2009-01-14 15:05:00 -07:00
|
|
|
// Copyright 2009 The Go Authors. All rights reserved.
|
|
|
|
// Use of this source code is governed by a BSD-style
|
|
|
|
// license that can be found in the LICENSE file.
|
|
|
|
|
2009-03-06 04:22:02 -07:00
|
|
|
// This package provides data and functions to test some properties of Unicode code points.
|
2009-01-14 15:05:00 -07:00
|
|
|
package unicode
|
|
|
|
|
2009-08-31 14:01:25 -06:00
|
|
|
const (
|
2009-12-15 16:41:46 -07:00
|
|
|
MaxRune = 0x10FFFF // Maximum valid Unicode code point.
|
|
|
|
ReplacementChar = 0xFFFD // Represents invalid code points.
|
2009-08-31 14:01:25 -06:00
|
|
|
)
|
|
|
|
|
|
|
|
|
2009-03-06 04:22:02 -07:00
|
|
|
// The representation of a range of Unicode code points. The range runs from Lo to Hi
|
|
|
|
// inclusive and has the specified stride.
|
2009-01-20 15:40:40 -07:00
|
|
|
type Range struct {
|
2009-12-15 16:41:46 -07:00
|
|
|
Lo int
|
|
|
|
Hi int
|
|
|
|
Stride int
|
2009-01-14 15:05:00 -07:00
|
|
|
}
|
|
|
|
|
2009-08-29 00:05:16 -06:00
|
|
|
// The representation of a range of Unicode code points for case conversion.
|
|
|
|
// The range runs from Lo to Hi inclusive, with a fixed stride of 1. Deltas
|
|
|
|
// are the number to add to the code point to reach the code point for a
|
|
|
|
// different case for that character. They may be negative. If zero, it
|
2009-08-30 15:02:42 -06:00
|
|
|
// means the character is in the corresponding case. There is a special
|
|
|
|
// case representing sequences of alternating corresponding Upper and Lower
|
2009-08-30 19:17:52 -06:00
|
|
|
// pairs. It appears with a fixed Delta of
|
|
|
|
// {UpperLower, UpperLower, UpperLower}
|
|
|
|
// The constant UpperLower has an otherwise impossible delta value.
|
2009-08-29 00:05:16 -06:00
|
|
|
type CaseRange struct {
|
2009-12-15 16:41:46 -07:00
|
|
|
Lo int
|
|
|
|
Hi int
|
|
|
|
Delta d
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
// Indices into the Delta arrays inside CaseRanges for case mapping.
|
|
|
|
const (
|
2009-12-15 16:41:46 -07:00
|
|
|
UpperCase = iota
|
|
|
|
LowerCase
|
|
|
|
TitleCase
|
|
|
|
MaxCase
|
2009-08-29 00:05:16 -06:00
|
|
|
)
|
2009-10-07 12:55:06 -06:00
|
|
|
|
2009-12-15 16:41:46 -07:00
|
|
|
type d [MaxCase]int32 // to make the CaseRanges text shorter
|
2009-08-29 00:05:16 -06:00
|
|
|
|
|
|
|
// If the Delta field of a CaseRange is UpperLower or LowerUpper, it means
|
|
|
|
// this CaseRange represents a sequence of the form (say)
|
|
|
|
// Upper Lower Upper Lower.
|
|
|
|
const (
|
2009-12-15 16:41:46 -07:00
|
|
|
UpperLower = MaxRune + 1 // (Cannot be a valid delta.)
|
2009-08-29 00:05:16 -06:00
|
|
|
)
|
|
|
|
|
2009-03-06 04:22:02 -07:00
|
|
|
// Is tests whether rune is in the specified table of ranges.
