func parseCombinator(rdr io.ByteScanner, p *Link) error {
rdr.UnreadByte()
for c, err := rdr.ReadByte(); err != io.EOF; c, err = rdr.ReadByte() {
if err != nil {
return err
}
switch c {
case '{':
rdr.UnreadByte()
return EOS
case ',':
return fmt.Errorf("Encountered ',' after combinator")
case ' ', '\t', '\n', '\r', '\f':
case '>', '+', '~':
if p.Combinator == Descendant {
p.Combinator = combinatorMap[c]
} else {
return fmt.Errorf("Can't combine multiple combinators")
}
default:
rdr.UnreadByte()
return nil
}
}
return nil
}
func (t *Terminal) Read(r io.ByteScanner) error {
c, err := r.ReadByte()
if err != nil {
return err
}
switch {
case c == 0x7: // bell
// ignore
case c == 0x8: // backspace
if t.Col > 0 {
t.Col--
}
case c == 0x1b:
return t.readEscape(r)
case c == '\r':
t.Col = 0
case c == '\n':
t.Col = 0
t.Row++
t.fixPosition()
case c == '\t':
t.Col += 8 - (t.Col % 8)
t.fixPosition()
case c >= ' ' && c <= '~':
t.writeRune(rune(c), t.Attr)
default:
r.UnreadByte()
return t.readUTF8(r)
}
return nil
}
func readString(buf io.ByteScanner) Literal {
strbuf := []byte{}
loop:
for {
b, err := buf.ReadByte()
if err != nil {
break
}
switch b {
case '"':
return String(string(strbuf))
case '\\':
b, err := buf.ReadByte()
if err != nil {
break loop
}
switch b {
case 't':
strbuf = append(strbuf, '\t')
case 'n':
strbuf = append(strbuf, '\n')
case '\\':
strbuf = append(strbuf, '\\')
case '"':
strbuf = append(strbuf, '"')
default:
panic("read: unknown escape")
}
default:
strbuf = append(strbuf, b)
}
}
panic("read: unterminated string")
return String("")
}
func read(buf io.ByteScanner) Literal {
skipws(buf)
b, err := buf.ReadByte()
if err == io.EOF {
panic("read: premature end of file")
}
if err != nil {
panic("read: input error: " + err.Error())
}
switch b {
case ')':
panic("read: unmatched close-parenthesis")
case '"':
return readString(buf)
case '\'':
return readQuote(buf)
case '`':
return readQuasi(buf)
case ',':
return readComma(buf)
case '&':
return readAmpersand(buf)
case '(':
return readList(buf)
}
buf.UnreadByte()
return readAtom(buf)
}
func readAtom(buf io.ByteScanner) Literal {
atomBuf := []byte{}
loop:
for {
b, err := buf.ReadByte()
if err != nil {
break
}
switch b {
case '(', ')', '\'', '"', ' ', '\t', '\n':
buf.UnreadByte()
break loop
}
atomBuf = append(atomBuf, b)
}
if len(atomBuf) == 0 {
panic("read: empty atom")
}
atom := string(atomBuf)
n, err := strconv.ParseFloat(atom, 64)
if err == nil {
return Number(n)
}
return Intern(atom)
}
func nextLine(buf io.ByteScanner) {
for {
b, err := buf.ReadByte()
if err != nil || b == '\n' {
break
}
}
}
func TestMockByteScanner(t *testing.T) {
var s io.ByteScanner
m := &ByteScannerMock{}
s = m
m.On("ReadByte").Return(byte('_'), nil)
b, err := s.ReadByte()
assert.Equal(t, byte('_'), b)
assert.Equal(t, nil, err)
}
func readTerminator(r io.ByteScanner, term byte) (bool, error) {
tok, err := r.ReadByte()
if err != nil {
return false, err
} else if tok == term {
return true, nil
}
return false, r.UnreadByte()
}
func (t *Terminal) expect(r io.ByteScanner, exp byte) (bool, error) {
c, err := r.ReadByte()
if err != nil {
