xref: /aosp_15_r20/external/icu/libandroidicu/include/unicode/uloc.h (revision 0e209d3975ff4a8c132096b14b0e9364a753506e)
1 // © 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
3 /*
4 **********************************************************************
5 *   Copyright (C) 1997-2016, International Business Machines
6 *   Corporation and others.  All Rights Reserved.
7 **********************************************************************
8 *
9 * File ULOC.H
10 *
11 * Modification History:
12 *
13 *   Date        Name        Description
14 *   04/01/97    aliu        Creation.
15 *   08/22/98    stephen     JDK 1.2 sync.
16 *   12/08/98    rtg         New C API for Locale
17 *   03/30/99    damiba      overhaul
18 *   03/31/99    helena      Javadoc for uloc functions.
19 *   04/15/99    Madhu       Updated Javadoc
20 ********************************************************************************
21 */
22 
23 #ifndef ULOC_H
24 #define ULOC_H
25 
26 #include "unicode/utypes.h"
27 #include "unicode/uenum.h"
28 
29 /**
30  * \file
31  * \brief C API: Locale ID functionality similar to C++ class Locale
32  *
33  * <h2> ULoc C API for Locale </h2>
34  * A <code>Locale</code> represents a specific geographical, political,
35  * or cultural region. An operation that requires a <code>Locale</code> to perform
36  * its task is called <em>locale-sensitive</em> and uses the <code>Locale</code>
37  * to tailor information for the user. For example, displaying a number
38  * is a locale-sensitive operation--the number should be formatted
39  * according to the customs/conventions of the user's native country,
40  * region, or culture.  In the C APIs, a locales is simply a const char string.
41  *
42  * <P>
43  * You create a <code>Locale</code> with one of the three options listed below.
44  * Each of the component is separated by '_' in the locale string.
45  * \htmlonly<blockquote>\endhtmlonly
46  * <pre>
47  * \code
48  *       newLanguage
49  *
50  *       newLanguage + newCountry
51  *
52  *       newLanguage + newCountry + newVariant
53  * \endcode
54  * </pre>
55  * \htmlonly</blockquote>\endhtmlonly
56  * The first option is a valid <STRONG>ISO
57  * Language Code.</STRONG> These codes are the lower-case two-letter
58  * codes as defined by ISO-639.
59  * You can find a full list of these codes at a number of sites, such as:
60  * <BR><a href ="http://www.ics.uci.edu/pub/ietf/http/related/iso639.txt">
61  * http://www.ics.uci.edu/pub/ietf/http/related/iso639.txt</a>
62  *
63  * <P>
64  * The second option includes an additional <STRONG>ISO Country
65  * Code.</STRONG> These codes are the upper-case two-letter codes
66  * as defined by ISO-3166.
67  * You can find a full list of these codes at a number of sites, such as:
68  * <BR><a href="http://www.chemie.fu-berlin.de/diverse/doc/ISO_3166.html">
69  * http://www.chemie.fu-berlin.de/diverse/doc/ISO_3166.html</a>
70  *
71  * <P>
72  * The third option requires another additional information--the
73  * <STRONG>Variant.</STRONG>
74  * The Variant codes are vendor and browser-specific.
75  * For example, use WIN for Windows, MAC for Macintosh, and POSIX for POSIX.
76  * Where there are two variants, separate them with an underscore, and
77  * put the most important one first. For
78  * example, a Traditional Spanish collation might be referenced, with
79  * "ES", "ES", "Traditional_WIN".
80  *
81  * <P>
82  * Because a <code>Locale</code> is just an identifier for a region,
83  * no validity check is performed when you specify a <code>Locale</code>.
84  * If you want to see whether particular resources are available for the
85  * <code>Locale</code> you asked for, you must query those resources. For
86  * example, ask the <code>UNumberFormat</code> for the locales it supports
87  * using its <code>getAvailable</code> method.
88  * <BR><STRONG>Note:</STRONG> When you ask for a resource for a particular
89  * locale, you get back the best available match, not necessarily
90  * precisely what you asked for. For more information, look at
91  * <code>UResourceBundle</code>.
92  *
93  * <P>
94  * The <code>Locale</code> provides a number of convenient constants
95  * that you can use to specify the commonly used
96  * locales. For example, the following refers to a locale
97  * for the United States:
98  * \htmlonly<blockquote>\endhtmlonly
99  * <pre>
100  * \code
101  *       ULOC_US
102  * \endcode
103  * </pre>
104  * \htmlonly</blockquote>\endhtmlonly
105  *
106  * <P>
107  * Once you've specified a locale you can query it for information about
108  * itself. Use <code>uloc_getCountry</code> to get the ISO Country Code and
109  * <code>uloc_getLanguage</code> to get the ISO Language Code. You can
110  * use <code>uloc_getDisplayCountry</code> to get the
111  * name of the country suitable for displaying to the user. Similarly,
112  * you can use <code>uloc_getDisplayLanguage</code> to get the name of
113  * the language suitable for displaying to the user. Interestingly,
114  * the <code>uloc_getDisplayXXX</code> methods are themselves locale-sensitive
115  * and have two versions: one that uses the default locale and one
116  * that takes a locale as an argument and displays the name or country in
117  * a language appropriate to that locale.
118  *
119  * <P>
120  * The ICU provides a number of services that perform locale-sensitive
121  * operations. For example, the <code>unum_xxx</code> functions format
122  * numbers, currency, or percentages in a locale-sensitive manner.
123  * </P>
124  * \htmlonly<blockquote>\endhtmlonly
125  * <pre>
126  * \code
127  *     UErrorCode success = U_ZERO_ERROR;
128  *     UNumberFormat *nf;
129  *     const char* myLocale = "fr_FR";
130  *
131  *     nf = unum_open( UNUM_DEFAULT, NULL, success );
132  *     unum_close(nf);
133  *     nf = unum_open( UNUM_CURRENCY, NULL, success );
134  *     unum_close(nf);
135  *     nf = unum_open( UNUM_PERCENT, NULL, success );
136  *     unum_close(nf);
137  * \endcode
138  * </pre>
139  * \htmlonly</blockquote>\endhtmlonly
140  * Each of these methods has two variants; one with an explicit locale
141  * and one without; the latter using the default locale.
142  * \htmlonly<blockquote>\endhtmlonly
143  * <pre>
144  * \code
145  *
146  *     nf = unum_open( UNUM_DEFAULT, myLocale, success );
147  *     unum_close(nf);
148  *     nf = unum_open( UNUM_CURRENCY, myLocale, success );
149  *     unum_close(nf);
150  *     nf = unum_open( UNUM_PERCENT, myLocale, success );
151  *     unum_close(nf);
152  * \endcode
153  * </pre>
154  * \htmlonly</blockquote>\endhtmlonly
155  * A <code>Locale</code> is the mechanism for identifying the kind of services
156  * (<code>UNumberFormat</code>) that you would like to get. The locale is
157  * <STRONG>just</STRONG> a mechanism for identifying these services.
158  *
159  * <P>
160  * Each international service that performs locale-sensitive operations
161  * allows you
162  * to get all the available objects of that type. You can sift
163  * through these objects by language, country, or variant,
164  * and use the display names to present a menu to the user.
