2021-08-12 20:03:24 +01:00
|
|
|
// Copyright (c) 2012-2020 Ugorji Nwoke. All rights reserved.
|
|
|
|
// Use of this source code is governed by a MIT license found in the LICENSE file.
|
|
|
|
|
|
|
|
package codec
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
import (
|
|
|
|
"bufio"
|
|
|
|
"bytes"
|
|
|
|
"io"
|
|
|
|
"strings"
|
|
|
|
)
|
2021-08-12 20:03:24 +01:00
|
|
|
|
|
|
|
// decReader abstracts the reading source, allowing implementations that can
|
|
|
|
// read from an io.Reader or directly off a byte slice with zero-copying.
|
|
|
|
type decReader interface {
|
2022-05-02 14:05:18 +01:00
|
|
|
// readx will return a view of the []byte if decoding from a []byte, OR
|
|
|
|
// read into the implementation scratch buffer if possible i.e. n < len(scratchbuf), OR
|
|
|
|
// create a new []byte and read into that
|
2021-08-12 20:03:24 +01:00
|
|
|
readx(n uint) []byte
|
2022-05-02 14:05:18 +01:00
|
|
|
|
2021-08-12 20:03:24 +01:00
|
|
|
readb([]byte)
|
|
|
|
|
|
|
|
readn1() byte
|
|
|
|
readn2() [2]byte
|
2023-02-25 12:12:40 +00:00
|
|
|
readn3() [3]byte
|
2021-08-12 20:03:24 +01:00
|
|
|
readn4() [4]byte
|
|
|
|
readn8() [8]byte
|
|
|
|
// readn1eof() (v uint8, eof bool)
|
|
|
|
|
|
|
|
// // read up to 8 bytes at a time
|
|
|
|
// readn(num uint8) (v [8]byte)
|
|
|
|
|
|
|
|
numread() uint // number of bytes read
|
|
|
|
|
|
|
|
// skip any whitespace characters, and return the first non-matching byte
|
|
|
|
skipWhitespace() (token byte)
|
|
|
|
|
|
|
|
// jsonReadNum will include last read byte in first element of slice,
|
|
|
|
// and continue numeric characters until it sees a non-numeric char
|
|
|
|
// or EOF. If it sees a non-numeric character, it will unread that.
|
|
|
|
jsonReadNum() []byte
|
|
|
|
|
|
|
|
// jsonReadAsisChars will read json plain characters (anything but " or \)
|
|
|
|
// and return a slice terminated by a non-json asis character.
|
|
|
|
jsonReadAsisChars() []byte
|
|
|
|
|
|
|
|
// skip will skip any byte that matches, and return the first non-matching byte
|
|
|
|
// skip(accept *bitset256) (token byte)
|
|
|
|
|
|
|
|
// readTo will read any byte that matches, stopping once no-longer matching.
|
|
|
|
// readTo(accept *bitset256) (out []byte)
|
|
|
|
|
|
|
|
// readUntil will read, only stopping once it matches the 'stop' byte (which it excludes).
|
|
|
|
readUntil(stop byte) (out []byte)
|
|
|
|
}
|
|
|
|
|
|
|
|
// ------------------------------------------------
|
|
|
|
|
|
|
|
type unreadByteStatus uint8
|
|
|
|
|
|
|
|
// unreadByteStatus goes from
|
|
|
|
// undefined (when initialized) -- (read) --> canUnread -- (unread) --> canRead ...
