Abhilash S.L | 3b49463 | 2019-07-16 15:51:09 +0530 | [diff] [blame] | 1 | // Copyright 2013 The Go Authors. All rights reserved. |
| 2 | // Use of this source code is governed by a BSD-style |
| 3 | // license that can be found in the LICENSE file. |
| 4 | |
| 5 | //go:generate go run gen.go gen_common.go -output tables.go |
| 6 | |
| 7 | package language // import "golang.org/x/text/internal/language" |
| 8 | |
| 9 | // TODO: Remove above NOTE after: |
| 10 | // - verifying that tables are dropped correctly (most notably matcher tables). |
| 11 | |
| 12 | import ( |
| 13 | "errors" |
| 14 | "fmt" |
| 15 | "strings" |
| 16 | ) |
| 17 | |
| 18 | const ( |
| 19 | // maxCoreSize is the maximum size of a BCP 47 tag without variants and |
| 20 | // extensions. Equals max lang (3) + script (4) + max reg (3) + 2 dashes. |
| 21 | maxCoreSize = 12 |
| 22 | |
| 23 | // max99thPercentileSize is a somewhat arbitrary buffer size that presumably |
| 24 | // is large enough to hold at least 99% of the BCP 47 tags. |
| 25 | max99thPercentileSize = 32 |
| 26 | |
| 27 | // maxSimpleUExtensionSize is the maximum size of a -u extension with one |
| 28 | // key-type pair. Equals len("-u-") + key (2) + dash + max value (8). |
| 29 | maxSimpleUExtensionSize = 14 |
| 30 | ) |
| 31 | |
| 32 | // Tag represents a BCP 47 language tag. It is used to specify an instance of a |
| 33 | // specific language or locale. All language tag values are guaranteed to be |
| 34 | // well-formed. The zero value of Tag is Und. |
| 35 | type Tag struct { |
| 36 | // TODO: the following fields have the form TagTypeID. This name is chosen |
| 37 | // to allow refactoring the public package without conflicting with its |
| 38 | // Base, Script, and Region methods. Once the transition is fully completed |
| 39 | // the ID can be stripped from the name. |
| 40 | |
| 41 | LangID Language |
| 42 | RegionID Region |
| 43 | // TODO: we will soon run out of positions for ScriptID. Idea: instead of |
| 44 | // storing lang, region, and ScriptID codes, store only the compact index and |
| 45 | // have a lookup table from this code to its expansion. This greatly speeds |
| 46 | // up table lookup, speed up common variant cases. |
| 47 | // This will also immediately free up 3 extra bytes. Also, the pVariant |
| 48 | // field can now be moved to the lookup table, as the compact index uniquely |
| 49 | // determines the offset of a possible variant. |
| 50 | ScriptID Script |
| 51 | pVariant byte // offset in str, includes preceding '-' |
| 52 | pExt uint16 // offset of first extension, includes preceding '-' |
| 53 | |
| 54 | // str is the string representation of the Tag. It will only be used if the |
| 55 | // tag has variants or extensions. |
| 56 | str string |
| 57 | } |
| 58 | |
| 59 | // Make is a convenience wrapper for Parse that omits the error. |
| 60 | // In case of an error, a sensible default is returned. |
| 61 | func Make(s string) Tag { |
| 62 | t, _ := Parse(s) |
| 63 | return t |
| 64 | } |
| 65 | |
| 66 | // Raw returns the raw base language, script and region, without making an |
| 67 | // attempt to infer their values. |
| 68 | // TODO: consider removing |
| 69 | func (t Tag) Raw() (b Language, s Script, r Region) { |
| 70 | return t.LangID, t.ScriptID, t.RegionID |
| 71 | } |
| 72 | |
| 73 | // equalTags compares language, script and region subtags only. |
| 74 | func (t Tag) equalTags(a Tag) bool { |
| 75 | return t.LangID == a.LangID && t.ScriptID == a.ScriptID && t.RegionID == a.RegionID |
| 76 | } |
| 77 | |
| 78 | // IsRoot returns true if t is equal to language "und". |
| 79 | func (t Tag) IsRoot() bool { |
| 80 | if int(t.pVariant) < len(t.str) { |