return false, err
}
ok := c == exp
if !ok {
log.Printf("expect %s failed, got %s", showChar(exp), showChar(c))
}
return ok, nil
}
func readAmpersand(buf io.ByteScanner) Literal {
skipws(buf)
b, err := buf.ReadByte()
if err != nil {
panic("read: incomplete input")
}
if b != '(' {
panic("read: ill-formed ampersand")
}
x := readList(buf)
return newListLiteral(Intern("ampersand"), x)
}
func parseSequence(rdr io.ByteScanner) (Sequence, error) {
seq := []SimpleSelector{}
rdr.UnreadByte()
for c, err := rdr.ReadByte(); err != io.EOF; c, err = rdr.ReadByte() {
if err != nil {
return nil, err
}
switch c {
case '*':
seq = append(seq, SimpleSelector{Type: Universal})
case '#':
sel := SimpleSelector{Type: Id, AttrName: "id"}
if err := parseSimpleSelector(rdr, &sel); err != nil {
return nil, err
}
seq = append(seq, sel)
case '.':
sel := SimpleSelector{Type: Class, AttrName: "class"}
if err := parseSimpleSelector(rdr, &sel); err != nil {
return nil, err
}
seq = append(seq, sel)
case ':':
sel := SimpleSelector{Type: PseudoClass}
if err := parseSimpleSelector(rdr, &sel); err != nil {
return nil, err
}
seq = append(seq, sel)
case '[':
sel := SimpleSelector{Type: Attr}
if err := parseSimpleAttr(rdr, &sel); err != nil {
return nil, err
}
seq = append(seq, sel)
case '{':
rdr.UnreadByte()
return seq, EOS
case ' ', '\t', '\n', '\r', '\f', '>', '+', '~':
rdr.UnreadByte()
return seq, nil
default:
sel := SimpleSelector{Type: Tag, Tag: string(c)}
if err := parseSimpleTag(rdr, &sel); err != nil {
return nil, err
}
seq = append(seq, sel)
}
}
return seq, nil
}
func (t *Terminal) readTo(r io.ByteScanner, end byte) ([]byte, error) {
var buf []byte
for i := 0; i < 1000; i++ {
c, err := r.ReadByte()
if err != nil {
return nil, err
}
if c == end {
return buf, nil
}
buf = append(buf, c)
}
return nil, fmt.Errorf("term: readTo(%s) overlong", showChar(end))
}
func readComma(buf io.ByteScanner) Literal {
b, err := buf.ReadByte()
if err != nil {
panic("read: incomplete comma")
}
tag := Intern("unquote")
if b == '@' {
tag = Intern("unquotesplicing")
} else {
buf.UnreadByte()
}
x := read(buf)
return newListLiteral(tag, x)
}
func scanSign(r io.ByteScanner) (neg bool, err error) {
var ch byte
if ch, err = r.ReadByte(); err != nil {
return false, err
}
switch ch {
case '-':
neg = true
case '+':
// nothing to do
default:
r.UnreadByte()
}
return
}
func (t *Terminal) readInt(r io.ByteScanner) (int, error) {
n := 0
for i := 0; i < 20; i++ {
c, err := r.ReadByte()
if err != nil {
return -1, err
}
if c >= '0' && c <= '9' {
n = n*10 + int(c) - '0'
} else {
r.UnreadByte()
return n, err
}
}
return -1, fmt.Errorf("term: readInt overlong")
}
func skipws(buf io.ByteScanner) {
for {
b, err := buf.ReadByte()
if err != nil {
break
}
switch b {
case ' ', '\t', '\n':
continue
case ';':
nextLine(buf)
continue
}
buf.UnreadByte()
break
}
}
func readList(buf io.ByteScanner) Literal {
skipws(buf)
items := []Literal{}
for {
b, err := buf.ReadByte()
if err != nil {
panic("read: premature end of file")
}
if b == ')' {
break
}
buf.UnreadByte()
items = append(items, read(buf))
skipws(buf)
}
return newListLiteral(items...)
