Golang transform.RemoveFunc函數代碼示例

func NormalizeTitle(title string) string {
	normalizedTitle := title
	normalizedTitle = strings.ToLower(normalizedTitle)
	normalizedTitle = RomanizeHepburn(title)
	normalizedTitle = strings.ToLower(normalizedTitle)
	normalizedTitle = RemoveTrailingApostrophe(normalizedTitle)
	normalizedTitle, _, _ = transform.String(transform.Chain(
		norm.NFD,
		transform.RemoveFunc(func(r rune) bool {
			return unicode.Is(unicode.Mn, r)
		}),
		norm.NFC), normalizedTitle)
	normalizedTitle = strings.ToLower(normalizedTitle)
	normalizedTitle = regexp.MustCompile(`\(\d+\)`).ReplaceAllString(normalizedTitle, " ")
	normalizedTitle = strings.Map(func(r rune) rune {
		if !unicode.IsLetter(r) && !unicode.IsDigit(r) && r != '.' {
			return ' '
		}
		return r
	}, normalizedTitle)
	normalizedTitle = regexp.MustCompile(`\s+`).ReplaceAllString(normalizedTitle, " ")
	normalizedTitle = strings.TrimSpace(normalizedTitle)

	return normalizedTitle
}

func init() {
	words = make(map[string]int)
	count = 0

	amz = "./amazon_cells_labelled.txt"
	ylp = "./yelp_labelled.txt"
	imd = "./imdb_labelled.txt"

	sanitize = transform.RemoveFunc(func(r rune) bool {
		switch {
		case r >= 'A' && r <= 'Z':
			return false
		case r >= 'a' && r <= 'z':
			return false
		case r >= '0' && r <= '1':
			return false
		case r == ' ':
			return false
		case r == '\t':
			return false
		default:
			return true
		}
	})

	rand.Seed(42)
}

func main() {
	t := transform.Chain(norm.NFD, transform.RemoveFunc(isMn), norm.NFC)
	r := transform.NewReader(os.Stdin, t)
	if _, err := io.Copy(os.Stdout, r); err != nil {
		log.Fatal(err)
	}
}

func removeNlChars(str string) string {
	isOk := func(r rune) bool {
		return r < 32 || r >= 127
	}
	t := transform.Chain(norm.NFKD, transform.RemoveFunc(isOk))
	str, _, _ = transform.String(t, str)
	return str
}

func TestLettersShouldPass1(t *testing.T) {
	s, _, _ := transform.String(transform.RemoveFunc(OnlyLetters), "THIS iz A L337 aNd Un'Sani~~~~tized sentence")
	sanitized := []rune(s)

	for i := range sanitized {
		assert.False(t, OnlyLetters(sanitized[i]), "Letter %v should be sanitized", sanitized[i])
	}
}

// NewNaiveBayes returns a NaiveBayes model the
// given number of classes instantiated, ready
// to learn off the given data stream. The sanitization
// function is set to the given function. It must
// comply with the transform.RemoveFunc interface
func NewNaiveBayes(stream <-chan base.TextDatapoint, classes uint8, sanitize func(rune) bool) *NaiveBayes {
	return &NaiveBayes{
		Words:         make(map[string]Word),
		Count:         make([]uint64, classes),
		Probabilities: make([]float64, classes),

		sanitize: transform.RemoveFunc(sanitize),
		stream:   stream,
	}
}

// RestoreWithFuncs takes raw JSON data of a model and
// restores a model from it. The tokenizer and sanitizer
// passed in will be assigned to the restored model.
func (b *NaiveBayes) RestoreWithFuncs(data io.Reader, sanitizer func(rune) bool, tokenizer Tokenizer) error {
	if b == nil {
		return errors.New("Cannot restore a model to a nil pointer")
	}
	err := json.NewDecoder(data).Decode(b)
	if err != nil {
		return err
	}
	b.sanitize = transform.RemoveFunc(sanitizer)
	b.Tokenizer = tokenizer
	return nil
}

//function to sanitize input
//from: http://rosettacode.org/wiki/Strip_control_codes_and_extended_characters_from_a_string#Go
func stripCtlAndExtFromUnicode(str string) string {
	isOk := func(r rune) bool {
		return r < 32 || r >= 127
	}
	// The isOk filter is such that there is no need to chain to norm.NFC
	t := transform.Chain(norm.NFKD, transform.RemoveFunc(isOk))
	// This Transformer could also trivially be applied as an io.Reader
	// or io.Writer filter to automatically do such filtering when reading
	// or writing data anywhere.
	str, _, _ = transform.String(t, str)
	return str
}

// NewNaiveBayes returns a NaiveBayes model the
// given number of classes instantiated, ready
// to learn off the given data stream. The sanitization
// function is set to the given function. It must
// comply with the transform.RemoveFunc interface
func NewNaiveBayes(stream <-chan base.TextDatapoint, classes uint8, sanitize func(rune) bool) *NaiveBayes {
	return &NaiveBayes{
		Words:         concurrentMap{sync.RWMutex{}, make(map[string]Word)},
		Count:         make([]uint64, classes),
		Probabilities: make([]float64, classes),

		sanitize:  transform.RemoveFunc(sanitize),
		stream:    stream,
		Tokenizer: &SimpleTokenizer{SplitOn: " "},

