shamoji

README

shamoji

About

The shamoji (杓文字) is word filtering package.

Install

$ go get github.com/osamingo/shamoji@latest

Usage

package main

import (
	"fmt"
	"sync"

	"github.com/osamingo/shamoji"
	"github.com/osamingo/shamoji/filter"
	"github.com/osamingo/shamoji/tokenizer"
	"golang.org/x/text/unicode/norm"
)

var (
	o sync.Once
	s *shamoji.Serve
)

func main() {
	yes, word := Contains("我が生涯に一片の悔い無し")
	fmt.Printf("Result: %v, Word: %s", yes, word)
}

func Contains(sentence string) (bool, string) {
	o.Do(func() {
		tok, err := tokenizer.NewKagomeTokenizer(norm.NFKC)
		if err != nil {
			panic(err)
		}
		s = &shamoji.Serve{
			Tokenizer: tok,
			Filer:     filter.NewCuckooFilter("涯に", "悔い"),
		}
	})
	return s.Do(sentence)
}

License

Released under the MIT License.

Documentation

Index

• type Filter
• type Serve
  • func (s *Serve) Do(sentence string) (bool, string)
  • func (s *Serve) DoAsync(ctx context.Context, sentence string) (bool, string)
• type Tokenizer

Types

type Filter

type Filter interface {
	Test(src []byte) (result bool)
}

Filter implements Test method.

type Serve

type Serve struct {
	Tokenizer Tokenizer
	Filer     Filter
}

Serve has Tokenizer and Filter interfaces.

func (*Serve) Do

func (s *Serve) Do(sentence string) (bool, string)

Do filtering sentence.

func (*Serve) DoAsync

func (s *Serve) DoAsync(ctx context.Context, sentence string) (bool, string)

DoAsync filtering sentence.

type Tokenizer

type Tokenizer interface {
	Tokenize(sentence string) (tokens [][]byte)
}

Tokenizer implements Tokenize method.