|
|
|
|
const (
|
|
|
|
unreadByteUndefined unreadByteStatus = iota
|
|
|
|
unreadByteCanRead
|
|
|
|
unreadByteCanUnread
|
|
|
|
)
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
// const defBufReaderSize = 4096
|
|
|
|
|
2021-08-12 20:03:24 +01:00
|
|
|
// --------------------
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
// ioReaderByteScanner contains the io.Reader and io.ByteScanner interfaces
|
|
|
|
type ioReaderByteScanner interface {
|
|
|
|
io.Reader
|
|
|
|
io.ByteScanner
|
|
|
|
// ReadByte() (byte, error)
|
|
|
|
// UnreadByte() error
|
|
|
|
// Read(p []byte) (n int, err error)
|
|
|
|
}
|
2021-08-12 20:03:24 +01:00
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
// ioReaderByteScannerT does a simple wrapper of a io.ByteScanner
|
|
|
|
// over a io.Reader
|
|
|
|
type ioReaderByteScannerT struct {
|
|
|
|
r io.Reader
|
2021-08-12 20:03:24 +01:00
|
|
|
|
|
|
|
l byte // last byte
|
|
|
|
ls unreadByteStatus // last byte status
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
_ [2]byte // padding
|
|
|
|
b [4]byte // tiny buffer for reading single bytes
|
2021-08-12 20:03:24 +01:00
|
|
|
}
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
func (z *ioReaderByteScannerT) ReadByte() (c byte, err error) {
|
|
|
|
if z.ls == unreadByteCanRead {
|
|
|
|
z.ls = unreadByteCanUnread
|
|
|
|
c = z.l
|
|
|
|
} else {
|
|
|
|
_, err = z.Read(z.b[:1])
|
|
|
|
c = z.b[0]
|
|
|
|
}
|
|
|
|
return
|
2021-08-12 20:03:24 +01:00
|
|
|
}
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
func (z *ioReaderByteScannerT) UnreadByte() (err error) {
|
|
|
|
switch z.ls {
|
|
|
|
case unreadByteCanUnread:
|
|
|
|
z.ls = unreadByteCanRead
|
|
|
|
case unreadByteCanRead:
|
|
|
|
err = errDecUnreadByteLastByteNotRead
|
|
|
|
case unreadByteUndefined:
|
|
|
|
err = errDecUnreadByteNothingToRead
|
|
|
|
default:
|
|
|
|
err = errDecUnreadByteUnknown
|
|
|
|
}
|
|
|
|
return
|
2021-08-12 20:03:24 +01:00
|
|
|
}
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
func (z *ioReaderByteScannerT) Read(p []byte) (n int, err error) {
|
2021-08-12 20:03:24 +01:00
|
|
|
if len(p) == 0 {
|
|
|
|
return
|
|
|
|
}
|
|
|
|
var firstByte bool
|
|
|
|
if z.ls == unreadByteCanRead {
|
|
|
|
z.ls = unreadByteCanUnread
|
|
|
|
p[0] = z.l
|
|
|
|
if len(p) == 1 {
|
|
|
|
n = 1
|
|
|
|
return
|
|
|
|
}
|
|
|
|
firstByte = true
|
|
|
|
p = p[1:]
|
|
|
|
}
|
|
|
|
n, err = z.r.Read(p)
|
|
|
|
if n > 0 {
|
|
|
|
if err == io.EOF && n == len(p) {
|
|
|
|
err = nil // read was successful, so postpone EOF (till next time)
|
|
|
|
}
|
|
|
|
z.l = p[n-1]
|
|
|
|
z.ls = unreadByteCanUnread
|
|
|
|
}
|
|
|
|
if firstByte {
|
|
|
|
n++
|
|
|
|
}
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
func (z *ioReaderByteScannerT) reset(r io.Reader) {
|
|
|
|
z.r = r
|
|
|
|
z.ls = unreadByteUndefined
|
|
|
|
z.l = 0
|
|
|
|
}
|
|
|
|
|
|
|
|
// ioDecReader is a decReader that reads off an io.Reader.
|
|
|
|
type ioDecReader struct {
|
|
|
|
rr ioReaderByteScannerT // the reader passed in, wrapped into a reader+bytescanner
|
|
|
|
|
|
|
|
n uint // num read
|
|
|
|
|
|
|
|
blist *bytesFreelist
|
|
|
|
|
|
|
|
bufr []byte // buffer for readTo/readUntil
|
|
|
|
br ioReaderByteScanner // main reader used for Read|ReadByte|UnreadByte
|
|
|
|
bb *bufio.Reader // created internally, and reused on reset if needed
|
|
|
|
|
|
|
|
x [64 + 40]byte // for: get struct field name, swallow valueTypeBytes, etc
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *ioDecReader) reset(r io.Reader, bufsize int, blist *bytesFreelist) {
|
|
|
|
z.blist = blist
|
|
|
|
z.n = 0
|
|
|
|
z.bufr = z.blist.check(z.bufr, 256)
|
|
|
|
z.br = nil
|
|
|
|
|
|
|
|
var ok bool
|
|
|
|
|
|
|
|
if bufsize <= 0 {
|
|
|
|
z.br, ok = r.(ioReaderByteScanner)
|
|
|
|
if !ok {
|
|
|
|
z.rr.reset(r)
|
|
|
|
z.br = &z.rr
|
2021-08-12 20:03:24 +01:00
|
|
|
}
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
// bufsize > 0 ...