		Output: os.Stdout,
	}
}

func TestAsciiLetters(t *testing.T) {
	tests := []testCase{
		{"THIS iz A L337 aNd Un'Sani~~~~tized sentence", "THISizALaNdUnSanitizedsentence"},
		{"here're some unicode letters: --Æ.ÒÑ", "hereresomeunicodeletters"},
		{")(*&^%[email protected][email protected]#$%^&*(*&^%$#@#$%", ""},
	}
	for _, test := range tests {
		s, _, _ := transform.String(transform.RemoveFunc(OnlyAsciiLetters), test.input)
		if s != test.expectedOutput {
			t.Errorf("got \"%s\" expected \"%s\"\n", s, test.expectedOutput)
		}
	}
}

// GetCompatibleString removes all the special characters
// from the string name to create a new string compatible
// with different file names.
func GetCompatibleString(name string) string {
	// Replace all the & signs with and text
	name = strings.Replace(name, "&", "and", -1)
	// Change all the characters to ASCII
	t := transform.Chain(norm.NFD, transform.RemoveFunc(isMn), norm.NFC)
	result, _, _ := transform.String(t, name)
	// Replace all the spaces with underscore
	s, _ := regexp.Compile(`\s+`)
	result = s.ReplaceAllString(result, "_")
	// Remove all the non alphanumeric characters
	r, _ := regexp.Compile(`\W`)
	result = r.ReplaceAllString(result, "")
	return result
}

func ExampleRemoveFunc() {
	input := []byte(`tschüß; до свидания`)

	b := make([]byte, len(input))

	t := transform.RemoveFunc(unicode.IsSpace)
	n, _, _ := t.Transform(b, input, true)
	fmt.Println(string(b[:n]))

	t = transform.RemoveFunc(func(r rune) bool {
		return !unicode.Is(unicode.Latin, r)
	})
	n, _, _ = t.Transform(b, input, true)
	fmt.Println(string(b[:n]))

	n, _, _ = t.Transform(b, norm.NFD.Bytes(input), true)
	fmt.Println(string(b[:n]))

	// Output:
	// tschüß;досвидания
	// tschüß
	// tschuß
}

func normalize(name, src string) (string, error) {
	if name == "" {
		name = baseWithoutExt(src)
	}
	t := transform.Chain(norm.NFD, transform.RemoveFunc(remove), norm.NFC)
	name = strings.TrimSpace(name)
	name, _, err := transform.String(t, name)
	if err != nil {
		return "", err
	}
	name = strings.ToLower(name)
	name = strings.Replace(name, " ", "_", -1)
	return name, nil
}

// normalize does unicode normalization.
func normalize(in []byte) ([]byte, error) {
	// We need a new transformer for each input as it cannot be reused.
	filter := func(r rune) bool {
		return unicode.Is(unicode.Mn, r) // Mn: nonspacing marks (to be removed)
	}
	transformer := transform.Chain(norm.NFD, transform.RemoveFunc(filter), norm.NFC)
	out, _, err := transform.Bytes(transformer, in)
	out = bytes.Map(func(r rune) rune {
		if unicode.IsPunct(r) { // Replace punctuations with spaces.
			return ' '
		}
		return unicode.ToLower(r) // Convert to lower case.
	}, out)
	return out, err
}

func replace(path string) {
	copy := []string{}
	r := `(<script(\s|\S)*?<\/script>)|(<style(\s|\S)*?<\/style>)|(<!--(\s|\S)*?-->)|(<\/?(\s|\S)*?>)|(nbsp;)|((?:\s)\s)|(png)|(jpeg)|(jpg)|(mpg)|(\\u0026)|(\n)|(\v)|(\r)|(\0)|(\t)|(n°)
		|(à)|(wbe)|(_)`
	regex, err := regexp.Compile(r)
	if err != nil {
		return // there was a problem with the regular expression.
	}
	c, _ := readLines(path)
	for _, v := range c {
		reg := regex.ReplaceAllString(v, " ")
		slug := utils.GenerateSlug(reg)
		regex1, _ := regexp.Compile(`((\-){1,})|(\b\w{1}\b)`)
		reg = regex1.ReplaceAllString(slug, " ")
		t := stripchars(reg, `?,.!/©*@#~()$+"'&}]|:;[{²`)
		s := strings.TrimSpace(t)
		// fmt.Println(s)

		normalize := transform.Chain(norm.NFD, transform.RemoveFunc(isMn), norm.NFC)
		normStr1, _, _ := transform.String(normalize, s)
		// fmt.Println(normStr1)

		if len(v) > 0 {
			copy = append(copy, normStr1)
		}
	}

	// fmt.Println(cleaned, "\n")

	j := strings.Replace(strings.Join((copy), " "), " ", ",", -1)
	// fmt.Println(j)
	regex2, err := regexp.Compile(`((\,){2,})`)
	j1 := regex2.ReplaceAllString(j, ",")
	// fmt.Println(j1)
	j2 := strings.Split(j1, ",")

	cleaned := []string{}

	for _, value := range j2 {
		if !stringInSlice(value, cleaned) {
			cleaned = append(cleaned, value)
		}
	}
	createCsv(path, filenameCsv, strings.Join(cleaned, ","))
}