLA" : true ,
"SMTP" : true ,
"SQL" : true ,
"SSH" : true ,
"TCP" : true ,
"TLS" : true ,
"TTL" : true ,
"UDP" : true ,
"UI" : true ,
"UID" : true ,
"UUID" : true ,
"URI" : true ,
"URL" : true ,
"UTF8" : true ,
"VM" : true ,
"XML" : true ,
"XMPP" : true ,
"XSRF" : true ,
"XSS" : true ,
}
// lintTypeDoc examines the doc comment on a type.
// It complains if they are missing from an exported type,
// or if they are not of the standard form.
func ( f * file ) lintTypeDoc ( t * ast . TypeSpec , doc * ast . CommentGroup ) {
if ! ast . IsExported ( t . Name . Name ) {
return
}
if doc == nil {
f . errorf ( t , 1 , link ( docCommentsLink ) , category ( "comments" ) , "exported type %v should have comment or be unexported" , t . Name )
return
}
s := doc . Text ( )
articles := [ ... ] string { "A" , "An" , "The" }
for _ , a := range articles {
if strings . HasPrefix ( s , a + " " ) {
s = s [ len ( a ) + 1 : ]
break
}
}
if ! strings . HasPrefix ( s , t . Name . Name + " " ) {
f . errorf ( doc , 1 , link ( docCommentsLink ) , category ( "comments" ) , ` comment on exported type %v should be of the form "%v ..." (with optional leading article) ` , t . Name , t . Name )
}
}
var commonMethods = map [ string ] bool {
"Error" : true ,
"Read" : true ,
"ServeHTTP" : true ,
"String" : true ,
"Write" : true ,
2021-05-15 10:08:31 +00:00
"Unwrap" : true ,
2019-05-16 22:23:05 +00:00
}
// lintFuncDoc examines doc comments on functions and methods.
// It complains if they are missing, or not of the right form.
// It has specific exclusions for well-known methods (see commonMethods above).
func ( f * file ) lintFuncDoc ( fn * ast . FuncDecl ) {
if ! ast . IsExported ( fn . Name . Name ) {
// func is unexported
return
}
kind := "function"
name := fn . Name . Name
if fn . Recv != nil && len ( fn . Recv . List ) > 0 {
// method
kind = "method"
recv := receiverType ( fn )
if ! ast . IsExported ( recv ) {
// receiver is unexported
return
}
if commonMethods [ name ] {
return
}
switch name {
case "Len" , "Less" , "Swap" :
if f . pkg . sortable [ recv ] {
return
}
}
name = recv + "." + name
}
if fn . Doc == nil {
f . errorf ( fn , 1 , link ( docCommentsLink ) , category ( "comments" ) , "exported %s %s should have comment or be unexported" , kind , name )
return
}
s := fn . Doc . Text ( )
prefix := fn . Name . Name + " "
if ! strings . HasPrefix ( s , prefix ) {
f . errorf ( fn . Doc , 1 , link ( docCommentsLink ) , category ( "comments" ) , ` comment on exported %s %s should be of the form "%s..." ` , kind , name , prefix )
}
}
// lintValueSpecDoc examines package-global variables and constants.
// It complains if they are not individually declared,
// or if they are not suitably documented in the right form (unless they are in a block that is commented).
func ( f * file ) lintValueSpecDoc ( vs * ast . ValueSpec , gd * ast . GenDecl , genDeclMissingComments map [ * ast . GenDecl ] bool ) {
kind := "var"
if gd . Tok == token . CONST {
kind = "const"
}
if len ( vs . Names ) > 1 {
// Check that none are exported except for the first.
for _ , n := range vs . Names [ 1 : ] {
if ast . IsExported ( n . Name ) {
f . errorf ( vs , 1 , category ( "comments" ) , "exported %s %s should have its own declaration" , kind , n . Name )
return
}
}
}
// Only one name.
