content/2011/lex/lex1.oldgo - talks - Git at Google

 // 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.

 // reformatted, slightly edited version of lex.go from weekly.11-06-23

 package template

 import (
 	"fmt"
 	"strings"
 	"unicode"
 	"utf8"
 )

 // item represents a token returned from the scanner.
 type item struct {
 	typ itemType  // Type, such as itemNumber.
 	val string    // Value, such as "23.2".
 }

 func (i item) String() string {
 	switch i.typ {
 	case itemEOF:
 		return "EOF"
 	case itemError:
 		return i.val
 	}
 	if len(i.val) > 10 {
 		return fmt.Sprintf("%.10q...", i.val)
 	}
 	return fmt.Sprintf("%q", i.val)
 }

 // itemType identifies the type of lex items.
 type itemType int

 const (
 	itemError itemType = iota // error occurred;
 	                          // value is text of error
 	itemDot                   // the cursor, spelled '.'
 	itemEOF
 	itemElse       // else keyword
 	itemEnd        // end keyword
 	itemField      // identifier, starting with '.'
 	itemIdentifier // identifier
 	itemIf         // if keyword
 	itemLeftMeta   // left meta-string
 	itemNumber     // number
 	itemPipe       // pipe symbol
 	itemRange      // range keyword
 	itemRawString  // raw quoted string (includes quotes)
 	itemRightMeta  // right meta-string
 	itemString     // quoted string (includes quotes)
 	itemText       // plain text
 )

 // Make the types prettyprint.
 var itemName = map[itemType]string{
 	itemError:      "error",
 	itemDot:        ".",
 	itemEOF:        "EOF",
 	itemElse:       "else",
 	itemEnd:        "end",
 	itemField:      "field",
 	itemIdentifier: "identifier",
 	itemIf:         "if",
 	itemLeftMeta:   "left meta",
 	itemNumber:     "number",
 	itemPipe:       "pipe",
 	itemRange:      "range",
 	itemRawString:  "raw string",
 	itemRightMeta:  "rightMeta",
 	itemString:     "string",
 	itemText:       "text",
 }

 func (i itemType) String() string {
 	s := itemName[i]
 	if s == "" {
 		return fmt.Sprintf("item%d", int(i))
 	}
 	return s
 }

 var key = map[string]itemType{
 	".":     itemDot,
 	"else":  itemElse,
 	"end":   itemEnd,
 	"if":    itemIf,
 	"range": itemRange,
 }

 const eof = -1

 // stateFn represents the state of the scanner
 // as a function that returns the next state.
 type stateFn func(*lexer) stateFn

 // lexer holds the state of the scanner.
 type lexer struct {
 	name  string    // used only for error reports.
 	input string    // the string being scanned.
 	start int       // start position of this item.
 	pos   int       // current position in the input.
 	width int       // width of last rune read from input.
 	items chan item // channel of scanned items.
 }

 // next returns the next rune in the input.
 func (l *lexer) next() (rune int) {
 	if l.pos >= len(l.input) {
 		l.width = 0
 		return eof
 	}
 	rune, l.width =
 		utf8.DecodeRuneInString(l.input[l.pos:])
 	l.pos += l.width
 	return rune
 }

 // peek returns but does not consume
 // the next rune in the input.
 func (l *lexer) peek() int {
 	rune := l.next()
 	l.backup()
 	return rune
 }

 // backup steps back one rune.
 // Can be called only once per call of next.
 func (l *lexer) backup() {
 	l.pos -= l.width
 }

 // emit passes an item back to the client.
 func (l *lexer) emit(t itemType) {
 	l.items <- item{t, l.input[l.start:l.pos]}
 	l.start = l.pos
 }

 // ignore skips over the pending input before this point.
 func (l *lexer) ignore() {
 	l.start = l.pos
 }

 // accept consumes the next rune
 // if it's from the valid set.
 func (l *lexer) accept(valid string) bool {
 	if strings.IndexRune(valid, l.next()) >= 0 {
 		return true
 	}
 	l.backup()
 	return false
 }

 // acceptRun consumes a run of runes from the valid set.
 func (l *lexer) acceptRun(valid string) {
 	for strings.IndexRune(valid, l.next()) >= 0 {
 	}
 	l.backup()
 }