| 81 | return false |
| 82 | } |
| 83 | return t.equalTags(Und) |
| 84 | } |
| 85 | |
| 86 | // IsPrivateUse reports whether the Tag consists solely of an IsPrivateUse use |
| 87 | // tag. |
| 88 | func (t Tag) IsPrivateUse() bool { |
| 89 | return t.str != "" && t.pVariant == 0 |
| 90 | } |
| 91 | |
| 92 | // RemakeString is used to update t.str in case lang, script or region changed. |
| 93 | // It is assumed that pExt and pVariant still point to the start of the |
| 94 | // respective parts. |
| 95 | func (t *Tag) RemakeString() { |
| 96 | if t.str == "" { |
| 97 | return |
| 98 | } |
| 99 | extra := t.str[t.pVariant:] |
| 100 | if t.pVariant > 0 { |
| 101 | extra = extra[1:] |
| 102 | } |
| 103 | if t.equalTags(Und) && strings.HasPrefix(extra, "x-") { |
| 104 | t.str = extra |
| 105 | t.pVariant = 0 |
| 106 | t.pExt = 0 |
| 107 | return |
| 108 | } |
| 109 | var buf [max99thPercentileSize]byte // avoid extra memory allocation in most cases. |
| 110 | b := buf[:t.genCoreBytes(buf[:])] |
| 111 | if extra != "" { |
| 112 | diff := len(b) - int(t.pVariant) |
| 113 | b = append(b, '-') |
| 114 | b = append(b, extra...) |
| 115 | t.pVariant = uint8(int(t.pVariant) + diff) |
| 116 | t.pExt = uint16(int(t.pExt) + diff) |
| 117 | } else { |
| 118 | t.pVariant = uint8(len(b)) |
| 119 | t.pExt = uint16(len(b)) |
| 120 | } |
| 121 | t.str = string(b) |
| 122 | } |
| 123 | |
| 124 | // genCoreBytes writes a string for the base languages, script and region tags |
| 125 | // to the given buffer and returns the number of bytes written. It will never |
| 126 | // write more than maxCoreSize bytes. |
| 127 | func (t *Tag) genCoreBytes(buf []byte) int { |
| 128 | n := t.LangID.StringToBuf(buf[:]) |
| 129 | if t.ScriptID != 0 { |
| 130 | n += copy(buf[n:], "-") |
| 131 | n += copy(buf[n:], t.ScriptID.String()) |
| 132 | } |
| 133 | if t.RegionID != 0 { |
| 134 | n += copy(buf[n:], "-") |
| 135 | n += copy(buf[n:], t.RegionID.String()) |
| 136 | } |
| 137 | return n |
| 138 | } |
| 139 | |
| 140 | // String returns the canonical string representation of the language tag. |
| 141 | func (t Tag) String() string { |
| 142 | if t.str != "" { |
| 143 | return t.str |
| 144 | } |
| 145 | if t.ScriptID == 0 && t.RegionID == 0 { |
| 146 | return t.LangID.String() |
| 147 | } |
| 148 | buf := [maxCoreSize]byte{} |
| 149 | return string(buf[:t.genCoreBytes(buf[:])]) |
| 150 | } |
| 151 | |
| 152 | // MarshalText implements encoding.TextMarshaler. |
| 153 | func (t Tag) MarshalText() (text []byte, err error) { |
| 154 | if t.str != "" { |
| 155 | text = append(text, t.str...) |
| 156 | } else if t.ScriptID == 0 && t.RegionID == 0 { |
| 157 | text = append(text, t.LangID.String()...) |
| 158 | } else { |
| 159 | buf := [maxCoreSize]byte{} |
| 160 | text = buf[:t.genCoreBytes(buf[:])] |
| 161 | } |
| 162 | return text, nil |
| 163 | } |
| 164 | |
| 165 | // UnmarshalText implements encoding.TextUnmarshaler. |
| 166 | func (t *Tag) UnmarshalText(text []byte) error { |
| 167 | tag, err := Parse(string(text)) |
| 168 | *t = tag |
| 169 | return err |
| 170 | } |
| 171 | |
| 172 | // Variants returns the part of the tag holding all variants or the empty string |
| 173 | // if there are no variants defined. |
| 174 | func (t Tag) Variants() string { |
| 175 | if t.pVariant == 0 { |
| 176 | return "" |
| 177 | } |
| 178 | return t.str[t.pVariant:t.pExt] |
| 179 | } |
| 180 | |
| 181 | // VariantOrPrivateUseTags returns variants or private use tags. |
| 182 | func (t Tag) VariantOrPrivateUseTags() string { |
| 183 | if t.pExt > 0 { |
| 184 | return t.str[t.pVariant:t.pExt] |
| 185 | } |
| 186 | return t.str[t.pVariant:] |
| 187 | } |
| 188 | |