name := vs . Names [ 0 ] . Name
if ! ast . IsExported ( name ) {
return
}
if vs . Doc == nil && gd . Doc == nil {
if genDeclMissingComments [ gd ] {
return
}
block := ""
if kind == "const" && gd . Lparen . IsValid ( ) {
block = " (or a comment on this block)"
}
f . errorf ( vs , 1 , link ( docCommentsLink ) , category ( "comments" ) , "exported %s %s should have comment%s or be unexported" , kind , name , block )
genDeclMissingComments [ gd ] = true
return
}
// If this GenDecl has parens and a comment, we don't check its comment form.
if gd . Lparen . IsValid ( ) && gd . Doc != nil {
return
}
// The relevant text to check will be on either vs.Doc or gd.Doc.
// Use vs.Doc preferentially.
doc := vs . Doc
if doc == nil {
doc = gd . Doc
}
prefix := name + " "
if ! strings . HasPrefix ( doc . Text ( ) , prefix ) {
f . errorf ( doc , 1 , link ( docCommentsLink ) , category ( "comments" ) , ` comment on exported %s %s should be of the form "%s..." ` , kind , name , prefix )
}
}
func ( f * file ) checkStutter ( id * ast . Ident , thing string ) {
pkg , name := f . f . Name . Name , id . Name
if ! ast . IsExported ( name ) {
// unexported name
return
}
// A name stutters if the package name is a strict prefix
// and the next character of the name starts a new word.
if len ( name ) <= len ( pkg ) {
// name is too short to stutter.
// This permits the name to be the same as the package name.
return
}
if ! strings . EqualFold ( pkg , name [ : len ( pkg ) ] ) {
return
}
// We can assume the name is well-formed UTF-8.
// If the next rune after the package name is uppercase or an underscore
// the it's starting a new word and thus this name stutters.
rem := name [ len ( pkg ) : ]
if next , _ := utf8 . DecodeRuneInString ( rem ) ; next == '_' || unicode . IsUpper ( next ) {
f . errorf ( id , 0.8 , link ( styleGuideBase + "#package-names" ) , category ( "naming" ) , "%s name will be used as %s.%s by other packages, and that stutters; consider calling this %s" , thing , pkg , name , rem )
}
}
// zeroLiteral is a set of ast.BasicLit values that are zero values.
// It is not exhaustive.
var zeroLiteral = map [ string ] bool {
"false" : true , // bool
// runes
` '\x00' ` : true ,
` '\000' ` : true ,
// strings
` "" ` : true ,
"``" : true ,
// numerics
"0" : true ,
"0." : true ,
"0.0" : true ,
"0i" : true ,
}
// lintElses examines else blocks. It complains about any else block whose if block ends in a return.
func ( f * file ) lintElses ( ) {
// We don't want to flag if { } else if { } else { } constructions.
// They will appear as an IfStmt whose Else field is also an IfStmt.
// Record such a node so we ignore it when we visit it.
ignore := make ( map [ * ast . IfStmt ] bool )
f . walk ( func ( node ast . Node ) bool {
ifStmt , ok := node . ( * ast . IfStmt )
if ! ok || ifStmt . Else == nil {
return true
}
if elseif , ok := ifStmt . Else . ( * ast . IfStmt ) ; ok {
ignore [ elseif ] = true
return true
}
if ignore [ ifStmt ] {
return true
}
if _ , ok := ifStmt . Else . ( * ast . BlockStmt ) ; ! ok {
// only care about elses without conditions
return true
}
if len ( ifStmt . Body . List ) == 0 {
return true
}
shortDecl := false // does the if statement have a ":=" initialization statement?
if ifStmt . Init != nil {
if as , ok := ifStmt . Init . ( * ast . AssignStmt ) ; ok && as . Tok == token . DEFINE {
shortDecl = true
}
}
lastStmt := ifStmt . Body . List [ len ( ifStmt . Body . List ) - 1 ]
if _ , ok := lastStmt . ( * ast . ReturnStmt ) ; ok {
extra := ""
if shortDecl {
extra = " (move short variable declaration to its own line if necessary)"
}
f . errorf ( ifStmt . Else , 1 , link ( styleGuideBase + "#indent-error-flow" ) , category ( "indent" ) , "if block ends with a return statement, so drop this else and outdent its block" + extra )
}
return true
} )
}
// lintRanges examines range clauses. It complains about redundant constructions.