 // lineNumber reports which line we're on. Doing it this way
 // means we don't have to worry about peek double counting.
 func (l *lexer) lineNumber() int {
 	return 1 + strings.Count(l.input[:l.pos], "\n")
 }

 // error returns an error token and terminates the scan
 // by passing back a nil pointer that will be the next
 // state, terminating l.run.
 func (l *lexer) errorf(format string, args ...interface{})
   stateFn {
 	l.items <- item{
 		itemError,
 		fmt.Sprintf(format, args...),
 	}
 	return nil
 }

 // run lexes the input by executing state functions until
 // the state is nil.
 func (l *lexer) run() {
 	for state := lexText; state != nil; {
 		state = state(l)
 	}
 	close(l.items) // No more tokens will be delivered.
 }

 // lex launches a new scanner and returns the channel of items.
 func lex(name, input string) (*lexer, chan item) {
 	l := &lexer{
 		name:  name,
 		input: input,
 		items: make(chan item),
 	}
 	go l.run()  // Concurrently run state machine.
 	return l, l.items
 }

 // state functions

 const leftMeta = "{{"
 const rightMeta = "}}"

 // lexText scans until a metacharacter
 func lexText(l *lexer) stateFn {
 	for {
 		if strings.HasPrefix(l.input[l.pos:], leftMeta) {
 			if l.pos > l.start {
 				l.emit(itemText)
 			}
 			return lexLeftMeta    // Next state.
 		}
 		if l.next() == eof { break }
 	}
 	// Correctly reached EOF.
 	if l.pos > l.start {
 		l.emit(itemText)
 	}
 	l.emit(itemEOF)  // Useful to make EOF a token.
 	return nil       // Stop the run loop.
 }

 // leftMeta scans the left "metacharacter", which is known to be present.
 func lexLeftMeta(l *lexer) stateFn {
 	l.pos += len(leftMeta)
 	l.emit(itemLeftMeta)
 	return lexInsideAction    // Now inside {{ }}.
 }

 // rightMeta scans the right "metacharacter", which is known to be present.
 func lexRightMeta(l *lexer) stateFn {
 	l.pos += len(rightMeta)
 	l.emit(itemRightMeta)
 	return lexText
 }

 // lexInsideAction scans the elements inside "metacharacters".
 func lexInsideAction(l *lexer) stateFn {
 	// Either number, quoted string, or identifier.
 	// Spaces separate and are ignored.
 	// Pipe symbols separate and are emitted.
 	for {
 		if strings.HasPrefix(l.input[l.pos:], rightMeta) {
 			return lexRightMeta
 		}
 		switch r := l.next(); {
 		case r == eof || r == '\n':
 			return l.errorf("unclosed action")
 		case isSpace(r):
 			l.ignore()
 		case r == '|':
 			l.emit(itemPipe)
 		case r == '"':
 			return lexQuote
 		case r == '`':
 			return lexRawQuote
 		case r == '.':
 			// special look-ahead for ".field" so we don't break l.backup().
 			if l.pos < len(l.input) {
 				r := l.input[l.pos]
 				if r < '0' || '9' < r {
 					return lexIdentifier // itemDot comes from the keyword table.
 				}
 			}
 			fallthrough // '.' can start a number.
 		case r == '+' || r == '-' || '0' <= r && r <= '9':
 			l.backup()
 			return lexNumber
 		case isAlphaNumeric(r):
 			l.backup()
 			return lexIdentifier
 		default:
 			return l.errorf("unrecognized character in action: %#U", r)
 		}
 	}
 	return nil
 }

 // lexIdentifier scans an alphanumeric or field.
 func lexIdentifier(l *lexer) stateFn {
 Loop:
 	for {
 		switch r := l.next(); {
 		case isAlphaNumeric(r):
 			// absorb
 		default:
 			l.backup()
 			word := l.input[l.start:l.pos]
 			switch {
 			case key[word] != itemError:
 				l.emit(key[word])
 			case word[0] == '.':
 				l.emit(itemField)
 			default:
 				l.emit(itemIdentifier)
 			}
 			break Loop
 		}
 	}
 	return lexInsideAction
 }

