Repository navigation
Expand file tree
/
Copy pathreadability.go
More file actions
740 lines (609 loc) · 18 KB
/
Copy pathreadability.go
File metadata and controls
740 lines (609 loc) · 18 KB
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
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
package readability
import (
"bytes"
"golang.org/x/net/html"
"fmt"
"github.com/PuerkitoBio/goquery"
"io"
"io/ioutil"
"log"
"math"
"net/url"
"regexp"
"sort"
"strings"
)
var replaceBrsRegexp = regexp.MustCompile(`(?i)(<br[^>]*>[ \n\r\t]*){2,}`)
var replaceFontsRegexp = regexp.MustCompile(`(?i)<(\/?)\s*font[^>]*?>`)
var blacklistCandidatesRegexp = regexp.MustCompile(`(?i)popupbody`)
var okMaybeItsACandidateRegexp = regexp.MustCompile(`(?i)and|article|body|column|main|shadow`)
var unlikelyCandidatesRegexp = regexp.MustCompile(`(?i)combx|comment|community|hidden|disqus|modal
10BC8
|extra|foot|header|menu|remark|rss|shoutbox|sidebar|sponsor|ad-break|agegate|pagination|pager|popup`)
var divToPElementsRegexp = regexp.MustCompile(`(?i)<(a|blockquote|dl|div|img|ol|p|pre|table|ul)`)
var negativeRegexp = regexp.MustCompile(`(?i)combx|comment|com-|foot|footer|footnote|masthead|media|meta|outbrain|promo|related|scroll|shoutbox|sidebar|sponsor|shopping|tags|tool|widget`)
var positiveRegexp = regexp.MustCompile(`(?i)article|body|content|entry|hentry|main|page|pagination|post|text|blog|story`)
var stripCommentRegexp = regexp.MustCompile(`(?s)\<\!\-{2}.+?-{2}\>`)
var sentenceRegexp = regexp.MustCompile(`[.。;；]( |$)`)
var normalizeWhitespaceRegexp = regexp.MustCompile(`[\r\n\f]+`)
type candidate struct {
selection *goquery.Selection
score float32
}
type byScore []*candidate
func (a byScore) Len() int { return len(a) }
func (a byScore) Swap(i, j int) { a[i], a[j] = a[j], a[i] }
func (a byScore) Less(i, j int) bool { return a[i].score > a[j].score }
func (c *candidate) Node() *html.Node {
return c.selection.Get(0)
}
type Document struct {
input string
density float64
content string
document *goquery.Document
candidates map[*html.Node]*candidate
// bestCandidate *candidate
sortedCandidates []*candidate
RemoveUnlikelyCandidates bool
WeightClasses bool
CleanConditionally bool
// BestCandidateHasImage bool
RetryLength int
MinTextLength int
ScoreThreshold float32
RemoveEmptyNodes bool
WhitelistTags []string
Url *url.URL
// Images []string
logger *log.Logger
}
func NewDocument(s, _url string) (*Document, error) {
u, err := url.Parse(_url)
if err != nil {
return nil, err
}
d := &Document{
input: s,
WhitelistTags: []string{
"div", "p", "a", "img",
"ul", "li", "ol", "h1",
"h2", "h3", "h4", "h5",
"h6", "dl", "dd", "strong",
"span", "br", "hr", "pre", "code",
"blockquote",
},
RemoveUnlikelyCandidates: true,
WeightClasses: true,
CleanConditionally: true,
RetryLength: 200,
MinTextLength: 20,
RemoveEmptyNodes: true,
logger: log.New(ioutil.Discard, "", log.LstdFlags),
Url: u,
}
err = d.initializeHtml(s)
if err != nil {
return nil, err
}
return d, nil
}
func (d *Document) SetLogger(l *log.Logger) {
d.logger = l
}
func (d *Document) initializeHtml(s string) error {
// replace consecutive <br>'s with p tags
s = replaceBrsRegexp.ReplaceAllString(s, "</p><p>")
// replace font tags
s = replaceFontsRegexp.ReplaceAllString(s, `<${1}span>`)
// manually strip regexps since html parser seems to miss some
s = stripCommentRegexp.ReplaceAllString(s, "")
doc, err := goquery.NewDocumentFromReader(strings.NewReader(s))
if err != nil {
return err
}