func ( f * file ) lintRanges ( ) {
f . walk ( func ( node ast . Node ) bool {
rs , ok := node . ( * ast . RangeStmt )
if ! ok {
return true
}
if isIdent ( rs . Key , "_" ) && ( rs . Value == nil || isIdent ( rs . Value , "_" ) ) {
p := f . errorf ( rs . Key , 1 , category ( "range-loop" ) , "should omit values from range; this loop is equivalent to `for range ...`" )
newRS := * rs // shallow copy
newRS . Value = nil
newRS . Key = nil
p . ReplacementLine = f . firstLineOf ( & newRS , rs )
return true
}
if isIdent ( rs . Value , "_" ) {
p := f . errorf ( rs . Value , 1 , category ( "range-loop" ) , "should omit 2nd value from range; this loop is equivalent to `for %s %s range ...`" , f . render ( rs . Key ) , rs . Tok )
newRS := * rs // shallow copy
newRS . Value = nil
p . ReplacementLine = f . firstLineOf ( & newRS , rs )
}
return true
} )
}
// lintErrorf examines errors.New and testing.Error calls. It complains if its only argument is an fmt.Sprintf invocation.
func ( f * file ) lintErrorf ( ) {
f . walk ( func ( node ast . Node ) bool {
ce , ok := node . ( * ast . CallExpr )
if ! ok || len ( ce . Args ) != 1 {
return true
}
isErrorsNew := isPkgDot ( ce . Fun , "errors" , "New" )
var isTestingError bool
se , ok := ce . Fun . ( * ast . SelectorExpr )
if ok && se . Sel . Name == "Error" {
if typ := f . pkg . typeOf ( se . X ) ; typ != nil {
isTestingError = typ . String ( ) == "*testing.T"
}
}
if ! isErrorsNew && ! isTestingError {
return true
}
if ! f . imports ( "errors" ) {
return true
}
arg := ce . Args [ 0 ]
ce , ok = arg . ( * ast . CallExpr )
if ! ok || ! isPkgDot ( ce . Fun , "fmt" , "Sprintf" ) {
return true
}
errorfPrefix := "fmt"
if isTestingError {
errorfPrefix = f . render ( se . X )
}
p := f . errorf ( node , 1 , category ( "errors" ) , "should replace %s(fmt.Sprintf(...)) with %s.Errorf(...)" , f . render ( se ) , errorfPrefix )
m := f . srcLineWithMatch ( ce , ` ^(.*) ` + f . render ( se ) + ` \(fmt\.Sprintf\((.*)\)\)(.*)$ ` )
if m != nil {
p . ReplacementLine = m [ 1 ] + errorfPrefix + ".Errorf(" + m [ 2 ] + ")" + m [ 3 ]
}
return true
} )
}
// lintErrors examines global error vars. It complains if they aren't named in the standard way.
func ( f * file ) lintErrors ( ) {
for _ , decl := range f . f . Decls {
gd , ok := decl . ( * ast . GenDecl )
if ! ok || gd . Tok != token . VAR {
continue
}
for _ , spec := range gd . Specs {
spec := spec . ( * ast . ValueSpec )
if len ( spec . Names ) != 1 || len ( spec . Values ) != 1 {
continue
}
ce , ok := spec . Values [ 0 ] . ( * ast . CallExpr )
if ! ok {
continue
}
if ! isPkgDot ( ce . Fun , "errors" , "New" ) && ! isPkgDot ( ce . Fun , "fmt" , "Errorf" ) {
continue
}
id := spec . Names [ 0 ]
prefix := "err"
if id . IsExported ( ) {
prefix = "Err"
}
if ! strings . HasPrefix ( id . Name , prefix ) {
f . errorf ( id , 0.9 , category ( "naming" ) , "error var %s should have name of the form %sFoo" , id . Name , prefix )
}
}
}
}
func lintErrorString ( s string ) ( isClean bool , conf float64 ) {
const basicConfidence = 0.8
const capConfidence = basicConfidence - 0.2
first , firstN := utf8 . DecodeRuneInString ( s )
last , _ := utf8 . DecodeLastRuneInString ( s )
if last == '.' || last == ':' || last == '!' || last == '\n' {
return false , basicConfidence
}
if unicode . IsUpper ( first ) {
// People use proper nouns and exported Go identifiers in error strings,
// so decrease the confidence of warnings for capitalization.