}
// parseExampleLine parses an example line. It does not parse any wiki syntax.
func (p *Parser) parseExampleLine(r io.ByteScanner) {
p.startElement("Example")
defer p.endElement()
p.startElement("Text")
defer p.endElement()
for {
c, err := r.ReadByte()
if err != nil {
return
}
if c == '\n' {
return
}
p.current.Text += string(c)
p.nextColumn()
}
}
func parseSimpleSelector(rdr io.ByteScanner, sel *SimpleSelector) error {
b, err := rdr.ReadByte()
if err != nil && err != EOS {
return err
}
bs, err := consumeValue(rdr)
if err != nil && err != EOS {
return err
}
bs = append([]byte{b}, bs...)
if sel.Type == PseudoClass && bs[0] == ':' {
sel.Type = PseudoElement
bs = bs[1:]
}
sel.Value = string(bs)
return err
}
func (t *Terminal) readUTF8(r io.ByteScanner) error {
c, err := r.ReadByte()
if err != nil {
return err
}
attr := t.Attr
var uc rune
n := 0
switch {
case c&0xE0 == 0xB0:
uc = rune(c & 0x1F)
n = 2
case c&0xF0 == 0xE0:
uc = rune(c & 0x0F)
n = 3
default:
if c&0xF0 == 0xF0 {
log.Printf("term: not yet implemented: utf8 start %#v", c)
}
attr.SetInverse(true)
t.writeRune('@', attr)
return nil
}
for i := 1; i < n; i++ {
c, err := r.ReadByte()
if err != nil {
return err
}
if c&0xC0 != 0x80 {
log.Printf("term: not yet implemented: utf8 continuation %#v", c)
attr.SetInverse(true)
uc = '@'
break
}
uc = uc<<6 | rune(c&0x3F)
}
t.writeRune(uc, attr)
return nil
}
func parseChain(rdr io.ByteScanner, chn *Chain) error {
for c, err := rdr.ReadByte(); err != io.EOF; c, err = rdr.ReadByte() {
if err != nil {
return err
}
switch c {
case ',':
return fmt.Errorf("Parser does not handle groups")
case ' ', '\t', '\n', '\r', '\f', '>', '+', '~':
if chn.Head == nil {
return fmt.Errorf("Starting selector chain with combinator %c", c)
}
part := Link{}
if err := parseCombinator(rdr, &part); err != nil {
return err
}
chn.Tail = append(chn.Tail, part)
default:
if chn.Head == nil {
chn.Head, err = parseSequence(rdr)
if err != nil && err != io.EOF {
return err
}
} else {
last := last(chn.Tail)
if last != nil {
last.Sequence, err = parseSequence(rdr)
if err != nil && err != io.EOF {
return err
}
} else {
return fmt.Errorf(
"Attempt to add tail seqence without combinator char: %c", c)
}
}
}
}
return nil
}
// parseHeaderBullet parses a bullet of the header.
func (p *Parser) parseHeaderBullet(r io.ByteScanner) {
level := 0
for {
c, err := r.ReadByte()
if err != nil {
p.addError("ParseHeaderBullet", "unexpected EOF")
return
}
if c == '*' {
level += 1
p.nextColumn()
} else if c == ' ' {
p.nextColumn()
break
} else {
p.nextColumn()
p.addError("ParseHeaderBullet", "* or space expected")
return
}
}
p.current.Attr["level"] = fmt.Sprint(level)
}
// parseDocument parses a document.
