blob: 99ead42988185a78e479631574f54a3cbd30208f [file] [log] [blame]
// Copyright 2015 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.
// Unicode table generator.
// Data read from the web.
//go:build ignore
// +build ignore
package main
import (
var outputFile = flag.String("output", "tables.go", "output file for generated tables; default tables.go")
var assigned, disallowedRunes *unicode.RangeTable
var runeCategory = map[rune]category{}
var overrides = map[category]category{
viramaModifier: viramaJoinT,
greek: greekJoinT,
hebrew: hebrewJoinT,
func setCategory(r rune, cat category) {
if c, ok := runeCategory[r]; ok {
if override, ok := overrides[c]; cat == joiningT && ok {
cat = override
} else {
log.Fatalf("%U: multiple categories for rune (%v and %v)", r, c, cat)
runeCategory[r] = cat
func init() {
if numCategories > 1<<propShift {
log.Fatalf("Number of categories is %d; may at most be %d", numCategories, 1<<propShift)
func main() {
// Load data
runes := []rune{}
// PrecisIgnorableProperties:
ucd.Parse(gen.OpenUCDFile("DerivedCoreProperties.txt"), func(p *ucd.Parser) {
if p.String(1) == "Default_Ignorable_Code_Point" {
runes = append(runes, p.Rune(0))
ucd.Parse(gen.OpenUCDFile("PropList.txt"), func(p *ucd.Parser) {
switch p.String(1) {
case "Noncharacter_Code_Point":
runes = append(runes, p.Rune(0))
// OldHangulJamo:
ucd.Parse(gen.OpenUCDFile("HangulSyllableType.txt"), func(p *ucd.Parser) {
switch p.String(1) {
case "L", "V", "T":
runes = append(runes, p.Rune(0))
disallowedRunes = rangetable.New(runes...)
assigned = rangetable.Assigned(unicode.Version)
// Load category data.
runeCategory['l'] = latinSmallL
ucd.Parse(gen.OpenUCDFile("UnicodeData.txt"), func(p *ucd.Parser) {
const cccVirama = 9
if p.Int(ucd.CanonicalCombiningClass) == cccVirama {
setCategory(p.Rune(0), viramaModifier)
ucd.Parse(gen.OpenUCDFile("Scripts.txt"), func(p *ucd.Parser) {
switch p.String(1) {
case "Greek":
setCategory(p.Rune(0), greek)
case "Hebrew":
setCategory(p.Rune(0), hebrew)
case "Hiragana", "Katakana", "Han":
setCategory(p.Rune(0), japanese)
// Set the rule categories associated with exceptions. This overrides any
// previously set categories. The original categories are manually
// reintroduced in the categoryTransitions table.
for r, e := range exceptions {
if != 0 {
runeCategory[r] =
cat := map[string]category{
"L": joiningL,
"D": joiningD,
"T": joiningT,
"R": joiningR,
ucd.Parse(gen.OpenUCDFile("extracted/DerivedJoiningType.txt"), func(p *ucd.Parser) {
switch v := p.String(1); v {
case "L", "D", "T", "R":
setCategory(p.Rune(0), cat[v])
gen.Repackage("gen_trieval.go", "trieval.go", "precis")
type exception struct {
prop property
cat category
func init() {
// Programmatically add the Arabic and Indic digits to the exceptions map.
// See comment in the exceptions map below why these are marked disallowed.
for i := rune(0); i <= 9; i++ {
exceptions[0x0660+i] = exception{
prop: disallowed,
cat: arabicIndicDigit,
exceptions[0x06F0+i] = exception{
prop: disallowed,
cat: extendedArabicIndicDigit,
// The Exceptions class as defined in RFC 5892
var exceptions = map[rune]exception{
0x00DF: {prop: pValid},
0x03C2: {prop: pValid},
0x06FD: {prop: pValid},
0x06FE: {prop: pValid},
0x0F0B: {prop: pValid},
0x3007: {prop: pValid},
// ContextO|J rules are marked as disallowed, taking a "guilty until proven
// innocent" approach. The main reason for this is that the check for
// whether a context rule should be applied can be moved to the logic for
// handing disallowed runes, taken it off the common path. The exception to
// this rule is for katakanaMiddleDot, as the rule logic is handled without
// using a rule function.
// ContextJ (Join control)
0x200C: {prop: disallowed, cat: zeroWidthNonJoiner},
0x200D: {prop: disallowed, cat: zeroWidthJoiner},
// ContextO
0x00B7: {prop: disallowed, cat: middleDot},
0x0375: {prop: disallowed, cat: greekLowerNumeralSign},
0x05F3: {prop: disallowed, cat: hebrewPreceding}, // punctuation Geresh
0x05F4: {prop: disallowed, cat: hebrewPreceding}, // punctuation Gershayim
0x30FB: {prop: pValid, cat: katakanaMiddleDot},
// These are officially ContextO, but the implementation does not require
// special treatment of these, so we simply mark them as valid.
