1 // © 2018 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
3 #ifndef __LOCALEBUILDER_H__
4 #define __LOCALEBUILDER_H__
5 
6 #include "unicode/utypes.h"
7 
8 #if U_SHOW_CPLUSPLUS_API
9 
10 #include "unicode/locid.h"
11 #include "unicode/localematcher.h"
12 #include "unicode/stringpiece.h"
13 #include "unicode/uobject.h"
14 
15 /**
16  * \file
17  * \brief C++ API: Builder API for Locale
18  */
19 
20 U_NAMESPACE_BEGIN
21 class CharString;
22 
23 /**
24  * <code>LocaleBuilder</code> is used to build instances of <code>Locale</code>
25  * from values configured by the setters.  Unlike the <code>Locale</code>
26  * constructors, the <code>LocaleBuilder</code> checks if a value configured by a
27  * setter satisfies the syntax requirements defined by the <code>Locale</code>
28  * class.  A <code>Locale</code> object created by a <code>LocaleBuilder</code> is
29  * well-formed and can be transformed to a well-formed IETF BCP 47 language tag
30  * without losing information.
31  *
32  * <p>The following example shows how to create a <code>Locale</code> object
33  * with the <code>LocaleBuilder</code>.
34  * <blockquote>
35  * <pre>
36  *     UErrorCode status = U_ZERO_ERROR;
37  *     Locale aLocale = LocaleBuilder()
38  *                          .setLanguage("sr")
39  *                          .setScript("Latn")
40  *                          .setRegion("RS")
41  *                          .build(status);
42  *     if (U_SUCCESS(status)) {
43  *       // ...
44  *     }
45  * </pre>
46  * </blockquote>
47  *
48  * <p>LocaleBuilders can be reused; <code>clear()</code> resets all
49  * fields to their default values.
50  *
51  * <p>LocaleBuilder tracks errors in an internal UErrorCode. For all setters,
52  * except setLanguageTag and setLocale, LocaleBuilder will return immediately
53  * if the internal UErrorCode is in error state.
54  * To reset internal state and error code, call clear method.
55  * The setLanguageTag and setLocale method will first clear the internal
56  * UErrorCode, then track the error of the validation of the input parameter
57  * into the internal UErrorCode.
58  *
59  * @stable ICU 64
60  */
61 class U_COMMON_API LocaleBuilder : public UObject {
62 public:
63     /**
64      * Constructs an empty LocaleBuilder. The default value of all
65      * fields, extensions, and private use information is the
66      * empty string.
67      *
68      * @stable ICU 64
69      */
70     LocaleBuilder();
71 
72     /**
73      * Destructor
74      * @stable ICU 64
75      */
76     virtual ~LocaleBuilder();
77 
78     /**
79      * Resets the <code>LocaleBuilder</code> to match the provided
80      * <code>locale</code>.  Existing state is discarded.
81      *
82      * <p>All fields of the locale must be well-formed.
83      * <p>This method clears the internal UErrorCode.
84      *
85      * @param locale the locale
86      * @return This builder.
87      *
88      * @stable ICU 64
89      */
90     LocaleBuilder& setLocale(const Locale& locale);
91 
92     /**
93      * Resets the LocaleBuilder to match the provided
94      * [Unicode Locale Identifier](http://www.unicode.org/reports/tr35/tr35.html#unicode_locale_id) .
95      * Discards the existing state.
96      * The empty string causes the builder to be reset, like {@link #clear}.
97      * Legacy language tags (marked as “Type: grandfathered” in BCP 47)
98      * are converted to their canonical form before being processed.
99      * Otherwise, the <code>language tag</code> must be well-formed,
100      * or else the build() method will later report an U_ILLEGAL_ARGUMENT_ERROR.
101      *
102      * <p>This method clears the internal UErrorCode.
103      *
104      * @param tag the language tag, defined as
105      *   [unicode_locale_id](http://www.unicode.org/reports/tr35/tr35.html#unicode_locale_id).
106      * @return This builder.