165  * For example, you can create a menu of all the collation objects
166  * suitable for a given language. Such classes implement these
167  * three class methods:
168  * \htmlonly<blockquote>\endhtmlonly
169  * <pre>
170  * \code
171  *       const char* uloc_getAvailable(int32_t index);
172  *       int32_t uloc_countAvailable();
173  *       int32_t
174  *       uloc_getDisplayName(const char* localeID,
175  *                 const char* inLocaleID,
176  *                 UChar* result,
177  *                 int32_t maxResultSize,
178  *                  UErrorCode* err);
179  *
180  * \endcode
181  * </pre>
182  * \htmlonly</blockquote>\endhtmlonly
183  * <P>
184  * Concerning POSIX/RFC1766 Locale IDs,
185  *  the getLanguage/getCountry/getVariant/getName functions do understand
186  * the POSIX type form of  language_COUNTRY.ENCODING\@VARIANT
187  * and if there is not an ICU-stype variant, uloc_getVariant() for example
188  * will return the one listed after the \@at sign. As well, the hyphen
189  * "-" is recognized as a country/variant separator similarly to RFC1766.
190  * So for example, "en-us" will be interpreted as en_US.
191  * As a result, uloc_getName() is far from a no-op, and will have the
192  * effect of converting POSIX/RFC1766 IDs into ICU form, although it does
193  * NOT map any of the actual codes (i.e. russian->ru) in any way.
194  * Applications should call uloc_getName() at the point where a locale ID
195  * is coming from an external source (user entry, OS, web browser)
196  * and pass the resulting string to other ICU functions.  For example,
197  * don't use de-de\@EURO as an argument to resourcebundle.
198  *
199  * @see UResourceBundle
200  */
201 
202 /** Useful constant for this language. @stable ICU 2.0 */
203 #define ULOC_CHINESE            "zh"
204 /** Useful constant for this language. @stable ICU 2.0 */
205 #define ULOC_ENGLISH            "en"
206 /** Useful constant for this language. @stable ICU 2.0 */
207 #define ULOC_FRENCH             "fr"
208 /** Useful constant for this language. @stable ICU 2.0 */
209 #define ULOC_GERMAN             "de"
210 /** Useful constant for this language. @stable ICU 2.0 */
211 #define ULOC_ITALIAN            "it"
212 /** Useful constant for this language. @stable ICU 2.0 */
213 #define ULOC_JAPANESE           "ja"
214 /** Useful constant for this language. @stable ICU 2.0 */
215 #define ULOC_KOREAN             "ko"
216 /** Useful constant for this language. @stable ICU 2.0 */
217 #define ULOC_SIMPLIFIED_CHINESE "zh_CN"
218 /** Useful constant for this language. @stable ICU 2.0 */
219 #define ULOC_TRADITIONAL_CHINESE "zh_TW"
220 
221 /** Useful constant for this country/region. @stable ICU 2.0 */
222 #define ULOC_CANADA         "en_CA"
223 /** Useful constant for this country/region. @stable ICU 2.0 */
224 #define ULOC_CANADA_FRENCH  "fr_CA"
225 /** Useful constant for this country/region. @stable ICU 2.0 */
226 #define ULOC_CHINA          "zh_CN"
227 /** Useful constant for this country/region. @stable ICU 2.0 */
228 #define ULOC_PRC            "zh_CN"
229 /** Useful constant for this country/region. @stable ICU 2.0 */
230 #define ULOC_FRANCE         "fr_FR"
231 /** Useful constant for this country/region. @stable ICU 2.0 */
232 #define ULOC_GERMANY        "de_DE"
233 /** Useful constant for this country/region. @stable ICU 2.0 */
234 #define ULOC_ITALY          "it_IT"
235 /** Useful constant for this country/region. @stable ICU 2.0 */
236 #define ULOC_JAPAN          "ja_JP"
237 /** Useful constant for this country/region. @stable ICU 2.0 */
238 #define ULOC_KOREA          "ko_KR"
239 /** Useful constant for this country/region. @stable ICU 2.0 */
240 #define ULOC_TAIWAN         "zh_TW"
241 /** Useful constant for this country/region. @stable ICU 2.0 */
242 #define ULOC_UK             "en_GB"
243 /** Useful constant for this country/region. @stable ICU 2.0 */
244 #define ULOC_US             "en_US"
245 
246 /**
247  * Useful constant for the maximum size of the language part of a locale ID.
248  * (including the terminating NULL).
249  * @stable ICU 2.0
250  */
251 #define ULOC_LANG_CAPACITY 12
252 
253 /**
254  * Useful constant for the maximum size of the country part of a locale ID
255  * (including the terminating NULL).
256  * @stable ICU 2.0
257  */
258 #define ULOC_COUNTRY_CAPACITY 4
259 /**
260  * Useful constant for the maximum size of the whole locale ID
261  * (including the terminating NULL and all keywords).
262  * @stable ICU 2.0
263  */
264 #define ULOC_FULLNAME_CAPACITY 157
265 
266 /**
267  * Useful constant for the maximum size of the script part of a locale ID
268  * (including the terminating NULL).
269  * @stable ICU 2.8
270  */
271 #define ULOC_SCRIPT_CAPACITY 6
272 
273 /**
274  * Useful constant for the maximum size of keywords in a locale
275  * @stable ICU 2.8
276  */
277 #define ULOC_KEYWORDS_CAPACITY 96
278 
279 /**
280  * Useful constant for the maximum total size of keywords and their values in a locale
281  * @stable ICU 2.8
282  */
283 #define ULOC_KEYWORD_AND_VALUES_CAPACITY 100
284 
285 /**
286  * Invariant character separating keywords from the locale string
287  * @stable ICU 2.8
288  */
289 #define ULOC_KEYWORD_SEPARATOR '@'
290 
291 /**
292   * Unicode code point for '@' separating keywords from the locale string.
293   * @see ULOC_KEYWORD_SEPARATOR
294   * @stable ICU 4.6
295   */
296 #define ULOC_KEYWORD_SEPARATOR_UNICODE 0x40
297 
298 /**
299  * Invariant character for assigning value to a keyword
300  * @stable ICU 2.8
301  */
302 #define ULOC_KEYWORD_ASSIGN '='
303 
304 /**
305   * Unicode code point for '=' for assigning value to a keyword.
306   * @see ULOC_KEYWORD_ASSIGN
307   * @stable ICU 4.6
308   */
309 #define ULOC_KEYWORD_ASSIGN_UNICODE 0x3D
310 
311 /**
312  * Invariant character separating keywords
313  * @stable ICU 2.8
314  */
315 #define ULOC_KEYWORD_ITEM_SEPARATOR ';'
316 
317 /**
318   * Unicode code point for ';' separating keywords
319   * @see ULOC_KEYWORD_ITEM_SEPARATOR
320   * @stable ICU 4.6
321   */
322 #define ULOC_KEYWORD_ITEM_SEPARATOR_UNICODE 0x3B
323 
324 /**
325  * Constants for *_getLocale()
326  * Allow user to select whether she wants information on
327  * requested, valid or actual locale.
328  * For example, a collator for "en_US_CALIFORNIA" was
329  * requested. In the current state of ICU (2.0),
330  * the requested locale is "en_US_CALIFORNIA",
331  * the valid locale is "en_US" (most specific locale supported by ICU)
332  * and the actual locale is "root" (the collation data comes unmodified
333  * from the UCA)
334  * The locale is considered supported by ICU if there is a core ICU bundle
335  * for that locale (although it may be empty).
336  * @stable ICU 2.1
337  */
338 typedef enum {
339   /** This is locale the data actually comes from
340    * @stable ICU 2.1
341    */
342   ULOC_ACTUAL_LOCALE    = 0,
343   /** This is the most specific locale supported by ICU
344    * @stable ICU 2.1
345    */
346   ULOC_VALID_LOCALE    = 1,
347 
348 #ifndef U_HIDE_DEPRECATED_API
349   /** This is the requested locale
350    *  @deprecated ICU 2.8
351    */
352   ULOC_REQUESTED_LOCALE = 2,
353 
354     /**
355      * One more than the highest normal ULocDataLocaleType value.