0x0660: {prop: pValid},
0x0661: {prop: pValid},
0x0662: {prop: pValid},
0x0663: {prop: pValid},
0x0664: {prop: pValid},
0x0665: {prop: pValid},
0x0666: {prop: pValid},
0x0667: {prop: pValid},
0x0668: {prop: pValid},
0x0669: {prop: pValid},
0x06F0: {prop: pValid},
0x06F1: {prop: pValid},
0x06F2: {prop: pValid},
0x06F3: {prop: pValid},
0x06F4: {prop: pValid},
0x06F5: {prop: pValid},
0x06F6: {prop: pValid},
0x06F7: {prop: pValid},
0x06F8: {prop: pValid},
0x06F9: {prop: pValid},
0x0640: {prop: disallowed},
0x07FA: {prop: disallowed},
0x302E: {prop: disallowed},
0x302F: {prop: disallowed},
0x3031: {prop: disallowed},
0x3032: {prop: disallowed},
0x3033: {prop: disallowed},
0x3034: {prop: disallowed},
0x3035: {prop: disallowed},
0x303B: {prop: disallowed},
// LetterDigits:
// r in {Ll, Lu, Lo, Nd, Lm, Mn, Mc}.
func isLetterDigits(r rune) bool {
return unicode.In(r,
unicode.Ll, unicode.Lu, unicode.Lm, unicode.Lo, // Letters
unicode.Mn, unicode.Mc, // Modifiers
unicode.Nd, // Digits
func isIdDisAndFreePVal(r rune) bool {
return unicode.In(r,
// OtherLetterDigits:
// r in in {Lt, Nl, No, Me}
unicode.Lt, unicode.Nl, unicode.No, // Other letters / numbers
unicode.Me, // Modifiers
// Spaces:
// r in in {Zs}
// Symbols:
// r in {Sm, Sc, Sk, So}
unicode.Sm, unicode.Sc, unicode.Sk, unicode.So,
// Punctuation:
// r in {Pc, Pd, Ps, Pe, Pi, Pf, Po}
unicode.Pc, unicode.Pd, unicode.Ps, unicode.Pe,
unicode.Pi, unicode.Pf, unicode.Po,
// HasCompat:
func hasCompat(r rune) bool {
return !norm.NFKC.IsNormalString(string(r))
// From
// If .cp. .in. Exceptions Then Exceptions(cp);
// Else If .cp. .in. BackwardCompatible Then BackwardCompatible(cp);
// Else If .cp. .in. Unassigned Then UNASSIGNED;
// Else If .cp. .in. ASCII7 Then PVALID;
// Else If .cp. .in. JoinControl Then CONTEXTJ;
// Else If .cp. .in. OldHangulJamo Then DISALLOWED;
// Else If .cp. .in. PrecisIgnorableProperties Then DISALLOWED;
// Else If .cp. .in. Controls Then DISALLOWED;
// Else If .cp. .in. HasCompat Then ID_DIS or FREE_PVAL;
// Else If .cp. .in. LetterDigits Then PVALID;
// Else If .cp. .in. OtherLetterDigits Then ID_DIS or FREE_PVAL;
// Else If .cp. .in. Spaces Then ID_DIS or FREE_PVAL;
// Else If .cp. .in. Symbols Then ID_DIS or FREE_PVAL;
// Else If .cp. .in. Punctuation Then ID_DIS or FREE_PVAL;
func writeTables() {
propTrie := triegen.NewTrie("derivedProperties")
w := gen.NewCodeWriter()
defer w.WriteVersionedGoFile(*outputFile, "precis")
// Iterate over all the runes...
for i := rune(0); i < unicode.MaxRune; i++ {
r := rune(i)
if !utf8.ValidRune(r) {
e, ok := exceptions[i]
p := e.prop
switch {
case ok:
case !unicode.In(r, assigned):
p = unassigned
case r >= 0x0021 && r <= 0x007e: // Is ASCII 7
p = pValid
case unicode.In(r, disallowedRunes, unicode.Cc):
p = disallowed
case hasCompat(r):
p = idDisOrFreePVal
case isLetterDigits(r):
p = pValid
case isIdDisAndFreePVal(r):
p = idDisOrFreePVal
p = disallowed
cat := runeCategory[r]
// Don't set category for runes that are disallowed.
if p == disallowed {
cat = exceptions[r].cat
propTrie.Insert(r, uint64(p)|uint64(cat))
sz, err := propTrie.Gen(w)
if err != nil {
w.Size += sz