summaryrefslogtreecommitdiffstats
path: root/src/go/scanner
diff options
context:
space:
mode:
Diffstat (limited to '')
-rw-r--r--src/go/scanner/errors.go124
-rw-r--r--src/go/scanner/example_test.go46
-rw-r--r--src/go/scanner/scanner.go987
-rw-r--r--src/go/scanner/scanner_test.go1126
4 files changed, 2283 insertions, 0 deletions
diff --git a/src/go/scanner/errors.go b/src/go/scanner/errors.go
new file mode 100644
index 0000000..3114f4b
--- /dev/null
+++ b/src/go/scanner/errors.go
@@ -0,0 +1,124 @@
+// 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.
+
+package scanner
+
+import (
+ "fmt"
+ "go/token"
+ "io"
+ "sort"
+)
+
+// In an ErrorList, an error is represented by an *Error.
+// The position Pos, if valid, points to the beginning of
+// the offending token, and the error condition is described
+// by Msg.
+//
+type Error struct {
+ Pos token.Position
+ Msg string
+}
+
+// Error implements the error interface.
+func (e Error) Error() string {
+ if e.Pos.Filename != "" || e.Pos.IsValid() {
+ // don't print "<unknown position>"
+ // TODO(gri) reconsider the semantics of Position.IsValid
+ return e.Pos.String() + ": " + e.Msg
+ }
+ return e.Msg
+}
+
+// ErrorList is a list of *Errors.
+// The zero value for an ErrorList is an empty ErrorList ready to use.
+//
+type ErrorList []*Error
+
+// Add adds an Error with given position and error message to an ErrorList.
+func (p *ErrorList) Add(pos token.Position, msg string) {
+ *p = append(*p, &Error{pos, msg})
+}
+
+// Reset resets an ErrorList to no errors.
+func (p *ErrorList) Reset() { *p = (*p)[0:0] }
+
+// ErrorList implements the sort Interface.
+func (p ErrorList) Len() int { return len(p) }
+func (p ErrorList) Swap(i, j int) { p[i], p[j] = p[j], p[i] }
+
+func (p ErrorList) Less(i, j int) bool {
+ e := &p[i].Pos
+ f := &p[j].Pos
+ // Note that it is not sufficient to simply compare file offsets because
+ // the offsets do not reflect modified line information (through //line
+ // comments).
+ if e.Filename != f.Filename {
+ return e.Filename < f.Filename
+ }
+ if e.Line != f.Line {
+ return e.Line < f.Line
+ }
+ if e.Column != f.Column {
+ return e.Column < f.Column
+ }
+ return p[i].Msg < p[j].Msg
+}
+
+// Sort sorts an ErrorList. *Error entries are sorted by position,
+// other errors are sorted by error message, and before any *Error
+// entry.
+//
+func (p ErrorList) Sort() {
+ sort.Sort(p)
+}
+
+// RemoveMultiples sorts an ErrorList and removes all but the first error per line.
+func (p *ErrorList) RemoveMultiples() {
+ sort.Sort(p)
+ var last token.Position // initial last.Line is != any legal error line
+ i := 0
+ for _, e := range *p {
+ if e.Pos.Filename != last.Filename || e.Pos.Line != last.Line {
+ last = e.Pos
+ (*p)[i] = e
+ i++
+ }
+ }
+ *p = (*p)[0:i]
+}
+
+// An ErrorList implements the error interface.
+func (p ErrorList) Error() string {
+ switch len(p) {
+ case 0:
+ return "no errors"
+ case 1:
+ return p[0].Error()
+ }
+ return fmt.Sprintf("%s (and %d more errors)", p[0], len(p)-1)
+}
+
+// Err returns an error equivalent to this error list.
+// If the list is empty, Err returns nil.
+func (p ErrorList) Err() error {
+ if len(p) == 0 {
+ return nil
+ }
+ return p
+}
+
+// PrintError is a utility function that prints a list of errors to w,
+// one error per line, if the err parameter is an ErrorList. Otherwise
+// it prints the err string.
+//
+func PrintError(w io.Writer, err error) {
+ if list, ok := err.(ErrorList); ok {
+ for _, e := range list {
+ fmt.Fprintf(w, "%s\n", e)
+ }
+ } else if err != nil {
+ fmt.Fprintf(w, "%s\n", err)
+ }
+}
diff --git a/src/go/scanner/example_test.go b/src/go/scanner/example_test.go
new file mode 100644
index 0000000..9004a4a
--- /dev/null
+++ b/src/go/scanner/example_test.go
@@ -0,0 +1,46 @@
+// Copyright 2012 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.
+
+package scanner_test
+
+import (
+ "fmt"
+ "go/scanner"
+ "go/token"
+)
+
+func ExampleScanner_Scan() {
+ // src is the input that we want to tokenize.
+ src := []byte("cos(x) + 1i*sin(x) // Euler")
+
+ // Initialize the scanner.
+ var s scanner.Scanner
+ fset := token.NewFileSet() // positions are relative to fset
+ file := fset.AddFile("", fset.Base(), len(src)) // register input "file"
+ s.Init(file, src, nil /* no error handler */, scanner.ScanComments)
+
+ // Repeated calls to Scan yield the token sequence found in the input.
+ for {
+ pos, tok, lit := s.Scan()
+ if tok == token.EOF {
+ break
+ }
+ fmt.Printf("%s\t%s\t%q\n", fset.Position(pos), tok, lit)
+ }
+
+ // output:
+ // 1:1 IDENT "cos"
+ // 1:4 ( ""
+ // 1:5 IDENT "x"
+ // 1:6 ) ""
+ // 1:8 + ""
+ // 1:10 IMAG "1i"
+ // 1:12 * ""
+ // 1:13 IDENT "sin"
+ // 1:16 ( ""
+ // 1:17 IDENT "x"
+ // 1:18 ) ""
+ // 1:20 ; "\n"
+ // 1:20 COMMENT "// Euler"
+}
diff --git a/src/go/scanner/scanner.go b/src/go/scanner/scanner.go
new file mode 100644
index 0000000..f08e28c
--- /dev/null
+++ b/src/go/scanner/scanner.go
@@ -0,0 +1,987 @@
+// 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.
+
+// Package scanner implements a scanner for Go source text.
+// It takes a []byte as source which can then be tokenized
+// through repeated calls to the Scan method.
+//
+package scanner
+
+import (
+ "bytes"
+ "fmt"
+ "go/token"
+ "path/filepath"
+ "strconv"
+ "unicode"
+ "unicode/utf8"
+)
+
+// An ErrorHandler may be provided to Scanner.Init. If a syntax error is
+// encountered and a handler was installed, the handler is called with a
+// position and an error message. The position points to the beginning of
+// the offending token.
+//
+type ErrorHandler func(pos token.Position, msg string)
+
+// A Scanner holds the scanner's internal state while processing
+// a given text. It can be allocated as part of another data
+// structure but must be initialized via Init before use.
+//
+type Scanner struct {
+ // immutable state
+ file *token.File // source file handle
+ dir string // directory portion of file.Name()
+ src []byte // source
+ err ErrorHandler // error reporting; or nil
+ mode Mode // scanning mode
+
+ // scanning state
+ ch rune // current character
+ offset int // character offset
+ rdOffset int // reading offset (position after current character)
+ lineOffset int // current line offset
+ insertSemi bool // insert a semicolon before next newline
+
+ // public state - ok to modify
+ ErrorCount int // number of errors encountered
+}
+
+const (
+ bom = 0xFEFF // byte order mark, only permitted as very first character
+ eof = -1 // end of file
+)
+
+// Read the next Unicode char into s.ch.
+// s.ch < 0 means end-of-file.
+//
+// For optimization, there is some overlap between this method and
+// s.scanIdentifier.
+func (s *Scanner) next() {
+ if s.rdOffset < len(s.src) {
+ s.offset = s.rdOffset
+ if s.ch == '\n' {
+ s.lineOffset = s.offset
+ s.file.AddLine(s.offset)
+ }
+ r, w := rune(s.src[s.rdOffset]), 1
+ switch {
+ case r == 0:
+ s.error(s.offset, "illegal character NUL")
+ case r >= utf8.RuneSelf:
+ // not ASCII
+ r, w = utf8.DecodeRune(s.src[s.rdOffset:])
+ if r == utf8.RuneError && w == 1 {
+ s.error(s.offset, "illegal UTF-8 encoding")
+ } else if r == bom && s.offset > 0 {
+ s.error(s.offset, "illegal byte order mark")
+ }
+ }
+ s.rdOffset += w
+ s.ch = r
+ } else {
+ s.offset = len(s.src)
+ if s.ch == '\n' {
+ s.lineOffset = s.offset
+ s.file.AddLine(s.offset)
+ }
+ s.ch = eof
+ }
+}
+
+// peek returns the byte following the most recently read character without
+// advancing the scanner. If the scanner is at EOF, peek returns 0.
+func (s *Scanner) peek() byte {
+ if s.rdOffset < len(s.src) {
+ return s.src[s.rdOffset]
+ }
+ return 0
+}
+
+// A mode value is a set of flags (or 0).
+// They control scanner behavior.
+//
+type Mode uint
+
+const (
+ ScanComments Mode = 1 << iota // return comments as COMMENT tokens
+ dontInsertSemis // do not automatically insert semicolons - for testing only
+)
+
+// Init prepares the scanner s to tokenize the text src by setting the
+// scanner at the beginning of src. The scanner uses the file set file
+// for position information and it adds line information for each line.
+// It is ok to re-use the same file when re-scanning the same file as
+// line information which is already present is ignored. Init causes a
+// panic if the file size does not match the src size.
+//
+// Calls to Scan will invoke the error handler err if they encounter a
+// syntax error and err is not nil. Also, for each error encountered,
+// the Scanner field ErrorCount is incremented by one. The mode parameter
+// determines how comments are handled.
+//
+// Note that Init may call err if there is an error in the first character
+// of the file.
+//
+func (s *Scanner) Init(file *token.File, src []byte, err ErrorHandler, mode Mode) {
+ // Explicitly initialize all fields since a scanner may be reused.