356      * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420.
357      */
358     ULOC_DATA_LOCALE_TYPE_LIMIT = 3
359 #endif  // U_HIDE_DEPRECATED_API
360 } ULocDataLocaleType;
361 
362 #ifndef U_HIDE_SYSTEM_API
363 /**
364  * Gets ICU's default locale.
365  * The returned string is a snapshot in time, and will remain valid
366  *   and unchanged even when uloc_setDefault() is called.
367  *   The returned storage is owned by ICU, and must not be altered or deleted
368  *   by the caller.
369  *
370  * @return the ICU default locale
371  * @system
372  * @stable ICU 2.0
373  */
374 U_CAPI const char* U_EXPORT2
375 uloc_getDefault(void);
376 
377 /**
378  * Sets ICU's default locale.
379  *    By default (without calling this function), ICU's default locale will be based
380  *    on information obtained from the underlying system environment.
381  *    <p>
382  *    Changes to ICU's default locale do not propagate back to the
383  *    system environment.
384  *    <p>
385  *    Changes to ICU's default locale to not affect any ICU services that
386  *    may already be open based on the previous default locale value.
387  *
388  * @param localeID the new ICU default locale. A value of NULL will try to get
389  *                 the system's default locale.
390  * @param status the error information if the setting of default locale fails
391  * @system
392  * @stable ICU 2.0
393  */
394 U_CAPI void U_EXPORT2
395 uloc_setDefault(const char* localeID,
396         UErrorCode*       status);
397 #endif  /* U_HIDE_SYSTEM_API */
398 
399 /**
400  * Gets the language code for the specified locale.
401  *
402  * This function may return with a failure error code for certain kinds of inputs
403  * but does not fully check for well-formed locale IDs / language tags.
404  *
405  * @param localeID the locale to get the ISO language code with
406  * @param language the language code for localeID
407  * @param languageCapacity the size of the language buffer to store the
408  * language code with
409  * @param err error information if retrieving the language code failed
410  * @return the actual buffer size needed for the language code.  If it's greater
411  * than languageCapacity, the returned language code will be truncated.
412  * @stable ICU 2.0
413  */
414 U_CAPI int32_t U_EXPORT2
415 uloc_getLanguage(const char*    localeID,
416          char* language,
417          int32_t languageCapacity,
418          UErrorCode* err);
419 
420 /**
421  * Gets the script code for the specified locale.
422  *
423  * This function may return with a failure error code for certain kinds of inputs
424  * but does not fully check for well-formed locale IDs / language tags.
425  *
426  * @param localeID the locale to get the ISO language code with
427  * @param script the language code for localeID
428  * @param scriptCapacity the size of the language buffer to store the
429  * language code with
430  * @param err error information if retrieving the language code failed
431  * @return the actual buffer size needed for the language code.  If it's greater
432  * than scriptCapacity, the returned language code will be truncated.
433  * @stable ICU 2.8
434  */
435 U_CAPI int32_t U_EXPORT2
436 uloc_getScript(const char*    localeID,
437          char* script,
438          int32_t scriptCapacity,
439          UErrorCode* err);
440 
441 /**
442  * Gets the  country code for the specified locale.
443  *
444  * This function may return with a failure error code for certain kinds of inputs
445  * but does not fully check for well-formed locale IDs / language tags.
446  *
447  * @param localeID the locale to get the country code with
448  * @param country the country code for localeID
449  * @param countryCapacity the size of the country buffer to store the
450  * country code with
451  * @param err error information if retrieving the country code failed
452  * @return the actual buffer size needed for the country code.  If it's greater
453  * than countryCapacity, the returned country code will be truncated.
454  * @stable ICU 2.0
455  */
456 U_CAPI int32_t U_EXPORT2
457 uloc_getCountry(const char*    localeID,
458         char* country,
459         int32_t countryCapacity,
460         UErrorCode* err);
461 
462 /**
463  * Gets the variant code for the specified locale.
464  *
465  * This function may return with a failure error code for certain kinds of inputs
466  * but does not fully check for well-formed locale IDs / language tags.
467  *
468  * @param localeID the locale to get the variant code with
469  * @param variant the variant code for localeID
470  * @param variantCapacity the size of the variant buffer to store the
471  * variant code with
472  * @param err error information if retrieving the variant code failed
473  * @return the actual buffer size needed for the variant code.  If it's greater
474  * than variantCapacity, the returned variant code will be truncated.
475  * @stable ICU 2.0
476  */
477 U_CAPI int32_t U_EXPORT2
478 uloc_getVariant(const char*    localeID,
479         char* variant,
480         int32_t variantCapacity,
481         UErrorCode* err);
482 
483 
484 /**
485  * Gets the full name for the specified locale.
486  *
487  * This function may return with a failure error code for certain kinds of inputs
488  * but does not fully check for well-formed locale IDs / language tags.
489  *
490  * Note: This has the effect of 'canonicalizing' the ICU locale ID to
491  * a certain extent. Upper and lower case are set as needed.
492  * It does NOT map aliased names in any way.
493  * See the top of this header file.
494  * This API supports preflighting.
495  *
496  * @param localeID the locale to get the full name with
497  * @param name fill in buffer for the name without keywords.
498  * @param nameCapacity capacity of the fill in buffer.
499  * @param err error information if retrieving the full name failed
500  * @return the actual buffer size needed for the full name.  If it's greater
501  * than nameCapacity, the returned full name will be truncated.
502  * @stable ICU 2.0
503  */
504 U_CAPI int32_t U_EXPORT2
505 uloc_getName(const char*    localeID,
506          char* name,
507          int32_t nameCapacity,
508          UErrorCode* err);
509 
510 /**
511  * Gets the full name for the specified locale.
512  * Note: This has the effect of 'canonicalizing' the string to
513  * a certain extent. Upper and lower case are set as needed,
514  * and if the components were in 'POSIX' format they are changed to
515  * ICU format.  It does NOT map aliased names in any way.
516  * See the top of this header file.
517  *
518  * @param localeID the locale to get the full name with
519  * @param name the full name for localeID
520  * @param nameCapacity the size of the name buffer to store the
521  * full name with
522  * @param err error information if retrieving the full name failed
523  * @return the actual buffer size needed for the full name.  If it's greater
524  * than nameCapacity, the returned full name will be truncated.
525  * @stable ICU 2.8
526  */
527 U_CAPI int32_t U_EXPORT2
528 uloc_canonicalize(const char*    localeID,
529          char* name,
530          int32_t nameCapacity,
531          UErrorCode* err);
532 
533 /**
534  * Gets the ISO language code for the specified locale.
535  *
536  * @param localeID the locale to get the ISO language code with
537  * @return language the ISO language code for localeID
538  * @stable ICU 2.0
539  */
540 U_CAPI const char* U_EXPORT2
541 uloc_getISO3Language(const char* localeID);
542 
543 
544 /**
545  * Gets the ISO country code for the specified locale.
546  *
547  * @param localeID the locale to get the ISO country code with
548  * @return country the ISO country code for localeID
549  * @stable ICU 2.0
550  */
551 U_CAPI const char* U_EXPORT2
552 uloc_getISO3Country(const char* localeID);
553 
554 /**
555  * Gets the Win32 LCID value for the specified locale.
556  * If the ICU locale is not recognized by Windows, 0 will be returned.