func (p *Parser) parseDocument(r io.ByteScanner) {
_, err := r.ReadByte()
if err != nil {
return
}
r.UnreadByte()
p.root = addElement(nil, "Document")
p.current = p.root
for {
_, err := r.ReadByte()
if err != nil {
break
}
r.UnreadByte()
p.parseLine(r)
}
}
func consumeValue(rdr io.ByteScanner) ([]byte, error) {
bs := []byte{}
for c, err := rdr.ReadByte(); err != io.EOF; c, err = rdr.ReadByte() {
if err != nil {
return nil, err
}
switch c {
case '{':
rdr.UnreadByte()
return bs, EOS
case '>', '+', '~', ' ', '\t', '\n', '\f', ',', '.', '#', '[', ':':
rdr.UnreadByte()
return bs, nil
default:
bs = append(bs, c)
}
}
return bs, nil
}
// ReadGrid reads a grid from s using the same format as WriteTo.
func ReadGrid(s io.ByteScanner) (Grid, error) {
g := NewGrid()
for r := 0; r < GridSize; r++ {
for c := 0; c < GridSize; c++ {
// read number
b, err := s.ReadByte()
if err != nil {
return g, fmt.Errorf("failed to read sudoku grid row %d: %s", r+1, err)
}
if b == '_' {
g[r][c] = Cell{}
} else if b >= '1' && b <= '9' {
g[r][c].resolve(b - '0')
} else {
return g, fmt.Errorf("fot a number %c at row %d", b, r+1)
}
if c != GridSize-1 {
// read space
b, err = s.ReadByte()
if err != nil {
return g, fmt.Errorf("failed to read sudoku grid row %d: %s", r+1, err)
}
if b != ' ' {
return g, fmt.Errorf("unexpected character '%c' at row %d", b, r+1)
}
} else {
// read newline
b, err = s.ReadByte()
if r == GridSize-1 && err == io.EOF {
break // TODO: return EOF here?
}
if err != nil {
return g, fmt.Errorf("failed to read sudoku grid row %d: %s", r+1, err)
}
if b != '\n' {
// TODO: support Windows and MAC new lines
return g, fmt.Errorf("unexpected character '%c' at row %d", b, r+1)
}
}
}
}
return g, nil
}
// Read from reader while f holds true, unread bytes if reset == true
func readWhile(reader io.ByteScanner, f func(b byte) bool, reset bool) string {
s := ""
for {
b, err := reader.ReadByte()
if err != nil {
return s
}
if !f(b) {
reader.UnreadByte()
return s
}
if reset {
defer reader.UnreadByte()
}
s += string(b)
}
return s
}
func parseSimpleAttr(rdr io.ByteScanner, sel *SimpleSelector) error {
var name []byte
var value []byte
var c1 byte = 0
for c2, err := rdr.ReadByte(); err != io.EOF; c2, err = rdr.ReadByte() {
if err != nil {
return err
}
switch c2 {
case ']':
sel.AttrName = string(name)
sel.Value = string(value)
return nil
case '=':
if c1 == '~' {
sel.AttrMatch = Contains
} else if c1 == '|' {
sel.AttrMatch = DashPrefix
} else {
sel.AttrMatch = Exactly
}
case '{':
rdr.UnreadByte()
return EOS
case '~':
case '|':
// TODO(jwall): Substring matchers
default:
if sel.AttrMatch == Presence {
name = append(name, c2)
} else {
value = append(value, c2)
}
}
c1 = c2
}
return fmt.Errorf("Didn't close Attribute Matcher")
}
// Match provides a simple sequential ASCII text matcher.
// It is specialized for processing well formed, structured printable ASCII.
// s yields input bytes and p yields pattern bytes.
// n and o correspond to the number of bytes read from s and p, respectively.
// If m is non-nil, len(m) should be greater than or equal to the number of
// capture-bytes in the pattern.
//
// Patterns:
// * Printable ASCII bytes will be matched literally.
// * All groups (specified by non-printable bytes in the pattern stream)
// are non-greedy, and match zero or more characters.
// * \x00: ASCII whitespace bytes.
// * \x01: non-whitespace ASCII bytes.
// * \xa0: lowercase letters.
// * \xa1: uppercase letters.
// * \xaf: letters.
// * \xd0: decimal digits.
// * \xd6: hexadecimal digits.
// * \xd8: octal digits.
// * \xd3: base-36 digits.
// * \xfe: printable ASCII bytes.