+ if file.Size() != len(src) {
+ panic(fmt.Sprintf("file size (%d) does not match src len (%d)", file.Size(), len(src)))
+ }
+ s.file = file
+ s.dir, _ = filepath.Split(file.Name())
+ s.src = src
+ s.err = err
+ s.mode = mode
+
+ s.ch = ' '
+ s.offset = 0
+ s.rdOffset = 0
+ s.lineOffset = 0
+ s.insertSemi = false
+ s.ErrorCount = 0
+
+ s.next()
+ if s.ch == bom {
+ s.next() // ignore BOM at file beginning
+ }
+}
+
+func (s *Scanner) error(offs int, msg string) {
+ if s.err != nil {
+ s.err(s.file.Position(s.file.Pos(offs)), msg)
+ }
+ s.ErrorCount++
+}
+
+func (s *Scanner) errorf(offs int, format string, args ...interface{}) {
+ s.error(offs, fmt.Sprintf(format, args...))
+}
+
+func (s *Scanner) scanComment() string {
+ // initial '/' already consumed; s.ch == '/' || s.ch == '*'
+ offs := s.offset - 1 // position of initial '/'
+ next := -1 // position immediately following the comment; < 0 means invalid comment
+ numCR := 0
+
+ if s.ch == '/' {
+ //-style comment
+ // (the final '\n' is not considered part of the comment)
+ s.next()
+ for s.ch != '\n' && s.ch >= 0 {
+ if s.ch == '\r' {
+ numCR++
+ }
+ s.next()
+ }
+ // if we are at '\n', the position following the comment is afterwards
+ next = s.offset
+ if s.ch == '\n' {
+ next++
+ }
+ goto exit
+ }
+
+ /*-style comment */
+ s.next()
+ for s.ch >= 0 {
+ ch := s.ch
+ if ch == '\r' {
+ numCR++
+ }
+ s.next()
+ if ch == '*' && s.ch == '/' {
+ s.next()
+ next = s.offset
+ goto exit
+ }
+ }
+
+ s.error(offs, "comment not terminated")
+
+exit:
+ lit := s.src[offs:s.offset]
+
+ // On Windows, a (//-comment) line may end in "\r\n".
+ // Remove the final '\r' before analyzing the text for
+ // line directives (matching the compiler). Remove any
+ // other '\r' afterwards (matching the pre-existing be-
+ // havior of the scanner).
+ if numCR > 0 && len(lit) >= 2 && lit[1] == '/' && lit[len(lit)-1] == '\r' {
+ lit = lit[:len(lit)-1]
+ numCR--
+ }
+
+ // interpret line directives
+ // (//line directives must start at the beginning of the current line)
+ if next >= 0 /* implies valid comment */ && (lit[1] == '*' || offs == s.lineOffset) && bytes.HasPrefix(lit[2:], prefix) {
+ s.updateLineInfo(next, offs, lit)
+ }
+
+ if numCR > 0 {
+ lit = stripCR(lit, lit[1] == '*')
+ }
+
+ return string(lit)
+}
+
+var prefix = []byte("line ")
+
+// updateLineInfo parses the incoming comment text at offset offs
+// as a line directive. If successful, it updates the line info table
+// for the position next per the line directive.
+func (s *Scanner) updateLineInfo(next, offs int, text []byte) {
+ // extract comment text
+ if text[1] == '*' {
+ text = text[:len(text)-2] // lop off trailing "*/"
+ }
+ text = text[7:] // lop off leading "//line " or "/*line "
+ offs += 7
+
+ i, n, ok := trailingDigits(text)
+ if i == 0 {
+ return // ignore (not a line directive)
+ }
+ // i > 0
+
+ if !ok {
+ // text has a suffix :xxx but xxx is not a number
+ s.error(offs+i, "invalid line number: "+string(text[i:]))
+ return
+ }
+
+ var line, col int
+ i2, n2, ok2 := trailingDigits(text[:i-1])
+ if ok2 {
+ //line filename:line:col
+ i, i2 = i2, i
+ line, col = n2, n
+ if col == 0 {
+ s.error(offs+i2, "invalid column number: "+string(text[i2:]))
+ return
+ }
+ text = text[:i2-1] // lop off ":col"
+ } else {
+ //line filename:line
+ line = n
+ }
+
+ if line == 0 {
+ s.error(offs+i, "invalid line number: "+string(text[i:]))
+ return
+ }
+
+ // If we have a column (//line filename:line:col form),
+ // an empty filename means to use the previous filename.
+ filename := string(text[:i-1]) // lop off ":line", and trim white space
+ if filename == "" && ok2 {
+ filename = s.file.Position(s.file.Pos(offs)).Filename
+ } else if filename != "" {
+ // Put a relative filename in the current directory.
+ // This is for compatibility with earlier releases.
+ // See issue 26671.
+ filename = filepath.Clean(filename)
+ if !filepath.IsAbs(filename) {
+ filename = filepath.Join(s.dir, filename)
+ }
+ }
+
+ s.file.AddLineColumnInfo(next, filename, line, col)
+}
+
+func trailingDigits(text []byte) (int, int, bool) {
+ i := bytes.LastIndexByte(text, ':') // look from right (Windows filenames may contain ':')
+ if i < 0 {
+ return 0, 0, false // no ":"
+ }
+ // i >= 0
+ n, err := strconv.ParseUint(string(text[i+1:]), 10, 0)
+ return i + 1, int(n), err == nil
+}
+
+func (s *Scanner) findLineEnd() bool {
+ // initial '/' already consumed
+
+ defer func(offs int) {
+ // reset scanner state to where it was upon calling findLineEnd
+ s.ch = '/'
+ s.offset = offs
+ s.rdOffset = offs + 1
+ s.next() // consume initial '/' again
+ }(s.offset - 1)
+
+ // read ahead until a newline, EOF, or non-comment token is found
+ for s.ch == '/' || s.ch == '*' {
+ if s.ch == '/' {
+ //-style comment always contains a newline
+ return true
+ }
+ /*-style comment: look for newline */
+ s.next()
+ for s.ch >= 0 {
+ ch := s.ch
+ if ch == '\n' {
+ return true
+ }
+ s.next()
+ if ch == '*' && s.ch == '/' {
+ s.next()
+ break
+ }
+ }
+ s.skipWhitespace() // s.insertSemi is set
+ if s.ch < 0 || s.ch == '\n' {
+ return true
+ }
+ if s.ch != '/' {
+ // non-comment token
+ return false
+ }
+ s.next() // consume '/'
+ }
+
+ return false
+}
+
+func isLetter(ch rune) bool {
+ return 'a' <= lower(ch) && lower(ch) <= 'z' || ch == '_' || ch >= utf8.RuneSelf && unicode.IsLetter(ch)
+}
+
+func isDigit(ch rune) bool {
+ return isDecimal(ch) || ch >= utf8.RuneSelf && unicode.IsDigit(ch)
+}
+
+// scanIdentifier reads the string of valid identifier characters at s.offset.
+// It must only be called when s.ch is known to be a valid letter.
+//
+// Be careful when making changes to this function: it is optimized and affects
+// scanning performance significantly.
+func (s *Scanner) scanIdentifier() string {
+ offs := s.offset
+
+ // Optimize for the common case of an ASCII identifier.
+ //
+ // Ranging over s.src[s.rdOffset:] lets us avoid some bounds checks, and
+ // avoids conversions to runes.
+ //
+ // In case we encounter a non-ASCII character, fall back on the slower path
+ // of calling into s.next().
+ for rdOffset, b := range s.src[s.rdOffset:] {
+ if 'a' <= b && b <= 'z' || 'A' <= b && b <= 'Z' || b == '_' || '0' <= b && b <= '9' {
+ // Avoid assigning a rune for the common case of an ascii character.
+ continue
+ }
+ s.rdOffset += rdOffset
+ if 0 < b && b < utf8.RuneSelf {
+ // Optimization: we've encountered an ASCII character that's not a letter
+ // or number. Avoid the call into s.next() and corresponding set up.
+ //
+ // Note that s.next() does some line accounting if s.ch is '\n', so this
+ // shortcut is only possible because we know that the preceding character
+ // is not '\n'.
+ s.ch = rune(b)
+ s.offset = s.rdOffset
+ s.rdOffset++
+ goto exit
+ }
+ // We know that the preceding character is valid for an identifier because
+ // scanIdentifier is only called when s.ch is a letter, so calling s.next()
+ // at s.rdOffset resets the scanner state.
+ s.next()
+ for isLetter(s.ch) || isDigit(s.ch) {
+ s.next()
+ }
+ goto exit
+ }
+ s.offset = len(s.src)
+ s.rdOffset = len(s.src)
+ s.ch = eof
+
+exit:
+ return string(s.src[offs:s.offset])
+}
+
+func digitVal(ch rune) int {
+ switch {
+ case '0' <= ch && ch <= '9':
+ return int(ch - '0')
+ case 'a' <= lower(ch) && lower(ch) <= 'f':
+ return int(lower(ch) - 'a' + 10)
+ }
+ return 16 // larger than any legal digit val
+}
+
+func lower(ch rune) rune { return ('a' - 'A') | ch } // returns lower-case ch iff ch is ASCII letter
+func isDecimal(ch rune) bool { return '0' <= ch && ch <= '9' }
+func isHex(ch rune) bool { return '0' <= ch && ch <= '9' || 'a' <= lower(ch) && lower(ch) <= 'f' }
+
+// digits accepts the sequence { digit | '_' }.
+// If base <= 10, digits accepts any decimal digit but records
+// the offset (relative to the source start) of a digit >= base
+// in *invalid, if *invalid < 0.
+// digits returns a bitset describing whether the sequence contained
+// digits (bit 0 is set), or separators '_' (bit 1 is set).