557  *
558  * LCIDs were deprecated with Windows Vista and Microsoft recommends
559  * that developers use BCP47 style tags instead (uloc_toLanguageTag).
560  *
561  * @param localeID the locale to get the Win32 LCID value with
562  * @return country the Win32 LCID for localeID
563  * @stable ICU 2.0
564  */
565 U_CAPI uint32_t U_EXPORT2
566 uloc_getLCID(const char* localeID);
567 
568 /**
569  * Gets the language name suitable for display for the specified locale.
570  *
571  * @param locale the locale to get the ISO language code with
572  * @param displayLocale Specifies the locale to be used to display the name. In
573  *                 other words, if the locale's language code is "en", passing
574  *                 Locale::getFrench() for inLocale would result in "Anglais",
575  *                 while passing Locale::getGerman() for inLocale would result
576  *                 in "Englisch".
577  * @param language the displayable language code for localeID
578  * @param languageCapacity the size of the language buffer to store the
579  *                 displayable language code with.
580  * @param status error information if retrieving the displayable language code
581  *                 failed. U_USING_DEFAULT_WARNING indicates that no data was
582  *                 found from the locale resources and a case canonicalized
583  *                 language code is placed into language as fallback.
584  * @return the actual buffer size needed for the displayable language code. If
585  *                 it's greater than languageCapacity, the returned language
586  *                 code will be truncated.
587  * @stable ICU 2.0
588  */
589 U_CAPI int32_t U_EXPORT2
590 uloc_getDisplayLanguage(const char* locale,
591             const char* displayLocale,
592             UChar* language,
593             int32_t languageCapacity,
594             UErrorCode* status);
595 
596 /**
597  * Gets the script name suitable for display for the specified locale.
598  *
599  * @param locale the locale to get the displayable script code with. NULL may be
600  *                 used to specify the default.
601  * @param displayLocale Specifies the locale to be used to display the name. In
602  *                 other words, if the locale's language code is "en", passing
603  *                 Locale::getFrench() for inLocale would result in "", while
604  *                 passing Locale::getGerman() for inLocale would result in "".
605  *                 NULL may be used to specify the default.
606  * @param script the displayable script for the localeID.
607  * @param scriptCapacity the size of the script buffer to store the displayable
608  *                 script code with.
609  * @param status error information if retrieving the displayable script code
610  *                 failed. U_USING_DEFAULT_WARNING indicates that no data was
611  *                 found from the locale resources and a case canonicalized
612  *                 script code is placed into script as fallback.
613  * @return the actual buffer size needed for the displayable script code. If
614  *                 it's greater than scriptCapacity, the returned displayable
615  *                 script code will be truncated.
616  * @stable ICU 2.8
617  */
618 U_CAPI int32_t U_EXPORT2
619 uloc_getDisplayScript(const char* locale,
620             const char* displayLocale,
621             UChar* script,
622             int32_t scriptCapacity,
623             UErrorCode* status);
624 
625 /**
626  * Gets the country name suitable for display for the specified locale.
627  * Warning: this is for the region part of a valid locale ID; it cannot just be
628  * the region code (like "FR"). To get the display name for a region alone, or
629  * for other options, use ULocaleDisplayNames instead.
630  *
631  * @param locale the locale to get the displayable country code with. NULL may
632  *                 be used to specify the default.
633  * @param displayLocale Specifies the locale to be used to display the name. In
634  *                 other words, if the locale's language code is "en", passing
635  *                 Locale::getFrench() for inLocale would result in "Anglais",
636  *                 while passing Locale::getGerman() for inLocale would result
637  *                 in "Englisch". NULL may be used to specify the default.
638  * @param country the displayable country code for localeID.
639  * @param countryCapacity the size of the country buffer to store the
640  *                 displayable country code with.
641  * @param status error information if retrieving the displayable country code
642  *                 failed. U_USING_DEFAULT_WARNING indicates that no data was
643  *                 found from the locale resources and a case canonicalized
644  *                 country code is placed into country as fallback.
645  * @return the actual buffer size needed for the displayable country code. If
646  *                 it's greater than countryCapacity, the returned displayable
647  *                 country code will be truncated.
648  * @stable ICU 2.0
649  */
650 U_CAPI int32_t U_EXPORT2
651 uloc_getDisplayCountry(const char* locale,
652                        const char* displayLocale,
653                        UChar* country,
654                        int32_t countryCapacity,
655                        UErrorCode* status);
656 
657 
658 /**
659  * Gets the variant name suitable for display for the specified locale.
660  *
661  * @param locale the locale to get the displayable variant code with. NULL may
662  *                 be used to specify the default.
663  * @param displayLocale Specifies the locale to be used to display the name. In
664  *                 other words, if the locale's language code is "en", passing
665  *                 Locale::getFrench() for inLocale would result in "Anglais",
666  *                 while passing Locale::getGerman() for inLocale would result
667  *                 in "Englisch". NULL may be used to specify the default.
668  * @param variant the displayable variant code for localeID.
669  * @param variantCapacity the size of the variant buffer to store the
670  *                 displayable variant code with.
671  * @param status error information if retrieving the displayable variant code
672  *                 failed. U_USING_DEFAULT_WARNING indicates that no data was
673  *                 found from the locale resources and a case canonicalized
674  *                 variant code is placed into variant as fallback.
675  * @return the actual buffer size needed for the displayable variant code. If
676  *                 it's greater than variantCapacity, the returned displayable
677  *                 variant code will be truncated.
678  * @stable ICU 2.0
679  */
680 U_CAPI int32_t U_EXPORT2
681 uloc_getDisplayVariant(const char* locale,
682                        const char* displayLocale,
683                        UChar* variant,
684                        int32_t variantCapacity,
685                        UErrorCode* status);
686 
687 /**
688  * Gets the keyword name suitable for display for the specified locale. E.g:
689  * for the locale string de_DE\@collation=PHONEBOOK, this API gets the display
690  * string for the keyword collation.
691  * Usage:
692  * <code>
693  *    UErrorCode status = U_ZERO_ERROR;
694  *    const char* keyword =NULL;
695  *    int32_t keywordLen = 0;
696  *    int32_t keywordCount = 0;
697  *    UChar displayKeyword[256];
698  *    int32_t displayKeywordLen = 0;
699  *    UEnumeration* keywordEnum = uloc_openKeywords("de_DE@collation=PHONEBOOK;calendar=TRADITIONAL", &status);
700  *    for(keywordCount = uenum_count(keywordEnum, &status); keywordCount > 0 ; keywordCount--){
701  *          if(U_FAILURE(status)){
702  *              ...something went wrong so handle the error...
703  *              break;
704  *          }
705  *          // the uenum_next returns NUL terminated string
706  *          keyword = uenum_next(keywordEnum, &keywordLen, &status);
707  *          displayKeywordLen = uloc_getDisplayKeyword(keyword, "en_US", displayKeyword, 256);
708  *          ... do something interesting .....
709  *    }
710  *    uenum_close(keywordEnum);
711  * </code>
712  * @param keyword           The keyword whose display string needs to be returned.
713  * @param displayLocale     Specifies the locale to be used to display the name.  In other words,
714  *                          if the locale's language code is "en", passing Locale::getFrench() for
715  *                          inLocale would result in "Anglais", while passing Locale::getGerman()
716  *                          for inLocale would result in "Englisch". NULL may be used to specify the default.
717  * @param dest              the buffer to which the displayable keyword should be written.
718  * @param destCapacity      The size of the buffer (number of UChars). If it is 0, then
719  *                          dest may be NULL and the function will only return the length of the
720  *                          result without writing any of the result string (pre-flighting).