// * \xff: 8-bit bytes.
// * All groups are capturing beside \x00.
// * Use of other non-printable or non-ASCII bytes is undefined.
func Match(s, p io.ByteScanner, m []string) (n, o int, err error) {
var (
quit = false
buf = make([]byte, 0, 1024)
a, b byte
c byte
v int
)
for !quit {
a, err = p.ReadByte()
switch {
case err == io.EOF:
err = nil
fallthrough
case err != nil:
return
}
o++
if tab[a]&prg == 0 {
c, err = s.ReadByte()
if err != nil {
return
}
n++
if a == c {
continue
} else if err = s.UnreadByte(); err == nil {
err = ErrByteMismatch
n--
}
return
}
a = tab[a]
b, err = p.ReadByte()
if err == io.EOF {
b = nop
} else if err != nil {
return
} else if err = p.UnreadByte(); err != nil {
o++
return
}
for {
c, err = s.ReadByte()
if err != nil {
quit = true
goto fill
}
n++
if tab[b]&prg == 0 && c == b ||
tab[b]&prg != 0 && tab[c]&tab[b] != 0 ||
a != any && tab[c]&a == 0 {
break
} else if a&^prg != ws {
buf = append(buf, c)
}
}
if err = s.UnreadByte(); err == nil {
err = io.EOF
n--
}
fill:
if a&^prg != ws {
if _, ok := s.(*rb); ok {
revbytes(buf)
}
m[v] = string(buf)
v++
buf = buf[:0]
}
}
return
}
// scanExponent scans the longest possible prefix of r representing a decimal
// ('e', 'E') or binary ('p') exponent, if any. It returns the exponent, the
// exponent base (10 or 2), or a read or syntax error, if any.
//
// exponent = ( "E" | "e" | "p" ) [ sign ] digits .
// sign = "+" | "-" .
// digits = digit { digit } .
// digit = "0" ... "9" .
//
// A binary exponent is only permitted if binExpOk is set.
func scanExponent(r io.ByteScanner, binExpOk bool) (exp int64, base int, err error) {
base = 10
var ch byte
if ch, err = r.ReadByte(); err != nil {
if err == io.EOF {
err = nil // no exponent; same as e0
}
return
}
switch ch {
case 'e', 'E':
// ok
case 'p':
if binExpOk {
base = 2
break // ok
}
fallthrough // binary exponent not permitted
default:
r.UnreadByte()
return // no exponent; same as e0
}
var neg bool
if neg, err = scanSign(r); err != nil {
return
}
var digits []byte
if neg {
digits = append(digits, '-')
}
// no need to use nat.scan for exponent digits
// since we only care about int64 values - the
// from-scratch scan is easy enough and faster
for i := 0; ; i++ {
if ch, err = r.ReadByte(); err != nil {
if err != io.EOF || i == 0 {
return
}
err = nil
break // i > 0
}
if ch < '0' || '9' < ch {
if i == 0 {
r.UnreadByte()
err = fmt.Errorf("invalid exponent (missing digits)")
return
}
break // i > 0
}
digits = append(digits, byte(ch))
}
// i > 0 => we have at least one digit
exp, err = strconv.ParseInt(string(digits), 10, 64)
return
}
// scan scans the number corresponding to the longest possible prefix
// from r representing an unsigned number in a given conversion base.
// It returns the corresponding natural number res, the actual base b,
// a digit count, and a read or syntax error err, if any.
//
// number = [ prefix ] mantissa .
// prefix = "0" [ "x" | "X" | "b" | "B" ] .
// mantissa = digits | digits "." [ digits ] | "." digits .
// digits = digit { digit } .
// digit = "0" ... "9" | "a" ... "z" | "A" ... "Z" .
//
// Unless fracOk is set, the base argument must be 0 or a value between
// 2 and MaxBase. If fracOk is set, the base argument must be one of
// 0, 2, 10, or 16. Providing an invalid base argument leads to a run-
// time panic.