|
|
|
|
|
|
|
|
// if bytes.[Buffer|Reader], no value in adding extra buffer
|
|
|
|
// if bufio.Reader, no value in extra buffer unless size changes
|
|
|
|
switch bb := r.(type) {
|
|
|
|
case *strings.Reader:
|
|
|
|
z.br = bb
|
|
|
|
case *bytes.Buffer:
|
|
|
|
z.br = bb
|
|
|
|
case *bytes.Reader:
|
|
|
|
z.br = bb
|
|
|
|
case *bufio.Reader:
|
|
|
|
if bb.Size() == bufsize {
|
|
|
|
z.br = bb
|
2021-08-12 20:03:24 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
if z.br == nil {
|
|
|
|
if z.bb != nil && z.bb.Size() == bufsize {
|
|
|
|
z.bb.Reset(r)
|
|
|
|
} else {
|
|
|
|
z.bb = bufio.NewReaderSize(r, bufsize)
|
2021-08-12 20:03:24 +01:00
|
|
|
}
|
2023-02-25 12:12:40 +00:00
|
|
|
z.br = z.bb
|
2021-08-12 20:03:24 +01:00
|
|
|
}
|
2023-02-25 12:12:40 +00:00
|
|
|
}
|
2021-08-12 20:03:24 +01:00
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
func (z *ioDecReader) numread() uint {
|
|
|
|
return z.n
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *ioDecReader) readn1() (b uint8) {
|
|
|
|
b, err := z.br.ReadByte()
|
|
|
|
halt.onerror(err)
|
|
|
|
z.n++
|
2021-08-12 20:03:24 +01:00
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *ioDecReader) readn2() (bs [2]byte) {
|
|
|
|
z.readb(bs[:])
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
func (z *ioDecReader) readn3() (bs [3]byte) {
|
|
|
|
z.readb(bs[:])
|
2021-08-12 20:03:24 +01:00
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *ioDecReader) readn4() (bs [4]byte) {
|
|
|
|
z.readb(bs[:])
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *ioDecReader) readn8() (bs [8]byte) {
|
|
|
|
z.readb(bs[:])
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *ioDecReader) readx(n uint) (bs []byte) {
|
|
|
|
if n == 0 {
|
2023-02-25 12:12:40 +00:00
|
|
|
return zeroByteSlice
|
2021-08-12 20:03:24 +01:00
|
|
|
}
|
|
|
|
if n < uint(len(z.x)) {
|
|
|
|
bs = z.x[:n]
|
|
|
|
} else {
|
|
|
|
bs = make([]byte, n)
|
|
|
|
}
|
2023-02-25 12:12:40 +00:00
|
|
|
nn, err := readFull(z.br, bs)
|
|
|
|
z.n += nn
|
2021-08-12 20:03:24 +01:00
|
|
|
halt.onerror(err)
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *ioDecReader) readb(bs []byte) {
|
|
|
|
if len(bs) == 0 {
|
|
|
|
return
|
|
|
|
}
|
2023-02-25 12:12:40 +00:00
|
|
|
nn, err := readFull(z.br, bs)
|
|
|
|
z.n += nn
|
2021-08-12 20:03:24 +01:00
|
|
|
halt.onerror(err)
|
|
|
|
}
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
// func (z *ioDecReader) readn1eof() (b uint8, eof bool) {
|
|
|
|
// b, err := z.br.ReadByte()
|
|
|
|
// if err == nil {
|
|
|
|
// z.n++
|
|
|
|
// } else if err == io.EOF {
|
|
|
|
// eof = true
|
|
|
|
// } else {
|
|
|
|
// halt.onerror(err)
|
|
|
|
// }
|
|
|
|
// return
