-
Notifications
You must be signed in to change notification settings - Fork 20
/
html2text.go
333 lines (288 loc) · 7.98 KB
/
html2text.go
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
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
package html2text
import (
"bytes"
"regexp"
"strconv"
"strings"
)
// Line break constants
// Deprecated: Please use HTML2TextWithOptions(text, WithUnixLineBreak())
const (
WIN_LBR = "\r\n"
UNIX_LBR = "\n"
)
var legacyLBR = WIN_LBR
var badTagnamesRE = regexp.MustCompile(`^(head|script|style|a)($|\s+)`)
var linkTagRE = regexp.MustCompile(`^(?i:a)(?:$|\s).*(?i:href)\s*=\s*('([^']*?)'|"([^"]*?)"|([^\s"'` + "`" + `=<>]+))`)
var badLinkHrefRE = regexp.MustCompile(`javascript:`)
var headersRE = regexp.MustCompile(`^(\/)?h[1-6]`)
var numericEntityRE = regexp.MustCompile(`(?i)^#(x?[a-f0-9]+)$`)
type options struct {
lbr string
linksInnerText bool
listPrefix string
}
func newOptions() *options {
// apply defaults
return &options{
lbr: WIN_LBR,
}
}
// Option is a functional option
type Option func(*options)
// WithUnixLineBreaks instructs the converter to use unix line breaks ("\n" instead of "\r\n" default)
func WithUnixLineBreaks() Option {
return func(o *options) {
o.lbr = UNIX_LBR
}
}
// WithLinksInnerText instructs the converter to retain link tag inner text and append href URLs in angle brackets after the text
// Example: click news <http://bit.ly/2n4wXRs>
func WithLinksInnerText() Option {
return func(o *options) {
o.linksInnerText = true
}
}
// WithListSupportPrefix formats <ul> and <li> lists with the specified prefix
func WithListSupportPrefix(prefix string) Option {
return func(o *options) {
o.listPrefix = prefix
}
}
// WithListSupport formats <ul> and <li> lists with " - " prefix
func WithListSupport() Option {
return WithListSupportPrefix(" - ")
}
func parseHTMLEntity(entName string) (string, bool) {
if r, ok := entity[entName]; ok {
return string(r), true
}
if match := numericEntityRE.FindStringSubmatch(entName); len(match) == 2 {
var (
err error
n int64
digits = match[1]
)
if digits != "" && (digits[0] == 'x' || digits[0] == 'X') {
n, err = strconv.ParseInt(digits[1:], 16, 64)
} else {
n, err = strconv.ParseInt(digits, 10, 64)
}
if err == nil && (n == 9 || n == 10 || n == 13 || n > 31) {
return string(rune(n)), true
}
}
return "", false
}
// SetUnixLbr with argument true sets Unix-style line-breaks in output ("\n")
// with argument false sets Windows-style line-breaks in output ("\r\n", the default)
// Deprecated: Please use HTML2TextWithOptions(text, WithUnixLineBreak())
func SetUnixLbr(b bool) {
if b {
legacyLBR = UNIX_LBR
} else {
legacyLBR = WIN_LBR
}
}
// HTMLEntitiesToText decodes HTML entities inside a provided
// string and returns decoded text
func HTMLEntitiesToText(htmlEntsText string) string {
outBuf := bytes.NewBufferString("")
inEnt := false
for i, r := range htmlEntsText {
switch {
case r == ';' && inEnt:
inEnt = false
continue
case r == '&': //possible html entity
entName := ""
isEnt := false
// parse the entity name - max 10 chars
chars := 0
for _, er := range htmlEntsText[i+1:] {
if er == ';' {
isEnt = true
break
} else {
entName += string(er)
}
chars++
if chars == 10 {
break
}
}
if isEnt {
if ent, isEnt := parseHTMLEntity(entName); isEnt {
outBuf.WriteString(ent)
inEnt = true
continue
}
}
}
if !inEnt {
outBuf.WriteRune(r)
}
}
return outBuf.String()
}
func writeSpace(outBuf *bytes.Buffer) {
bts := outBuf.Bytes()
if len(bts) > 0 && bts[len(bts)-1] != ' ' {
outBuf.WriteString(" ")
}
}
// HTML2Text converts html into a text form
func HTML2Text(html string) string {
var opts []Option
if legacyLBR == UNIX_LBR {
opts = append(opts, WithUnixLineBreaks())
}
return HTML2TextWithOptions(html, opts...)