+func (s *Scanner) digits(base int, invalid *int) (digsep int) {
+ if base <= 10 {
+ max := rune('0' + base)
+ for isDecimal(s.ch) || s.ch == '_' {
+ ds := 1
+ if s.ch == '_' {
+ ds = 2
+ } else if s.ch >= max && *invalid < 0 {
+ *invalid = s.offset // record invalid rune offset
+ }
+ digsep |= ds
+ s.next()
+ }
+ } else {
+ for isHex(s.ch) || s.ch == '_' {
+ ds := 1
+ if s.ch == '_' {
+ ds = 2
+ }
+ digsep |= ds
+ s.next()
+ }
+ }
+ return
+}
+
+func (s *Scanner) scanNumber() (token.Token, string) {
+ offs := s.offset
+ tok := token.ILLEGAL
+
+ base := 10 // number base
+ prefix := rune(0) // one of 0 (decimal), '0' (0-octal), 'x', 'o', or 'b'
+ digsep := 0 // bit 0: digit present, bit 1: '_' present
+ invalid := -1 // index of invalid digit in literal, or < 0
+
+ // integer part
+ if s.ch != '.' {
+ tok = token.INT
+ if s.ch == '0' {
+ s.next()
+ switch lower(s.ch) {
+ case 'x':
+ s.next()
+ base, prefix = 16, 'x'
+ case 'o':
+ s.next()
+ base, prefix = 8, 'o'
+ case 'b':
+ s.next()
+ base, prefix = 2, 'b'
+ default:
+ base, prefix = 8, '0'
+ digsep = 1 // leading 0
+ }
+ }
+ digsep |= s.digits(base, &invalid)
+ }
+
+ // fractional part
+ if s.ch == '.' {
+ tok = token.FLOAT
+ if prefix == 'o' || prefix == 'b' {
+ s.error(s.offset, "invalid radix point in "+litname(prefix))
+ }
+ s.next()
+ digsep |= s.digits(base, &invalid)
+ }
+
+ if digsep&1 == 0 {
+ s.error(s.offset, litname(prefix)+" has no digits")
+ }
+
+ // exponent
+ if e := lower(s.ch); e == 'e' || e == 'p' {
+ switch {
+ case e == 'e' && prefix != 0 && prefix != '0':
+ s.errorf(s.offset, "%q exponent requires decimal mantissa", s.ch)
+ case e == 'p' && prefix != 'x':
+ s.errorf(s.offset, "%q exponent requires hexadecimal mantissa", s.ch)
+ }
+ s.next()
+ tok = token.FLOAT
+ if s.ch == '+' || s.ch == '-' {
+ s.next()
+ }
+ ds := s.digits(10, nil)
+ digsep |= ds
+ if ds&1 == 0 {
+ s.error(s.offset, "exponent has no digits")
+ }
+ } else if prefix == 'x' && tok == token.FLOAT {
+ s.error(s.offset, "hexadecimal mantissa requires a 'p' exponent")
+ }
+
+ // suffix 'i'
+ if s.ch == 'i' {
+ tok = token.IMAG
+ s.next()
+ }
+
+ lit := string(s.src[offs:s.offset])
+ if tok == token.INT && invalid >= 0 {
+ s.errorf(invalid, "invalid digit %q in %s", lit[invalid-offs], litname(prefix))
+ }
+ if digsep&2 != 0 {
+ if i := invalidSep(lit); i >= 0 {
+ s.error(offs+i, "'_' must separate successive digits")
+ }
+ }
+
+ return tok, lit
+}
+
+func litname(prefix rune) string {
+ switch prefix {
+ case 'x':
+ return "hexadecimal literal"
+ case 'o', '0':
+ return "octal literal"
+ case 'b':
+ return "binary literal"
+ }
+ return "decimal literal"
+}
+
+// invalidSep returns the index of the first invalid separator in x, or -1.
+func invalidSep(x string) int {
+ x1 := ' ' // prefix char, we only care if it's 'x'
+ d := '.' // digit, one of '_', '0' (a digit), or '.' (anything else)
+ i := 0
+
+ // a prefix counts as a digit
+ if len(x) >= 2 && x[0] == '0' {
+ x1 = lower(rune(x[1]))
+ if x1 == 'x' || x1 == 'o' || x1 == 'b' {
+ d = '0'
+ i = 2
+ }
+ }
+
+ // mantissa and exponent
+ for ; i < len(x); i++ {
+ p := d // previous digit
+ d = rune(x[i])
+ switch {
+ case d == '_':
+ if p != '0' {
+ return i
+ }
+ case isDecimal(d) || x1 == 'x' && isHex(d):
+ d = '0'
+ default:
+ if p == '_' {
+ return i - 1
+ }
+ d = '.'
+ }
+ }
+ if d == '_' {
+ return len(x) - 1
+ }
+
+ return -1
+}
+
+// scanEscape parses an escape sequence where rune is the accepted
+// escaped quote. In case of a syntax error, it stops at the offending
+// character (without consuming it) and returns false. Otherwise
+// it returns true.
+func (s *Scanner) scanEscape(quote rune) bool {
+ offs := s.offset
+
+ var n int
+ var base, max uint32
+ switch s.ch {
+ case 'a', 'b', 'f', 'n', 'r', 't', 'v', '\\', quote:
+ s.next()
+ return true
+ case '0', '1', '2', '3', '4', '5', '6', '7':
+ n, base, max = 3, 8, 255
+ case 'x':
+ s.next()
+ n, base, max = 2, 16, 255
+ case 'u':
+ s.next()
+ n, base, max = 4, 16, unicode.MaxRune
+ case 'U':
+ s.next()
+ n, base, max = 8, 16, unicode.MaxRune
+ default:
+ msg := "unknown escape sequence"
+ if s.ch < 0 {
+ msg = "escape sequence not terminated"
+ }
+ s.error(offs, msg)
+ return false
+ }
+
+ var x uint32
+ for n > 0 {
+ d := uint32(digitVal(s.ch))
+ if d >= base {
+ msg := fmt.Sprintf("illegal character %#U in escape sequence", s.ch)
+ if s.ch < 0 {
+ msg = "escape sequence not terminated"
+ }
+ s.error(s.offset, msg)
+ return false
+ }
+ x = x*base + d
+ s.next()
+ n--
+ }
+
+ if x > max || 0xD800 <= x && x < 0xE000 {
+ s.error(offs, "escape sequence is invalid Unicode code point")
+ return false
+ }
+
+ return true
+}
+
+func (s *Scanner) scanRune() string {
+ // '\'' opening already consumed
+ offs := s.offset - 1
+
+ valid := true
+ n := 0
+ for {
+ ch := s.ch
+ if ch == '\n' || ch < 0 {
+ // only report error if we don't have one already
+ if valid {
+ s.error(offs, "rune literal not terminated")
+ valid = false
+ }
+ break
+ }
+ s.next()
+ if ch == '\'' {
+ break
+ }
+ n++
+ if ch == '\\' {
+ if !s.scanEscape('\'') {
+ valid = false
+ }
+ // continue to read to closing quote
+ }
+ }
+
+ if valid && n != 1 {
+ s.error(offs, "illegal rune literal")
+ }
+
+ return string(s.src[offs:s.offset])
+}
+
+func (s *Scanner) scanString() string {
+ // '"' opening already consumed
+ offs := s.offset - 1
+
+ for {
+ ch := s.ch
+ if ch == '\n' || ch < 0 {
+ s.error(offs, "string literal not terminated")
+ break
+ }
+ s.next()
+ if ch == '"' {
+ break
+ }
+ if ch == '\\' {
+ s.scanEscape('"')
+ }
+ }
+
+ return string(s.src[offs:s.offset])
+}
+
+func stripCR(b []byte, comment bool) []byte {
+ c := make([]byte, len(b))
+ i := 0
+ for j, ch := range b {
+ // In a /*-style comment, don't strip \r from *\r/ (incl.
+ // sequences of \r from *\r\r...\r/) since the resulting
+ // */ would terminate the comment too early unless the \r
+ // is immediately following the opening /* in which case
+ // it's ok because /*/ is not closed yet (issue #11151).
+ if ch != '\r' || comment && i > len("/*") && c[i-1] == '*' && j+1 < len(b) && b[j+1] == '/' {
+ c[i] = ch
+ i++
+ }
+ }
+ return c[:i]
+}
+
+func (s *Scanner) scanRawString() string {
+ // '`' opening already consumed
+ offs := s.offset - 1
+
+ hasCR := false
+ for {
+ ch := s.ch
+ if ch < 0 {
+ s.error(offs, "raw string literal not terminated")
+ break
+ }
+ s.next()
+ if ch == '`' {
+ break
+ }
+ if ch == '\r' {
+ hasCR = true
+ }
+ }
+
+ lit := s.src[offs:s.offset]
+ if hasCR {
+ lit = stripCR(lit, false)
+ }
+
+ return string(lit)
+}
+
+func (s *Scanner) skipWhitespace() {
+ for s.ch == ' ' || s.ch == '\t' || s.ch == '\n' && !s.insertSemi || s.ch == '\r' {
+ s.next()
+ }
+}
+
+// Helper functions for scanning multi-byte tokens such as >> += >>= .
+// Different routines recognize different length tok_i based on matches
+// of ch_i. If a token ends in '=', the result is tok1 or tok3
+// respectively. Otherwise, the result is tok0 if there was no other
+// matching character, or tok2 if the matching character was ch2.
+
+func (s *Scanner) switch2(tok0, tok1 token.Token) token.Token {
+ if s.ch == '=' {
+ s.next()
+ return tok1
+ }
+ return tok0
+}
+
+func (s *Scanner) switch3(tok0, tok1 token.Token, ch2 rune, tok2 token.Token) token.Token {
+ if s.ch == '=' {
+ s.next()
+ return tok1
+ }
+ if s.ch == ch2 {
+ s.next()
+ return tok2
+ }
+ return tok0
+}
+
+func (s *Scanner) switch4(tok0, tok1 token.Token, ch2 rune, tok2, tok3 token.Token) token.Token {
+ if s.ch == '=' {
+ s.next()
+ return tok1
+ }
+ if s.ch == ch2 {
+ s.next()
+ if s.ch == '=' {
+ s.next()
+ return tok3
+ }
+ return tok2
+ }
+ return tok0
+}
+
+// Scan scans the next token and returns the token position, the token,
+// and its literal string if applicable. The source end is indicated by
+// token.EOF.
+//
+// If the returned token is a literal (token.IDENT, token.INT, token.FLOAT,
+// token.IMAG, token.CHAR, token.STRING) or token.COMMENT, the literal string
+// has the corresponding value.
+//
+// If the returned token is a keyword, the literal string is the keyword.
+//
+// If the returned token is token.SEMICOLON, the corresponding
+// literal string is ";" if the semicolon was present in the source,
+// and "\n" if the semicolon was inserted because of a newline or
+// at EOF.
+//
+// If the returned token is token.ILLEGAL, the literal string is the
+// offending character.
+//
+// In all other cases, Scan returns an empty literal string.