| 189 | // HasString reports whether this tag defines more than just the raw |
| 190 | // components. |
| 191 | func (t Tag) HasString() bool { |
| 192 | return t.str != "" |
| 193 | } |
| 194 | |
| 195 | // Parent returns the CLDR parent of t. In CLDR, missing fields in data for a |
| 196 | // specific language are substituted with fields from the parent language. |
| 197 | // The parent for a language may change for newer versions of CLDR. |
| 198 | func (t Tag) Parent() Tag { |
| 199 | if t.str != "" { |
| 200 | // Strip the variants and extensions. |
| 201 | b, s, r := t.Raw() |
| 202 | t = Tag{LangID: b, ScriptID: s, RegionID: r} |
| 203 | if t.RegionID == 0 && t.ScriptID != 0 && t.LangID != 0 { |
| 204 | base, _ := addTags(Tag{LangID: t.LangID}) |
| 205 | if base.ScriptID == t.ScriptID { |
| 206 | return Tag{LangID: t.LangID} |
| 207 | } |
| 208 | } |
| 209 | return t |
| 210 | } |
| 211 | if t.LangID != 0 { |
| 212 | if t.RegionID != 0 { |
| 213 | maxScript := t.ScriptID |
| 214 | if maxScript == 0 { |
| 215 | max, _ := addTags(t) |
| 216 | maxScript = max.ScriptID |
| 217 | } |
| 218 | |
| 219 | for i := range parents { |
| 220 | if Language(parents[i].lang) == t.LangID && Script(parents[i].maxScript) == maxScript { |
| 221 | for _, r := range parents[i].fromRegion { |
| 222 | if Region(r) == t.RegionID { |
| 223 | return Tag{ |
| 224 | LangID: t.LangID, |
| 225 | ScriptID: Script(parents[i].script), |
| 226 | RegionID: Region(parents[i].toRegion), |
| 227 | } |
| 228 | } |
| 229 | } |
| 230 | } |
| 231 | } |
| 232 | |
| 233 | // Strip the script if it is the default one. |
| 234 | base, _ := addTags(Tag{LangID: t.LangID}) |
| 235 | if base.ScriptID != maxScript { |
| 236 | return Tag{LangID: t.LangID, ScriptID: maxScript} |
| 237 | } |
| 238 | return Tag{LangID: t.LangID} |
| 239 | } else if t.ScriptID != 0 { |
| 240 | // The parent for an base-script pair with a non-default script is |
| 241 | // "und" instead of the base language. |
| 242 | base, _ := addTags(Tag{LangID: t.LangID}) |
| 243 | if base.ScriptID != t.ScriptID { |
| 244 | return Und |
| 245 | } |
| 246 | return Tag{LangID: t.LangID} |
| 247 | } |
| 248 | } |
| 249 | return Und |
| 250 | } |
| 251 | |
| 252 | // ParseExtension parses s as an extension and returns it on success. |
| 253 | func ParseExtension(s string) (ext string, err error) { |
| 254 | scan := makeScannerString(s) |
| 255 | var end int |
| 256 | if n := len(scan.token); n != 1 { |
| 257 | return "", ErrSyntax |
| 258 | } |
| 259 | scan.toLower(0, len(scan.b)) |
| 260 | end = parseExtension(&scan) |
| 261 | if end != len(s) { |
| 262 | return "", ErrSyntax |
| 263 | } |
| 264 | return string(scan.b), nil |
| 265 | } |
| 266 | |
| 267 | // HasVariants reports whether t has variants. |
| 268 | func (t Tag) HasVariants() bool { |
| 269 | return uint16(t.pVariant) < t.pExt |
| 270 | } |
| 271 | |
| 272 | // HasExtensions reports whether t has extensions. |
| 273 | func (t Tag) HasExtensions() bool { |
| 274 | return int(t.pExt) < len(t.str) |
| 275 | } |
| 276 | |
| 277 | // Extension returns the extension of type x for tag t. It will return |
| 278 | // false for ok if t does not have the requested extension. The returned |
| 279 | // extension will be invalid in this case. |
| 280 | func (t Tag) Extension(x byte) (ext string, ok bool) { |
| 281 | for i := int(t.pExt); i < len(t.str)-1; { |
| 282 | var ext string |
| 283 | i, ext = getExtension(t.str, i) |
| 284 | if ext[0] == x { |
| 285 | return ext, true |
| 286 | } |
| 287 | } |
| 288 | return "", false |
| 289 | } |
| 290 | |
| 291 | // Extensions returns all extensions of t. |
| 292 | func (t Tag) Extensions() []string { |
| 293 | e := []string{} |