if len ( s ) <= firstN {
return false , capConfidence
}
// Flag strings starting with something that doesn't look like an initialism.
if second , _ := utf8 . DecodeRuneInString ( s [ firstN : ] ) ; ! unicode . IsUpper ( second ) {
return false , capConfidence
}
}
return true , 0
}
// lintErrorStrings examines error strings.
// It complains if they are capitalized or end in punctuation or a newline.
func ( f * file ) lintErrorStrings ( ) {
f . walk ( func ( node ast . Node ) bool {
ce , ok := node . ( * ast . CallExpr )
if ! ok {
return true
}
if ! isPkgDot ( ce . Fun , "errors" , "New" ) && ! isPkgDot ( ce . Fun , "fmt" , "Errorf" ) {
return true
}
if len ( ce . Args ) < 1 {
return true
}
str , ok := ce . Args [ 0 ] . ( * ast . BasicLit )
if ! ok || str . Kind != token . STRING {
return true
}
s , _ := strconv . Unquote ( str . Value ) // can assume well-formed Go
if s == "" {
return true
}
clean , conf := lintErrorString ( s )
if clean {
return true
}
f . errorf ( str , conf , link ( styleGuideBase + "#error-strings" ) , category ( "errors" ) ,
"error strings should not be capitalized or end with punctuation or a newline" )
return true
} )
}
// lintReceiverNames examines receiver names. It complains about inconsistent
// names used for the same type and names such as "this".
func ( f * file ) lintReceiverNames ( ) {
typeReceiver := map [ string ] string { }
f . walk ( func ( n ast . Node ) bool {
fn , ok := n . ( * ast . FuncDecl )
if ! ok || fn . Recv == nil || len ( fn . Recv . List ) == 0 {
return true
}
names := fn . Recv . List [ 0 ] . Names
if len ( names ) < 1 {
return true
}
name := names [ 0 ] . Name
const ref = styleGuideBase + "#receiver-names"
if name == "_" {
f . errorf ( n , 1 , link ( ref ) , category ( "naming" ) , ` receiver name should not be an underscore, omit the name if it is unused ` )
return true
}
if name == "this" || name == "self" {
f . errorf ( n , 1 , link ( ref ) , category ( "naming" ) , ` receiver name should be a reflection of its identity; don't use generic names such as "this" or "self" ` )
return true
}
recv := receiverType ( fn )
if prev , ok := typeReceiver [ recv ] ; ok && prev != name {
f . errorf ( n , 1 , link ( ref ) , category ( "naming" ) , "receiver name %s should be consistent with previous receiver name %s for %s" , name , prev , recv )
return true
}
typeReceiver [ recv ] = name
return true
} )
}
// lintIncDec examines statements that increment or decrement a variable.
// It complains if they don't use x++ or x--.
func ( f * file ) lintIncDec ( ) {
f . walk ( func ( n ast . Node ) bool {
as , ok := n . ( * ast . AssignStmt )
if ! ok {
return true
}
if len ( as . Lhs ) != 1 {
return true
}
if ! isOne ( as . Rhs [ 0 ] ) {
return true
}
var suffix string
switch as . Tok {
case token . ADD_ASSIGN :
suffix = "++"
case token . SUB_ASSIGN :
suffix = "--"
default :
return true
}
f . errorf ( as , 0.8 , category ( "unary-op" ) , "should replace %s with %s%s" , f . render ( as ) , f . render ( as . Lhs [ 0 ] ) , suffix )
return true
} )
}
// lintErrorReturn examines function declarations that return an error.