|
|
|
|
// }
|
2021-08-12 20:03:24 +01:00
|
|
|
|
|
|
|
func (z *ioDecReader) jsonReadNum() (bs []byte) {
|
|
|
|
z.unreadn1()
|
|
|
|
z.bufr = z.bufr[:0]
|
|
|
|
LOOP:
|
2023-02-25 12:12:40 +00:00
|
|
|
// i, eof := z.readn1eof()
|
|
|
|
i, err := z.br.ReadByte()
|
|
|
|
if err == io.EOF {
|
2021-08-12 20:03:24 +01:00
|
|
|
return z.bufr
|
|
|
|
}
|
2023-02-25 12:12:40 +00:00
|
|
|
if err != nil {
|
|
|
|
halt.onerror(err)
|
|
|
|
}
|
|
|
|
z.n++
|
2021-08-12 20:03:24 +01:00
|
|
|
if isNumberChar(i) {
|
|
|
|
z.bufr = append(z.bufr, i)
|
|
|
|
goto LOOP
|
|
|
|
}
|
|
|
|
z.unreadn1()
|
|
|
|
return z.bufr
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *ioDecReader) jsonReadAsisChars() (bs []byte) {
|
|
|
|
z.bufr = z.bufr[:0]
|
|
|
|
LOOP:
|
|
|
|
i := z.readn1()
|
|
|
|
z.bufr = append(z.bufr, i)
|
|
|
|
if i == '"' || i == '\\' {
|
|
|
|
return z.bufr
|
|
|
|
}
|
|
|
|
goto LOOP
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *ioDecReader) skipWhitespace() (token byte) {
|
|
|
|
LOOP:
|
|
|
|
token = z.readn1()
|
|
|
|
if isWhitespaceChar(token) {
|
|
|
|
goto LOOP
|
|
|
|
}
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
// func (z *ioDecReader) readUntil(stop byte) []byte {
|
|
|
|
// z.bufr = z.bufr[:0]
|
|
|
|
// LOOP:
|
|
|
|
// token := z.readn1()
|
|
|
|
// z.bufr = append(z.bufr, token)
|
|
|
|
// if token == stop {
|
|
|
|
// return z.bufr[:len(z.bufr)-1]
|
|
|
|
// }
|
|
|
|
// goto LOOP
|
|
|
|
// }
|
|
|
|
|
2021-08-12 20:03:24 +01:00
|
|
|
func (z *ioDecReader) readUntil(stop byte) []byte {
|
|
|
|
z.bufr = z.bufr[:0]
|
|
|
|
LOOP:
|
|
|
|
token := z.readn1()
|
|
|
|
if token == stop {
|
2023-02-25 12:12:40 +00:00
|
|
|
return z.bufr
|
2021-08-12 20:03:24 +01:00
|
|
|
}
|
2023-02-25 12:12:40 +00:00
|
|
|
z.bufr = append(z.bufr, token)
|
2021-08-12 20:03:24 +01:00
|
|
|
goto LOOP
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *ioDecReader) unreadn1() {
|
2023-02-25 12:12:40 +00:00
|
|
|
err := z.br.UnreadByte()
|
2021-08-12 20:03:24 +01:00
|
|
|
halt.onerror(err)
|
|
|
|
z.n--
|
|
|
|
}
|
|
|
|
|
|
|
|
// ------------------------------------
|
|
|
|
|
|
|
|
// bytesDecReader is a decReader that reads off a byte slice with zero copying
|
|
|
|
//
|
2024-03-11 10:12:06 +00:00
|
|
|
// Note: we do not try to convert index'ing out of bounds to an io error.
|
2021-08-12 20:03:24 +01:00
|
|
|
// instead, we let it bubble up to the exported Encode/Decode method
|
2024-03-11 10:12:06 +00:00
|
|
|
// and recover it as an io error.
|
2021-08-12 20:03:24 +01:00
|
|
|
//
|
2023-02-25 12:12:40 +00:00
|
|
|
// Every function here MUST defensively check bounds either explicitly
|
|
|
|
// or via a bounds check.