107      * @stable ICU 64
108      */
109     LocaleBuilder& setLanguageTag(StringPiece tag);
110 
111     /**
112      * Sets the language.  If <code>language</code> is the empty string, the
113      * language in this <code>LocaleBuilder</code> is removed. Otherwise, the
114      * <code>language</code> must be well-formed, or else the build() method will
115      * later report an U_ILLEGAL_ARGUMENT_ERROR.
116      *
117      * <p>The syntax of language value is defined as
118      * [unicode_language_subtag](http://www.unicode.org/reports/tr35/tr35.html#unicode_language_subtag).
119      *
120      * @param language the language
121      * @return This builder.
122      * @stable ICU 64
123      */
124     LocaleBuilder& setLanguage(StringPiece language);
125 
126     /**
127      * Sets the script. If <code>script</code> is the empty string, the script in
128      * this <code>LocaleBuilder</code> is removed.
129      * Otherwise, the <code>script</code> must be well-formed, or else the build()
130      * method will later report an U_ILLEGAL_ARGUMENT_ERROR.
131      *
132      * <p>The script value is a four-letter script code as
133      * [unicode_script_subtag](http://www.unicode.org/reports/tr35/tr35.html#unicode_script_subtag)
134      * defined by ISO 15924
135      *
136      * @param script the script
137      * @return This builder.
138      * @stable ICU 64
139      */
140     LocaleBuilder& setScript(StringPiece script);
141 
142     /**
143      * Sets the region.  If region is the empty string, the region in this
144      * <code>LocaleBuilder</code> is removed. Otherwise, the <code>region</code>
145      * must be well-formed, or else the build() method will later report an
146      * U_ILLEGAL_ARGUMENT_ERROR.
147      *
148      * <p>The region value is defined by
149      *  [unicode_region_subtag](http://www.unicode.org/reports/tr35/tr35.html#unicode_region_subtag)
150      * as a two-letter ISO 3166 code or a three-digit UN M.49 area code.
151      *
152      * <p>The region value in the <code>Locale</code> created by the
153      * <code>LocaleBuilder</code> is always normalized to upper case.
154      *
155      * @param region the region
156      * @return This builder.
157      * @stable ICU 64
158      */
159     LocaleBuilder& setRegion(StringPiece region);
160 
161     /**
162      * Sets the variant.  If variant is the empty string, the variant in this
163      * <code>LocaleBuilder</code> is removed.  Otherwise, the <code>variant</code>
164      * must be well-formed, or else the build() method will later report an
165      * U_ILLEGAL_ARGUMENT_ERROR.
166      *
167      * <p><b>Note:</b> This method checks if <code>variant</code>
168      * satisfies the
169      * [unicode_variant_subtag](http://www.unicode.org/reports/tr35/tr35.html#unicode_variant_subtag)
170      * syntax requirements, and normalizes the value to lowercase letters. However,
171      * the <code>Locale</code> class does not impose any syntactic
172      * restriction on variant. To set an ill-formed variant, use a Locale constructor.
173      * If there are multiple unicode_variant_subtag, the caller must concatenate
174      * them with '-' as separator (ex: "foobar-fibar").
175      *
176      * @param variant the variant
177      * @return This builder.
178      * @stable ICU 64
179      */
180     LocaleBuilder& setVariant(StringPiece variant);
181 
182     /**
183      * Sets the extension for the given key. If the value is the empty string,
184      * the extension is removed.  Otherwise, the <code>key</code> and
185      * <code>value</code> must be well-formed, or else the build() method will
186      * later report an U_ILLEGAL_ARGUMENT_ERROR.
187      *
188      * <p><b>Note:</b> The key ('u') is used for the Unicode locale extension.
189      * Setting a value for this key replaces any existing Unicode locale key/type
190      * pairs with those defined in the extension.
191      *
192      * <p><b>Note:</b> The key ('x') is used for the private use code. To be
193      * well-formed, the value for this key needs only to have subtags of one to
194      * eight alphanumeric characters, not two to eight as in the general case.
195      *
196      * @param key the extension key
197      * @param value the extension value
198      * @return This builder.