//
// For base 0, the number prefix determines the actual base: A prefix of
// ``0x'' or ``0X'' selects base 16; if fracOk is not set, the ``0'' prefix
// selects base 8, and a ``0b'' or ``0B'' prefix selects base 2. Otherwise
// the selected base is 10 and no prefix is accepted.
//
// If fracOk is set, an octal prefix is ignored (a leading ``0'' simply
// stands for a zero digit), and a period followed by a fractional part
// is permitted. The result value is computed as if there were no period
// present; and the count value is used to determine the fractional part.
//
// A result digit count > 0 corresponds to the number of (non-prefix) digits
// parsed. A digit count <= 0 indicates the presence of a period (if fracOk
// is set, only), and -count is the number of fractional digits found.
// In this case, the actual value of the scanned number is res * b**count.
//
func (z nat) scan(r io.ByteScanner, base int, fracOk bool) (res nat, b, count int, err error) {
// reject illegal bases
baseOk := base == 0 ||
!fracOk && 2 <= base && base <= MaxBase ||
fracOk && (base == 2 || base == 10 || base == 16)
if !baseOk {
panic(fmt.Sprintf("illegal number base %d", base))
}
// one char look-ahead
ch, err := r.ReadByte()
if err != nil {
return
}
// determine actual base
b = base
if base == 0 {
// actual base is 10 unless there's a base prefix
b = 10
if ch == '0' {
count = 1
switch ch, err = r.ReadByte(); err {
case nil:
// possibly one of 0x, 0X, 0b, 0B
if !fracOk {
b = 8
}
switch ch {
case 'x', 'X':
b = 16
case 'b', 'B':
b = 2
}
switch b {
case 16, 2:
count = 0 // prefix is not counted
if ch, err = r.ReadByte(); err != nil {
// io.EOF is also an error in this case
return
}
case 8:
count = 0 // prefix is not counted
}
case io.EOF:
// input is "0"
res = z[:0]
err = nil
return
default:
// read error
return
}
}
}
// convert string
// Algorithm: Collect digits in groups of at most n digits in di
// and then use mulAddWW for every such group to add them to the
// result.
z = z[:0]
b1 := Word(b)
bn, n := maxPow(b1) // at most n digits in base b1 fit into Word
di := Word(0) // 0 <= di < b1**i < bn
i := 0 // 0 <= i < n
dp := -1 // position of decimal point
for {
if fracOk && ch == '.' {
fracOk = false
dp = count
// advance
if ch, err = r.ReadByte(); err != nil {
if err == io.EOF {
err = nil
break
}
return
}
}
// convert rune into digit value d1
var d1 Word
switch {
case '0' <= ch && ch <= '9':
d1 = Word(ch - '0')
case 'a' <= ch && ch <= 'z':
d1 = Word(ch - 'a' + 10)
case 'A' <= ch && ch <= 'Z':
d1 = Word(ch - 'A' + 10)
default:
d1 = MaxBase + 1
}
if d1 >= b1 {
r.UnreadByte() // ch does not belong to number anymore
break
}
count++
// collect d1 in di
di = di*b1 + d1
i++
// if di is "full", add it to the result
if i == n {
z = z.mulAddWW(z, bn, di)
di = 0
i = 0
}
// advance
if ch, err = r.ReadByte(); err != nil {
if err == io.EOF {
err = nil
break
}
return
}
}
if count == 0 {
// no digits found
switch {
case base == 0 && b == 8:
// there was only the octal prefix 0 (possibly followed by digits > 7);
// count as one digit and return base 10, not 8
count = 1
b = 10
case base != 0 || b != 8:
// there was neither a mantissa digit nor the octal prefix 0
err = errors.New("syntax error scanning number")
}
return
}
// count > 0
// add remaining digits to result
if i > 0 {
z = z.mulAddWW(z, pow(b1, i), di)
}
res = z.norm()
// adjust for fraction, if any
if dp >= 0 {
// 0 <= dp <= count > 0
count = dp - count
}
return
}