// It complains if the error isn't the last parameter.
func ( f * file ) lintErrorReturn ( ) {
f . walk ( func ( n ast . Node ) bool {
fn , ok := n . ( * ast . FuncDecl )
if ! ok || fn . Type . Results == nil {
return true
}
ret := fn . Type . Results . List
if len ( ret ) <= 1 {
return true
}
if isIdent ( ret [ len ( ret ) - 1 ] . Type , "error" ) {
return true
}
// An error return parameter should be the last parameter.
// Flag any error parameters found before the last.
for _ , r := range ret [ : len ( ret ) - 1 ] {
if isIdent ( r . Type , "error" ) {
f . errorf ( fn , 0.9 , category ( "arg-order" ) , "error should be the last type when returning multiple items" )
break // only flag one
}
}
return true
} )
}
// lintUnexportedReturn examines exported function declarations.
// It complains if any return an unexported type.
func ( f * file ) lintUnexportedReturn ( ) {
f . walk ( func ( n ast . Node ) bool {
fn , ok := n . ( * ast . FuncDecl )
if ! ok {
return true
}
if fn . Type . Results == nil {
return false
}
if ! fn . Name . IsExported ( ) {
return false
}
thing := "func"
if fn . Recv != nil && len ( fn . Recv . List ) > 0 {
thing = "method"
if ! ast . IsExported ( receiverType ( fn ) ) {
// Don't report exported methods of unexported types,
// such as private implementations of sort.Interface.
return false
}
}
for _ , ret := range fn . Type . Results . List {
typ := f . pkg . typeOf ( ret . Type )
if exportedType ( typ ) {
continue
}
f . errorf ( ret . Type , 0.8 , category ( "unexported-type-in-api" ) ,
"exported %s %s returns unexported type %s, which can be annoying to use" ,
thing , fn . Name . Name , typ )
break // only flag one
}
return false
} )
}
// exportedType reports whether typ is an exported type.
// It is imprecise, and will err on the side of returning true,
// such as for composite types.
func exportedType ( typ types . Type ) bool {
switch T := typ . ( type ) {
case * types . Named :
// Builtin types have no package.
return T . Obj ( ) . Pkg ( ) == nil || T . Obj ( ) . Exported ( )
case * types . Map :
return exportedType ( T . Key ( ) ) && exportedType ( T . Elem ( ) )
case interface {
Elem ( ) types . Type
} : // array, slice, pointer, chan
return exportedType ( T . Elem ( ) )
}
// Be conservative about other types, such as struct, interface, etc.
return true
}
// timeSuffixes is a list of name suffixes that imply a time unit.
// This is not an exhaustive list.
var timeSuffixes = [ ] string {
"Sec" , "Secs" , "Seconds" ,
"Msec" , "Msecs" ,
"Milli" , "Millis" , "Milliseconds" ,
"Usec" , "Usecs" , "Microseconds" ,
"MS" , "Ms" ,
}
func ( f * file ) lintTimeNames ( ) {
f . walk ( func ( node ast . Node ) bool {
v , ok := node . ( * ast . ValueSpec )
if ! ok {
return true
}
for _ , name := range v . Names {
origTyp := f . pkg . typeOf ( name )
// Look for time.Duration or *time.Duration;
// the latter is common when using flag.Duration.
typ := origTyp
if pt , ok := typ . ( * types . Pointer ) ; ok {
typ = pt . Elem ( )
}
if ! f . pkg . isNamedType ( typ , "time" , "Duration" ) {
continue
}
suffix := ""
for _ , suf := range timeSuffixes {
if strings . HasSuffix ( name . Name , suf ) {
suffix = suf
break
}
}
if suffix == "" {
continue
}
f . errorf ( v , 0.9 , category ( "time" ) , "var %s is of type %v; don't use unit-specific suffix %q" , name . Name , origTyp , suffix )
}
return true
} )
}
// lintContextKeyTypes checks for call expressions to context.WithValue with
// basic types used for the key argument.