+//
+// For more tolerant parsing, Scan will return a valid token if
+// possible even if a syntax error was encountered. Thus, even
+// if the resulting token sequence contains no illegal tokens,
+// a client may not assume that no error occurred. Instead it
+// must check the scanner's ErrorCount or the number of calls
+// of the error handler, if there was one installed.
+//
+// Scan adds line information to the file added to the file
+// set with Init. Token positions are relative to that file
+// and thus relative to the file set.
+//
+func (s *Scanner) Scan() (pos token.Pos, tok token.Token, lit string) {
+scanAgain:
+ s.skipWhitespace()
+
+ // current token start
+ pos = s.file.Pos(s.offset)
+
+ // determine token value
+ insertSemi := false
+ switch ch := s.ch; {
+ case isLetter(ch):
+ lit = s.scanIdentifier()
+ if len(lit) > 1 {
+ // keywords are longer than one letter - avoid lookup otherwise
+ tok = token.Lookup(lit)
+ switch tok {
+ case token.IDENT, token.BREAK, token.CONTINUE, token.FALLTHROUGH, token.RETURN:
+ insertSemi = true
+ }
+ } else {
+ insertSemi = true
+ tok = token.IDENT
+ }
+ case isDecimal(ch) || ch == '.' && isDecimal(rune(s.peek())):
+ insertSemi = true
+ tok, lit = s.scanNumber()
+ default:
+ s.next() // always make progress
+ switch ch {
+ case -1:
+ if s.insertSemi {
+ s.insertSemi = false // EOF consumed
+ return pos, token.SEMICOLON, "\n"
+ }
+ tok = token.EOF
+ case '\n':
+ // we only reach here if s.insertSemi was
+ // set in the first place and exited early
+ // from s.skipWhitespace()
+ s.insertSemi = false // newline consumed
+ return pos, token.SEMICOLON, "\n"
+ case '"':
+ insertSemi = true
+ tok = token.STRING
+ lit = s.scanString()
+ case '\'':
+ insertSemi = true
+ tok = token.CHAR
+ lit = s.scanRune()
+ case '`':
+ insertSemi = true
+ tok = token.STRING
+ lit = s.scanRawString()
+ case ':':
+ tok = s.switch2(token.COLON, token.DEFINE)
+ case '.':
+ // fractions starting with a '.' are handled by outer switch
+ tok = token.PERIOD
+ if s.ch == '.' && s.peek() == '.' {
+ s.next()
+ s.next() // consume last '.'
+ tok = token.ELLIPSIS
+ }
+ case ',':
+ tok = token.COMMA
+ case ';':
+ tok = token.SEMICOLON
+ lit = ";"
+ case '(':
+ tok = token.LPAREN
+ case ')':
+ insertSemi = true
+ tok = token.RPAREN
+ case '[':
+ tok = token.LBRACK
+ case ']':
+ insertSemi = true
+ tok = token.RBRACK
+ case '{':
+ tok = token.LBRACE
+ case '}':
+ insertSemi = true
+ tok = token.RBRACE
+ case '+':
+ tok = s.switch3(token.ADD, token.ADD_ASSIGN, '+', token.INC)
+ if tok == token.INC {
+ insertSemi = true
+ }
+ case '-':
+ tok = s.switch3(token.SUB, token.SUB_ASSIGN, '-', token.DEC)
+ if tok == token.DEC {
+ insertSemi = true
+ }
+ case '*':
+ tok = s.switch2(token.MUL, token.MUL_ASSIGN)
+ case '/':
+ if s.ch == '/' || s.ch == '*' {
+ // comment
+ if s.insertSemi && s.findLineEnd() {
+ // reset position to the beginning of the comment
+ s.ch = '/'
+ s.offset = s.file.Offset(pos)
+ s.rdOffset = s.offset + 1
+ s.insertSemi = false // newline consumed
+ return pos, token.SEMICOLON, "\n"
+ }
+ comment := s.scanComment()
+ if s.mode&ScanComments == 0 {
+ // skip comment
+ s.insertSemi = false // newline consumed
+ goto scanAgain
+ }
+ tok = token.COMMENT
+ lit = comment
+ } else {
+ tok = s.switch2(token.QUO, token.QUO_ASSIGN)
+ }
+ case '%':
+ tok = s.switch2(token.REM, token.REM_ASSIGN)
+ case '^':
+ tok = s.switch2(token.XOR, token.XOR_ASSIGN)
+ case '<':
+ if s.ch == '-' {
+ s.next()
+ tok = token.ARROW
+ } else {
+ tok = s.switch4(token.LSS, token.LEQ, '<', token.SHL, token.SHL_ASSIGN)
+ }
+ case '>':
+ tok = s.switch4(token.GTR, token.GEQ, '>', token.SHR, token.SHR_ASSIGN)
+ case '=':
+ tok = s.switch2(token.ASSIGN, token.EQL)
+ case '!':
+ tok = s.switch2(token.NOT, token.NEQ)
+ case '&':
+ if s.ch == '^' {
+ s.next()
+ tok = s.switch2(token.AND_NOT, token.AND_NOT_ASSIGN)
+ } else {
+ tok = s.switch3(token.AND, token.AND_ASSIGN, '&', token.LAND)
+ }
+ case '|':
+ tok = s.switch3(token.OR, token.OR_ASSIGN, '|', token.LOR)
+ default:
+ // next reports unexpected BOMs - don't repeat
+ if ch != bom {
+ s.errorf(s.file.Offset(pos), "illegal character %#U", ch)
+ }
+ insertSemi = s.insertSemi // preserve insertSemi info
+ tok = token.ILLEGAL
+ lit = string(ch)
+ }
+ }
+ if s.mode&dontInsertSemis == 0 {
+ s.insertSemi = insertSemi
+ }
+
+ return
+}
diff --git a/src/go/scanner/scanner_test.go b/src/go/scanner/scanner_test.go
new file mode 100644
index 0000000..db123c3
--- /dev/null
+++ b/src/go/scanner/scanner_test.go
@@ -0,0 +1,1126 @@
+// 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.
+
+package scanner
+
+import (
+ "go/token"
+ "os"
+ "path/filepath"
+ "runtime"
+ "strings"
+ "testing"
+)
+
+var fset = token.NewFileSet()
+
+const /* class */ (
+ special = iota
+ literal
+ operator
+ keyword
+)
+
+func tokenclass(tok token.Token) int {
+ switch {
+ case tok.IsLiteral():
+ return literal
+ case tok.IsOperator():
+ return operator
+ case tok.IsKeyword():
+ return keyword
+ }
+ return special
+}
+
+type elt struct {
+ tok token.Token
+ lit string
+ class int
+}
+
+var tokens = [...]elt{
+ // Special tokens
+ {token.COMMENT, "/* a comment */", special},
+ {token.COMMENT, "// a comment \n", special},
+ {token.COMMENT, "/*\r*/", special},
+ {token.COMMENT, "/**\r/*/", special}, // issue 11151
+ {token.COMMENT, "/**\r\r/*/", special},
+ {token.COMMENT, "//\r\n", special},
+
+ // Identifiers and basic type literals
+ {token.IDENT, "foobar", literal},
+ {token.IDENT, "a۰۱۸", literal},
+ {token.IDENT, "foo६४", literal},
+ {token.IDENT, "bar９８７６", literal},
+ {token.IDENT, "ŝ", literal}, // was bug (issue 4000)
+ {token.IDENT, "ŝfoo", literal}, // was bug (issue 4000)
+ {token.INT, "0", literal},
+ {token.INT, "1", literal},
+ {token.INT, "123456789012345678890", literal},
+ {token.INT, "01234567", literal},
+ {token.INT, "0xcafebabe", literal},
+ {token.FLOAT, "0.", literal},
+ {token.FLOAT, ".0", literal},
+ {token.FLOAT, "3.14159265", literal},
+ {token.FLOAT, "1e0", literal},
+ {token.FLOAT, "1e+100", literal},
+ {token.FLOAT, "1e-100", literal},
+ {token.FLOAT, "2.71828e-1000", literal},
+ {token.IMAG, "0i", literal},
+ {token.IMAG, "1i", literal},
+ {token.IMAG, "012345678901234567889i", literal},
+ {token.IMAG, "123456789012345678890i", literal},
+ {token.IMAG, "0.i", literal},
+ {token.IMAG, ".0i", literal},
+ {token.IMAG, "3.14159265i", literal},
+ {token.IMAG, "1e0i", literal},
+ {token.IMAG, "1e+100i", literal},
+ {token.IMAG, "1e-100i", literal},
+ {token.IMAG, "2.71828e-1000i", literal},
+ {token.CHAR, "'a'", literal},
+ {token.CHAR, "'\\000'", literal},
+ {token.CHAR, "'\\xFF'", literal},
+ {token.CHAR, "'\\uff16'", literal},
+ {token.CHAR, "'\\U0000ff16'", literal},
+ {token.STRING, "`foobar`", literal},
+ {token.STRING, "`" + `foo
+ bar` +
+ "`",
+ literal,
+ },
+ {token.STRING, "`\r`", literal},
+ {token.STRING, "`foo\r\nbar`", literal},
+
+ // Operators and delimiters
+ {token.ADD, "+", operator},
+ {token.SUB, "-", operator},
+ {token.MUL, "*", operator},
+ {token.QUO, "/", operator},
+ {token.REM, "%", operator},
+
+ {token.AND, "&", operator},
+ {token.OR, "|", operator},
+ {token.XOR, "^", operator},
+ {token.SHL, "<<", operator},
+ {token.SHR, ">>", operator},
+ {token.AND_NOT, "&^", operator},
+
+ {token.ADD_ASSIGN, "+=", operator},
+ {token.SUB_ASSIGN, "-=", operator},
+ {token.MUL_ASSIGN, "*=", operator},
+ {token.QUO_ASSIGN, "/=", operator},
+ {token.REM_ASSIGN, "%=", operator},
+
+ {token.AND_ASSIGN, "&=", operator},
+ {token.OR_ASSIGN, "|=", operator},
+ {token.XOR_ASSIGN, "^=", operator},
+ {token.SHL_ASSIGN, "<<=", operator},
+ {token.SHR_ASSIGN, ">>=", operator},
+ {token.AND_NOT_ASSIGN, "&^=", operator},
+
+ {token.LAND, "&&", operator},
+ {token.LOR, "||", operator},
+ {token.ARROW, "<-", operator},
+ {token.INC, "++", operator},
+ {token.DEC, "--", operator},
+
+ {token.EQL, "==", operator},
+ {token.LSS, "<", operator},
+ {token.GTR, ">", operator},
+ {token.ASSIGN, "=", operator},
+ {token.NOT, "!", operator},
+
+ {token.NEQ, "!=", operator},
+ {token.LEQ, "<=", operator},
+ {token.GEQ, ">=", operator},
+ {token.DEFINE, ":=", operator},
+ {token.ELLIPSIS, "...", operator},
+
+ {token.LPAREN, "(", operator},
+ {token.LBRACK, "[", operator},
+ {token.LBRACE, "{", operator},
+ {token.COMMA, ",", operator},
+ {token.PERIOD, ".", operator},
+
+ {token.RPAREN, ")", operator},
+ {token.RBRACK, "]", operator},
+ {token.RBRACE, "}", operator},
+ {token.SEMICOLON, ";", operator},
+ {token.COLON, ":", operator},
+
+ // Keywords
+ {token.BREAK, "break", keyword},
+ {token.CASE, "case", keyword},
+ {token.CHAN, "chan", keyword},
+ {token.CONST, "const", keyword},
+ {token.CONTINUE, "continue", keyword},
+
+ {token.DEFAULT, "default", keyword},
+ {token.DEFER, "defer", keyword},
+ {token.ELSE, "else", keyword},
+ {token.FALLTHROUGH, "fallthrough", keyword},
+ {token.FOR, "for", keyword},
+
+ {token.FUNC, "func", keyword},
+ {token.GO, "go", keyword},
+ {token.GOTO, "goto", keyword},
+ {token.IF, "if", keyword},
+ {token.IMPORT, "import", keyword},
+
+ {token.INTERFACE, "interface", keyword},
+ {token.MAP, "map", keyword},
+ {token.PACKAGE, "package", keyword},
+ {token.RANGE, "range", keyword},
+ {token.RETURN, "return", keyword},
+
+ {token.SELECT, "select", keyword},
+ {token.STRUCT, "struct", keyword},
+ {token.SWITCH, "switch", keyword},
+ {token.TYPE, "type", keyword},
+ {token.VAR, "var", keyword},
+}
+
+const whitespace = " \t \n\n\n" // to separate tokens
+
+var source = func() []byte {
+ var src []byte
+ for _, t := range tokens {
+ src = append(src, t.lit...)