199      * @stable ICU 64
200      */
201     LocaleBuilder& setExtension(char key, StringPiece value);
202 
203     /**
204      * Sets the Unicode locale keyword type for the given key. If the type
205      * StringPiece is constructed with a nullptr, the keyword is removed.
206      * If the type is the empty string, the keyword is set without type subtags.
207      * Otherwise, the key and type must be well-formed, or else the build()
208      * method will later report an U_ILLEGAL_ARGUMENT_ERROR.
209      *
210      * <p>Keys and types are converted to lower case.
211      *
212      * <p><b>Note</b>:Setting the 'u' extension via {@link #setExtension}
213      * replaces all Unicode locale keywords with those defined in the
214      * extension.
215      *
216      * @param key the Unicode locale key
217      * @param type the Unicode locale type
218      * @return This builder.
219      * @stable ICU 64
220      */
221     LocaleBuilder& setUnicodeLocaleKeyword(
222         StringPiece key, StringPiece type);
223 
224     /**
225      * Adds a unicode locale attribute, if not already present, otherwise
226      * has no effect.  The attribute must not be empty string and must be
227      * well-formed or U_ILLEGAL_ARGUMENT_ERROR will be set to status
228      * during the build() call.
229      *
230      * @param attribute the attribute
231      * @return This builder.
232      * @stable ICU 64
233      */
234     LocaleBuilder& addUnicodeLocaleAttribute(StringPiece attribute);
235 
236     /**
237      * Removes a unicode locale attribute, if present, otherwise has no
238      * effect.  The attribute must not be empty string and must be well-formed
239      * or U_ILLEGAL_ARGUMENT_ERROR will be set to status during the build() call.
240      *
241      * <p>Attribute comparison for removal is case-insensitive.
242      *
243      * @param attribute the attribute
244      * @return This builder.
245      * @stable ICU 64
246      */
247     LocaleBuilder& removeUnicodeLocaleAttribute(StringPiece attribute);
248 
249     /**
250      * Resets the builder to its initial, empty state.
251      * <p>This method clears the internal UErrorCode.
252      *
253      * @return this builder
254      * @stable ICU 64
255      */
256     LocaleBuilder& clear();
257 
258     /**
259      * Resets the extensions to their initial, empty state.
260      * Language, script, region and variant are unchanged.
261      *
262      * @return this builder
263      * @stable ICU 64
264      */
265     LocaleBuilder& clearExtensions();
266 
267     /**
268      * Returns an instance of <code>Locale</code> created from the fields set
269      * on this builder.
270      * If any set methods or during the build() call require memory allocation
271      * but fail U_MEMORY_ALLOCATION_ERROR will be set to status.
272      * If any of the fields set by the setters are not well-formed, the status
273      * will be set to U_ILLEGAL_ARGUMENT_ERROR. The state of the builder will
274      * not change after the build() call and the caller is free to keep using
275      * the same builder to build more locales.
276      *
277      * @return a new Locale
278      * @stable ICU 64
279      */
280     Locale build(UErrorCode& status);
281 
282     /**
283      * Sets the UErrorCode if an error occurred while recording sets.
284      * Preserves older error codes in the outErrorCode.
285      * @param outErrorCode Set to an error code that occurred while setting subtags.
286      *                  Unchanged if there is no such error or if outErrorCode
287      *                  already contained an error.
288      * @return true if U_FAILURE(outErrorCode)
289      * @stable ICU 65
290      */
291     UBool copyErrorTo(UErrorCode &outErrorCode) const;
292 
293 private:
294     friend class LocaleMatcher::Result;
295 
296     void copyExtensionsFrom(const Locale& src, UErrorCode& errorCode);
297 
298     UErrorCode status_;
299     char language_[9];
300     char script_[5];
301     char region_[4];
302     CharString *variant_;  // Pointer not object so we need not #include internal charstr.h.
303     icu::Locale *extensions_;  // Pointer not object. Storage for all other fields.
304 
305 };
306 
307 U_NAMESPACE_END
308 
309 #endif /* U_SHOW_CPLUSPLUS_API */
310 
311 #endif  // __LOCALEBUILDER_H__
312