435 lines
9.5 KiB
Go
435 lines
9.5 KiB
Go
package qcode
|
|
|
|
import (
|
|
"fmt"
|
|
"strings"
|
|
"unicode"
|
|
"unicode/utf8"
|
|
)
|
|
|
|
// Pos represents a byte position in the original input text from which
|
|
// this template was parsed.
|
|
type Pos int
|
|
|
|
func (p Pos) Position() Pos {
|
|
return p
|
|
}
|
|
|
|
// item represents a token or text string returned from the scanner.
|
|
type item struct {
|
|
typ itemType // The type of this item.
|
|
pos Pos // The starting position, in bytes, of this item in the input string.
|
|
val string // The value of this item.
|
|
line int // The line number at the start of this item.
|
|
}
|
|
|
|
func (i *item) String() string {
|
|
var v string
|
|
|
|
switch i.typ {
|
|
case itemEOF:
|
|
v = "EOF"
|
|
case itemError:
|
|
v = "error"
|
|
case itemName:
|
|
v = "name"
|
|
case itemQuery:
|
|
v = "query"
|
|
case itemMutation:
|
|
v = "mutation"
|
|
case itemSub:
|
|
v = "subscription"
|
|
case itemPunctuator:
|
|
v = "punctuator"
|
|
case itemDirective:
|
|
v = "directive"
|
|
case itemVariable:
|
|
v = "variable"
|
|
case itemIntVal:
|
|
v = "int"
|
|
case itemFloatVal:
|
|
v = "float"
|
|
case itemStringVal:
|
|
v = "string"
|
|
}
|
|
return fmt.Sprintf("%s %q", v, i.val)
|
|
}
|
|
|
|
// itemType identifies the type of lex items.
|
|
type itemType int
|
|
|
|
const (
|
|
itemError itemType = iota // error occurred; value is text of error
|
|
itemEOF
|
|
itemName
|
|
itemQuery
|
|
itemMutation
|
|
itemSub
|
|
itemPunctuator
|
|
itemArgsOpen
|
|
itemArgsClose
|
|
itemListOpen
|
|
itemListClose
|
|
itemObjOpen
|
|
itemObjClose
|
|
itemColon
|
|
itemEquals
|
|
itemDirective
|
|
itemVariable
|
|
itemSpread
|
|
itemIntVal
|
|
itemFloatVal
|
|
itemStringVal
|
|
itemBoolVal
|
|
)
|
|
|
|
// !$():=@[]{|}
|
|
var punctuators = map[rune]itemType{
|
|
'{': itemObjOpen,
|
|
'}': itemObjClose,
|
|
'[': itemListOpen,
|
|
']': itemListClose,
|
|
'(': itemArgsOpen,
|
|
')': itemArgsClose,
|
|
':': itemColon,
|
|
'=': itemEquals,
|
|
}
|
|
|
|
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 // the name of the input; used only for error reports
|
|
input string // the string being scanned
|
|
pos Pos // current position in the input
|
|
start Pos // start position of this item
|
|
width Pos // width of last rune read from input
|
|
items []item // array of scanned items
|
|
line int // 1+number of newlines seen
|
|
}
|
|
|
|
// next returns the next rune in the input.
|
|
func (l *lexer) next() rune {
|
|
if int(l.pos) >= len(l.input) {
|
|
l.width = 0
|
|
return eof
|
|
}
|
|
r, w := utf8.DecodeRuneInString(l.input[l.pos:])
|
|
l.width = Pos(w)
|
|
l.pos += l.width
|
|
if r == '\n' {
|
|
l.line++
|
|
}
|
|
return r
|
|
}
|
|
|
|
// peek returns but does not consume the next rune in the input.
|
|
func (l *lexer) peek() rune {
|
|
r := l.next()
|
|
l.backup()
|
|
return r
|
|
}
|
|
|
|
// backup steps back one rune. Can only be called once per call of next.
|
|
func (l *lexer) backup() {
|
|
l.pos -= l.width
|
|
// Correct newline count.
|
|
if l.width == 1 && l.input[l.pos] == '\n' {
|
|
l.line--
|
|
}
|
|
}
|
|
|
|
func (l *lexer) current() string {
|
|
return l.input[l.start:l.pos]
|
|
}
|
|
|
|
// emit passes an item back to the client.