// if no body (like from a redirect or empty string)
if doc.Find("body").Length() == 0 {
s = "<body/>"
return d.initializeHtml(s)
}
d.document = doc
return nil
}
func (d *Document) Content() string {
if d.content == "" {
d.prepareCandidates()
article := d.getArticle()
// d.logger.Println(":::::::::article:::::::::", article)
articleText := d.sanitize(article)
// d.logger.Println("::::::::::article_text:::::::::", articleText)
length := len(strings.TrimSpace(articleText))
if length < d.RetryLength {
retry := true
if d.RemoveUnlikelyCandidates {
d.RemoveUnlikelyCandidates = false
} else if d.WeightClasses {
d.WeightClasses = false
} else if d.CleanConditionally {
d.CleanConditionally = false
} else {
d.content = articleText
retry = false
}
if retry {
d.logger.Printf("Retrying with length %d < retry length %d\n", length, d.RetryLength)
d.initializeHtml(d.input)
articleText = d.Content()
}
}
d.content = articleText
}
return d.content
}
func (d *Document) Title() string {
return d.document.Find("title").Text()
}
func (d *Document) Text() string {
content := d.Content()
doc, err := goquery.NewDocumentFromReader(strings.NewReader(content))
if err != nil {
d.logger.Println("Unable to create document", err)
return ""
}
return doc.Text()
}
func (d *Document) Density() float64 {
content := d.Content()
return float64(len(content)) / float64(len(d.input))
}
func (d *Document) prepareCandidates() {
// noscript might be valid, but probably not so we'll just remove it
d.document.Find("script, style,noscript").Each(func(i int, s *goquery.Selection) {
removeNodes(s)
})
if d.RemoveUnlikelyCandidates {
d.removeUnlikelyCandidates()
}
d.transformMisusedDivsIntoParagraphs()
d.scoreParagraphs(d.MinTextLength)
d.selectBestCandidate()
}
func (d *Document) selectBestCandidate() {
// var best *candidate
// for _, c := range d.candidates {
// if best == nil {
// best = c
// } else if best.score < c.score {
// best = c
// }
// }
d.sortedCandidates = make([]*candidate, 0, len(d.candidates))
if len(d.candidates) > 0 {
for _, c := range d.candidates {
d.sortedCandidates = append(d.sortedCandidates, c)
}
sort.Sort(byScore(d.sortedCandidates))
// best = d.sortedCandidates[0]
for _, c := range d.sortedCandidates {
d.Debug(fmt.Sprintf(":::::::::::::::candidate:%v:::::::::::", c.score), c.selection)
}
} else {
d.sortedCandidates = append(d.sortedCandidates, &candidate{d.document.Find("body"), 0})
}
// d.bestCandidate = best
}
func (d *Document) cleanCandidates(s *goquery.Selection) {
s.Children().Each(func(i int, child *goquery.Selection) {
n := child.Get(0)
if c, ok := d.candidates[n]; ok {
// d.logger.Println("====================================",)
if c.score < d.ScoreThreshold {
d.Debug("CLEANED", child)
removeNodes(child)
}
} else {
d.cleanCandidates(child)
}
})
}
func (d *Document) getArticle() string {
output := bytes.NewBufferString("<div>")
d.ScoreThreshold = float32(math.Max(10, float64(d.sortedCandidates[0].score*.2)))
var f func(i int, s *goquery.Selection)
f = func(i int, s *goquery.Selection) {
n := s.Get(0)
append := false
if c, ok := d.candidates[n]; ok {
if c.score < d.ScoreThreshold {
// d.Debug("ignored", s)
// d.logger.Println("score:", c.score)
} else {
// d.cleanCandidates(s)
append = true
}
} else if s.Is("p") {
linkDensity := d.getLinkDensity(s)
content := s.Text()
contentLength := len(content)
if contentLength >= 80 && linkDensity < .25 {
append = true
} else if contentLength < 80 && linkDensity == 0 {
