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
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
|
package transform
import (
"bytes"
"io"
"net/url"
"strings"
"unicode/utf8"
)
type matchState int
const (
matchStateNone matchState = iota
matchStateWhitespace
matchStatePartial
matchStateFull
)
const (
matchPrefixSrc int = iota
matchPrefixHref
)
type contentlexer struct {
content []byte
pos int // input position
start int // item start position
width int // width of last element
matchers []absURLMatcher
state stateFunc
prefixLookup *prefixes
w io.Writer
}
type stateFunc func(*contentlexer) stateFunc
type prefixRunes []rune
type prefixes struct {
pr []prefixRunes
curr prefixRunes // current prefix lookup table
i int // current index
// first rune in potential match
first rune
// match-state:
// none, whitespace, partial, full
ms matchState
}
// match returns partial and full match for the prefix in play
// - it's a full match if all prefix runes has checked out in row
// - it's a partial match if it's on its way towards a full match
func (l *contentlexer) match(r rune) {
p := l.prefixLookup
if p.curr == nil {
// assumes prefixes all start off on a different rune
// works in this special case: href, src
p.i = 0
for _, pr := range p.pr {
if pr[p.i] == r {
fullMatch := len(p.pr) == 1
p.first = r
if !fullMatch {
p.curr = pr
l.prefixLookup.ms = matchStatePartial
} else {
l.prefixLookup.ms = matchStateFull
}
return
}
}
} else {
p.i++
if p.curr[p.i] == r {
fullMatch := len(p.curr) == p.i+1
if fullMatch {
p.curr = nil
l.prefixLookup.ms = matchStateFull
} else {
l.prefixLookup.ms = matchStatePartial
}
return
}
p.curr = nil
}
l.prefixLookup.ms = matchStateNone
}
func (l *contentlexer) emit() {
l.w.Write(l.content[l.start:l.pos])
l.start = l.pos
}
var mainPrefixRunes = []prefixRunes{{'s', 'r', 'c', '='}, {'h', 'r', 'e', 'f', '='}}
type absURLMatcher struct {
prefix int
match []byte
replacement []byte
}
func (a absURLMatcher) isSourceType() bool {
return a.prefix == matchPrefixSrc
}
func checkCandidate(l *contentlexer) {
isSource := l.prefixLookup.first == 's'
for _, m := range l.matchers {
if isSource && !m.isSourceType() || !isSource && m.isSourceType() {
continue
}
if bytes.HasPrefix(l.content[l.pos:], m.match) {
// check for schemaless URLs
posAfter := l.pos + len(m.match)
if int(posAfter) >= len(l.content) {
return
}
r, _ := utf8.DecodeRune(l.content[posAfter:])
if r == '/' {
// schemaless: skip
return
}
if l.pos > l.start {
l.emit()
}
l.pos += len(m.match)
l.w.Write(m.replacement)
l.start = l.pos
return
}
}
}
func (l *contentlexer) replace() {
contentLength := len(l.content)
var r rune
for {
if int(l.pos) >= contentLength {
l.width = 0
break
}
var width int = 1
r = rune(l.content[l.pos])
if r >= utf8.RuneSelf {
r, width = utf8.DecodeRune(l.content[l.pos:])
}
l.width = width
l.pos += l.width
if r == ' ' {
l.prefixLookup.ms = matchStateWhitespace
} else if l.prefixLookup.ms != matchStateNone {
l.match(r)
if l.prefixLookup.ms == matchStateFull {
checkCandidate(l)
}
}
}
// Done!
if l.pos > l.start {
l.emit()
}
}
func doReplace(rw contentRewriter, matchers []absURLMatcher) {
lexer := &contentlexer{
content: rw.Content(),
w: rw,
prefixLookup: &prefixes{pr: mainPrefixRunes},
matchers: matchers}
lexer.replace()
}
type absURLReplacer struct {
htmlMatchers []absURLMatcher
xmlMatchers []absURLMatcher
}
func newAbsURLReplacer(baseURL string) *absURLReplacer {
u, _ := url.Parse(baseURL)
base := strings.TrimRight(u.String(), "/")
// HTML
dqHTMLMatch := []byte("\"/")
sqHTMLMatch := []byte("'/")
// XML
dqXMLMatch := []byte(""/")
sqXMLMatch := []byte("'/")
dqHTML := []byte("\"" + base + "/")
sqHTML := []byte("'" + base + "/")
dqXML := []byte(""" + base + "/")
sqXML := []byte("'" + base + "/")
return &absURLReplacer{
htmlMatchers: []absURLMatcher{
{matchPrefixSrc, dqHTMLMatch, dqHTML},
{matchPrefixSrc, sqHTMLMatch, sqHTML},
{matchPrefixHref, dqHTMLMatch, dqHTML},
{matchPrefixHref, sqHTMLMatch, sqHTML}},
xmlMatchers: []absURLMatcher{
{matchPrefixSrc, dqXMLMatch, dqXML},
{matchPrefixSrc, sqXMLMatch, sqXML},
{matchPrefixHref, dqXMLMatch, dqXML},
{matchPrefixHref, sqXMLMatch, sqXML},
}}
}
func (au *absURLReplacer) replaceInHTML(rw contentRewriter) {
doReplace(rw, au.htmlMatchers)
}
func (au *absURLReplacer) replaceInXML(rw contentRewriter) {
doReplace(rw, au.xmlMatchers)
}
|