721  * @param status            error information if retrieving the displayable string failed.
722  *                          Should not be NULL and should not indicate failure on entry.
723  *                          U_USING_DEFAULT_WARNING indicates that no data was found from the locale
724  *                          resources and the keyword is placed into dest as fallback.
725  * @return the actual buffer size needed for the displayable variant code.
726  * @see #uloc_openKeywords
727  * @stable ICU 2.8
728  */
729 U_CAPI int32_t U_EXPORT2
730 uloc_getDisplayKeyword(const char* keyword,
731                        const char* displayLocale,
732                        UChar* dest,
733                        int32_t destCapacity,
734                        UErrorCode* status);
735 /**
736  * Gets the value of the keyword suitable for display for the specified locale.
737  * E.g: for the locale string de_DE\@collation=PHONEBOOK, this API gets the display
738  * string for PHONEBOOK, in the display locale, when "collation" is specified as the keyword.
739  *
740  * @param locale            The locale to get the displayable variant code with. NULL may be used to specify the default.
741  * @param keyword           The keyword for whose value should be used.
742  * @param displayLocale     Specifies the locale to be used to display the name.  In other words,
743  *                          if the locale's language code is "en", passing Locale::getFrench() for
744  *                          inLocale would result in "Anglais", while passing Locale::getGerman()
745  *                          for inLocale would result in "Englisch". NULL may be used to specify the default.
746  * @param dest              the buffer to which the displayable keyword should be written.
747  * @param destCapacity      The size of the buffer (number of UChars). If it is 0, then
748  *                          dest may be NULL and the function will only return the length of the
749  *                          result without writing any of the result string (pre-flighting).
750  * @param status            error information if retrieving the displayable string failed.
751  *                          Should not be NULL and must not indicate failure on entry.
752  *                          U_USING_DEFAULT_WARNING indicates that no data was found from the locale
753  *                          resources and the value of the keyword is placed into dest as fallback.
754  * @return the actual buffer size needed for the displayable variant code.
755  * @stable ICU 2.8
756  */
757 U_CAPI int32_t U_EXPORT2
758 uloc_getDisplayKeywordValue(   const char* locale,
759                                const char* keyword,
760                                const char* displayLocale,
761                                UChar* dest,
762                                int32_t destCapacity,
763                                UErrorCode* status);
764 /**
765  * Gets the full name suitable for display for the specified locale.
766  *
767  * @param localeID the locale to get the displayable name with. NULL may be used to specify the default.
768  * @param inLocaleID Specifies the locale to be used to display the name.  In other words,
769  *                   if the locale's language code is "en", passing Locale::getFrench() for
770  *                   inLocale would result in "Anglais", while passing Locale::getGerman()
771  *                   for inLocale would result in "Englisch". NULL may be used to specify the default.
772  * @param result the displayable name for localeID
773  * @param maxResultSize the size of the name buffer to store the
774  * displayable full name with
775  * @param err error information if retrieving the displayable name failed
776  * @return the actual buffer size needed for the displayable name.  If it's greater
777  * than maxResultSize, the returned displayable name will be truncated.
778  * @stable ICU 2.0
779  */
780 U_CAPI int32_t U_EXPORT2
781 uloc_getDisplayName(const char* localeID,
782             const char* inLocaleID,
783             UChar* result,
784             int32_t maxResultSize,
785             UErrorCode* err);
786 
787 
788 /**
789  * Gets the specified locale from a list of available locales.
790  *
791  * This method corresponds to uloc_openAvailableByType called with the
792  * ULOC_AVAILABLE_DEFAULT type argument.
793  *
794  * The return value is a pointer to an item of a locale name array. Both this
795  * array and the pointers it contains are owned by ICU and should not be
796  * deleted or written through by the caller. The locale name is terminated by
797  * a null pointer.
798  *
799  * @param n the specific locale name index of the available locale list;
800  *     should not exceed the number returned by uloc_countAvailable.
801  * @return a specified locale name of all available locales
802  * @stable ICU 2.0
803  */
804 U_CAPI const char* U_EXPORT2
805 uloc_getAvailable(int32_t n);
806 
807 /**
808  * Gets the size of the all available locale list.
809  *
810  * @return the size of the locale list
811  * @stable ICU 2.0
812  */
813 U_CAPI int32_t U_EXPORT2 uloc_countAvailable(void);
814 
815 /**
816  * Types for uloc_getAvailableByType and uloc_countAvailableByType.
817  *
818  * @stable ICU 65
819  */
820 typedef enum ULocAvailableType {
821   /**
822    * Locales that return data when passed to ICU APIs,
823    * but not including legacy or alias locales.
824    *
825    * @stable ICU 65
826    */
827   ULOC_AVAILABLE_DEFAULT,
828 
829   /**
830    * Legacy or alias locales that return data when passed to ICU APIs.
831    * Examples of supported legacy or alias locales:
832    *
833    * - iw (alias to he)
834    * - mo (alias to ro)
835    * - zh_CN (alias to zh_Hans_CN)
836    * - sr_BA (alias to sr_Cyrl_BA)
837    * - ars (alias to ar_SA)
838    *
839    * The locales in this set are disjoint from the ones in
840    * ULOC_AVAILABLE_DEFAULT. To get both sets at the same time, use
841    * ULOC_AVAILABLE_WITH_LEGACY_ALIASES.
842    *
843    * @stable ICU 65
844    */
845   ULOC_AVAILABLE_ONLY_LEGACY_ALIASES,
846 
847   /**
848    * The union of the locales in ULOC_AVAILABLE_DEFAULT and
849    * ULOC_AVAILABLE_ONLY_LEGACY_ALIAS.
850    *
851    * @stable ICU 65
852    */
853   ULOC_AVAILABLE_WITH_LEGACY_ALIASES,
854 
855 #ifndef U_HIDE_INTERNAL_API
856   /**
857    * @internal
858    */
859   ULOC_AVAILABLE_COUNT
860 #endif  /* U_HIDE_INTERNAL_API */
861 } ULocAvailableType;
862 
863 /**
864  * Gets a list of available locales according to the type argument, allowing
865  * the user to access different sets of supported locales in ICU.
866  *
867  * The returned UEnumeration must be closed by the caller.
868  *
869  * @param type Type choice from ULocAvailableType.
870  * @param status Set if an error occurred.
871  * @return a UEnumeration owned by the caller, or nullptr on failure.
872  * @stable ICU 65
873  */
874 U_CAPI UEnumeration* U_EXPORT2
875 uloc_openAvailableByType(ULocAvailableType type, UErrorCode* status);
876 
877 /**
878  *
879  * Gets a list of all available 2-letter language codes defined in ISO 639,
880  * plus additional 3-letter codes determined to be useful for locale generation as
881  * defined by Unicode CLDR. This is a pointer
882  * to an array of pointers to arrays of char.  All of these pointers are owned
883  * by ICU-- do not delete them, and do not write through them.  The array is
884  * terminated with a null pointer.
885  * @return a list of all available language codes
886  * @stable ICU 2.0
887  */
888 U_CAPI const char* const* U_EXPORT2
889 uloc_getISOLanguages(void);
890 
891 /**
892  *
893  * Gets a list of all available 2-letter country codes defined in ISO 639.  This is a
894  * pointer to an array of pointers to arrays of char.  All of these pointers are
895  * owned by ICU-- do not delete them, and do not write through them.  The array is
896  * terminated with a null pointer.
897  * @return a list of all available country codes
898  * @stable ICU 2.0
899  */
900 U_CAPI const char* const* U_EXPORT2
901 uloc_getISOCountries(void);
902 
903 /**
904  * Truncate the locale ID string to get the parent locale ID.