append = sentenceRegexp.MatchString(content)
}
}
if append {
appendDoc(output, s)
} else {
s.Children().Each(f)
}
}
if len(d.sortedCandidates) == 1 {
// d.bestCandidate.selection.Siblings().Union(d.bestCandidate.selection).Each(func(i int, s *goquery.Selection) {
// d.sortedCandidates[0].selection.Parent().Children().Each(f)
appendDoc(output, d.sortedCandidates[0].selection)
} else {
s := findLCA(d.document.Find("body"), d.sortedCandidates[0].selection, d.sortedCandidates[1].selection)
d.Debug("findLCA", s)
s.Children().Each(f)
}
output.Write([]byte("</div>"))
return output.String()
}
func (d *Document) removeUnlikelyCandidates() {
d.document.Find("*").Not("html,body").Each(func(i int, s *goquery.Selection) {
class, _ := s.Attr("class")
id, _ := s.Attr("id")
str := class + id
if blacklistCandidatesRegexp.MatchString(str) || (unlikelyCandidatesRegexp.MatchString(str) && !okMaybeItsACandidateRegexp.MatchString(str)) {
d.logger.Printf("Removing unlikely candidate -%s\n", str)
removeNodes(s)
}
})
}
func (d *Document) transformMisusedDivsIntoParagraphs() {
d.document.Find("div").Each(func(i int, s *goquery.Selection) {
html, err := s.Html()
if err != nil {
d.logger.Printf("Unable to transform div to p %s\n", err)
return
}
// transform <div>s that do not contain other block elements into <p>s
if !divToPElementsRegexp.MatchString(html) {
class, _ := s.Attr("class")
id, _ := s.Attr("id")
d.logger.Printf("Altering div(#%s.%s) to p\n", id, class)
node := s.Get(0)
node.Data = "p"
}
})
}
func (d *Document) scoreParagraphs(minimumTextLength int) {
candidates := make(map[*html.Node]*candidate)
htmlNode := d.document.Find("html").Get(0)
d.document.Find("p,td").Each(func(i int, s *goquery.Selection) {
text := s.Text()
// if this paragraph is less than x chars, don't count it
if len(text) < minimumTextLength {
return
}
parent := s.Parent()
parentNode := parent.Get(0)
grandparent := parent.Parent()
var grandparentNode *html.Node
if grandparent.Length() > 0 {
grandparentNode = grandparent.Get(0)
}
if _, ok := candidates[parentNode]; !ok {
candidates[parentNode] = d.scoreNode(parent)
}
if grandparentNode != nil && grandparentNode != htmlNode {
if _, ok := candidates[grandparentNode]; !ok {
candidates[grandparentNode] = d.scoreNode(grandparent)
}
}
contentScore := float32(1.0)
contentScore += float32(strings.Count(text, ",") + strings.Count(text, "，") + 1)
contentScore += float32(math.Min(float64(int(len(text)/100.0)), 3))
candidates[parentNode].score += contentScore
if grandparentNode != nil && grandparentNode != htmlNode {
candidates[grandparentNode].score += contentScore / 2.0
}
})
// scale the final candidates score based on link density. Good content
// should have a relatively small link density (5% or less) and be mostly
// unaffected by this operation
for _, candidate := range candidates {
candidate.score = candidate.score * (1 - d.getLinkDensity(candidate.selection))
}
d.candidates = candidates
}
func (d *Document) getLinkDensity(s *goquery.Selection) float32 {
linkLength := len(s.Find("a").Text())
textLength := len(s.Text())
if textLength == 0 {
return 0
}
return float32(linkLength) / float32(textLength)
}
func (d *Document) classWeight(s *goquery.Selection) int {
weight := 0
if !d.WeightClasses {
return weight
}
class, _ := s.Attr("class")
id, _ := s.Attr("id")
if class != "" {
if negativeRegexp.MatchString(class) {
weight -= 25
}
if positiveRegexp.MatchString(class) {
weight += 25
}
}
if id != "" {