|
2009-01-20 15:40:40 -07:00
|
|
|
func Is(ranges []Range, rune int) bool {
|
2009-01-14 15:05:00 -07:00
|
|
|
// common case: rune is ASCII or Latin-1
|
|
|
|
if rune < 0x100 {
|
2009-09-15 10:41:59 -06:00
|
|
|
for _, r := range ranges {
|
2009-03-06 04:22:02 -07:00
|
|
|
if rune > r.Hi {
|
2009-11-09 13:07:39 -07:00
|
|
|
continue
|
2009-01-14 15:05:00 -07:00
|
|
|
}
|
2009-03-06 04:22:02 -07:00
|
|
|
if rune < r.Lo {
|
2009-11-09 13:07:39 -07:00
|
|
|
return false
|
2009-01-14 15:05:00 -07:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return (rune-r.Lo)%r.Stride == 0
|
2009-01-14 15:05:00 -07:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return false
|
2009-01-14 15:05:00 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
// binary search over ranges
|
2009-12-15 16:41:46 -07:00
|
|
|
lo := 0
|
|
|
|
hi := len(ranges)
|
2009-01-14 15:05:00 -07:00
|
|
|
for lo < hi {
|
2009-12-15 16:41:46 -07:00
|
|
|
m := lo + (hi-lo)/2
|
|
|
|
r := ranges[m]
|
2009-03-06 04:22:02 -07:00
|
|
|
if r.Lo <= rune && rune <= r.Hi {
|
2009-11-09 22:23:52 -07:00
|
|
|
return (rune-r.Lo)%r.Stride == 0
|
2009-01-14 15:05:00 -07:00
|
|
|
}
|
2009-03-06 04:22:02 -07:00
|
|
|
if rune < r.Lo {
|
2009-11-09 13:07:39 -07:00
|
|
|
hi = m
|
2009-01-14 15:05:00 -07:00
|
|
|
} else {
|
2009-11-09 22:23:52 -07:00
|
|
|
lo = m + 1
|
2009-01-14 15:05:00 -07:00
|
|
|
}
|
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return false
|
2009-01-14 15:05:00 -07:00
|
|
|
}
|
|
|
|
|
2009-08-26 17:01:31 -06:00
|
|
|
// IsUpper reports whether the rune is an upper case letter.
|
2009-01-20 15:40:40 -07:00
|
|
|
func IsUpper(rune int) bool {
|
2009-12-15 16:41:46 -07:00
|
|
|
if rune < 0x80 { // quick ASCII check
|
2009-11-09 13:07:39 -07:00
|
|
|
return 'A' <= rune && rune <= 'Z'
|
2009-08-26 17:53:07 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return Is(Upper, rune)
|
2009-01-14 15:05:00 -07:00
|
|
|
}
|
|
|
|
|
2009-08-26 17:01:31 -06:00
|
|
|
// IsLower reports whether the rune is a lower case letter.
|
|
|
|
func IsLower(rune int) bool {
|
2009-12-15 16:41:46 -07:00
|
|
|
if rune < 0x80 { // quick ASCII check
|
2009-11-09 13:07:39 -07:00
|
|
|
return 'a' <= rune && rune <= 'z'
|
2009-08-26 17:53:07 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return Is(Lower, rune)
|
2009-08-26 17:01:31 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
// IsTitle reports whether the rune is a title case letter.
|
|
|
|
func IsTitle(rune int) bool {
|
2009-12-15 16:41:46 -07:00
|
|
|
if rune < 0x80 { // quick ASCII check
|
2009-11-09 13:07:39 -07:00
|
|
|
return false
|
2009-08-26 17:53:07 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return Is(Title, rune)
|
2009-08-26 17:01:31 -06:00
|
|
|
}
|
|
|
|
|
2009-03-06 04:22:02 -07:00
|
|
|
// IsLetter reports whether the rune is a letter.
|
2009-01-20 15:40:40 -07:00
|
|
|
func IsLetter(rune int) bool {
|
2009-12-15 16:41:46 -07:00
|
|
|
if rune < 0x80 { // quick ASCII check
|
|
|
|
rune &^= 'a' - 'A'
|
|
|
|
return 'A' <= rune && rune <= 'Z'
|
2009-08-26 17:53:07 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return Is(Letter, rune)
|
2009-01-14 15:05:00 -07:00
|
|
|
}
|
2009-08-29 00:05:16 -06:00
|
|
|
|
2009-08-31 19:12:40 -06:00
|
|
|
// IsSpace reports whether the rune is a white space character.