|
|
func (l *lexer) emit(t itemType) {
|
|
l.items = append(l.items, item{t, l.start, l.input[l.start:l.pos], l.line})
|
|
// Some items contain text internally. If so, count their newlines.
|
|
switch t {
|
|
case itemName:
|
|
l.line += strings.Count(l.input[l.start:l.pos], "\n")
|
|
}
|
|
l.start = l.pos
|
|
}
|
|
|
|
// ignore skips over the pending input before this point.
|
|
func (l *lexer) ignore() {
|
|
l.line += strings.Count(l.input[l.start:l.pos], "\n")
|
|
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.ContainsRune(valid, l.next()) {
|
|
return true
|
|
}
|
|
l.backup()
|
|
return false
|
|
}
|
|
|
|
// acceptAlphaNum consumes a run of runes while they are alpha nums
|
|
func (l *lexer) acceptAlphaNum() bool {
|
|
n := 0
|
|
for r := l.next(); isAlphaNumeric(r); r = l.next() {
|
|
n++
|
|
}
|
|
l.backup()
|
|
return (n != 0)
|
|
}
|
|
|
|
// acceptComment consumes a run of runes while till the end of line
|
|
func (l *lexer) acceptComment() {
|
|
n := 0
|
|
for r := l.next(); !isEndOfLine(r); r = l.next() {
|
|
n++
|
|
}
|
|
}
|
|
|
|
// acceptRun consumes a run of runes from the valid set.
|
|
func (l *lexer) acceptRun(valid string) {
|
|
for strings.ContainsRune(valid, l.next()) {
|
|
}
|
|
l.backup()
|
|
}
|
|
|
|
// errorf returns an error token and terminates the scan by passing
|
|
// back a nil pointer that will be the next state, terminating l.nextItem.
|
|
func (l *lexer) errorf(format string, args ...interface{}) stateFn {
|
|
l.items = append(l.items, item{itemError, l.start,
|
|
fmt.Sprintf(format, args...), l.line})
|
|
return nil
|
|
}
|
|
|
|
// lex creates a new scanner for the input string.
|
|
func lex(input string) (*lexer, error) {
|
|
l := &lexer{
|
|
input: input,
|
|
items: make([]item, 0, 100),
|
|
line: 1,
|
|
}
|
|
l.run()
|
|
|
|
if last := l.items[len(l.items)-1]; last.typ == itemError {
|
|
return nil, fmt.Errorf(last.val)
|
|
}
|
|
return l, nil
|
|
}
|
|
|
|
// run runs the state machine for the lexer.
|
|
func (l *lexer) run() {
|
|
for state := lexRoot; state != nil; {
|
|
state = state(l)
|
|
}
|
|
}
|
|
|
|
// lexInsideAction scans the elements inside action delimiters.
|
|
func lexRoot(l *lexer) stateFn {
|
|
r := l.next()
|
|
|
|
switch {
|
|
case r == eof:
|
|
l.emit(itemEOF)
|
|
return nil
|
|
case isEndOfLine(r):
|
|
l.ignore()
|
|
case isSpace(r):
|
|
l.ignore()
|
|
case r == '#':
|
|
l.ignore()
|
|
l.acceptComment()
|
|
l.ignore()
|
|
case r == '@':
|
|
l.ignore()
|
|
if l.acceptAlphaNum() {
|
|
l.emit(itemDirective)
|
|
}
|
|
case r == '$':
|
|
l.ignore()
|
|
if l.acceptAlphaNum() {
|
|
l.emit(itemVariable)
|
|
}
|
|
case strings.ContainsRune("!():=[]{|}", r):
|
|
if item, ok := punctuators[r]; ok {
|
|
l.emit(item)
|
|
} else {
|
|
l.emit(itemPunctuator)
|
|
}
|
|
case r == '"' || r == '\'':
|
|
l.backup()
|
|
return lexString
|
|
case r == '.':
|
|
if len(l.input) >= 3 {
|
|
if strings.HasSuffix(l.input[:l.pos], "...") {
|
|
l.emit(itemSpread)
|
|
return lexRoot
|
|
}
|
|
}
|
|
fallthrough // '.' can start a number.
|
|
case r == '+' || r == '-' || ('0' <= r && r <= '9'):
|
|
l.backup()
|
|
return lexNumber
|
|
case isAlphaNumeric(r):
|
|
l.backup()
|
|
return lexName
|
|
default:
|
|
return l.errorf("unrecognized character in action: %#U", r)
|
|
}
|
|
return lexRoot
|
|
}
|
|
|
|
// lexName scans a name.