| 294 | for i := int(t.pExt); i < len(t.str)-1; { |
| 295 | var ext string |
| 296 | i, ext = getExtension(t.str, i) |
| 297 | e = append(e, ext) |
| 298 | } |
| 299 | return e |
| 300 | } |
| 301 | |
| 302 | // TypeForKey returns the type associated with the given key, where key and type |
| 303 | // are of the allowed values defined for the Unicode locale extension ('u') in |
| 304 | // https://www.unicode.org/reports/tr35/#Unicode_Language_and_Locale_Identifiers. |
| 305 | // TypeForKey will traverse the inheritance chain to get the correct value. |
| 306 | func (t Tag) TypeForKey(key string) string { |
| 307 | if start, end, _ := t.findTypeForKey(key); end != start { |
| 308 | return t.str[start:end] |
| 309 | } |
| 310 | return "" |
| 311 | } |
| 312 | |
| 313 | var ( |
| 314 | errPrivateUse = errors.New("cannot set a key on a private use tag") |
| 315 | errInvalidArguments = errors.New("invalid key or type") |
| 316 | ) |
| 317 | |
| 318 | // SetTypeForKey returns a new Tag with the key set to type, where key and type |
| 319 | // are of the allowed values defined for the Unicode locale extension ('u') in |
| 320 | // https://www.unicode.org/reports/tr35/#Unicode_Language_and_Locale_Identifiers. |
| 321 | // An empty value removes an existing pair with the same key. |
| 322 | func (t Tag) SetTypeForKey(key, value string) (Tag, error) { |
| 323 | if t.IsPrivateUse() { |
| 324 | return t, errPrivateUse |
| 325 | } |
| 326 | if len(key) != 2 { |
| 327 | return t, errInvalidArguments |
| 328 | } |
| 329 | |
| 330 | // Remove the setting if value is "". |
| 331 | if value == "" { |
| 332 | start, end, _ := t.findTypeForKey(key) |
| 333 | if start != end { |
| 334 | // Remove key tag and leading '-'. |
| 335 | start -= 4 |
| 336 | |
| 337 | // Remove a possible empty extension. |
| 338 | if (end == len(t.str) || t.str[end+2] == '-') && t.str[start-2] == '-' { |
| 339 | start -= 2 |
| 340 | } |
| 341 | if start == int(t.pVariant) && end == len(t.str) { |
| 342 | t.str = "" |
| 343 | t.pVariant, t.pExt = 0, 0 |
| 344 | } else { |
| 345 | t.str = fmt.Sprintf("%s%s", t.str[:start], t.str[end:]) |
| 346 | } |
| 347 | } |
| 348 | return t, nil |
| 349 | } |
| 350 | |
| 351 | if len(value) < 3 || len(value) > 8 { |
| 352 | return t, errInvalidArguments |
| 353 | } |
| 354 | |
| 355 | var ( |
| 356 | buf [maxCoreSize + maxSimpleUExtensionSize]byte |
| 357 | uStart int // start of the -u extension. |
| 358 | ) |
| 359 | |
| 360 | // Generate the tag string if needed. |
| 361 | if t.str == "" { |
| 362 | uStart = t.genCoreBytes(buf[:]) |
| 363 | buf[uStart] = '-' |
| 364 | uStart++ |
| 365 | } |
| 366 | |
| 367 | // Create new key-type pair and parse it to verify. |
| 368 | b := buf[uStart:] |
| 369 | copy(b, "u-") |
| 370 | copy(b[2:], key) |
| 371 | b[4] = '-' |
| 372 | b = b[:5+copy(b[5:], value)] |
| 373 | scan := makeScanner(b) |
| 374 | if parseExtensions(&scan); scan.err != nil { |
| 375 | return t, scan.err |
| 376 | } |
| 377 | |
| 378 | // Assemble the replacement string. |
| 379 | if t.str == "" { |
| 380 | t.pVariant, t.pExt = byte(uStart-1), uint16(uStart-1) |
| 381 | t.str = string(buf[:uStart+len(b)]) |
| 382 | } else { |
| 383 | s := t.str |
| 384 | start, end, hasExt := t.findTypeForKey(key) |
| 385 | if start == end { |
| 386 | if hasExt { |
| 387 | b = b[2:] |
| 388 | } |
| 389 | t.str = fmt.Sprintf("%s-%s%s", s[:start], b, s[end:]) |
| 390 | } else { |
| 391 | t.str = fmt.Sprintf("%s%s%s", s[:start], value, s[end:]) |
| 392 | } |
| 393 | } |
| 394 | return t, nil |
| 395 | } |
| 396 | |
| 397 | // findKeyAndType returns the start and end position for the type corresponding |
| 398 | // to key or the point at which to insert the key-value pair if the type |