// See: https://golang.org/issue/17293
func ( f * file ) lintContextKeyTypes ( ) {
f . walk ( func ( node ast . Node ) bool {
switch node := node . ( type ) {
case * ast . CallExpr :
f . checkContextKeyType ( node )
}
return true
} )
}
// checkContextKeyType reports an error if the call expression calls
// context.WithValue with a key argument of basic type.
func ( f * file ) checkContextKeyType ( x * ast . CallExpr ) {
sel , ok := x . Fun . ( * ast . SelectorExpr )
if ! ok {
return
}
pkg , ok := sel . X . ( * ast . Ident )
if ! ok || pkg . Name != "context" {
return
}
if sel . Sel . Name != "WithValue" {
return
}
// key is second argument to context.WithValue
if len ( x . Args ) != 3 {
return
}
key := f . pkg . typesInfo . Types [ x . Args [ 1 ] ]
if ktyp , ok := key . Type . ( * types . Basic ) ; ok && ktyp . Kind ( ) != types . Invalid {
f . errorf ( x , 1.0 , category ( "context" ) , fmt . Sprintf ( "should not use basic type %s as key in context.WithValue" , key . Type ) )
}
}
// lintContextArgs examines function declarations that contain an
// argument with a type of context.Context
// It complains if that argument isn't the first parameter.
func ( f * file ) lintContextArgs ( ) {
f . walk ( func ( n ast . Node ) bool {
fn , ok := n . ( * ast . FuncDecl )
if ! ok || len ( fn . Type . Params . List ) <= 1 {
return true
}
// A context.Context should be the first parameter of a function.
// Flag any that show up after the first.
for _ , arg := range fn . Type . Params . List [ 1 : ] {
if isPkgDot ( arg . Type , "context" , "Context" ) {
f . errorf ( fn , 0.9 , link ( "https://golang.org/pkg/context/" ) , category ( "arg-order" ) , "context.Context should be the first parameter of a function" )
break // only flag one
}
}
return true
} )
}
// containsComments returns whether the interval [start, end) contains any
// comments without "// MATCH " prefix.
func ( f * file ) containsComments ( start , end token . Pos ) bool {
for _ , cgroup := range f . f . Comments {
comments := cgroup . List
if comments [ 0 ] . Slash >= end {
// All comments starting with this group are after end pos.
return false
}
if comments [ len ( comments ) - 1 ] . Slash < start {
// Comments group ends before start pos.
continue
}
for _ , c := range comments {
if start <= c . Slash && c . Slash < end && ! strings . HasPrefix ( c . Text , "// MATCH " ) {
return true
}
}
}
return false
}
// receiverType returns the named type of the method receiver, sans "*",
// or "invalid-type" if fn.Recv is ill formed.
func receiverType ( fn * ast . FuncDecl ) string {
switch e := fn . Recv . List [ 0 ] . Type . ( type ) {
case * ast . Ident :
return e . Name
case * ast . StarExpr :
if id , ok := e . X . ( * ast . Ident ) ; ok {
return id . Name
}
}
// The parser accepts much more than just the legal forms.
return "invalid-type"
}
func ( f * file ) walk ( fn func ( ast . Node ) bool ) {
ast . Walk ( walker ( fn ) , f . f )
}
func ( f * file ) render ( x interface { } ) string {
var buf bytes . Buffer
if err := printer . Fprint ( & buf , f . fset , x ) ; err != nil {
panic ( err )
}
return buf . String ( )
}
func ( f * file ) debugRender ( x interface { } ) string {
var buf bytes . Buffer
if err := ast . Fprint ( & buf , f . fset , x , nil ) ; err != nil {
panic ( err )
}
return buf . String ( )
}
// walker adapts a function to satisfy the ast.Visitor interface.