+ src = append(src, whitespace...)
+ }
+ return src
+}()
+
+func newlineCount(s string) int {
+ n := 0
+ for i := 0; i < len(s); i++ {
+ if s[i] == '\n' {
+ n++
+ }
+ }
+ return n
+}
+
+func checkPos(t *testing.T, lit string, p token.Pos, expected token.Position) {
+ pos := fset.Position(p)
+ // Check cleaned filenames so that we don't have to worry about
+ // different os.PathSeparator values.
+ if pos.Filename != expected.Filename && filepath.Clean(pos.Filename) != filepath.Clean(expected.Filename) {
+ t.Errorf("bad filename for %q: got %s, expected %s", lit, pos.Filename, expected.Filename)
+ }
+ if pos.Offset != expected.Offset {
+ t.Errorf("bad position for %q: got %d, expected %d", lit, pos.Offset, expected.Offset)
+ }
+ if pos.Line != expected.Line {
+ t.Errorf("bad line for %q: got %d, expected %d", lit, pos.Line, expected.Line)
+ }
+ if pos.Column != expected.Column {
+ t.Errorf("bad column for %q: got %d, expected %d", lit, pos.Column, expected.Column)
+ }
+}
+
+// Verify that calling Scan() provides the correct results.
+func TestScan(t *testing.T) {
+ whitespace_linecount := newlineCount(whitespace)
+
+ // error handler
+ eh := func(_ token.Position, msg string) {
+ t.Errorf("error handler called (msg = %s)", msg)
+ }
+
+ // verify scan
+ var s Scanner
+ s.Init(fset.AddFile("", fset.Base(), len(source)), source, eh, ScanComments|dontInsertSemis)
+
+ // set up expected position
+ epos := token.Position{
+ Filename: "",
+ Offset: 0,
+ Line: 1,
+ Column: 1,
+ }
+
+ index := 0
+ for {
+ pos, tok, lit := s.Scan()
+
+ // check position
+ if tok == token.EOF {
+ // correction for EOF
+ epos.Line = newlineCount(string(source))
+ epos.Column = 2
+ }
+ checkPos(t, lit, pos, epos)
+
+ // check token
+ e := elt{token.EOF, "", special}
+ if index < len(tokens) {
+ e = tokens[index]
+ index++
+ }
+ if tok != e.tok {
+ t.Errorf("bad token for %q: got %s, expected %s", lit, tok, e.tok)
+ }
+
+ // check token class
+ if tokenclass(tok) != e.class {
+ t.Errorf("bad class for %q: got %d, expected %d", lit, tokenclass(tok), e.class)
+ }
+
+ // check literal
+ elit := ""
+ switch e.tok {
+ case token.COMMENT:
+ // no CRs in comments
+ elit = string(stripCR([]byte(e.lit), e.lit[1] == '*'))
+ //-style comment literal doesn't contain newline
+ if elit[1] == '/' {
+ elit = elit[0 : len(elit)-1]
+ }
+ case token.IDENT:
+ elit = e.lit
+ case token.SEMICOLON:
+ elit = ";"
+ default:
+ if e.tok.IsLiteral() {
+ // no CRs in raw string literals
+ elit = e.lit
+ if elit[0] == '`' {
+ elit = string(stripCR([]byte(elit), false))
+ }
+ } else if e.tok.IsKeyword() {
+ elit = e.lit
+ }
+ }
+ if lit != elit {
+ t.Errorf("bad literal for %q: got %q, expected %q", lit, lit, elit)
+ }
+
+ if tok == token.EOF {
+ break
+ }
+
+ // update position
+ epos.Offset += len(e.lit) + len(whitespace)
+ epos.Line += newlineCount(e.lit) + whitespace_linecount
+
+ }
+
+ if s.ErrorCount != 0 {
+ t.Errorf("found %d errors", s.ErrorCount)
+ }
+}
+
+func TestStripCR(t *testing.T) {
+ for _, test := range []struct{ have, want string }{
+ {"//\n", "//\n"},
+ {"//\r\n", "//\n"},
+ {"//\r\r\r\n", "//\n"},
+ {"//\r*\r/\r\n", "//*/\n"},
+ {"/**/", "/**/"},
+ {"/*\r/*/", "/*/*/"},
+ {"/*\r*/", "/**/"},
+ {"/**\r/*/", "/**\r/*/"},
+ {"/*\r/\r*\r/*/", "/*/*\r/*/"},
+ {"/*\r\r\r\r*/", "/**/"},
+ } {
+ got := string(stripCR([]byte(test.have), len(test.have) >= 2 && test.have[1] == '*'))
+ if got != test.want {
+ t.Errorf("stripCR(%q) = %q; want %q", test.have, got, test.want)
+ }
+ }
+}
+
+func checkSemi(t *testing.T, line string, mode Mode) {
+ var S Scanner
+ file := fset.AddFile("TestSemis", fset.Base(), len(line))
+ S.Init(file, []byte(line), nil, mode)
+ pos, tok, lit := S.Scan()
+ for tok != token.EOF {
+ if tok == token.ILLEGAL {
+ // the illegal token literal indicates what
+ // kind of semicolon literal to expect
+ semiLit := "\n"
+ if lit[0] == '#' {
+ semiLit = ";"
+ }
+ // next token must be a semicolon
+ semiPos := file.Position(pos)
+ semiPos.Offset++
+ semiPos.Column++
+ pos, tok, lit = S.Scan()
+ if tok == token.SEMICOLON {
+ if lit != semiLit {
+ t.Errorf(`bad literal for %q: got %q, expected %q`, line, lit, semiLit)
+ }
+ checkPos(t, line, pos, semiPos)
+ } else {
+ t.Errorf("bad token for %q: got %s, expected ;", line, tok)
+ }
+ } else if tok == token.SEMICOLON {
+ t.Errorf("bad token for %q: got ;, expected no ;", line)
+ }
+ pos, tok, lit = S.Scan()
+ }
+}
+
+var lines = []string{
+ // # indicates a semicolon present in the source
+ // $ indicates an automatically inserted semicolon
+ "",
+ "\ufeff#;", // first BOM is ignored
+ "#;",
+ "foo$\n",
+ "123$\n",
+ "1.2$\n",
+ "'x'$\n",
+ `"x"` + "$\n",
+ "`x`$\n",
+
+ "+\n",
+ "-\n",
+ "*\n",
+ "/\n",
+ "%\n",
+
+ "&\n",
+ "|\n",
+ "^\n",
+ "<<\n",
+ ">>\n",
+ "&^\n",
+
+ "+=\n",
+ "-=\n",
+ "*=\n",
+ "/=\n",
+ "%=\n",
+
+ "&=\n",
+ "|=\n",
+ "^=\n",
+ "<<=\n",
+ ">>=\n",
+ "&^=\n",
+
+ "&&\n",
+ "||\n",
+ "<-\n",
+ "++$\n",
+ "--$\n",
+
+ "==\n",
+ "<\n",
+ ">\n",
+ "=\n",
+ "!\n",
+
+ "!=\n",
+ "<=\n",
+ ">=\n",
+ ":=\n",
+ "...\n",
+
+ "(\n",
+ "[\n",
+ "{\n",
+ ",\n",
+ ".\n",
+
+ ")$\n",
+ "]$\n",
+ "}$\n",
+ "#;\n",
+ ":\n",
+
+ "break$\n",
+ "case\n",
+ "chan\n",
+ "const\n",
+ "continue$\n",
+
+ "default\n",
+ "defer\n",
+ "else\n",
+ "fallthrough$\n",
+ "for\n",
+
+ "func\n",
+ "go\n",
+ "goto\n",
+ "if\n",
+ "import\n",
+
+ "interface\n",
+ "map\n",
+ "package\n",
+ "range\n",
+ "return$\n",
+
+ "select\n",
+ "struct\n",
+ "switch\n",
+ "type\n",
+ "var\n",
+
+ "foo$//comment\n",
+ "foo$//comment",
+ "foo$/*comment*/\n",
+ "foo$/*\n*/",
+ "foo$/*comment*/ \n",
+ "foo$/*\n*/ ",
+
+ "foo $// comment\n",
+ "foo $// comment",
+ "foo $/*comment*/\n",
+ "foo $/*\n*/",
+ "foo $/* */ /* \n */ bar$/**/\n",
+ "foo $/*0*/ /*1*/ /*2*/\n",
+
+ "foo $/*comment*/ \n",
