1 # The set of ICU4J Javadoc replacements. 2 3 # This is a version of the upstream UCharacter docs from ICU 58 with some changes: 4 # Removal of paragraphs that make no sense on Android. 5 --type:android.icu.lang.UCharacter 6 /** 7 * {@icuenhanced java.lang.Character}.{@icu _usage_} 8 * 9 * <p>The UCharacter class provides extensions to the {@link java.lang.Character} class. 10 * These extensions provide support for more Unicode properties. 11 * Each ICU release supports the latest version of Unicode available at that time. 12 * 13 * <p>For some time before Java 5 added support for supplementary Unicode code points, 14 * The ICU UCharacter class and many other ICU classes already supported them. 15 * Some UCharacter methods and constants were widened slightly differently than 16 * how the Character class methods and constants were widened later. 17 * In particular, {@link Character#MAX_VALUE} is still a char with the value U+FFFF, 18 * while the {@link UCharacter#MAX_VALUE} is an int with the value U+10FFFF. 19 * 20 * <p>Code points are represented in these API using ints. While it would be 21 * more convenient in Java to have a separate primitive datatype for them, 22 * ints suffice in the meantime. 23 * 24 * <p>Aside from the additions for UTF-16 support, and the updated Unicode 25 * properties, the main differences between UCharacter and Character are: 26 * <ul> 27 * <li> UCharacter is not designed to be a char wrapper and does not have 28 * APIs to which involves management of that single char.<br> 29 * These include: 30 * <ul> 31 * <li> char charValue(), 32 * <li> int compareTo(java.lang.Character, java.lang.Character), etc. 33 * </ul> 34 * <li> UCharacter does not include Character APIs that are deprecated, nor 35 * does it include the Java-specific character information, such as 36 * boolean isJavaIdentifierPart(char ch). 37 * <li> Character maps characters 'A' - 'Z' and 'a' - 'z' to the numeric 38 * values '10' - '35'. UCharacter also does this in digit and 39 * getNumericValue, to adhere to the java semantics of these 40 * methods. New methods unicodeDigit, and 41 * getUnicodeNumericValue do not treat the above code points 42 * as having numeric values. This is a semantic change from ICU4J 1.3.1. 43 * </ul> 44 * <p> 45 * In addition to Java compatibility functions, which calculate derived properties, 46 * this API provides low-level access to the Unicode Character Database. 47 * <p> 48 * Unicode assigns each code point (not just assigned character) values for 49 * many properties. 50 * Most of them are simple boolean flags, or constants from a small enumerated list. 51 * For some properties, values are strings or other relatively more complex types. 52 * <p> 53 * For more information see 54 * <a href="http://www.unicode/org/ucd/">"About the Unicode Character Database"</a> 55 * (http://www.unicode.org/ucd/) 56 * and the <a href="http://www.icu-project.org/userguide/properties.html">ICU 57 * User Guide chapter on Properties</a> 58 * (http://www.icu-project.org/userguide/properties.html). 59 * <p> 60 * There are also functions that provide easy migration from C/POSIX functions 61 * like isblank(). Their use is generally discouraged because the C/POSIX 62 * standards do not define their semantics beyond the ASCII range, which means 63 * that different implementations exhibit very different behavior. 64 * Instead, Unicode properties should be used directly. 65 * <p> 66 * There are also only a few, broad C/POSIX character classes, and they tend 67 * to be used for conflicting purposes. For example, the "isalpha()" class 68 * is sometimes used to determine word boundaries, while a more sophisticated 69 * approach would at least distinguish initial letters from continuation 70 * characters (the latter including combining marks). 