|
|
|
|
//
|
2021-08-12 20:03:24 +01:00
|
|
|
// see panicValToErr(...) function in helper.go.
|
|
|
|
type bytesDecReader struct {
|
|
|
|
b []byte // data
|
|
|
|
c uint // cursor
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *bytesDecReader) reset(in []byte) {
|
|
|
|
z.b = in[:len(in):len(in)] // reslicing must not go past capacity
|
|
|
|
z.c = 0
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *bytesDecReader) numread() uint {
|
|
|
|
return z.c
|
|
|
|
}
|
|
|
|
|
|
|
|
// Note: slicing from a non-constant start position is more expensive,
|
|
|
|
// as more computation is required to decipher the pointer start position.
|
|
|
|
// However, we do it only once, and it's better than reslicing both z.b and return value.
|
|
|
|
|
|
|
|
func (z *bytesDecReader) readx(n uint) (bs []byte) {
|
2023-02-25 12:12:40 +00:00
|
|
|
// x := z.c + n
|
|
|
|
// bs = z.b[z.c:x]
|
|
|
|
// z.c = x
|
|
|
|
bs = z.b[z.c : z.c+n]
|
|
|
|
z.c += n
|
2021-08-12 20:03:24 +01:00
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *bytesDecReader) readb(bs []byte) {
|
|
|
|
copy(bs, z.readx(uint(len(bs))))
|
|
|
|
}
|
|
|
|
|
|
|
|
// MARKER: do not use this - as it calls into memmove (as the size of data to move is unknown)
|
|
|
|
// func (z *bytesDecReader) readnn(bs []byte, n uint) {
|
|
|
|
// x := z.c
|
|
|
|
// copy(bs, z.b[x:x+n])
|
|
|
|
// z.c += n
|
|
|
|
// }
|
|
|
|
|
|
|
|
// func (z *bytesDecReader) readn(num uint8) (bs [8]byte) {
|
|
|
|
// x := z.c + uint(num)
|
|
|
|
// copy(bs[:], z.b[z.c:x]) // slice z.b completely, so we get bounds error if past
|
|
|
|
// z.c = x
|
|
|
|
// return
|
|
|
|
// }
|
|
|
|
|
|
|
|
// func (z *bytesDecReader) readn1() uint8 {
|
|
|
|
// z.c++
|
|
|
|
// return z.b[z.c-1]
|
|
|
|
// }
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
// MARKER: readn{1,2,3,4,8} should throw an out of bounds error if past length.
|
|
|
|
// MARKER: readn1: explicitly ensure bounds check is done
|
|
|
|
// MARKER: readn{2,3,4,8}: ensure you slice z.b completely so we get bounds error if past end.
|
|
|
|
|
2021-08-12 20:03:24 +01:00
|
|
|
func (z *bytesDecReader) readn1() (v uint8) {
|
|
|
|
v = z.b[z.c]
|
|
|
|
z.c++
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *bytesDecReader) readn2() (bs [2]byte) {
|
|
|
|
// copy(bs[:], z.b[z.c:z.c+2])
|
2023-02-25 12:12:40 +00:00
|
|
|
// bs[1] = z.b[z.c+1]
|
|
|
|
// bs[0] = z.b[z.c]
|
|
|
|
bs = okBytes2(z.b[z.c : z.c+2])
|
2021-08-12 20:03:24 +01:00
|
|
|
z.c += 2
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
func (z *bytesDecReader) readn3() (bs [3]byte) {
|
2021-08-12 20:03:24 +01:00
|
|
|
// copy(bs[1:], z.b[z.c:z.c+3])
|
|
|
|
bs = okBytes3(z.b[z.c : z.c+3])
|
|
|
|
z.c += 3
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *bytesDecReader) readn4() (bs [4]byte) {
|
|
|
|
// copy(bs[:], z.b[z.c:z.c+4])
|
|
|
|
bs = okBytes4(z.b[z.c : z.c+4])
|
|
|
|
z.c += 4
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *bytesDecReader) readn8() (bs [8]byte) {
|
|
|
|
// copy(bs[:], z.b[z.c:z.c+8])
|