}
// HTML2TextWithOptions converts html into a text form with additional options
func HTML2TextWithOptions(html string, reqOpts ...Option) string {
opts := newOptions()
for _, opt := range reqOpts {
opt(opts)
}
inLen := len(html)
tagStart := 0
inEnt := false
badTagStackDepth := 0 // if == 1 it means we are inside <head>...</head>
shouldOutput := true
// maintain a stack of <a> tag href links and output it after the tag's inner text (for opts.linksInnerText only)
hrefs := []string{}
// new line cannot be printed at the beginning or
// for <p> after a new line created by previous <p></p>
canPrintNewline := false
outBuf := bytes.NewBufferString("")
for i, r := range html {
if inLen > 0 && i == inLen-1 {
// prevent new line at the end of the document
canPrintNewline = false
}
switch {
// skip new lines and spaces adding a single space if not there yet
case r <= 0xD, r == 0x85, r == 0x2028, r == 0x2029, // new lines
r == ' ', r >= 0x2008 && r <= 0x200B: // spaces
if shouldOutput && badTagStackDepth == 0 && !inEnt {
//outBuf.WriteString(fmt.Sprintf("{DBG r:%c, inEnt:%t, tag:%s}", r, inEnt, html[tagStart:i]))
writeSpace(outBuf)
}
continue
case r == ';' && inEnt: // end of html entity
inEnt = false
continue
case r == '&' && shouldOutput: // possible html entity
entName := ""
isEnt := false
// parse the entity name - max 10 chars
chars := 0
for _, er := range html[i+1:] {
if er == ';' {
isEnt = true
break
} else {
entName += string(er)
}
chars++
if chars == 10 {
break
}
}
if isEnt {
if ent, isEnt := parseHTMLEntity(entName); isEnt {
outBuf.WriteString(ent)
inEnt = true
continue
}
}
case r == '<': // start of a tag
tagStart = i + 1
shouldOutput = false
continue
case r == '>': // end of a tag
shouldOutput = true
tag := html[tagStart:i]
tagNameLowercase := strings.ToLower(tag)
if tagNameLowercase == "/ul" || tagNameLowercase == "/ol" {
outBuf.WriteString(opts.lbr)
} else if tagNameLowercase == "li" || tagNameLowercase == "li/" {
if opts.listPrefix != "" {
outBuf.WriteString(opts.lbr + opts.listPrefix)
} else {
outBuf.WriteString(opts.lbr)
}
} else if headersRE.MatchString(tagNameLowercase) {
if canPrintNewline {
outBuf.WriteString(opts.lbr + opts.lbr)
}
canPrintNewline = false
} else if tagNameLowercase == "br" || tagNameLowercase == "br/" {
// new line
outBuf.WriteString(opts.lbr)
} else if tagNameLowercase == "p" || tagNameLowercase == "/p" {
if canPrintNewline {
outBuf.WriteString(opts.lbr + opts.lbr)
}
canPrintNewline = false
} else if opts.linksInnerText && tagNameLowercase == "/a" {
// end of link
// links can be empty can happen if the link matches the badLinkHrefRE
if len(hrefs) > 0 {
outBuf.WriteString(" <")
outBuf.WriteString(HTMLEntitiesToText(hrefs[0]))
outBuf.WriteString(">")
hrefs = hrefs[1:]
}
} else if opts.linksInnerText && linkTagRE.MatchString(tagNameLowercase) {
// parse link href
// add special handling for a tags
m := linkTagRE.FindStringSubmatch(tag)
if len(m) == 5 {
link := m[2]
if len(link) == 0 {
link = m[3]
if len(link) == 0 {
link = m[4]
}
}
if opts.linksInnerText && !badLinkHrefRE.MatchString(link) {
hrefs = append(hrefs, link)
}
}
} else if badTagnamesRE.MatchString(tagNameLowercase) {
// unwanted block
badTagStackDepth++
// if link inner text preservation is not enabled
// and the current tag is a link tag, parse its href and output that
if !opts.linksInnerText {
// parse link href
m := linkTagRE.FindStringSubmatch(tag)
if len(m) == 5 {
link := m[2]
if len(link) == 0 {
link = m[3]
if len(link) == 0 {
link = m[4]
}
}
if !badLinkHrefRE.MatchString(link) {
outBuf.WriteString(HTMLEntitiesToText(link))
}
}
}
} else if len(tagNameLowercase) > 0 && tagNameLowercase[0] == '/' &&
badTagnamesRE.MatchString(tagNameLowercase[1:]) {
// end of unwanted block
badTagStackDepth--
}
continue
} // switch end
if shouldOutput && badTagStackDepth == 0 && !inEnt {
canPrintNewline = true
outBuf.WriteRune(r)
}
}
return outBuf.String()
}