+ "foo $/*0*/ /*1*/ /*2*/ \n",
+ "foo $/**/ /*-------------*/ /*----\n*/bar $/* \n*/baa$\n",
+ "foo $/* an EOF terminates a line */",
+ "foo $/* an EOF terminates a line */ /*",
+ "foo $/* an EOF terminates a line */ //",
+
+ "package main$\n\nfunc main() {\n\tif {\n\t\treturn /* */ }$\n}$\n",
+ "package main$",
+}
+
+func TestSemis(t *testing.T) {
+ for _, line := range lines {
+ checkSemi(t, line, 0)
+ checkSemi(t, line, ScanComments)
+
+ // if the input ended in newlines, the input must tokenize the
+ // same with or without those newlines
+ for i := len(line) - 1; i >= 0 && line[i] == '\n'; i-- {
+ checkSemi(t, line[0:i], 0)
+ checkSemi(t, line[0:i], ScanComments)
+ }
+ }
+}
+
+type segment struct {
+ srcline string // a line of source text
+ filename string // filename for current token; error message for invalid line directives
+ line, column int // line and column for current token; error position for invalid line directives
+}
+
+var segments = []segment{
+ // exactly one token per line since the test consumes one token per segment
+ {" line1", "TestLineDirectives", 1, 3},
+ {"\nline2", "TestLineDirectives", 2, 1},
+ {"\nline3 //line File1.go:100", "TestLineDirectives", 3, 1}, // bad line comment, ignored
+ {"\nline4", "TestLineDirectives", 4, 1},
+ {"\n//line File1.go:100\n line100", "File1.go", 100, 0},
+ {"\n//line \t :42\n line1", " \t ", 42, 0},
+ {"\n//line File2.go:200\n line200", "File2.go", 200, 0},
+ {"\n//line foo\t:42\n line42", "foo\t", 42, 0},
+ {"\n //line foo:42\n line43", "foo\t", 44, 0}, // bad line comment, ignored (use existing, prior filename)
+ {"\n//line foo 42\n line44", "foo\t", 46, 0}, // bad line comment, ignored (use existing, prior filename)
+ {"\n//line /bar:42\n line45", "/bar", 42, 0},
+ {"\n//line ./foo:42\n line46", "foo", 42, 0},
+ {"\n//line a/b/c/File1.go:100\n line100", "a/b/c/File1.go", 100, 0},
+ {"\n//line c:\\bar:42\n line200", "c:\\bar", 42, 0},
+ {"\n//line c:\\dir\\File1.go:100\n line201", "c:\\dir\\File1.go", 100, 0},
+
+ // tests for new line directive syntax
+ {"\n//line :100\na1", "", 100, 0}, // missing filename means empty filename
+ {"\n//line bar:100\nb1", "bar", 100, 0},
+ {"\n//line :100:10\nc1", "bar", 100, 10}, // missing filename means current filename
+ {"\n//line foo:100:10\nd1", "foo", 100, 10},
+
+ {"\n/*line :100*/a2", "", 100, 0}, // missing filename means empty filename
+ {"\n/*line bar:100*/b2", "bar", 100, 0},
+ {"\n/*line :100:10*/c2", "bar", 100, 10}, // missing filename means current filename
+ {"\n/*line foo:100:10*/d2", "foo", 100, 10},
+ {"\n/*line foo:100:10*/ e2", "foo", 100, 14}, // line-directive relative column
+ {"\n/*line foo:100:10*/\n\nf2", "foo", 102, 1}, // absolute column since on new line
+}
+
+var dirsegments = []segment{
+ // exactly one token per line since the test consumes one token per segment
+ {" line1", "TestLineDir/TestLineDirectives", 1, 3},
+ {"\n//line File1.go:100\n line100", "TestLineDir/File1.go", 100, 0},
+}
+
+var dirUnixSegments = []segment{
+ {"\n//line /bar:42\n line42", "/bar", 42, 0},
+}
+
+var dirWindowsSegments = []segment{
+ {"\n//line c:\\bar:42\n line42", "c:\\bar", 42, 0},
+}
+
+// Verify that line directives are interpreted correctly.
+func TestLineDirectives(t *testing.T) {
+ testSegments(t, segments, "TestLineDirectives")
+ testSegments(t, dirsegments, "TestLineDir/TestLineDirectives")
+ if runtime.GOOS == "windows" {
+ testSegments(t, dirWindowsSegments, "TestLineDir/TestLineDirectives")
+ } else {
+ testSegments(t, dirUnixSegments, "TestLineDir/TestLineDirectives")
+ }
+}
+
+func testSegments(t *testing.T, segments []segment, filename string) {
+ var src string
+ for _, e := range segments {
+ src += e.srcline
+ }
+
+ // verify scan
+ var S Scanner
+ file := fset.AddFile(filename, fset.Base(), len(src))
+ S.Init(file, []byte(src), func(pos token.Position, msg string) { t.Error(Error{pos, msg}) }, dontInsertSemis)
+ for _, s := range segments {
+ p, _, lit := S.Scan()
+ pos := file.Position(p)
+ checkPos(t, lit, p, token.Position{
+ Filename: s.filename,
+ Offset: pos.Offset,
+ Line: s.line,
+ Column: s.column,
+ })
+ }
+
+ if S.ErrorCount != 0 {
+ t.Errorf("got %d errors", S.ErrorCount)
+ }
+}
+
+// The filename is used for the error message in these test cases.
+// The first line directive is valid and used to control the expected error line.
+var invalidSegments = []segment{
+ {"\n//line :1:1\n//line foo:42 extra text\ndummy", "invalid line number: 42 extra text", 1, 12},
+ {"\n//line :2:1\n//line foobar:\ndummy", "invalid line number: ", 2, 15},
+ {"\n//line :5:1\n//line :0\ndummy", "invalid line number: 0", 5, 9},
+ {"\n//line :10:1\n//line :1:0\ndummy", "invalid column number: 0", 10, 11},
+ {"\n//line :1:1\n//line :foo:0\ndummy", "invalid line number: 0", 1, 13}, // foo is considered part of the filename
+}
+
+// Verify that invalid line directives get the correct error message.
+func TestInvalidLineDirectives(t *testing.T) {
+ // make source
+ var src string
+ for _, e := range invalidSegments {
+ src += e.srcline
+ }
+
+ // verify scan
+ var S Scanner
+ var s segment // current segment
+ file := fset.AddFile(filepath.Join("dir", "TestInvalidLineDirectives"), fset.Base(), len(src))
+ S.Init(file, []byte(src), func(pos token.Position, msg string) {
+ if msg != s.filename {
+ t.Errorf("got error %q; want %q", msg, s.filename)
+ }
+ if pos.Line != s.line || pos.Column != s.column {
+ t.Errorf("got position %d:%d; want %d:%d", pos.Line, pos.Column, s.line, s.column)
+ }
+ }, dontInsertSemis)
+ for _, s = range invalidSegments {
+ S.Scan()
+ }
+
+ if S.ErrorCount != len(invalidSegments) {
+ t.Errorf("got %d errors; want %d", S.ErrorCount, len(invalidSegments))
+ }
+}
+
+// Verify that initializing the same scanner more than once works correctly.
