1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
|
package svg
import (
"github.com/tdewolff/parse/v2"
"github.com/tdewolff/parse/v2/xml"
minifyXML "github.com/tdewolff/minify/v2/xml"
)
// Token is a single token unit with an attribute value (if given) and hash of the data.
type Token struct {
xml.TokenType
Hash Hash
Data []byte
Text []byte
AttrVal []byte
Offset int
}
// TokenBuffer is a buffer that allows for token look-ahead.
type TokenBuffer struct {
r *parse.Input
l *xml.Lexer
buf []Token
pos int
attrBuffer []*Token
}
// NewTokenBuffer returns a new TokenBuffer.
func NewTokenBuffer(r *parse.Input, l *xml.Lexer) *TokenBuffer {
return &TokenBuffer{
r: r,
l: l,
buf: make([]Token, 0, 8),
}
}
func (z *TokenBuffer) read(t *Token) {
t.Offset = z.r.Offset()
t.TokenType, t.Data = z.l.Next()
t.Text = z.l.Text()
if t.TokenType == xml.AttributeToken {
t.Offset += 1 + len(t.Text) + 1
t.AttrVal = z.l.AttrVal()
if len(t.AttrVal) > 1 && (t.AttrVal[0] == '"' || t.AttrVal[0] == '\'') {
t.Offset++
t.AttrVal = t.AttrVal[1 : len(t.AttrVal)-1] // quotes will be readded in attribute loop if necessary
t.AttrVal = parse.ReplaceMultipleWhitespaceAndEntities(t.AttrVal, minifyXML.EntitiesMap, nil)
t.AttrVal = parse.TrimWhitespace(t.AttrVal)
}
t.Hash = ToHash(t.Text)
} else if t.TokenType == xml.StartTagToken || t.TokenType == xml.EndTagToken {
t.AttrVal = nil
t.Hash = ToHash(t.Text)
} else {
t.AttrVal = nil
t.Hash = 0
}
}
// Peek returns the ith element and possibly does an allocation.
// Peeking past an error will panic.
func (z *TokenBuffer) Peek(pos int) *Token {
pos += z.pos
if pos >= len(z.buf) {
if len(z.buf) > 0 && z.buf[len(z.buf)-1].TokenType == xml.ErrorToken {
return &z.buf[len(z.buf)-1]
}
c := cap(z.buf)
d := len(z.buf) - z.pos
p := pos - z.pos + 1 // required peek length
var buf []Token
if 2*p > c {
buf = make([]Token, 0, 2*c+p)
} else {
buf = z.buf
}
copy(buf[:d], z.buf[z.pos:])
buf = buf[:p]
pos -= z.pos
for i := d; i < p; i++ {
z.read(&buf[i])
if buf[i].TokenType == xml.ErrorToken {
buf = buf[:i+1]
pos = i
break
}
}
z.pos, z.buf = 0, buf
}
return &z.buf[pos]
}
// Shift returns the first element and advances position.
func (z *TokenBuffer) Shift() *Token {
if z.pos >= len(z.buf) {
t := &z.buf[:1][0]
z.read(t)
return t
}
t := &z.buf[z.pos]
z.pos++
return t
}
// Attributes extracts the gives attribute hashes from a tag.
// It returns in the same order pointers to the requested token data or nil.
func (z *TokenBuffer) Attributes(hashes ...Hash) []*Token {
n := 0
for {
if t := z.Peek(n); t.TokenType != xml.AttributeToken {
break
}
n++
}
if len(hashes) > cap(z.attrBuffer) {
z.attrBuffer = make([]*Token, len(hashes))
} else {
z.attrBuffer = z.attrBuffer[:len(hashes)]
for i := range z.attrBuffer {
z.attrBuffer[i] = nil
}
}
for i := z.pos; i < z.pos+n; i++ {
attr := &z.buf[i]
for j, hash := range hashes {
if hash == attr.Hash {
z.attrBuffer[j] = attr
}
}
}
return z.attrBuffer
}
|