if negativeRegexp.MatchString(id) {
weight -= 25
}
if positiveRegexp.MatchString(id) {
weight += 25
}
}
return weight
}
func (d *Document) scoreNode(s *goquery.Selection) *candidate {
contentScore := d.classWeight(s)
if s.Is("div") {
contentScore += 5
} else if s.Is("blockquote") {
contentScore += 3
} else if s.Is("form") {
contentScore -= 3
} else if s.Is("th") {
contentScore -= 5
}
return &candidate{s, float32(contentScore)}
}
func (d *Document) sanitize(article string) string {
doc, err := goquery.NewDocumentFromReader(strings.NewReader(article))
if err != nil {
d.logger.Println("Unable to create document", err)
return ""
}
s := doc.Find("body")
s.Find("h1,h2,h3,h4,h5,h6").Each(func(i int, header *goquery.Selection) {
if d.classWeight(header) < 0 || d.getLinkDensity(header) > 0.33 {
removeNodes(header)
}
})
//fix img
s.Find("img").Each(func(i int, s *goquery.Selection) {
node := s.Get(0)
d.fixImageSrc(node.Attr, d.Url.Host)
})
s.Find("input,select,textarea,button,object,iframe,embed").Each(func(i int, s *goquery.Selection) {
removeNodes(s)
})
if d.RemoveEmptyNodes {
s.Find("p").Each(func(i int, s *goquery.Selection) {
html, _ := s.Html()
if len(strings.TrimSpace(html)) == 0 {
removeNodes(s)
}
})
}
// d.Debug("before clean...", s)
// s.Find("*").Each(func(i int, s *goquery.Selection) {
// //.......
// node := s.Get(0)
// if c, ok := d.candidates[node]; ok && c.score < d.ScoreThreshold {
// d.Debug("ingored in sanitize", c.selection)
// removeNodes(c.selection)
// return
// }
// })
// d.Debug("after clean...", s)
// d.Debug("before clean...", s)
d.cleanConditionally(s, "table,ul,div")
// d.Debug("after clean...", s)
// we'll sanitize all elements using a whitelist
replaceWithWhitespace := map[string]bool{
"br": true,
"hr": true,
"h1": true,
"h2": true,
"h3": true,
"h4": true,
"h5": true,
"h6": true,
"dl": true,
"dd": true,
"ol": true,
"li": true,
"ul": true,
"address": true,
"blockquote": true,
"center": true,
}
whitelist := make(map[string]bool)
for _, tag := range d.WhitelistTags {
tag = strings.ToLower(tag)
whitelist[tag] = true
delete(replaceWithWhitespace, tag)
}
var text string
s.Find("*").Each(func(i int, s *goquery.Selection) {
if text != "" {
return
}
// only look at element nodes
node := s.Get(0)
if node.Type != html.ElementNode {
return
}
// if element is in whitelist, delete all its attributes
if _, ok := whitelist[node.Data]; ok {
// node.Attr = make([]html.Attribute, 0)
node.Attr = cleanStyle(node.Attr)
} else {
if _, ok := replaceWithWhitespace[node.Data]; ok {
// just replace with a text node and add whitesp
1798
ace
node.Data = fmt.Sprintf(" %s ", s.Text())
node.Type = html.TextNode
node.FirstChild = nil
node.LastChild = nil
} else {
if node.Parent == nil {
text = s.Text()
return
} else {
// replace node with children
replaceNodeWithChildren(node)
}
}
}
})
if text == "" {
text, _ = doc.Find("body").Html()
}
return normalizeWhitespaceRegexp.ReplaceAllString(text, "\n")
}
func (d *Document) cleanConditionally(s *goquery.Selection, selector string) {
if !d.CleanConditionally {
return
}
s.Find(selector).Each(func(i int, s *goquery.Selection) {
node := s.Get(0)
weight := float32(d.classWeight(s))
contentScore := float32(0)
if c, ok := d.candidates[node]; ok {
contentScore = c.score
}
if weight+contentScore < 0 {
removeNodes(s)
d.logger.Printf("Conditionally cleaned %s%s with weight %f and content score %f\n", node.Data, getName(s), weight, contentScore)
return
}
text := s.Text()