905  * Copies the part of the string before the last underscore.
906  * The parent locale ID will be an empty string if there is no
907  * underscore, or if there is only one underscore at localeID[0].
908  *
909  * @param localeID Input locale ID string.
910  * @param parent   Output string buffer for the parent locale ID.
911  * @param parentCapacity Size of the output buffer.
912  * @param err A UErrorCode value.
913  * @return The length of the parent locale ID.
914  * @stable ICU 2.0
915  */
916 U_CAPI int32_t U_EXPORT2
917 uloc_getParent(const char*    localeID,
918                  char* parent,
919                  int32_t parentCapacity,
920                  UErrorCode* err);
921 
922 
923 
924 
925 /**
926  * Gets the full name for the specified locale, like uloc_getName(),
927  * but without keywords.
928  *
929  * Note: This has the effect of 'canonicalizing' the string to
930  * a certain extent. Upper and lower case are set as needed,
931  * and if the components were in 'POSIX' format they are changed to
932  * ICU format.  It does NOT map aliased names in any way.
933  * See the top of this header file.
934  *
935  * This API strips off the keyword part, so "de_DE\@collation=phonebook"
936  * will become "de_DE".
937  * This API supports preflighting.
938  *
939  * @param localeID the locale to get the full name with
940  * @param name fill in buffer for the name without keywords.
941  * @param nameCapacity capacity of the fill in buffer.
942  * @param err error information if retrieving the full name failed
943  * @return the actual buffer size needed for the full name.  If it's greater
944  * than nameCapacity, the returned full name will be truncated.
945  * @stable ICU 2.8
946  */
947 U_CAPI int32_t U_EXPORT2
948 uloc_getBaseName(const char*    localeID,
949          char* name,
950          int32_t nameCapacity,
951          UErrorCode* err);
952 
953 /**
954  * Gets an enumeration of keywords for the specified locale. Enumeration
955  * must get disposed of by the client using uenum_close function.
956  *
957  * @param localeID the locale to get the variant code with
958  * @param status error information if retrieving the keywords failed
959  * @return enumeration of keywords or NULL if there are no keywords.
960  * @stable ICU 2.8
961  */
962 U_CAPI UEnumeration* U_EXPORT2
963 uloc_openKeywords(const char* localeID,
964                         UErrorCode* status);
965 
966 /**
967  * Get the value for a keyword. Locale name does not need to be normalized.
968  *
969  * @param localeID locale name containing the keyword ("de_DE@currency=EURO;collation=PHONEBOOK")
970  * @param keywordName name of the keyword for which we want the value; must not be
971  *  NULL or empty, and must consist only of [A-Za-z0-9]. Case insensitive.
972  * @param buffer receiving buffer
973  * @param bufferCapacity capacity of receiving buffer
974  * @param status containing error code: e.g. buffer not big enough or ill-formed localeID
975  *  or keywordName parameters.
976  * @return the length of keyword value
977  * @stable ICU 2.8
978  */
979 U_CAPI int32_t U_EXPORT2
980 uloc_getKeywordValue(const char* localeID,
981                      const char* keywordName,
982                      char* buffer, int32_t bufferCapacity,
983                      UErrorCode* status);
984 
985 
986 /**
987  * Sets or removes the value of the specified keyword.
988  *
989  * For removing all keywords, use uloc_getBaseName().
990  *
991  * NOTE: Unlike almost every other ICU function which takes a
992  * buffer, this function will NOT truncate the output text, and will
993  * not update the buffer with unterminated text setting a status of
994  * U_STRING_NOT_TERMINATED_WARNING. If a BUFFER_OVERFLOW_ERROR is received,
995  * it means a terminated version of the updated locale ID would not fit
996  * in the buffer, and the original buffer is untouched. This is done to
997  * prevent incorrect or possibly even malformed locales from being generated
998  * and used.
999  *
1000  * @param keywordName name of the keyword to be set; must not be
1001  *  NULL or empty, and must consist only of [A-Za-z0-9]. Case insensitive.
1002  * @param keywordValue value of the keyword to be set. If 0-length or
1003  *  NULL, will result in the keyword being removed; no error is given if
1004  *  that keyword does not exist. Otherwise, must consist only of
1005  *  [A-Za-z0-9] and [/_+-].
1006  * @param buffer input buffer containing well-formed locale ID to be
1007  *  modified.
1008  * @param bufferCapacity capacity of receiving buffer
1009  * @param status containing error code: e.g. buffer not big enough
1010  *  or ill-formed keywordName or keywordValue parameters, or ill-formed
1011  *  locale ID in buffer on input.
1012  * @return the length needed for the buffer
1013  * @see uloc_getKeywordValue
1014  * @stable ICU 3.2
1015  */
1016 U_CAPI int32_t U_EXPORT2
1017 uloc_setKeywordValue(const char* keywordName,
1018                      const char* keywordValue,
1019                      char* buffer, int32_t bufferCapacity,
1020                      UErrorCode* status);
1021 
1022 /**
1023  * Returns whether the locale's script is written right-to-left.
1024  * If there is no script subtag, then the likely script is used, see uloc_addLikelySubtags().
1025  * If no likely script is known, then false is returned.
1026  *
1027  * A script is right-to-left according to the CLDR script metadata
1028  * which corresponds to whether the script's letters have Bidi_Class=R or AL.
1029  *
1030  * Returns true for "ar" and "en-Hebr", false for "zh" and "fa-Cyrl".
1031  *
1032  * @param locale input locale ID
1033  * @return true if the locale's script is written right-to-left
1034  * @stable ICU 54
1035  */
1036 U_CAPI UBool U_EXPORT2
1037 uloc_isRightToLeft(const char *locale);
1038 
1039 /**
1040  * enums for the  return value for the character and line orientation
1041  * functions.
1042  * @stable ICU 4.0
1043  */
1044 typedef enum {
1045   ULOC_LAYOUT_LTR   = 0,  /* left-to-right. */
1046   ULOC_LAYOUT_RTL    = 1,  /* right-to-left. */
1047   ULOC_LAYOUT_TTB    = 2,  /* top-to-bottom. */
1048   ULOC_LAYOUT_BTT    = 3,   /* bottom-to-top. */
1049   ULOC_LAYOUT_UNKNOWN
1050 } ULayoutType;
1051 
1052 /**
1053  * Get the layout character orientation for the specified locale.
1054  *
1055  * @param localeId locale name
1056  * @param status Error status
1057  * @return an enum indicating the layout orientation for characters.
1058  * @stable ICU 4.0
1059  */
1060 U_CAPI ULayoutType U_EXPORT2
1061 uloc_getCharacterOrientation(const char* localeId,
1062                              UErrorCode *status);
1063 
1064 /**
1065  * Get the layout line orientation for the specified locale.
1066  *
1067  * @param localeId locale name
1068  * @param status Error status
1069  * @return an enum indicating the layout orientation for lines.
1070  * @stable ICU 4.0
1071  */
1072 U_CAPI ULayoutType U_EXPORT2
1073 uloc_getLineOrientation(const char* localeId,
1074                         UErrorCode *status);
1075 
1076 /**
1077  * Output values which uloc_acceptLanguage() writes to the 'outResult' parameter.
1078  *
1079  * @see uloc_acceptLanguageFromHTTP
1080  * @see uloc_acceptLanguage
1081  * @stable ICU 3.2
1082  */
1083 typedef enum {
1084     /**
1085      * No exact match was found.
1086      * @stable ICU 3.2
1087      */
1088     ULOC_ACCEPT_FAILED   = 0,
1089     /**
1090      * An exact match was found.