|
|
|
bs = okBytes8(z.b[z.c : z.c+8])
|
|
|
|
z.c += 8
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *bytesDecReader) jsonReadNum() []byte {
|
2023-02-25 12:12:40 +00:00
|
|
|
z.c-- // unread
|
2021-08-12 20:03:24 +01:00
|
|
|
i := z.c
|
|
|
|
LOOP:
|
2023-02-25 12:12:40 +00:00
|
|
|
// gracefully handle end of slice, as end of stream is meaningful here
|
2021-08-12 20:03:24 +01:00
|
|
|
if i < uint(len(z.b)) && isNumberChar(z.b[i]) {
|
|
|
|
i++
|
|
|
|
goto LOOP
|
|
|
|
}
|
|
|
|
z.c, i = i, z.c
|
2023-02-25 12:12:40 +00:00
|
|
|
// MARKER: 20230103: byteSliceOf here prevents inlining of jsonReadNum
|
|
|
|
// return byteSliceOf(z.b, i, z.c)
|
2021-08-12 20:03:24 +01:00
|
|
|
return z.b[i:z.c]
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *bytesDecReader) jsonReadAsisChars() []byte {
|
|
|
|
i := z.c
|
|
|
|
LOOP:
|
|
|
|
token := z.b[i]
|
|
|
|
i++
|
|
|
|
if token == '"' || token == '\\' {
|
|
|
|
z.c, i = i, z.c
|
2023-02-25 12:12:40 +00:00
|
|
|
return byteSliceOf(z.b, i, z.c)
|
|
|
|
// return z.b[i:z.c]
|
2021-08-12 20:03:24 +01:00
|
|
|
}
|
|
|
|
goto LOOP
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *bytesDecReader) skipWhitespace() (token byte) {
|
|
|
|
i := z.c
|
|
|
|
LOOP:
|
|
|
|
token = z.b[i]
|
|
|
|
if isWhitespaceChar(token) {
|
|
|
|
i++
|
|
|
|
goto LOOP
|
|
|
|
}
|
|
|
|
z.c = i + 1
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
func (z *bytesDecReader) readUntil(stop byte) (out []byte) {
|
|
|
|
i := z.c
|
|
|
|
LOOP:
|
|
|
|
if z.b[i] == stop {
|
2023-02-25 12:12:40 +00:00
|
|
|
out = byteSliceOf(z.b, z.c, i)
|
|
|
|
// out = z.b[z.c:i]
|
2021-08-12 20:03:24 +01:00
|
|
|
z.c = i + 1
|
|
|
|
return
|
|
|
|
}
|
|
|
|
i++
|
|
|
|
goto LOOP
|
|
|
|
}
|
|
|
|
|
|
|
|
// --------------
|
|
|
|
|
|
|
|
type decRd struct {
|
2023-02-25 12:12:40 +00:00
|
|
|
rb bytesDecReader
|
|
|
|
ri *ioDecReader
|
|
|
|
|
|
|
|
decReader
|
|
|
|
|
|
|
|
bytes bool // is bytes reader
|
|
|
|
|
|
|
|
// MARKER: these fields below should belong directly in Encoder.
|
|
|
|
// we pack them here for space efficiency and cache-line optimization.
|
|
|
|
|
2021-08-12 20:03:24 +01:00
|
|
|
mtr bool // is maptype a known type?
|
|
|
|
str bool // is slicetype a known type?
|
|
|
|
|
|
|
|
be bool // is binary encoding
|
|
|
|
js bool // is json handle
|
|
|
|
jsms bool // is json handle, and MapKeyAsString
|
|
|
|
cbor bool // is cbor handle
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
cbreak bool // is a check breaker
|
2021-08-12 20:03:24 +01:00
|
|
|
|
|
|
|
}
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
// From out benchmarking, we see the following impact performance:
|
2021-08-12 20:03:24 +01:00
|
|
|
//
|
2023-02-25 12:12:40 +00:00
|
|
|
// - functions that are too big to inline
|
|
|
|
// - interface calls (as no inlining can occur)
|
2021-08-12 20:03:24 +01:00
|
|
|
//
|
2023-02-25 12:12:40 +00:00
|
|
|
// decRd is designed to embed a decReader, and then re-implement some of the decReader
|
|
|
|
// methods using a conditional branch.