| 399 | // wasn't found. The hasExt return value reports whether an -u extension was present. |
| 400 | // Note: the extensions are typically very small and are likely to contain |
| 401 | // only one key-type pair. |
| 402 | func (t Tag) findTypeForKey(key string) (start, end int, hasExt bool) { |
| 403 | p := int(t.pExt) |
| 404 | if len(key) != 2 || p == len(t.str) || p == 0 { |
| 405 | return p, p, false |
| 406 | } |
| 407 | s := t.str |
| 408 | |
| 409 | // Find the correct extension. |
| 410 | for p++; s[p] != 'u'; p++ { |
| 411 | if s[p] > 'u' { |
| 412 | p-- |
| 413 | return p, p, false |
| 414 | } |
| 415 | if p = nextExtension(s, p); p == len(s) { |
| 416 | return len(s), len(s), false |
| 417 | } |
| 418 | } |
| 419 | // Proceed to the hyphen following the extension name. |
| 420 | p++ |
| 421 | |
| 422 | // curKey is the key currently being processed. |
| 423 | curKey := "" |
| 424 | |
| 425 | // Iterate over keys until we get the end of a section. |
| 426 | for { |
| 427 | // p points to the hyphen preceding the current token. |
| 428 | if p3 := p + 3; s[p3] == '-' { |
| 429 | // Found a key. |
| 430 | // Check whether we just processed the key that was requested. |
| 431 | if curKey == key { |
| 432 | return start, p, true |
| 433 | } |
| 434 | // Set to the next key and continue scanning type tokens. |
| 435 | curKey = s[p+1 : p3] |
| 436 | if curKey > key { |
| 437 | return p, p, true |
| 438 | } |
| 439 | // Start of the type token sequence. |
| 440 | start = p + 4 |
| 441 | // A type is at least 3 characters long. |
| 442 | p += 7 // 4 + 3 |
| 443 | } else { |
| 444 | // Attribute or type, which is at least 3 characters long. |
| 445 | p += 4 |
| 446 | } |
| 447 | // p points past the third character of a type or attribute. |
| 448 | max := p + 5 // maximum length of token plus hyphen. |
| 449 | if len(s) < max { |
| 450 | max = len(s) |
| 451 | } |
| 452 | for ; p < max && s[p] != '-'; p++ { |
| 453 | } |
| 454 | // Bail if we have exhausted all tokens or if the next token starts |
| 455 | // a new extension. |
| 456 | if p == len(s) || s[p+2] == '-' { |
| 457 | if curKey == key { |
| 458 | return start, p, true |
| 459 | } |
| 460 | return p, p, true |
| 461 | } |
| 462 | } |
| 463 | } |
| 464 | |
| 465 | // ParseBase parses a 2- or 3-letter ISO 639 code. |
| 466 | // It returns a ValueError if s is a well-formed but unknown language identifier |
| 467 | // or another error if another error occurred. |
| 468 | func ParseBase(s string) (Language, error) { |
| 469 | if n := len(s); n < 2 || 3 < n { |
| 470 | return 0, ErrSyntax |
| 471 | } |
| 472 | var buf [3]byte |
| 473 | return getLangID(buf[:copy(buf[:], s)]) |
| 474 | } |
| 475 | |
| 476 | // ParseScript parses a 4-letter ISO 15924 code. |
| 477 | // It returns a ValueError if s is a well-formed but unknown script identifier |
| 478 | // or another error if another error occurred. |
| 479 | func ParseScript(s string) (Script, error) { |
| 480 | if len(s) != 4 { |
| 481 | return 0, ErrSyntax |
| 482 | } |
| 483 | var buf [4]byte |
| 484 | return getScriptID(script, buf[:copy(buf[:], s)]) |
| 485 | } |
| 486 | |
| 487 | // EncodeM49 returns the Region for the given UN M.49 code. |
| 488 | // It returns an error if r is not a valid code. |
| 489 | func EncodeM49(r int) (Region, error) { |
| 490 | return getRegionM49(r) |
| 491 | } |
| 492 | |
| 493 | // ParseRegion parses a 2- or 3-letter ISO 3166-1 or a UN M.49 code. |
| 494 | // It returns a ValueError if s is a well-formed but unknown region identifier |
| 495 | // or another error if another error occurred. |
| 496 | func ParseRegion(s string) (Region, error) { |