|
|
func lexName(l *lexer) stateFn {
|
|
for {
|
|
r := l.next()
|
|
if r == eof {
|
|
l.emit(itemEOF)
|
|
return nil
|
|
}
|
|
if !isAlphaNumeric(r) {
|
|
l.backup()
|
|
v := l.current()
|
|
|
|
if len(v) == 0 {
|
|
switch {
|
|
case strings.EqualFold(v, "query"):
|
|
l.emit(itemQuery)
|
|
break
|
|
case strings.EqualFold(v, "mutation"):
|
|
l.emit(itemMutation)
|
|
break
|
|
case strings.EqualFold(v, "subscription"):
|
|
l.emit(itemSub)
|
|
break
|
|
}
|
|
}
|
|
|
|
switch {
|
|
case strings.EqualFold(v, "true"):
|
|
l.emit(itemBoolVal)
|
|
case strings.EqualFold(v, "false"):
|
|
l.emit(itemBoolVal)
|
|
default:
|
|
l.emit(itemName)
|
|
}
|
|
break
|
|
}
|
|
}
|
|
return lexRoot
|
|
}
|
|
|
|
// lexString scans a string.
|
|
func lexString(l *lexer) stateFn {
|
|
if l.accept("\"'") {
|
|
l.ignore()
|
|
|
|
for {
|
|
r := l.next()
|
|
if r == eof {
|
|
l.emit(itemEOF)
|
|
return nil
|
|
}
|
|
if r == '\'' || r == '"' {
|
|
l.backup()
|
|
l.emit(itemStringVal)
|
|
if l.accept("\"'") {
|
|
l.ignore()
|
|
}
|
|
break
|
|
}
|
|
}
|
|
}
|
|
return lexRoot
|
|
}
|
|
|
|
// 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.
|
|
func lexNumber(l *lexer) stateFn {
|
|
var it itemType
|
|
// Optional leading sign.
|
|
l.accept("+-")
|
|
|
|
// Is it integer
|
|
digits := "0123456789"
|
|
if l.accept(digits) {
|
|
l.acceptRun(digits)
|
|
it = itemIntVal
|
|
}
|
|
|
|
// Is it float
|
|
if l.peek() == '.' {
|
|
if l.accept(".") {
|
|
if l.accept(digits) {
|
|
l.acceptRun(digits)
|
|
it = itemFloatVal
|
|
}
|
|
} else {
|
|
l.backup()
|
|
}
|
|
}
|
|
|
|
// 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])
|
|
}
|
|
|
|
if it != 0 {
|
|
l.emit(it)
|
|
}
|
|
|
|
return lexRoot
|
|
}
|
|
|
|
// isSpace reports whether r is a space character.
|
|
func isSpace(r rune) bool {
|
|
return r == ',' || r == ' ' || r == '\t'
|
|
}
|
|
|
|
// isEndOfLine reports whether r is an end-of-line character.
|
|
func isEndOfLine(r rune) bool {
|
|
return r == '\r' || r == '\n'
|
|
}
|
|
|
|
// isAlphaNumeric reports whether r is an alphabetic, digit, or underscore.
|
|
func isAlphaNumeric(r rune) bool {
|
|
return r == '_' || unicode.IsLetter(r) || unicode.IsDigit(r)
|
|
}
|
|
|
|
/*
|
|
|
|
Copyright (c) 2009 The Go Authors. All rights reserved.
|
|
|
|
Redistribution and use in source and binary forms, with or without
|
|
modification, are permitted provided that the following conditions are
|
|
met:
|
|
|
|
* Redistributions of source code must retain the above copyright
|
|
notice, this list of conditions and the following disclaimer.
|
|
* Redistributions in binary form must reproduce the above
|
|
copyright notice, this list of conditions and the following disclaimer
|
|
in the documentation and/or other materials provided with the
|
|
distribution.
|
|
* Neither the name of Google Inc. nor the names of its
|
|
contributors may be used to endorse or promote products derived from
|
|
this software without specific prior written permission.
|
|
|
|
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
|
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
|
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
|
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
|
|
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
|
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
|
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
|
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
|
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
|
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
|
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
*/
|