lbcd/claimtrie/normalization/case_folder.go

61 lines
1.3 KiB
Go

package normalization
import (
"bytes"
_ "embed"
"regexp"
"strconv"
"strings"
"unicode/utf8"
)
//go:embed CaseFolding_v11.txt
var v11 string
var foldMap map[rune][]rune
func init() {
foldMap = map[rune][]rune{}
r, _ := regexp.Compile(`([[:xdigit:]]+?); (.); ([[:xdigit:] ]+?);`)
matches := r.FindAllStringSubmatch(v11, 1000000000)
for i := range matches {
if matches[i][2] == "C" || matches[i][2] == "F" {
key, err := strconv.ParseUint(matches[i][1], 16, len(matches[i][1])*4)
if err != nil {
panic(err)
}
splits := strings.Split(matches[i][3], " ")
var values []rune
for j := range splits {
value, err := strconv.ParseUint(splits[j], 16, len(splits[j])*4)
if err != nil {
panic(err)
}
values = append(values, rune(value))
}
foldMap[rune(key)] = values
}
}
}
func caseFold(name []byte) []byte {
var b bytes.Buffer
b.Grow(len(name))
for i := 0; i < len(name); {
r, w := utf8.DecodeRune(name[i:])
if r == utf8.RuneError && w < 2 {
// HACK: their RuneError is actually a valid character if coming from a width of 2 or more
return name
}
replacements := foldMap[r]
if len(replacements) > 0 {
for j := range replacements {
b.WriteRune(replacements[j])
}
} else {
b.WriteRune(r)
}
i += w
}
return b.Bytes()
}