+func TestInit(t *testing.T) {
+ var s Scanner
+
+ // 1st init
+ src1 := "if true { }"
+ f1 := fset.AddFile("src1", fset.Base(), len(src1))
+ s.Init(f1, []byte(src1), nil, dontInsertSemis)
+ if f1.Size() != len(src1) {
+ t.Errorf("bad file size: got %d, expected %d", f1.Size(), len(src1))
+ }
+ s.Scan() // if
+ s.Scan() // true
+ _, tok, _ := s.Scan() // {
+ if tok != token.LBRACE {
+ t.Errorf("bad token: got %s, expected %s", tok, token.LBRACE)
+ }
+
+ // 2nd init
+ src2 := "go true { ]"
+ f2 := fset.AddFile("src2", fset.Base(), len(src2))
+ s.Init(f2, []byte(src2), nil, dontInsertSemis)
+ if f2.Size() != len(src2) {
+ t.Errorf("bad file size: got %d, expected %d", f2.Size(), len(src2))
+ }
+ _, tok, _ = s.Scan() // go
+ if tok != token.GO {
+ t.Errorf("bad token: got %s, expected %s", tok, token.GO)
+ }
+
+ if s.ErrorCount != 0 {
+ t.Errorf("found %d errors", s.ErrorCount)
+ }
+}
+
+func TestStdErrorHander(t *testing.T) {
+ const src = "@\n" + // illegal character, cause an error
+ "@ @\n" + // two errors on the same line
+ "//line File2:20\n" +
+ "@\n" + // different file, but same line
+ "//line File2:1\n" +
+ "@ @\n" + // same file, decreasing line number
+ "//line File1:1\n" +
+ "@ @ @" // original file, line 1 again
+
+ var list ErrorList
+ eh := func(pos token.Position, msg string) { list.Add(pos, msg) }
+
+ var s Scanner
+ s.Init(fset.AddFile("File1", fset.Base(), len(src)), []byte(src), eh, dontInsertSemis)
+ for {
+ if _, tok, _ := s.Scan(); tok == token.EOF {
+ break
+ }
+ }
+
+ if len(list) != s.ErrorCount {
+ t.Errorf("found %d errors, expected %d", len(list), s.ErrorCount)
+ }
+
+ if len(list) != 9 {
+ t.Errorf("found %d raw errors, expected 9", len(list))
+ PrintError(os.Stderr, list)
+ }
+
+ list.Sort()
+ if len(list) != 9 {
+ t.Errorf("found %d sorted errors, expected 9", len(list))
+ PrintError(os.Stderr, list)
+ }
+
+ list.RemoveMultiples()
+ if len(list) != 4 {
+ t.Errorf("found %d one-per-line errors, expected 4", len(list))
+ PrintError(os.Stderr, list)
+ }
+}
+
+type errorCollector struct {
+ cnt int // number of errors encountered
+ msg string // last error message encountered
+ pos token.Position // last error position encountered
+}
+
+func checkError(t *testing.T, src string, tok token.Token, pos int, lit, err string) {
+ var s Scanner
+ var h errorCollector
+ eh := func(pos token.Position, msg string) {
+ h.cnt++
+ h.msg = msg
+ h.pos = pos
+ }
+ s.Init(fset.AddFile("", fset.Base(), len(src)), []byte(src), eh, ScanComments|dontInsertSemis)
+ _, tok0, lit0 := s.Scan()
+ if tok0 != tok {
+ t.Errorf("%q: got %s, expected %s", src, tok0, tok)
+ }
+ if tok0 != token.ILLEGAL && lit0 != lit {
+ t.Errorf("%q: got literal %q, expected %q", src, lit0, lit)
+ }
+ cnt := 0
+ if err != "" {
+ cnt = 1
+ }
+ if h.cnt != cnt {
+ t.Errorf("%q: got cnt %d, expected %d", src, h.cnt, cnt)
+ }
+ if h.msg != err {
+ t.Errorf("%q: got msg %q, expected %q", src, h.msg, err)
+ }
+ if h.pos.Offset != pos {
+ t.Errorf("%q: got offset %d, expected %d", src, h.pos.Offset, pos)
+ }
+}
+
+var errors = []struct {
+ src string
+ tok token.Token
+ pos int
+ lit string
+ err string
+}{
+ {"\a", token.ILLEGAL, 0, "", "illegal character U+0007"},
+ {`#`, token.ILLEGAL, 0, "", "illegal character U+0023 '#'"},
+ {`…`, token.ILLEGAL, 0, "", "illegal character U+2026 '…'"},
+ {"..", token.PERIOD, 0, "", ""}, // two periods, not invalid token (issue #28112)
+ {`' '`, token.CHAR, 0, `' '`, ""},
+ {`''`, token.CHAR, 0, `''`, "illegal rune literal"},
+ {`'12'`, token.CHAR, 0, `'12'`, "illegal rune literal"},
+ {`'123'`, token.CHAR, 0, `'123'`, "illegal rune literal"},
+ {`'\0'`, token.CHAR, 3, `'\0'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\07'`, token.CHAR, 4, `'\07'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\8'`, token.CHAR, 2, `'\8'`, "unknown escape sequence"},
+ {`'\08'`, token.CHAR, 3, `'\08'`, "illegal character U+0038 '8' in escape sequence"},
+ {`'\x'`, token.CHAR, 3, `'\x'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\x0'`, token.CHAR, 4, `'\x0'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\x0g'`, token.CHAR, 4, `'\x0g'`, "illegal character U+0067 'g' in escape sequence"},
+ {`'\u'`, token.CHAR, 3, `'\u'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\u0'`, token.CHAR, 4, `'\u0'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\u00'`, token.CHAR, 5, `'\u00'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\u000'`, token.CHAR, 6, `'\u000'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\u000`, token.CHAR, 6, `'\u000`, "escape sequence not terminated"},
+ {`'\u0000'`, token.CHAR, 0, `'\u0000'`, ""},
+ {`'\U'`, token.CHAR, 3, `'\U'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\U0'`, token.CHAR, 4, `'\U0'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\U00'`, token.CHAR, 5, `'\U00'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\U000'`, token.CHAR, 6, `'\U000'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\U0000'`, token.CHAR, 7, `'\U0000'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\U00000'`, token.CHAR, 8, `'\U00000'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\U000000'`, token.CHAR, 9, `'\U000000'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\U0000000'`, token.CHAR, 10, `'\U0000000'`, "illegal character U+0027 ''' in escape sequence"},
+ {`'\U0000000`, token.CHAR, 10, `'\U0000000`, "escape sequence not terminated"},
+ {`'\U00000000'`, token.CHAR, 0, `'\U00000000'`, ""},
+ {`'\Uffffffff'`, token.CHAR, 2, `'\Uffffffff'`, "escape sequence is invalid Unicode code point"},
+ {`'`, token.CHAR, 0, `'`, "rune literal not terminated"},
+ {`'\`, token.CHAR, 2, `'\`, "escape sequence not terminated"},
+ {"'\n", token.CHAR, 0, "'", "rune literal not terminated"},
+ {"'\n ", token.CHAR, 0, "'", "rune literal not terminated"},
+ {`""`, token.STRING, 0, `""`, ""},
+ {`"abc`, token.STRING, 0, `"abc`, "string literal not terminated"},
+ {"\"abc\n", token.STRING, 0, `"abc`, "string literal not terminated"},
+ {"\"abc\n ", token.STRING, 0, `"abc`, "string literal not terminated"},
+ {"``", token.STRING, 0, "``", ""},
+ {"`", token.STRING, 0, "`", "raw string literal not terminated"},
+ {"/**/", token.COMMENT, 0, "/**/", ""},
+ {"/*", token.COMMENT, 0, "/*", "comment not terminated"},
+ {"077", token.INT, 0, "077", ""},
+ {"078.", token.FLOAT, 0, "078.", ""},
+ {"07801234567.", token.FLOAT, 0, "07801234567.", ""},
+ {"078e0", token.FLOAT, 0, "078e0", ""},
+ {"0E", token.FLOAT, 2, "0E", "exponent has no digits"}, // issue 17621
+ {"078", token.INT, 2, "078", "invalid digit '8' in octal literal"},
+ {"07090000008", token.INT, 3, "07090000008", "invalid digit '9' in octal literal"},
+ {"0x", token.INT, 2, "0x", "hexadecimal literal has no digits"},
+ {"\"abc\x00def\"", token.STRING, 4, "\"abc\x00def\"", "illegal character NUL"},
+ {"\"abc\x80def\"", token.STRING, 4, "\"abc\x80def\"", "illegal UTF-8 encoding"},
+ {"\ufeff\ufeff", token.ILLEGAL, 3, "\ufeff\ufeff", "illegal byte order mark"}, // only first BOM is ignored
+ {"//\ufeff", token.COMMENT, 2, "//\ufeff", "illegal byte order mark"}, // only first BOM is ignored
+ {"'\ufeff" + `'`, token.CHAR, 1, "'\ufeff" + `'`, "illegal byte order mark"}, // only first BOM is ignored
+ {`"` + "abc\ufeffdef" + `"`, token.STRING, 4, `"` + "abc\ufeffdef" + `"`, "illegal byte order mark"}, // only first BOM is ignored
+ {"abc\x00def", token.IDENT, 3, "abc", "illegal character NUL"},
+ {"abc\x00", token.IDENT, 3, "abc", "illegal character NUL"},
+}
+
+func TestScanErrors(t *testing.T) {
+ for _, e := range errors {
+ checkError(t, e.src, e.tok, e.pos, e.lit, e.err)
+ }
+}
+
+// Verify that no comments show up as literal values when skipping comments.
+func TestIssue10213(t *testing.T) {
+ const src = `
+ var (
+ A = 1 // foo
+ )
+
+ var (
+ B = 2
+ // foo
+ )
+
+ var C = 3 // foo
+
+ var D = 4
+ // foo
+
+ func anycode() {
+ // foo
+ }
+ `
+ var s Scanner
+ s.Init(fset.AddFile("", fset.Base(), len(src)), []byte(src), nil, 0)
+ for {
+ pos, tok, lit := s.Scan()
+ class := tokenclass(tok)
+ if lit != "" && class != keyword && class != literal && tok != token.SEMICOLON {
+ t.Errorf("%s: tok = %s, lit = %q", fset.Position(pos), tok, lit)
+ }
+ if tok <= token.EOF {
+ break
+ }
+ }
+}
+
+func TestIssue28112(t *testing.T) {
+ const src = "... .. 0.. .." // make sure to have stand-alone ".." immediately before EOF to test EOF behavior
+ tokens := []token.Token{token.ELLIPSIS, token.PERIOD, token.PERIOD, token.FLOAT, token.PERIOD, token.PERIOD, token.PERIOD, token.EOF}
+ var s Scanner
+ s.Init(fset.AddFile("", fset.Base(), len(src)), []byte(src), nil, 0)
+ for _, want := range tokens {
+ pos, got, lit := s.Scan()
+ if got != want {
+ t.Errorf("%s: got %s, want %s", fset.Position(pos), got, want)
+ }
+ // literals expect to have a (non-empty) literal string and we don't care about other tokens for this test
+ if tokenclass(got) == literal && lit == "" {
+ t.Errorf("%s: for %s got empty literal string", fset.Position(pos), got)
+ }
+ }
+}
+
+func BenchmarkScan(b *testing.B) {
+ b.StopTimer()
+ fset := token.NewFileSet()
+ file := fset.AddFile("", fset.Base(), len(source))
+ var s Scanner
+ b.StartTimer()
+ for i := 0; i < b.N; i++ {
+ s.Init(file, source, nil, ScanComments)
+ for {
+ _, tok, _ := s.Scan()
+ if tok == token.EOF {
+ break
+ }
+ }
+ }
+}
+
+func BenchmarkScanFiles(b *testing.B) {
+ // Scan a few arbitrary large files, and one small one, to provide some
+ // variety in benchmarks.