1091      * @stable ICU 3.2
1092      */
1093     ULOC_ACCEPT_VALID    = 1,
1094     /**
1095      * A fallback was found. For example, the Accept-Language list includes 'ja_JP'
1096      * and is matched with available locale 'ja'.
1097      * @stable ICU 3.2
1098      */
1099     ULOC_ACCEPT_FALLBACK = 2   /*  */
1100 } UAcceptResult;
1101 
1102 /**
1103  * Based on a HTTP header from a web browser and a list of available locales,
1104  * determine an acceptable locale for the user.
1105  *
1106  * This is a thin wrapper over C++ class LocaleMatcher.
1107  *
1108  * @param result - buffer to accept the result locale
1109  * @param resultAvailable the size of the result buffer.
1110  * @param outResult - An out parameter that contains the fallback status
1111  * @param httpAcceptLanguage - "Accept-Language:" header as per HTTP.
1112  * @param availableLocales - list of available locales to match
1113  * @param status ICU error code. Its input value must pass the U_SUCCESS() test,
1114  *               or else the function returns immediately. Check for U_FAILURE()
1115  *               on output or use with function chaining. (See User Guide for details.)
1116  * @return length needed for the locale.
1117  * @stable ICU 3.2
1118  */
1119 U_CAPI int32_t U_EXPORT2
1120 uloc_acceptLanguageFromHTTP(char *result, int32_t resultAvailable,
1121                             UAcceptResult *outResult,
1122                             const char *httpAcceptLanguage,
1123                             UEnumeration* availableLocales,
1124                             UErrorCode *status);
1125 
1126 /**
1127  * Based on a list of available locales,
1128  * determine an acceptable locale for the user.
1129  *
1130  * This is a thin wrapper over C++ class LocaleMatcher.
1131  *
1132  * @param result - buffer to accept the result locale
1133  * @param resultAvailable the size of the result buffer.
1134  * @param outResult - An out parameter that contains the fallback status
1135  * @param acceptList - list of acceptable languages
1136  * @param acceptListCount - count of acceptList items
1137  * @param availableLocales - list of available locales to match
1138  * @param status ICU error code. Its input value must pass the U_SUCCESS() test,
1139  *               or else the function returns immediately. Check for U_FAILURE()
1140  *               on output or use with function chaining. (See User Guide for details.)
1141  * @return length needed for the locale.
1142  * @stable ICU 3.2
1143  */
1144 U_CAPI int32_t U_EXPORT2
1145 uloc_acceptLanguage(char *result, int32_t resultAvailable,
1146                     UAcceptResult *outResult, const char **acceptList,
1147                     int32_t acceptListCount,
1148                     UEnumeration* availableLocales,
1149                     UErrorCode *status);
1150 
1151 
1152 /**
1153  * Gets the ICU locale ID for the specified Win32 LCID value.
1154  *
1155  * @param hostID the Win32 LCID to translate
1156  * @param locale the output buffer for the ICU locale ID, which will be NUL-terminated
1157  *  if there is room.
1158  * @param localeCapacity the size of the output buffer
1159  * @param status an error is returned if the LCID is unrecognized or the output buffer
1160  *  is too small
1161  * @return actual the actual size of the locale ID, not including NUL-termination
1162  * @stable ICU 3.8
1163  */
1164 U_CAPI int32_t U_EXPORT2
1165 uloc_getLocaleForLCID(uint32_t hostID, char *locale, int32_t localeCapacity,
1166                     UErrorCode *status);
1167 
1168 
1169 /**
1170  * Add the likely subtags for a provided locale ID, per the algorithm described
1171  * in the following CLDR technical report:
1172  *
1173  *   http://www.unicode.org/reports/tr35/#Likely_Subtags
1174  *
1175  * If localeID is already in the maximal form, or there is no data available
1176  * for maximization, it will be copied to the output buffer.  For example,
1177  * "sh" cannot be maximized, since there is no reasonable maximization.
1178  *
1179  * Examples:
1180  *
1181  * "und_Zzzz" maximizes to "en_Latn_US"
1182  *
1183  * "en" maximizes to "en_Latn_US"
1184  *
1185  * "de" maximizes to "de_Latn_DE"
1186  *
1187  * "sr" maximizes to "sr_Cyrl_RS"
1188  *
1189  * "zh_Hani" maximizes to "zh_Hani_CN"
1190  *
1191  *
1192  * @param localeID The locale to maximize
1193  * @param maximizedLocaleID The maximized locale
1194  * @param maximizedLocaleIDCapacity The capacity of the maximizedLocaleID buffer
1195  * @param err Error information if maximizing the locale failed.  If the length
1196  * of the localeID and the null-terminator is greater than the maximum allowed size,
1197  * or the localeId is not well-formed, the error code is U_ILLEGAL_ARGUMENT_ERROR.
1198  * @return The actual buffer size needed for the maximized locale.  If it's
1199  * greater than maximizedLocaleIDCapacity, the returned ID will be truncated.
1200  * On error, the return value is -1.
1201  * @stable ICU 4.0
1202  */
1203 U_CAPI int32_t U_EXPORT2
1204 uloc_addLikelySubtags(const char*    localeID,
1205          char* maximizedLocaleID,
1206          int32_t maximizedLocaleIDCapacity,
1207          UErrorCode* err);
1208 
1209 
1210 /**
1211  * Minimize the subtags for a provided locale ID, per the algorithm described
1212  * in the following CLDR technical report:
1213  *
1214  *   http://www.unicode.org/reports/tr35/#Likely_Subtags
1215  *
1216  * If localeID is already in the minimal form, or there is no data available
1217  * for minimization, it will be copied to the output buffer.  Since the
1218  * minimization algorithm relies on proper maximization, see the comments
1219  * for uloc_addLikelySubtags for reasons why there might not be any data.
1220  *
1221  * Examples:
1222  *
1223  * "en_Latn_US" minimizes to "en"
1224  *
1225  * "de_Latn_US" minimizes to "de"
1226  *
1227  * "sr_Cyrl_RS" minimizes to "sr"
1228  *
1229  * "zh_Hant_TW" minimizes to "zh_TW" (The region is preferred to the
1230  * script, and minimizing to "zh" would imply "zh_Hans_CN".)
1231  *
1232  * @param localeID The locale to minimize
1233  * @param minimizedLocaleID The minimized locale
1234  * @param minimizedLocaleIDCapacity The capacity of the minimizedLocaleID buffer
1235  * @param err Error information if minimizing the locale failed.  If the length
1236  * of the localeID and the null-terminator is greater than the maximum allowed size,
1237  * or the localeId is not well-formed, the error code is U_ILLEGAL_ARGUMENT_ERROR.
1238  * @return The actual buffer size needed for the minimized locale.  If it's
1239  * greater than minimizedLocaleIDCapacity, the returned ID will be truncated.
1240  * On error, the return value is -1.
1241  * @stable ICU 4.0
1242  */
1243 U_CAPI int32_t U_EXPORT2
1244 uloc_minimizeSubtags(const char*    localeID,
1245          char* minimizedLocaleID,
1246          int32_t minimizedLocaleIDCapacity,
1247          UErrorCode* err);
1248 
1249 /**
1250  * Returns a locale ID for the specified BCP47 language tag string.
1251  * If the specified language tag contains any ill-formed subtags,
1252  * the first such subtag and all following subtags are ignored.
1253  * <p>
1254  * This implements the 'Language-Tag' production of BCP 47, and so
1255  * supports legacy language tags (marked as “Type: grandfathered” in BCP 47)
1256  * (regular and irregular) as well as private use language tags.