// The function return whether the walk should proceed into the node's children.
type walker func ( ast . Node ) bool
func ( w walker ) Visit ( node ast . Node ) ast . Visitor {
if w ( node ) {
return w
}
return nil
}
func isIdent ( expr ast . Expr , ident string ) bool {
id , ok := expr . ( * ast . Ident )
return ok && id . Name == ident
}
// isBlank returns whether id is the blank identifier "_".
// If id == nil, the answer is false.
func isBlank ( id * ast . Ident ) bool { return id != nil && id . Name == "_" }
func isPkgDot ( expr ast . Expr , pkg , name string ) bool {
sel , ok := expr . ( * ast . SelectorExpr )
return ok && isIdent ( sel . X , pkg ) && isIdent ( sel . Sel , name )
}
func isOne ( expr ast . Expr ) bool {
lit , ok := expr . ( * ast . BasicLit )
return ok && lit . Kind == token . INT && lit . Value == "1"
}
func isCgoExported ( f * ast . FuncDecl ) bool {
if f . Recv != nil || f . Doc == nil {
return false
}
cgoExport := regexp . MustCompile ( fmt . Sprintf ( "(?m)^//export %s$" , regexp . QuoteMeta ( f . Name . Name ) ) )
for _ , c := range f . Doc . List {
if cgoExport . MatchString ( c . Text ) {
return true
}
}
return false
}
var basicTypeKinds = map [ types . BasicKind ] string {
types . UntypedBool : "bool" ,
types . UntypedInt : "int" ,
types . UntypedRune : "rune" ,
types . UntypedFloat : "float64" ,
types . UntypedComplex : "complex128" ,
types . UntypedString : "string" ,
}
// isUntypedConst reports whether expr is an untyped constant,
// and indicates what its default type is.
// scope may be nil.
func ( f * file ) isUntypedConst ( expr ast . Expr ) ( defType string , ok bool ) {
// Re-evaluate expr outside of its context to see if it's untyped.
// (An expr evaluated within, for example, an assignment context will get the type of the LHS.)
exprStr := f . render ( expr )
tv , err := types . Eval ( f . fset , f . pkg . typesPkg , expr . Pos ( ) , exprStr )
if err != nil {
return "" , false
}
if b , ok := tv . Type . ( * types . Basic ) ; ok {
if dt , ok := basicTypeKinds [ b . Kind ( ) ] ; ok {
return dt , true
}
}
return "" , false
}
// firstLineOf renders the given node and returns its first line.
// It will also match the indentation of another node.
func ( f * file ) firstLineOf ( node , match ast . Node ) string {
line := f . render ( node )
if i := strings . Index ( line , "\n" ) ; i >= 0 {
line = line [ : i ]
}
return f . indentOf ( match ) + line
}
func ( f * file ) indentOf ( node ast . Node ) string {
line := srcLine ( f . src , f . fset . Position ( node . Pos ( ) ) )
for i , r := range line {
switch r {
case ' ' , '\t' :
default :
return line [ : i ]
}
}
return line // unusual or empty line
}
func ( f * file ) srcLineWithMatch ( node ast . Node , pattern string ) ( m [ ] string ) {
line := srcLine ( f . src , f . fset . Position ( node . Pos ( ) ) )
line = strings . TrimSuffix ( line , "\n" )
rx := regexp . MustCompile ( pattern )
return rx . FindStringSubmatch ( line )
}
// imports returns true if the current file imports the specified package path.
func ( f * file ) imports ( importPath string ) bool {
all := astutil . Imports ( f . fset , f . f )
for _ , p := range all {
for _ , i := range p {
uq , err := strconv . Unquote ( i . Path . Value )
if err == nil && importPath == uq {
return true
}
}
}
return false
}
// srcLine returns the complete line at p, including the terminating newline.
func srcLine ( src [ ] byte , p token . Position ) string {
// Run to end of line in both directions if not at line start/end.
lo , hi := p . Offset , p . Offset + 1
for lo > 0 && src [ lo - 1 ] != '\n' {
lo --
}
for hi < len ( src ) && src [ hi - 1 ] != '\n' {
hi ++
}
return string ( src [ lo : hi ] )
}