if strings.Count(text, ",")+strings.Count(text, "，") < 10 {
counts := map[string]int{
"p": s.Find("p").Length(),
"img": s.Find("img").Length(),
"li": s.Find("li").Length() - 100,
"a": s.Find("a").Length(),
"embed": s.Find("embed").Length(),
"input": s.Find("input").Length(),
}
contentLength := len(strings.TrimSpace(text))
linkDensity := d.getLinkDensity(s)
remove := false
reason := ""
if counts["img"] > counts["p"] {
reason = "too many images"
remove = true
} else if counts["li"] > counts["p"] && !s.Is("ul,ol") {
reason = "more <li>s than <p>s"
remove = true
} else if counts["input"] > int(counts["p"]/3.0) {
reason = "less than 3x <p>s than <input>s"
remove = true
} else if contentLength < d.MinTextLength && (counts["img"] == 0 || counts["img"] > 2) {
reason = "too short content length without a single image"
remove = true
} else if weight < 25 && linkDensity > 0.2 {
reason = fmt.Sprintf("too many links for its weight (%f)", weight)
remove = true
} else if weight >= 25 && linkDensity > 0.5 {
reason = fmt.Sprintf("too many links for its weight (%f)", weight)
remove = true
} else if (counts["embed"] == 1 && contentLength < 75) || counts["embed"] > 1 {
reason = "<embed>s with too short a content length, or too many <embed>s"
remove = true
}
if remove {
d.logger.Printf("Conditionally cleaned %s%s with weight %f and content score %f because it has %s\n", node.Data, getName(s), weight, contentScore, reason)
removeNodes(s)
}
}
})
}
func (d *Document) Debug(tag string, s *goquery.Selection) {
d.logger.Println("############################################", tag, "##########################################")
html, _ := s.Html()
d.logger.Println(html)
d.logger.Println("############################################", tag, "####################################################")
}
func getName(s *goquery.Selection) string {
class, _ := s.Attr("class")
id, _ := s.Attr("id")
return fmt.Sprintf("#%s.%s", id, class)
}
func removeNodes(s *goquery.Selection) {
s.Each(func(i int, s *goquery.Selection) {
parent := s.Parent()
if parent.Length() == 0 {
// TODO???
} else {
parent.Get(0).RemoveChild(s.Get(0))
}
})
}
func replaceNodeWithChildren(n *html.Node) {
var next *html.Node
parent := n.Parent
for c := n.FirstChild; c != nil; c = next {
next = c.NextSibling
n.RemoveChild(c)
parent.InsertBefore(c, n)
}
parent.RemoveChild(n)
}
func cleanStyle(attrs []html.Attribute) []html.Attribute {
ret := make([]html.Attribute, 0, len(attrs))
for _, attr := range attrs {
if attr.Key == "id" || attr.Key == "class" || attr.Key == "style" {
continue
}
ret = append(ret, attr)
}
return ret
}
func (d *Document) fixImageSrc(attrs []html.Attribute, host string) {
for i, attr := range attrs {
if attr.Key == "src" {
if strings.HasPrefix(attr.Val, "https://") || strings.HasPrefix(attr.Val, "http://") {
continue
}
attr.Val = fmt.Sprintf("http://%s%s", host, attr.Val)
attrs[i] = attr
return
}
}
}
func appendDoc(w io.Writer, s *goquery.Selection) {
tag := "div"
if s.Is("p") {
tag = s.Get(0).Data
}
html, _ := s.Html()
fmt.Fprintf(w, "<%s>%s</%s>", tag, html, tag)
}
func findLCA(root, n, m *goquery.Selection) *goquery.Selection {
i := 0
j := 0
for c := n; c.Get(0) != root.Get(0); c = c.Parent() {
i++
}
for c := m; c.Get(0) != root.Get(0); c = c.Parent() {
j++
}
if i > j {
for k := 0; k < i-j; k++ {
n = n.Parent()
}
} else {
for k := 0; k < j-i; k++ {
m = m.Parent()
}
}
for ; n.Get(0) != m.Get(0); n, m = n.Parent(), m.Parent() {
}
return n
}