1257  *
1258  * Private use tags are represented as 'x-whatever',
1259  * and legacy tags are converted to their canonical replacements where they exist.
1260  *
1261  * Note that a few legacy tags have no modern replacement;
1262  * these will be converted using the fallback described in
1263  * the first paragraph, so some information might be lost.
1264  *
1265  * @param langtag   the input BCP47 language tag.
1266  * @param localeID  the output buffer receiving a locale ID for the
1267  *                  specified BCP47 language tag.
1268  * @param localeIDCapacity  the size of the locale ID output buffer.
1269  * @param parsedLength  if not NULL, successfully parsed length
1270  *                      for the input language tag is set.
1271  * @param err       error information if receiving the locald ID
1272  *                  failed.
1273  * @return          the length of the locale ID.
1274  * @stable ICU 4.2
1275  */
1276 U_CAPI int32_t U_EXPORT2
1277 uloc_forLanguageTag(const char* langtag,
1278                     char* localeID,
1279                     int32_t localeIDCapacity,
1280                     int32_t* parsedLength,
1281                     UErrorCode* err);
1282 
1283 /**
1284  * Returns a well-formed language tag for this locale ID.
1285  * <p>
1286  * <b>Note</b>: When <code>strict</code> is false, any locale
1287  * fields which do not satisfy the BCP47 syntax requirement will
1288  * be omitted from the result.  When <code>strict</code> is
1289  * true, this function sets U_ILLEGAL_ARGUMENT_ERROR to the
1290  * <code>err</code> if any locale fields do not satisfy the
1291  * BCP47 syntax requirement.
1292  * @param localeID  the input locale ID
1293  * @param langtag   the output buffer receiving BCP47 language
1294  *                  tag for the locale ID.
1295  * @param langtagCapacity   the size of the BCP47 language tag
1296  *                          output buffer.
1297  * @param strict    boolean value indicating if the function returns
1298  *                  an error for an ill-formed input locale ID.
1299  * @param err       error information if receiving the language
1300  *                  tag failed.
1301  * @return          The length of the BCP47 language tag.
1302  * @stable ICU 4.2
1303  */
1304 U_CAPI int32_t U_EXPORT2
1305 uloc_toLanguageTag(const char* localeID,
1306                    char* langtag,
1307                    int32_t langtagCapacity,
1308                    UBool strict,
1309                    UErrorCode* err);
1310 
1311 /**
1312  * Converts the specified keyword (legacy key, or BCP 47 Unicode locale
1313  * extension key) to the equivalent BCP 47 Unicode locale extension key.
1314  * For example, BCP 47 Unicode locale extension key "co" is returned for
1315  * the input keyword "collation".
1316  * <p>
1317  * When the specified keyword is unknown, but satisfies the BCP syntax,
1318  * then the pointer to the input keyword itself will be returned.
1319  * For example,
1320  * <code>uloc_toUnicodeLocaleKey("ZZ")</code> returns "ZZ".
1321  *
1322  * @param keyword       the input locale keyword (either legacy key
1323  *                      such as "collation" or BCP 47 Unicode locale extension
1324  *                      key such as "co").
1325  * @return              the well-formed BCP 47 Unicode locale extension key,
1326  *                      or NULL if the specified locale keyword cannot be
1327  *                      mapped to a well-formed BCP 47 Unicode locale extension
1328  *                      key.
1329  * @see uloc_toLegacyKey
1330  * @stable ICU 54
1331  */
1332 U_CAPI const char* U_EXPORT2
1333 uloc_toUnicodeLocaleKey(const char* keyword);
1334 
1335 /**
1336  * Converts the specified keyword value (legacy type, or BCP 47
1337  * Unicode locale extension type) to the well-formed BCP 47 Unicode locale
1338  * extension type for the specified keyword (category). For example, BCP 47
1339  * Unicode locale extension type "phonebk" is returned for the input
1340  * keyword value "phonebook", with the keyword "collation" (or "co").
1341  * <p>
1342  * When the specified keyword is not recognized, but the specified value
1343  * satisfies the syntax of the BCP 47 Unicode locale extension type,
1344  * or when the specified keyword allows 'variable' type and the specified
1345  * value satisfies the syntax,  then the pointer to the input type value itself
1346  * will be returned.
1347  * For example,
1348  * <code>uloc_toUnicodeLocaleType("Foo", "Bar")</code> returns "Bar",
1349  * <code>uloc_toUnicodeLocaleType("variableTop", "00A4")</code> returns "00A4".
1350  *
1351  * @param keyword       the locale keyword (either legacy key such as
1352  *                      "collation" or BCP 47 Unicode locale extension
1353  *                      key such as "co").
1354  * @param value         the locale keyword value (either legacy type
1355  *                      such as "phonebook" or BCP 47 Unicode locale extension
1356  *                      type such as "phonebk").
1357  * @return              the well-formed BCP47 Unicode locale extension type,
1358  *                      or NULL if the locale keyword value cannot be mapped to
1359  *                      a well-formed BCP 47 Unicode locale extension type.
1360  * @see uloc_toLegacyType
1361  * @stable ICU 54
1362  */
1363 U_CAPI const char* U_EXPORT2
1364 uloc_toUnicodeLocaleType(const char* keyword, const char* value);
1365 
1366 /**
1367  * Converts the specified keyword (BCP 47 Unicode locale extension key, or
1368  * legacy key) to the legacy key. For example, legacy key "collation" is
1369  * returned for the input BCP 47 Unicode locale extension key "co".
1370  *
1371  * @param keyword       the input locale keyword (either BCP 47 Unicode locale
1372  *                      extension key or legacy key).
1373  * @return              the well-formed legacy key, or NULL if the specified
1374  *                      keyword cannot be mapped to a well-formed legacy key.
1375  * @see toUnicodeLocaleKey
1376  * @stable ICU 54
1377  */
1378 U_CAPI const char* U_EXPORT2
1379 uloc_toLegacyKey(const char* keyword);
1380 
1381 /**
1382  * Converts the specified keyword value (BCP 47 Unicode locale extension type,
1383  * or legacy type or type alias) to the canonical legacy type. For example,
1384  * the legacy type "phonebook" is returned for the input BCP 47 Unicode
1385  * locale extension type "phonebk" with the keyword "collation" (or "co").
1386  * <p>
1387  * When the specified keyword is not recognized, but the specified value
1388  * satisfies the syntax of legacy key, or when the specified keyword
1389  * allows 'variable' type and the specified value satisfies the syntax,
1390  * then the pointer to the input type value itself will be returned.
1391  * For example,
1392  * <code>uloc_toLegacyType("Foo", "Bar")</code> returns "Bar",
1393  * <code>uloc_toLegacyType("vt", "00A4")</code> returns "00A4".
1394  *
1395  * @param keyword       the locale keyword (either legacy keyword such as
1396  *                      "collation" or BCP 47 Unicode locale extension
1397  *                      key such as "co").
1398  * @param value         the locale keyword value (either BCP 47 Unicode locale
1399  *                      extension type such as "phonebk" or legacy keyword value
1400  *                      such as "phonebook").
1401  * @return              the well-formed legacy type, or NULL if the specified
1402  *                      keyword value cannot be mapped to a well-formed legacy
1403  *                      type.
1404  * @see toUnicodeLocaleType
1405  * @stable ICU 54
1406  */
1407 U_CAPI const char* U_EXPORT2
1408 uloc_toLegacyType(const char* keyword, const char* value);
1409 
1410 #endif /*_ULOC*/
1411