From: Marvin Stenger Date: Sun, 20 Sep 2015 22:45:21 +0000 (+0200) Subject: encoding/json: scanner: use byte, more consistent X-Git-Tag: go1.6beta1~1020 X-Git-Url: http://www.git.cypherpunks.su/?a=commitdiff_plain;h=fcf8143d638dbc34fdd418f45f8e8b45de802916;p=gostls13.git encoding/json: scanner: use byte, more consistent The fields step and redoState of struct scanner are now defined as `func(s *scanner, c byte) int` instead of `func(s *scanner, c int) int`, since bytes are sufficient. Further changes improve the consistency in the scanner.go file. Change-Id: Ifb85f2130d728d2b936d79914d87a1f0b5c6ee7d Reviewed-on: https://go-review.googlesource.com/14801 Reviewed-by: Andrew Gerrand Run-TryBot: Andrew Gerrand TryBot-Result: Gobot Gobot --- diff --git a/src/encoding/json/decode.go b/src/encoding/json/decode.go index 530e8521dc..0f5bb4d574 100644 --- a/src/encoding/json/decode.go +++ b/src/encoding/json/decode.go @@ -241,7 +241,7 @@ func (d *decodeState) scanWhile(op int) int { newOp = d.scan.eof() d.off = len(d.data) + 1 // mark processed EOF with len+1 } else { - c := int(d.data[d.off]) + c := d.data[d.off] d.off++ newOp = d.scan.step(&d.scan, c) } diff --git a/src/encoding/json/decode_test.go b/src/encoding/json/decode_test.go index 8aa158f08c..8c3b92b823 100644 --- a/src/encoding/json/decode_test.go +++ b/src/encoding/json/decode_test.go @@ -716,7 +716,7 @@ func TestErrorMessageFromMisusedString(t *testing.T) { } func noSpace(c rune) rune { - if isSpace(c) { + if isSpace(byte(c)) { //only used for ascii return -1 } return c diff --git a/src/encoding/json/indent.go b/src/encoding/json/indent.go index e1bacafd6b..153109fae6 100644 --- a/src/encoding/json/indent.go +++ b/src/encoding/json/indent.go @@ -36,7 +36,7 @@ func compact(dst *bytes.Buffer, src []byte, escape bool) error { dst.WriteByte(hex[src[i+2]&0xF]) start = i + 3 } - v := scan.step(&scan, int(c)) + v := scan.step(&scan, c) if v >= scanSkipSpace { if v == scanError { break @@ -80,7 +80,7 @@ func Indent(dst *bytes.Buffer, src []byte, prefix, indent string) error { depth := 0 for _, c := range src { scan.bytes++ - v := scan.step(&scan, int(c)) + v := scan.step(&scan, c) if v == scanSkipSpace { continue } diff --git a/src/encoding/json/scanner.go b/src/encoding/json/scanner.go index 38d0b0802b..ee6622e8cf 100644 --- a/src/encoding/json/scanner.go +++ b/src/encoding/json/scanner.go @@ -21,7 +21,7 @@ func checkValid(data []byte, scan *scanner) error { scan.reset() for _, c := range data { scan.bytes++ - if scan.step(scan, int(c)) == scanError { + if scan.step(scan, c) == scanError { return scan.err } } @@ -37,7 +37,7 @@ func checkValid(data []byte, scan *scanner) error { func nextValue(data []byte, scan *scanner) (value, rest []byte, err error) { scan.reset() for i, c := range data { - v := scan.step(scan, int(c)) + v := scan.step(scan, c) if v >= scanEndObject { switch v { // probe the scanner with a space to determine whether we will @@ -50,7 +50,7 @@ func nextValue(data []byte, scan *scanner) (value, rest []byte, err error) { case scanError: return nil, nil, scan.err case scanEnd: - return data[0:i], data[i:], nil + return data[:i], data[i:], nil } } } @@ -85,7 +85,7 @@ type scanner struct { // Also tried using an integer constant and a single func // with a switch, but using the func directly was 10% faster // on a 64-bit Mac Mini, and it's nicer to read. - step func(*scanner, int) int + step func(*scanner, byte) int // Reached end of top-level value. endTop bool @@ -99,7 +99,7 @@ type scanner struct { // 1-byte redo (see undo method) redo bool redoCode int - redoState func(*scanner, int) int + redoState func(*scanner, byte) int // total bytes consumed, updated by decoder.Decode bytes int64 @@ -188,13 +188,13 @@ func (s *scanner) popParseState() { } } -func isSpace(c rune) bool { +func isSpace(c byte) bool { return c == ' ' || c == '\t' || c == '\r' || c == '\n' } // stateBeginValueOrEmpty is the state after reading `[`. -func stateBeginValueOrEmpty(s *scanner, c int) int { - if c <= ' ' && isSpace(rune(c)) { +func stateBeginValueOrEmpty(s *scanner, c byte) int { + if c <= ' ' && isSpace(c) { return scanSkipSpace } if c == ']' { @@ -204,8 +204,8 @@ func stateBeginValueOrEmpty(s *scanner, c int) int { } // stateBeginValue is the state at the beginning of the input. -func stateBeginValue(s *scanner, c int) int { - if c <= ' ' && isSpace(rune(c)) { +func stateBeginValue(s *scanner, c byte) int { + if c <= ' ' && isSpace(c) { return scanSkipSpace } switch c { @@ -244,8 +244,8 @@ func stateBeginValue(s *scanner, c int) int { } // stateBeginStringOrEmpty is the state after reading `{`. -func stateBeginStringOrEmpty(s *scanner, c int) int { - if c <= ' ' && isSpace(rune(c)) { +func stateBeginStringOrEmpty(s *scanner, c byte) int { + if c <= ' ' && isSpace(c) { return scanSkipSpace } if c == '}' { @@ -257,8 +257,8 @@ func stateBeginStringOrEmpty(s *scanner, c int) int { } // stateBeginString is the state after reading `{"key": value,`. -func stateBeginString(s *scanner, c int) int { - if c <= ' ' && isSpace(rune(c)) { +func stateBeginString(s *scanner, c byte) int { + if c <= ' ' && isSpace(c) { return scanSkipSpace } if c == '"' { @@ -270,7 +270,7 @@ func stateBeginString(s *scanner, c int) int { // stateEndValue is the state after completing a value, // such as after reading `{}` or `true` or `["x"`. -func stateEndValue(s *scanner, c int) int { +func stateEndValue(s *scanner, c byte) int { n := len(s.parseState) if n == 0 { // Completed top-level before the current byte. @@ -278,7 +278,7 @@ func stateEndValue(s *scanner, c int) int { s.endTop = true return stateEndTop(s, c) } - if c <= ' ' && isSpace(rune(c)) { + if c <= ' ' && isSpace(c) { s.step = stateEndValue return scanSkipSpace } @@ -319,7 +319,7 @@ func stateEndValue(s *scanner, c int) int { // stateEndTop is the state after finishing the top-level value, // such as after reading `{}` or `[1,2,3]`. // Only space characters should be seen now. -func stateEndTop(s *scanner, c int) int { +func stateEndTop(s *scanner, c byte) int { if c != ' ' && c != '\t' && c != '\r' && c != '\n' { // Complain about non-space byte on next call. s.error(c, "after top-level value") @@ -328,7 +328,7 @@ func stateEndTop(s *scanner, c int) int { } // stateInString is the state after reading `"`. -func stateInString(s *scanner, c int) int { +func stateInString(s *scanner, c byte) int { if c == '"' { s.step = stateEndValue return scanContinue @@ -344,13 +344,12 @@ func stateInString(s *scanner, c int) int { } // stateInStringEsc is the state after reading `"\` during a quoted string. -func stateInStringEsc(s *scanner, c int) int { +func stateInStringEsc(s *scanner, c byte) int { switch c { case 'b', 'f', 'n', 'r', 't', '\\', '/', '"': s.step = stateInString return scanContinue - } - if c == 'u' { + case 'u': s.step = stateInStringEscU return scanContinue } @@ -358,7 +357,7 @@ func stateInStringEsc(s *scanner, c int) int { } // stateInStringEscU is the state after reading `"\u` during a quoted string. -func stateInStringEscU(s *scanner, c int) int { +func stateInStringEscU(s *scanner, c byte) int { if '0' <= c && c <= '9' || 'a' <= c && c <= 'f' || 'A' <= c && c <= 'F' { s.step = stateInStringEscU1 return scanContinue @@ -368,7 +367,7 @@ func stateInStringEscU(s *scanner, c int) int { } // stateInStringEscU1 is the state after reading `"\u1` during a quoted string. -func stateInStringEscU1(s *scanner, c int) int { +func stateInStringEscU1(s *scanner, c byte) int { if '0' <= c && c <= '9' || 'a' <= c && c <= 'f' || 'A' <= c && c <= 'F' { s.step = stateInStringEscU12 return scanContinue @@ -378,7 +377,7 @@ func stateInStringEscU1(s *scanner, c int) int { } // stateInStringEscU12 is the state after reading `"\u12` during a quoted