|
|
|
|
func IsSpace(rune int) bool {
|
2009-12-15 16:41:46 -07:00
|
|
|
if rune <= 0xFF { // quick Latin-1 check
|
2009-08-31 19:12:40 -06:00
|
|
|
switch rune {
|
|
|
|
case '\t', '\n', '\v', '\f', '\r', ' ', 0x85, 0xA0:
|
2009-11-09 13:07:39 -07:00
|
|
|
return true
|
2009-08-31 19:12:40 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return false
|
2009-08-31 19:12:40 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return Is(White_Space, rune)
|
2009-08-31 19:12:40 -06:00
|
|
|
}
|
|
|
|
|
2009-08-31 14:01:25 -06:00
|
|
|
// To maps the rune to the specified case: UpperCase, LowerCase, or TitleCase
|
2009-08-29 00:05:16 -06:00
|
|
|
func To(_case int, rune int) int {
|
|
|
|
if _case < 0 || MaxCase <= _case {
|
2009-12-15 16:41:46 -07:00
|
|
|
return ReplacementChar // as reasonable an error as any
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
|
|
|
// binary search over ranges
|
2009-12-15 16:41:46 -07:00
|
|
|
lo := 0
|
|
|
|
hi := len(CaseRanges)
|
2009-08-29 00:05:16 -06:00
|
|
|
for lo < hi {
|
2009-12-15 16:41:46 -07:00
|
|
|
m := lo + (hi-lo)/2
|
|
|
|
r := CaseRanges[m]
|
2009-08-29 00:05:16 -06:00
|
|
|
if r.Lo <= rune && rune <= r.Hi {
|
2009-12-15 16:41:46 -07:00
|
|
|
delta := int(r.Delta[_case])
|
2009-08-31 14:01:25 -06:00
|
|
|
if delta > MaxRune {
|
2009-08-30 19:17:52 -06:00
|
|
|
// In an Upper-Lower sequence, which always starts with
|
|
|
|
// an UpperCase letter, the real deltas always look like:
|
|
|
|
// {0, 1, 0} UpperCase (Lower is next)
|
|
|
|
// {-1, 0, -1} LowerCase (Upper, Title are previous)
|
|
|
|
// The characters at even offsets from the beginning of the
|
|
|
|
// sequence are upper case; the ones at odd offsets are lower.
|
|
|
|
// The correct mapping can be done by clearing or setting the low
|
2009-09-15 10:41:59 -06:00
|
|
|
// bit in the sequence offset.
|
2009-08-30 19:17:52 -06:00
|
|
|
// The constants UpperCase and TitleCase are even while LowerCase
|
|
|
|
// is odd so we take the low bit from _case.
|
2009-11-09 22:23:52 -07:00
|
|
|
return r.Lo + ((rune-r.Lo)&^1 | _case&1)
|
2009-08-30 15:02:42 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return rune + delta
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
|
|
|
if rune < r.Lo {
|
2009-11-09 13:07:39 -07:00
|
|
|
hi = m
|
2009-08-29 00:05:16 -06:00
|
|
|
} else {
|
2009-11-09 22:23:52 -07:00
|
|
|
lo = m + 1
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return rune
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
// ToUpper maps the rune to upper case
|
|
|
|
func ToUpper(rune int) int {
|
2009-12-15 16:41:46 -07:00
|
|
|
if rune < 0x80 { // quick ASCII check
|
2009-08-29 00:05:16 -06:00
|
|
|
if 'a' <= rune && rune <= 'z' {
|
2009-11-09 22:23:52 -07:00
|
|
|
rune -= 'a' - 'A'
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return rune
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return To(UpperCase, rune)
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
// ToLower maps the rune to lower case
|
|
|
|
func ToLower(rune int) int {
|
2009-12-15 16:41:46 -07:00
|
|
|
if rune < 0x80 { // quick ASCII check
|
2009-08-29 00:05:16 -06:00
|
|
|
if 'A' <= rune && rune <= 'Z' {
|
2009-11-09 22:23:52 -07:00
|
|
|
rune += 'a' - 'A'
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return rune
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return To(LowerCase, rune)
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
// ToTitle maps the rune to title case
|
|
|
|
func ToTitle(rune int) int {
|
2009-12-15 16:41:46 -07:00
|
|
|
if rune < 0x80 { // quick ASCII check
|
|
|
|
if 'a' <= rune && rune <= 'z' { // title case is upper case for ASCII
|
2009-11-09 22:23:52 -07:00
|
|
|
rune -= 'a' - 'A'
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return rune
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|
2009-12-15 16:41:46 -07:00
|
|
|
return To(TitleCase, rune)
|
2009-08-29 00:05:16 -06:00
|
|
|
}
|