File: detector.go

package info (click to toggle)
golang-github-gogits-chardet 0.0~git20150115.0.2404f77+dfsg-4
  • links: PTS, VCS
  • area: main
  • in suites: bullseye, buster, sid
  • size: 172 kB
  • sloc: makefile: 3
file content (136 lines) | stat: -rw-r--r-- 3,667 bytes parent folder | download
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 chardet ports character set detection from ICU.
package chardet

import (
	"errors"
	"sort"
)

// Result contains all the information that charset detector gives.
type Result struct {
	// IANA name of the detected charset.
	Charset string
	// IANA name of the detected language. It may be empty for some charsets.
	Language string
	// Confidence of the Result. Scale from 1 to 100. The bigger, the more confident.
	Confidence int
}

// Detector implements charset detection.
type Detector struct {
	recognizers []recognizer
	stripTag    bool
}

// List of charset recognizers
var recognizers = []recognizer{
	newRecognizer_utf8(),
	newRecognizer_utf16be(),
	newRecognizer_utf16le(),
	newRecognizer_utf32be(),
	newRecognizer_utf32le(),
	newRecognizer_8859_1_en(),
	newRecognizer_8859_1_da(),
	newRecognizer_8859_1_de(),
	newRecognizer_8859_1_es(),
	newRecognizer_8859_1_fr(),
	newRecognizer_8859_1_it(),
	newRecognizer_8859_1_nl(),
	newRecognizer_8859_1_no(),
	newRecognizer_8859_1_pt(),
	newRecognizer_8859_1_sv(),
	newRecognizer_8859_2_cs(),
	newRecognizer_8859_2_hu(),
	newRecognizer_8859_2_pl(),
	newRecognizer_8859_2_ro(),
	newRecognizer_8859_5_ru(),
	newRecognizer_8859_6_ar(),
	newRecognizer_8859_7_el(),
	newRecognizer_8859_8_I_he(),
	newRecognizer_8859_8_he(),
	newRecognizer_windows_1251(),
	newRecognizer_windows_1256(),
	newRecognizer_KOI8_R(),
	newRecognizer_8859_9_tr(),

	newRecognizer_sjis(),
	newRecognizer_gb_18030(),
	newRecognizer_euc_jp(),
	newRecognizer_euc_kr(),
	newRecognizer_big5(),

	newRecognizer_2022JP(),
	newRecognizer_2022KR(),
	newRecognizer_2022CN(),

	newRecognizer_IBM424_he_rtl(),
	newRecognizer_IBM424_he_ltr(),
	newRecognizer_IBM420_ar_rtl(),
	newRecognizer_IBM420_ar_ltr(),
}

// NewTextDetector creates a Detector for plain text.
func NewTextDetector() *Detector {
	return &Detector{recognizers, false}
}

// NewHtmlDetector creates a Detector for Html.
func NewHtmlDetector() *Detector {
	return &Detector{recognizers, true}
}

var (
	NotDetectedError = errors.New("Charset not detected.")
)

// DetectBest returns the Result with highest Confidence.
func (d *Detector) DetectBest(b []byte) (r *Result, err error) {
	var all []Result
	if all, err = d.DetectAll(b); err == nil {
		r = &all[0]
	}
	return
}

// DetectAll returns all Results which have non-zero Confidence. The Results are sorted by Confidence in descending order.
func (d *Detector) DetectAll(b []byte) ([]Result, error) {
	input := newRecognizerInput(b, d.stripTag)
	outputChan := make(chan recognizerOutput)
	for _, r := range d.recognizers {
		go matchHelper(r, input, outputChan)
	}
	outputs := make([]recognizerOutput, 0, len(d.recognizers))
	for i := 0; i < len(d.recognizers); i++ {
		o := <-outputChan
		if o.Confidence > 0 {
			outputs = append(outputs, o)
		}
	}
	if len(outputs) == 0 {
		return nil, NotDetectedError
	}

	sort.Sort(recognizerOutputs(outputs))
	dedupOutputs := make([]Result, 0, len(outputs))
	foundCharsets := make(map[string]struct{}, len(outputs))
	for _, o := range outputs {
		if _, found := foundCharsets[o.Charset]; !found {
			dedupOutputs = append(dedupOutputs, Result(o))
			foundCharsets[o.Charset] = struct{}{}
		}
	}
	if len(dedupOutputs) == 0 {
		return nil, NotDetectedError
	}
	return dedupOutputs, nil
}

func matchHelper(r recognizer, input *recognizerInput, outputChan chan<- recognizerOutput) {
	outputChan <- r.Match(input)
}

type recognizerOutputs []recognizerOutput

func (r recognizerOutputs) Len() int           { return len(r) }
func (r recognizerOutputs) Less(i, j int) bool { return r[i].Confidence > r[j].Confidence }
func (r recognizerOutputs) Swap(i, j int)      { r[i], r[j] = r[j], r[i] }