string. -func stateInStringEscU12(s *scanner, c int) int { +func stateInStringEscU12(s *scanner, c byte) int { if '0' <= c && c <= '9' || 'a' <= c && c <= 'f' || 'A' <= c && c <= 'F' { s.step = stateInStringEscU123 return scanContinue @@ -388,7 +387,7 @@ func stateInStringEscU12(s *scanner, c int) int { } // stateInStringEscU123 is the state after reading `"\u123` during a quoted string. -func stateInStringEscU123(s *scanner, c int) int { +func stateInStringEscU123(s *scanner, c byte) int { if '0' <= c && c <= '9' || 'a' <= c && c <= 'f' || 'A' <= c && c <= 'F' { s.step = stateInString return scanContinue @@ -398,7 +397,7 @@ func stateInStringEscU123(s *scanner, c int) int { } // stateNeg is the state after reading `-` during a number. -func stateNeg(s *scanner, c int) int { +func stateNeg(s *scanner, c byte) int { if c == '0' { s.step = state0 return scanContinue @@ -412,7 +411,7 @@ func stateNeg(s *scanner, c int) int { // state1 is the state after reading a non-zero integer during a number, // such as after reading `1` or `100` but not `0`. -func state1(s *scanner, c int) int { +func state1(s *scanner, c byte) int { if '0' <= c && c <= '9' { s.step = state1 return scanContinue @@ -421,7 +420,7 @@ func state1(s *scanner, c int) int { } // state0 is the state after reading `0` during a number. -func state0(s *scanner, c int) int { +func state0(s *scanner, c byte) int { if c == '.' { s.step = stateDot return scanContinue @@ -435,7 +434,7 @@ func state0(s *scanner, c int) int { // stateDot is the state after reading the integer and decimal point in a number, // such as after reading `1.`. -func stateDot(s *scanner, c int) int { +func stateDot(s *scanner, c byte) int { if '0' <= c && c <= '9' { s.step = stateDot0 return scanContinue @@ -445,9 +444,8 @@ func stateDot(s *scanner, c int) int { // stateDot0 is the state after reading the integer, decimal point, and subsequent // digits of a number, such as after reading `3.14`. -func stateDot0(s *scanner, c int) int { +func stateDot0(s *scanner, c byte) int { if '0' <= c && c <= '9' { - s.step = stateDot0 return scanContinue } if c == 'e' || c == 'E' { @@ -459,12 +457,8 @@ func stateDot0(s *scanner, c int) int { // stateE is the state after reading the mantissa and e in a number, // such as after reading `314e` or `0.314e`. -func stateE(s *scanner, c int) int { - if c == '+' { - s.step = stateESign - return scanContinue - } - if c == '-' { +func stateE(s *scanner, c byte) int { + if c == '+' || c == '-' { s.step = stateESign return scanContinue } @@ -473,7 +467,7 @@ func stateE(s *scanner, c int) int { // stateESign is the state after reading the mantissa, e, and sign in a number, // such as after reading `314e-` or `0.314e+`. -func stateESign(s *scanner, c int) int { +func stateESign(s *scanner, c byte) int { if '0' <= c && c <= '9' { s.step = stateE0 return scanContinue @@ -484,16 +478,15 @@ func stateESign(s *scanner, c int) int { // stateE0 is the state after reading the mantissa, e, optional sign, // and at least one digit of the exponent in a number, // such as after reading `314e-2` or `0.314e+1` or `3.14e0`. -func stateE0(s *scanner, c int) int { +func stateE0(s *scanner, c byte) int { if '0' <= c && c <= '9' { - s.step = stateE0 return scanContinue } return stateEndValue(s, c) } // stateT is the state after reading `t`. -func stateT(s *scanner, c int) int { +func stateT(s *scanner, c byte) int { if c == 'r' { s.step = stateTr return scanContinue @@ -502,7 +495,7 @@ func stateT(s *scanner, c int) int { } // stateTr is the state after reading `tr`. -func stateTr(s *scanner, c int) int { +func stateTr(s *scanner, c byte) int { if c == 'u' { s.step = stateTru return scanContinue @@ -511,7 +504,7 @@ func stateTr(s *scanner, c int) int { } // stateTru is the state after reading `tru`. -func stateTru(s *scanner, c int) int { +func stateTru(s *scanner, c byte) int { if c == 'e' { s.step = stateEndValue return scanContinue @@ -520,7 +513,7 @@ func stateTru(s *scanner, c int) int { } // stateF is the state after reading `f`. -func stateF(s *scanner, c int) int { +func stateF(s *scanner, c byte) int { if c == 'a' { s.step = stateFa return scanContinue @@ -529,7 +522,7 @@ func stateF(s *scanner, c int) int { } // stateFa is the state after reading `fa`. -func stateFa(s *scanner, c int) int { +func stateFa(s *scanner, c byte) int { if c == 'l' { s.step = stateFal return scanContinue @@ -538,7 +531,7 @@ func stateFa(s *scanner, c int) int { } // stateFal is the state after reading `fal`. -func stateFal(s *scanner, c int) int { +func stateFal(s *scanner, c byte) int { if c == 's' { s.step = stateFals return scanContinue @@ -547,7 +540,7 @@ func stateFal(s *scanner, c int) int { } // stateFals is the state after reading `fals`. -func stateFals(s *scanner, c int) int { +func stateFals(s *scanner, c byte) int { if c == 'e' { s.step = stateEndValue return scanContinue @@ -556,7 +549,7 @@ func stateFals(s *scanner, c int) int { } // stateN is the state after reading `n`. -func stateN(s *scanner, c int) int { +func stateN(s *scanner, c byte) int { if c == 'u' { s.step = stateNu return scanContinue @@ -565,7 +558,7 @@ func stateN(s *scanner, c int) int { } // stateNu is the state after reading `nu`. -func stateNu(s *scanner, c int) int { +func stateNu(s *scanner, c byte) int { if c == 'l' { s.step = stateNul return scanContinue @@ -574,7 +567,7 @@ func stateNu(s *scanner, c int) int { } // stateNul is the state after reading `nul`. -func stateNul(s *scanner, c int) int { +func stateNul(s *scanner, c byte) int { if c == 'l' { s.step = stateEndValue return scanContinue @@ -584,19 +577,19 @@ func stateNul(s *scanner, c int) int { // stateError is the state after reaching a syntax error, // such as after reading `[1}` or `5.1.2`. -func stateError(s *scanner, c int) int { +func stateError(s *scanner, c byte) int { return scanError } // error records an error and switches to the error state. -func (s *scanner) error(c int, context string) int { +func (s *scanner) error(c byte, context string) int { s.step = stateError s.err = &SyntaxError{"invalid character " + quoteChar(c) + " " + context, s.bytes} return scanError } // quoteChar formats c as a quoted character literal -func quoteChar(c int) string { +func quoteChar(c byte) string { // special cases - different from quoted strings if c == '\'' { return `'\''` @@ -623,7 +616,7 @@ func (s *scanner) undo(scanCode int) { } // stateRedo helps implement the scanner's 1-byte undo. -func stateRedo(s *scanner, c int) int { +func stateRedo(s *scanner, c byte) int { s.redo = false s.step = s.redoState return s.redoCode diff --git a/src/encoding/json/stream.go b/src/encoding/json/stream.go index dc53bceff8..8ddcf4d279 100644 --- a/src/encoding/json/stream.go +++ b/src/encoding/json/stream.go @@ -90,7 +90,7 @@ Input: // Look in the buffer for a new value. for i, c := range dec.buf[scanp:] { dec.scan.bytes++ - v := dec.scan.step(&dec.scan, int(c)) + v := dec.scan.step(&dec.scan, c) if v == scanEnd { scanp += i break Input @@ -157,7 +157,7 @@ func (dec *Decoder) refill() error { func nonSpace(b []byte) bool { for _, c := range b { - if !isSpace(rune(c)) { + if !isSpace(c) { return true } } @@ -433,7 +433,7 @@ func (dec *Decoder) tokenError(c byte) (Token, error) { case tokenObjectComma: context = " after object key:value pair" } - return nil, &SyntaxError{"invalid character " + quoteChar(int(c)) + " " + context, 0} + return nil, &SyntaxError{"invalid character " + quoteChar(c) + " " + context, 0} } // More reports whether there is another element in the @@ -448,7 +448,7 @@ func (dec *Decoder) peek() (byte, error) { for { for i := dec.scanp; i < len(dec.buf); i++ { c := dec.buf[i] - if isSpace(rune(c)) { + if isSpace(c) { continue } dec.scanp = i