 // lexNumber scans a number: decimal, octal, hex, float, or imaginary.  This
 // isn't a perfect number scanner - for instance it accepts "." and "0x0.2"
 // and "089" - but when it's wrong the input is invalid and the parser (via
 // strconv) will notice.
 // TODO: without expressions you can do imaginary but not complex.
 func lexNumber(l *lexer) stateFn {
 	// Optional leading sign.
 	l.accept("+-")
 	// Is it hex?
 	digits := "0123456789"
 	if l.accept("0") && l.accept("xX") {
 		digits = "0123456789abcdefABCDEF"
 	}
 	l.acceptRun(digits)
 	if l.accept(".") {
 		l.acceptRun(digits)
 	}
 	if l.accept("eE") {
 		l.accept("+-")
 		l.acceptRun("0123456789")
 	}
 	// Is it imaginary?
 	l.accept("i")
 	// Next thing mustn't be alphanumeric.
 	if isAlphaNumeric(l.peek()) {
 		l.next()
 		return l.errorf("bad number syntax: %q",
 			l.input[l.start:l.pos])
 	}
 	l.emit(itemNumber)
 	return lexInsideAction
 }

 // lexQuote scans a quoted string.
 func lexQuote(l *lexer) stateFn {
 Loop:
 	for {
 		switch l.next() {
 		case '\\':
 			if r := l.next(); r != eof && r != '\n' {
 				break
 			}
 			fallthrough
 		case eof, '\n':
 			return l.errorf("unterminated quoted string")
 		case '"':
 			break Loop
 		}
 	}
 	l.emit(itemString)
 	return lexInsideAction
 }

 // lexRawQuote scans a raw quoted string.
 func lexRawQuote(l *lexer) stateFn {
 Loop:
 	for {
 		switch l.next() {
 		case eof, '\n':
 			return l.errorf("unterminated raw quoted string")
 		case '`':
 			break Loop
 		}
 	}
 	l.emit(itemRawString)
 	return lexInsideAction
 }

 // isSpace reports whether r is a space character.
 func isSpace(r int) bool {
 	switch r {
 	case ' ', '\t', '\n', '\r':
 		return true
 	}
 	return false
 }

 // isAlphaNumeric reports whether r is an alphabetic, digit, or underscore.
 func isAlphaNumeric(r int) bool {
 	return r == '_' || unicode.IsLetter(r) || unicode.IsDigit(r)
 }
	// 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.

	// reformatted, slightly edited version of lex.go from weekly.11-06-23

	package template

	import (
	"fmt"
	"strings"
	"unicode"
	"utf8"
	)

	// item represents a token returned from the scanner.
	type item struct {
	typ itemType // Type, such as itemNumber.
	val string // Value, such as "23.2".
	}

	func (i item) String() string {
	switch i.typ {
	case itemEOF:
	return "EOF"
	case itemError:
	return i.val
	}
	if len(i.val) > 10 {
	return fmt.Sprintf("%.10q...", i.val)
	}
	return fmt.Sprintf("%q", i.val)
	}

	// itemType identifies the type of lex items.
	type itemType int

	const (
	itemError itemType = iota // error occurred;
	// value is text of error
	itemDot // the cursor, spelled '.'
	itemEOF
	itemElse // else keyword
	itemEnd // end keyword
	itemField // identifier, starting with '.'
	itemIdentifier // identifier
	itemIf // if keyword
	itemLeftMeta // left meta-string
	itemNumber // number
	itemPipe // pipe symbol
	itemRange // range keyword
	itemRawString // raw quoted string (includes quotes)
	itemRightMeta // right meta-string
	itemString // quoted string (includes quotes)
	itemText // plain text
	)