|
2021-08-12 20:03:24 +01:00
|
|
|
//
|
2023-02-25 12:12:40 +00:00
|
|
|
// We only override the ones where the bytes version is inlined AND the wrapper method
|
|
|
|
// (containing the bytes version alongside a conditional branch) is also inlined.
|
2021-08-12 20:03:24 +01:00
|
|
|
//
|
2023-02-25 12:12:40 +00:00
|
|
|
// We use ./run.sh -z to check.
|
2021-08-12 20:03:24 +01:00
|
|
|
//
|
2023-02-25 12:12:40 +00:00
|
|
|
// Right now, only numread and "carefully crafted" readn1 can be inlined.
|
2021-08-12 20:03:24 +01:00
|
|
|
|
|
|
|
func (z *decRd) numread() uint {
|
|
|
|
if z.bytes {
|
|
|
|
return z.rb.numread()
|
|
|
|
}
|
2023-02-25 12:12:40 +00:00
|
|
|
return z.ri.numread()
|
2021-08-12 20:03:24 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
func (z *decRd) readn1() (v uint8) {
|
|
|
|
if z.bytes {
|
|
|
|
// return z.rb.readn1()
|
2023-02-25 12:12:40 +00:00
|
|
|
// MARKER: calling z.rb.readn1() prevents decRd.readn1 from being inlined.
|
|
|
|
// copy code, to manually inline and explicitly return here.
|
|
|
|
// Keep in sync with bytesDecReader.readn1
|
2021-08-12 20:03:24 +01:00
|
|
|
v = z.rb.b[z.rb.c]
|
|
|
|
z.rb.c++
|
2023-02-25 12:12:40 +00:00
|
|
|
return
|
2021-08-12 20:03:24 +01:00
|
|
|
}
|
|
|
|
return z.ri.readn1()
|
|
|
|
}
|
|
|
|
|
2023-02-25 12:12:40 +00:00
|
|
|
// func (z *decRd) readn4() [4]byte {
|
|
|
|
// if z.bytes {
|
|
|
|
// return z.rb.readn4()
|
|
|
|
// }
|
|
|
|
// return z.ri.readn4()
|
|
|
|
// }
|
|
|
|
|
|
|
|
// func (z *decRd) readn3() [3]byte {
|
|
|
|
// if z.bytes {
|
|
|
|
// return z.rb.readn3()
|
|
|
|
// }
|
|
|
|
// return z.ri.readn3()
|
|
|
|
// }
|
|
|
|
|
|
|
|
// func (z *decRd) skipWhitespace() byte {
|
|
|
|
// if z.bytes {
|
|
|
|
// return z.rb.skipWhitespace()
|
|
|
|
// }
|
|
|
|
// return z.ri.skipWhitespace()
|
|
|
|
// }
|
|
|
|
|
2021-08-12 20:03:24 +01:00
|
|
|
type devNullReader struct{}
|
|
|
|
|
|
|
|
func (devNullReader) Read(p []byte) (int, error) { return 0, io.EOF }
|
|
|
|
func (devNullReader) Close() error { return nil }
|
|
|
|
|
|
|
|
func readFull(r io.Reader, bs []byte) (n uint, err error) {
|
|
|
|
var nn int
|
|
|
|
for n < uint(len(bs)) && err == nil {
|
|
|
|
nn, err = r.Read(bs[n:])
|
|
|
|
if nn > 0 {
|
|
|
|
if err == io.EOF {
|
|
|
|
// leave EOF for next time
|
|
|
|
err = nil
|
|
|
|
}
|
|
|
|
n += uint(nn)
|
|
|
|
}
|
|
|
|
}
|
|
|
|
// do not do this below - it serves no purpose
|
|
|
|
// if n != len(bs) && err == io.EOF { err = io.ErrUnexpectedEOF }
|
|
|
|
return
|
|
|
|
}
|
|
|
|
|
|
|
|
var _ decReader = (*decRd)(nil)
|