+ for _, p := range []string{
+ "go/types/expr.go",
+ "go/parser/parser.go",
+ "net/http/server.go",
+ "go/scanner/errors.go",
+ } {
+ b.Run(p, func(b *testing.B) {
+ b.StopTimer()
+ filename := filepath.Join("..", "..", filepath.FromSlash(p))
+ src, err := os.ReadFile(filename)
+ if err != nil {
+ b.Fatal(err)
+ }
+ fset := token.NewFileSet()
+ file := fset.AddFile(filename, fset.Base(), len(src))
+ b.SetBytes(int64(len(src)))
+ var s Scanner
+ b.StartTimer()
+ for i := 0; i < b.N; i++ {
+ s.Init(file, src, nil, ScanComments)
+ for {
+ _, tok, _ := s.Scan()
+ if tok == token.EOF {
+ break
+ }
+ }
+ }
+ })
+ }
+}
+
+func TestNumbers(t *testing.T) {
+ for _, test := range []struct {
+ tok token.Token
+ src, tokens, err string
+ }{
+ // binaries
+ {token.INT, "0b0", "0b0", ""},
+ {token.INT, "0b1010", "0b1010", ""},
+ {token.INT, "0B1110", "0B1110", ""},
+
+ {token.INT, "0b", "0b", "binary literal has no digits"},
+ {token.INT, "0b0190", "0b0190", "invalid digit '9' in binary literal"},
+ {token.INT, "0b01a0", "0b01 a0", ""}, // only accept 0-9
+
+ {token.FLOAT, "0b.", "0b.", "invalid radix point in binary literal"},
+ {token.FLOAT, "0b.1", "0b.1", "invalid radix point in binary literal"},
+ {token.FLOAT, "0b1.0", "0b1.0", "invalid radix point in binary literal"},
+ {token.FLOAT, "0b1e10", "0b1e10", "'e' exponent requires decimal mantissa"},
+ {token.FLOAT, "0b1P-1", "0b1P-1", "'P' exponent requires hexadecimal mantissa"},
+
+ {token.IMAG, "0b10i", "0b10i", ""},
+ {token.IMAG, "0b10.0i", "0b10.0i", "invalid radix point in binary literal"},
+
+ // octals
+ {token.INT, "0o0", "0o0", ""},
+ {token.INT, "0o1234", "0o1234", ""},
+ {token.INT, "0O1234", "0O1234", ""},
+
+ {token.INT, "0o", "0o", "octal literal has no digits"},
+ {token.INT, "0o8123", "0o8123", "invalid digit '8' in octal literal"},
+ {token.INT, "0o1293", "0o1293", "invalid digit '9' in octal literal"},
+ {token.INT, "0o12a3", "0o12 a3", ""}, // only accept 0-9
+
+ {token.FLOAT, "0o.", "0o.", "invalid radix point in octal literal"},
+ {token.FLOAT, "0o.2", "0o.2", "invalid radix point in octal literal"},
+ {token.FLOAT, "0o1.2", "0o1.2", "invalid radix point in octal literal"},
+ {token.FLOAT, "0o1E+2", "0o1E+2", "'E' exponent requires decimal mantissa"},
+ {token.FLOAT, "0o1p10", "0o1p10", "'p' exponent requires hexadecimal mantissa"},
+
+ {token.IMAG, "0o10i", "0o10i", ""},
+ {token.IMAG, "0o10e0i", "0o10e0i", "'e' exponent requires decimal mantissa"},
+
+ // 0-octals
+ {token.INT, "0", "0", ""},
+ {token.INT, "0123", "0123", ""},
+
+ {token.INT, "08123", "08123", "invalid digit '8' in octal literal"},
+ {token.INT, "01293", "01293", "invalid digit '9' in octal literal"},
+ {token.INT, "0F.", "0 F .", ""}, // only accept 0-9
+ {token.INT, "0123F.", "0123 F .", ""},
+ {token.INT, "0123456x", "0123456 x", ""},
+
+ // decimals
+ {token.INT, "1", "1", ""},
+ {token.INT, "1234", "1234", ""},
+
+ {token.INT, "1f", "1 f", ""}, // only accept 0-9
+
+ {token.IMAG, "0i", "0i", ""},
+ {token.IMAG, "0678i", "0678i", ""},
+
+ // decimal floats
+ {token.FLOAT, "0.", "0.", ""},
+ {token.FLOAT, "123.", "123.", ""},
+ {token.FLOAT, "0123.", "0123.", ""},
+
+ {token.FLOAT, ".0", ".0", ""},
+ {token.FLOAT, ".123", ".123", ""},
+ {token.FLOAT, ".0123", ".0123", ""},
+
+ {token.FLOAT, "0.0", "0.0", ""},
+ {token.FLOAT, "123.123", "123.123", ""},
+ {token.FLOAT, "0123.0123", "0123.0123", ""},
+
+ {token.FLOAT, "0e0", "0e0", ""},
+ {token.FLOAT, "123e+0", "123e+0", ""},
+ {token.FLOAT, "0123E-1", "0123E-1", ""},
+
+ {token.FLOAT, "0.e+1", "0.e+1", ""},
+ {token.FLOAT, "123.E-10", "123.E-10", ""},
+ {token.FLOAT, "0123.e123", "0123.e123", ""},
+
+ {token.FLOAT, ".0e-1", ".0e-1", ""},
+ {token.FLOAT, ".123E+10", ".123E+10", ""},
+ {token.FLOAT, ".0123E123", ".0123E123", ""},
+
+ {token.FLOAT, "0.0e1", "0.0e1", ""},
+ {token.FLOAT, "123.123E-10", "123.123E-10", ""},
+ {token.FLOAT, "0123.0123e+456", "0123.0123e+456", ""},
+
+ {token.FLOAT, "0e", "0e", "exponent has no digits"},
+ {token.FLOAT, "0E+", "0E+", "exponent has no digits"},
+ {token.FLOAT, "1e+f", "1e+ f", "exponent has no digits"},
+ {token.FLOAT, "0p0", "0p0", "'p' exponent requires hexadecimal mantissa"},
+ {token.FLOAT, "1.0P-1", "1.0P-1", "'P' exponent requires hexadecimal mantissa"},
+
+ {token.IMAG, "0.i", "0.i", ""},
+ {token.IMAG, ".123i", ".123i", ""},
+ {token.IMAG, "123.123i", "123.123i", ""},
+ {token.IMAG, "123e+0i", "123e+0i", ""},
+ {token.IMAG, "123.E-10i", "123.E-10i", ""},
+ {token.IMAG, ".123E+10i", ".123E+10i", ""},
+
+ // hexadecimals
+ {token.INT, "0x0", "0x0", ""},
+ {token.INT, "0x1234", "0x1234", ""},
+ {token.INT, "0xcafef00d", "0xcafef00d", ""},
+ {token.INT, "0XCAFEF00D", "0XCAFEF00D", ""},
+
+ {token.INT, "0x", "0x", "hexadecimal literal has no digits"},
+ {token.INT, "0x1g", "0x1 g", ""},
+
+ {token.IMAG, "0xf00i", "0xf00i", ""},
+
+ // hexadecimal floats
+ {token.FLOAT, "0x0p0", "0x0p0", ""},
+ {token.FLOAT, "0x12efp-123", "0x12efp-123", ""},
+ {token.FLOAT, "0xABCD.p+0", "0xABCD.p+0", ""},
+ {token.FLOAT, "0x.0189P-0", "0x.0189P-0", ""},
+ {token.FLOAT, "0x1.ffffp+1023", "0x1.ffffp+1023", ""},
+
+ {token.FLOAT, "0x.", "0x.", "hexadecimal literal has no digits"},
+ {token.FLOAT, "0x0.", "0x0.", "hexadecimal mantissa requires a 'p' exponent"},
+ {token.FLOAT, "0x.0", "0x.0", "hexadecimal mantissa requires a 'p' exponent"},
+ {token.FLOAT, "0x1.1", "0x1.1", "hexadecimal mantissa requires a 'p' exponent"},
+ {token.FLOAT, "0x1.1e0", "0x1.1e0", "hexadecimal mantissa requires a 'p' exponent"},
+ {token.FLOAT, "0x1.2gp1a", "0x1.2 gp1a", "hexadecimal mantissa requires a 'p' exponent"},
+ {token.FLOAT, "0x0p", "0x0p", "exponent has no digits"},
+ {token.FLOAT, "0xeP-", "0xeP-", "exponent has no digits"},
+ {token.FLOAT, "0x1234PAB", "0x1234P AB", "exponent has no digits"},
+ {token.FLOAT, "0x1.2p1a", "0x1.2p1 a", ""},
+
+ {token.IMAG, "0xf00.bap+12i", "0xf00.bap+12i", ""},
+
+ // separators
+ {token.INT, "0b_1000_0001", "0b_1000_0001", ""},
+ {token.INT, "0o_600", "0o_600", ""},
+ {token.INT, "0_466", "0_466", ""},
+ {token.INT, "1_000", "1_000", ""},
+ {token.FLOAT, "1_000.000_1", "1_000.000_1", ""},
+ {token.IMAG, "10e+1_2_3i", "10e+1_2_3i", ""},
+ {token.INT, "0x_f00d", "0x_f00d", ""},
+ {token.FLOAT, "0x_f00d.0p1_2", "0x_f00d.0p1_2", ""},
+
+ {token.INT, "0b__1000", "0b__1000", "'_' must separate successive digits"},
+ {token.INT, "0o60___0", "0o60___0", "'_' must separate successive digits"},
+ {token.INT, "0466_", "0466_", "'_' must separate successive digits"},
+ {token.FLOAT, "1_.", "1_.", "'_' must separate successive digits"},
+ {token.FLOAT, "0._1", "0._1", "'_' must separate successive digits"},
+ {token.FLOAT, "2.7_e0", "2.7_e0", "'_' must separate successive digits"},
+ {token.IMAG, "10e+12_i", "10e+12_i", "'_' must separate successive digits"},
+ {token.INT, "0x___0", "0x___0", "'_' must separate successive digits"},
+ {token.FLOAT, "0x1.0_p0", "0x1.0_p0", "'_' must separate successive digits"},
+ } {
+ var s Scanner
+ var err string
+ s.Init(fset.AddFile("", fset.Base(), len(test.src)), []byte(test.src), func(_ token.Position, msg string) {
+ if err == "" {
+ err = msg
+ }
+ }, 0)
+ for i, want := range strings.Split(test.tokens, " ") {
+ err = ""
+ _, tok, lit := s.Scan()
+
+ // compute lit where for tokens where lit is not defined
+ switch tok {
+ case token.PERIOD:
+ lit = "."
+ case token.ADD:
+ lit = "+"
+ case token.SUB:
+ lit = "-"
+ }
+
+ if i == 0 {
+ if tok != test.tok {
+ t.Errorf("%q: got token %s; want %s", test.src, tok, test.tok)
+ }
+ if err != test.err {
+ t.Errorf("%q: got error %q; want %q", test.src, err, test.err)
+ }
+ }
+
+ if lit != want {
+ t.Errorf("%q: got literal %q (%s); want %s", test.src, lit, tok, want)
+ }
+ }
+
+ // make sure we read all
+ _, tok, _ := s.Scan()
+ if tok == token.SEMICOLON {
+ _, tok, _ = s.Scan()
+ }
+ if tok != token.EOF {
+ t.Errorf("%q: got %s; want EOF", test.src, tok)
+ }
+ }
+}
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-01 03:50:02 +0000