	// Make the types prettyprint.
	var itemName = map[itemType]string{
	itemError: "error",
	itemDot: ".",
	itemEOF: "EOF",
	itemElse: "else",
	itemEnd: "end",
	itemField: "field",
	itemIdentifier: "identifier",
	itemIf: "if",
	itemLeftMeta: "left meta",
	itemNumber: "number",
	itemPipe: "pipe",
	itemRange: "range",
	itemRawString: "raw string",
	itemRightMeta: "rightMeta",
	itemString: "string",
	itemText: "text",
	}

	func (i itemType) String() string {
	s := itemName[i]
	if s == "" {
	return fmt.Sprintf("item%d", int(i))
	}
	return s
	}

	var key = map[string]itemType{
	".": itemDot,
	"else": itemElse,
	"end": itemEnd,
	"if": itemIf,
	"range": itemRange,
	}

	const eof = -1

	// stateFn represents the state of the scanner
	// as a function that returns the next state.
	type stateFn func(*lexer) stateFn

	// lexer holds the state of the scanner.
	type lexer struct {
	name string // used only for error reports.
	input string // the string being scanned.
	start int // start position of this item.
	pos int // current position in the input.
	width int // width of last rune read from input.
	items chan item // channel of scanned items.
	}

	// next returns the next rune in the input.
	func (l *lexer) next() (rune int) {
	if l.pos >= len(l.input) {
	l.width = 0
	return eof
	}
	rune, l.width =
	utf8.DecodeRuneInString(l.input[l.pos:])
	l.pos += l.width
	return rune
	}

	// peek returns but does not consume
	// the next rune in the input.
	func (l *lexer) peek() int {
	rune := l.next()
	l.backup()
	return rune
	}

	// backup steps back one rune.
	// Can be called only once per call of next.
	func (l *lexer) backup() {
	l.pos -= l.width
	}

	// emit passes an item back to the client.
	func (l *lexer) emit(t itemType) {
	l.items <- item{t, l.input[l.start:l.pos]}
	l.start = l.pos
	}

	// ignore skips over the pending input before this point.
	func (l *lexer) ignore() {
	l.start = l.pos
	}

	// accept consumes the next rune
	// if it's from the valid set.
	func (l *lexer) accept(valid string) bool {
	if strings.IndexRune(valid, l.next()) >= 0 {
	return true
	}
	l.backup()
	return false
	}

	// acceptRun consumes a run of runes from the valid set.
	func (l *lexer) acceptRun(valid string) {
	for strings.IndexRune(valid, l.next()) >= 0 {
	}
	l.backup()
	}

	// lineNumber reports which line we're on. Doing it this way
	// means we don't have to worry about peek double counting.
	func (l *lexer) lineNumber() int {
	return 1 + strings.Count(l.input[:l.pos], "\n")
	}

	// error returns an error token and terminates the scan
	// by passing back a nil pointer that will be the next
	// state, terminating l.run.
	func (l *lexer) errorf(format string, args ...interface{})
	stateFn {
	l.items <- item{
	itemError,
	fmt.Sprintf(format, args...),
	}
	return nil
	}

	// run lexes the input by executing state functions until
	// the state is nil.
	func (l *lexer) run() {
	for state := lexText; state != nil; {
	state = state(l)
	}
	close(l.items) // No more tokens will be delivered.
	}

	// lex launches a new scanner and returns the channel of items.
	func lex(name, input string) (*lexer, chan item) {
	l := &lexer{
	name: name,
	input: input,
	items: make(chan item),
	}
	go l.run() // Concurrently run state machine.
	return l, l.items
	}

	// state functions

	const leftMeta = "{{"
	const rightMeta = "}}"

	// lexText scans until a metacharacter
	func lexText(l *lexer) stateFn {
	for {
	if strings.HasPrefix(l.input[l.pos:], leftMeta) {
	if l.pos > l.start {
	l.emit(itemText)
	}
	return lexLeftMeta // Next state.
	}
	if l.next() == eof { break }
	}
	// Correctly reached EOF.
	if l.pos > l.start {
	l.emit(itemText)
	}
	l.emit(itemEOF) // Useful to make EOF a token.
	return nil // Stop the run loop.
	}

	// leftMeta scans the left "metacharacter", which is known to be present.
	func lexLeftMeta(l *lexer) stateFn {
	l.pos += len(leftMeta)
	l.emit(itemLeftMeta)
	return lexInsideAction // Now inside {{ }}.
	}

	// rightMeta scans the right "metacharacter", which is known to be present.
	func lexRightMeta(l *lexer) stateFn {
	l.pos += len(rightMeta)
	l.emit(itemRightMeta)
	return lexText
	}