| 497 | if n := len(s); n < 2 || 3 < n { |
| 498 | return 0, ErrSyntax |
| 499 | } |
| 500 | var buf [3]byte |
| 501 | return getRegionID(buf[:copy(buf[:], s)]) |
| 502 | } |
| 503 | |
| 504 | // IsCountry returns whether this region is a country or autonomous area. This |
| 505 | // includes non-standard definitions from CLDR. |
| 506 | func (r Region) IsCountry() bool { |
| 507 | if r == 0 || r.IsGroup() || r.IsPrivateUse() && r != _XK { |
| 508 | return false |
| 509 | } |
| 510 | return true |
| 511 | } |
| 512 | |
| 513 | // IsGroup returns whether this region defines a collection of regions. This |
| 514 | // includes non-standard definitions from CLDR. |
| 515 | func (r Region) IsGroup() bool { |
| 516 | if r == 0 { |
| 517 | return false |
| 518 | } |
| 519 | return int(regionInclusion[r]) < len(regionContainment) |
| 520 | } |
| 521 | |
| 522 | // Contains returns whether Region c is contained by Region r. It returns true |
| 523 | // if c == r. |
| 524 | func (r Region) Contains(c Region) bool { |
| 525 | if r == c { |
| 526 | return true |
| 527 | } |
| 528 | g := regionInclusion[r] |
| 529 | if g >= nRegionGroups { |
| 530 | return false |
| 531 | } |
| 532 | m := regionContainment[g] |
| 533 | |
| 534 | d := regionInclusion[c] |
| 535 | b := regionInclusionBits[d] |
| 536 | |
| 537 | // A contained country may belong to multiple disjoint groups. Matching any |
| 538 | // of these indicates containment. If the contained region is a group, it |
| 539 | // must strictly be a subset. |
| 540 | if d >= nRegionGroups { |
| 541 | return b&m != 0 |
| 542 | } |
| 543 | return b&^m == 0 |
| 544 | } |
| 545 | |
| 546 | var errNoTLD = errors.New("language: region is not a valid ccTLD") |
| 547 | |
| 548 | // TLD returns the country code top-level domain (ccTLD). UK is returned for GB. |
| 549 | // In all other cases it returns either the region itself or an error. |
| 550 | // |
| 551 | // This method may return an error for a region for which there exists a |
| 552 | // canonical form with a ccTLD. To get that ccTLD canonicalize r first. The |
| 553 | // region will already be canonicalized it was obtained from a Tag that was |
| 554 | // obtained using any of the default methods. |
| 555 | func (r Region) TLD() (Region, error) { |
| 556 | // See http://en.wikipedia.org/wiki/Country_code_top-level_domain for the |
| 557 | // difference between ISO 3166-1 and IANA ccTLD. |
| 558 | if r == _GB { |
| 559 | r = _UK |
| 560 | } |
| 561 | if (r.typ() & ccTLD) == 0 { |
| 562 | return 0, errNoTLD |
| 563 | } |
| 564 | return r, nil |
| 565 | } |
| 566 | |
| 567 | // Canonicalize returns the region or a possible replacement if the region is |
| 568 | // deprecated. It will not return a replacement for deprecated regions that |
| 569 | // are split into multiple regions. |
| 570 | func (r Region) Canonicalize() Region { |
| 571 | if cr := normRegion(r); cr != 0 { |
| 572 | return cr |
| 573 | } |
| 574 | return r |
| 575 | } |
| 576 | |
| 577 | // Variant represents a registered variant of a language as defined by BCP 47. |
| 578 | type Variant struct { |
| 579 | ID uint8 |
| 580 | str string |
| 581 | } |
| 582 | |
| 583 | // ParseVariant parses and returns a Variant. An error is returned if s is not |
| 584 | // a valid variant. |
| 585 | func ParseVariant(s string) (Variant, error) { |
| 586 | s = strings.ToLower(s) |
| 587 | if id, ok := variantIndex[s]; ok { |
| 588 | return Variant{id, s}, nil |
| 589 | } |
| 590 | return Variant{}, NewValueError([]byte(s)) |
| 591 | } |
| 592 | |
| 593 | // String returns the string representation of the variant. |
| 594 | func (v Variant) String() string { |
| 595 | return v.str |
| 596 | } |