	// lexInsideAction scans the elements inside "metacharacters".
	func lexInsideAction(l *lexer) stateFn {
	// Either number, quoted string, or identifier.
	// Spaces separate and are ignored.
	// Pipe symbols separate and are emitted.
	for {
	if strings.HasPrefix(l.input[l.pos:], rightMeta) {
	return lexRightMeta
	}
	switch r := l.next(); {
	case r == eof \|\| r == '\n':
	return l.errorf("unclosed action")
	case isSpace(r):
	l.ignore()
	case r == '\|':
	l.emit(itemPipe)
	case r == '"':
	return lexQuote
	case r == '`':
	return lexRawQuote
	case r == '.':
	// special look-ahead for ".field" so we don't break l.backup().
	if l.pos < len(l.input) {
	r := l.input[l.pos]
	if r < '0' \|\| '9' < r {
	return lexIdentifier // itemDot comes from the keyword table.
	}
	}
	fallthrough // '.' can start a number.
	case r == '+' \|\| r == '-' \|\| '0' <= r && r <= '9':
	l.backup()
	return lexNumber
	case isAlphaNumeric(r):
	l.backup()
	return lexIdentifier
	default:
	return l.errorf("unrecognized character in action: %#U", r)
	}
	}
	return nil
	}

	// lexIdentifier scans an alphanumeric or field.
	func lexIdentifier(l *lexer) stateFn {
	Loop:
	for {
	switch r := l.next(); {
	case isAlphaNumeric(r):
	// absorb
	default:
	l.backup()
	word := l.input[l.start:l.pos]
	switch {
	case key[word] != itemError:
	l.emit(key[word])
	case word[0] == '.':
	l.emit(itemField)
	default:
	l.emit(itemIdentifier)
	}
	break Loop
	}
	}
	return lexInsideAction
	}

	// lexNumber scans a number: decimal, octal, hex, float, or imaginary. This
	// isn't a perfect number scanner - for instance it accepts "." and "0x0.2"
	// and "089" - but when it's wrong the input is invalid and the parser (via
	// strconv) will notice.
	// TODO: without expressions you can do imaginary but not complex.
	func lexNumber(l *lexer) stateFn {
	// Optional leading sign.
	l.accept("+-")
	// Is it hex?
	digits := "0123456789"
	if l.accept("0") && l.accept("xX") {
	digits = "0123456789abcdefABCDEF"
	}
	l.acceptRun(digits)
	if l.accept(".") {
	l.acceptRun(digits)
	}
	if l.accept("eE") {
	l.accept("+-")
	l.acceptRun("0123456789")
	}
	// Is it imaginary?
	l.accept("i")
	// Next thing mustn't be alphanumeric.
	if isAlphaNumeric(l.peek()) {
	l.next()
	return l.errorf("bad number syntax: %q",
	l.input[l.start:l.pos])
	}
	l.emit(itemNumber)
	return lexInsideAction
	}

	// lexQuote scans a quoted string.
	func lexQuote(l *lexer) stateFn {
	Loop:
	for {
	switch l.next() {
	case '\\':
	if r := l.next(); r != eof && r != '\n' {
	break
	}
	fallthrough
	case eof, '\n':
	return l.errorf("unterminated quoted string")
	case '"':
	break Loop
	}
	}
	l.emit(itemString)
	return lexInsideAction
	}

	// lexRawQuote scans a raw quoted string.
	func lexRawQuote(l *lexer) stateFn {
	Loop:
	for {
	switch l.next() {
	case eof, '\n':
	return l.errorf("unterminated raw quoted string")
	case '`':
	break Loop
	}
	}
	l.emit(itemRawString)
	return lexInsideAction
	}

	// isSpace reports whether r is a space character.
	func isSpace(r int) bool {
	switch r {
	case ' ', '\t', '\n', '\r':
	return true
	}
	return false
	}

	// isAlphaNumeric reports whether r is an alphabetic, digit, or underscore.
	func isAlphaNumeric(r int) bool {
	return r == '_' \|\| unicode.IsLetter(r) \|\| unicode.IsDigit(r)
	}