71 * (In ICU, BreakIterator is the most sophisticated API for word boundaries.) 72 * Another example: There is no "istitle()" class for titlecase characters. 73 * <p> 74 * ICU 3.4 and later provides API access for all twelve C/POSIX character classes. 75 * ICU implements them according to the Standard Recommendations in 76 * Annex C: Compatibility Properties of UTS #18 Unicode Regular Expressions 77 * (http://www.unicode.org/reports/tr18/#Compatibility_Properties). 78 * <p> 79 * API access for C/POSIX character classes is as follows: 80 * <pre>{@code 81 * - alpha: isUAlphabetic(c) or hasBinaryProperty(c, UProperty.ALPHABETIC) 82 * - lower: isULowercase(c) or hasBinaryProperty(c, UProperty.LOWERCASE) 83 * - upper: isUUppercase(c) or hasBinaryProperty(c, UProperty.UPPERCASE) 84 * - punct: ((1<<getType(c)) & ((1<<DASH_PUNCTUATION)|(1<<START_PUNCTUATION)| 85 * (1<<END_PUNCTUATION)|(1<<CONNECTOR_PUNCTUATION)|(1<<OTHER_PUNCTUATION)| 86 * (1<<INITIAL_PUNCTUATION)|(1<<FINAL_PUNCTUATION)))!=0 87 * - digit: isDigit(c) or getType(c)==DECIMAL_DIGIT_NUMBER 88 * - xdigit: hasBinaryProperty(c, UProperty.POSIX_XDIGIT) 89 * - alnum: hasBinaryProperty(c, UProperty.POSIX_ALNUM) 90 * - space: isUWhiteSpace(c) or hasBinaryProperty(c, UProperty.WHITE_SPACE) 91 * - blank: hasBinaryProperty(c, UProperty.POSIX_BLANK) 92 * - cntrl: getType(c)==CONTROL 93 * - graph: hasBinaryProperty(c, UProperty.POSIX_GRAPH) 94 * - print: hasBinaryProperty(c, UProperty.POSIX_PRINT)}</pre> 95 * <p> 96 * The C/POSIX character classes are also available in UnicodeSet patterns, 97 * using patterns like [:graph:] or \p{graph}. 98 * 99 * <p>{@icunote} There are several ICU (and Java) whitespace functions. 100 * Comparison:<ul> 101 * <li> isUWhiteSpace=UCHAR_WHITE_SPACE: Unicode White_Space property; 102 * most of general categories "Z" (separators) + most whitespace ISO controls 103 * (including no-break spaces, but excluding IS1..IS4 and ZWSP) 104 * <li> isWhitespace: Java isWhitespace; Z + whitespace ISO controls but excluding no-break spaces 105 * <li> isSpaceChar: just Z (including no-break spaces)</ul> 106 * 107 * <p> 108 * This class is not subclassable. 109 * 110 * @author Syn Wee Quek 111 * @stable ICU 2.1 112 * @see com.ibm.icu.lang.UCharacterEnums 113 */ 114 -- 115 116 # Remove ICU class docs that refer to provisional methods. 117 --type:android.icu.util.IslamicCalendar 118 /** 119 * <code>IslamicCalendar</code> is a subclass of <code>Calendar</code> 120 * that that implements the Islamic civil and religious calendars. It 121 * is used as the civil calendar in most of the Arab world and the 122 * liturgical calendar of the Islamic faith worldwide. This calendar 123 * is also known as the "Hijri" calendar, since it starts at the time 124 * of Mohammed's emigration (or "hijra") to Medinah on Thursday, 125 * July 15, 622 AD (Julian). 126 * <p> 127 * The Islamic calendar is strictly lunar, and thus an Islamic year of twelve 128 * lunar months does not correspond to the solar year used by most other 129 * calendar systems, including the Gregorian. An Islamic year is, on average, 130 * about 354 days long, so each successive Islamic year starts about 11 days 131 * earlier in the corresponding Gregorian year. 132 * <p> 133 * Each month of the calendar starts when the new moon's crescent is visible 134 * at sunset. However, in order to keep the time fields in this class 135 * synchronized with those of the other calendars and with local clock time, 136 * we treat days and months as beginning at midnight, 137 * roughly 6 hours after the corresponding sunset. 138 * <p> 139 * There are three main variants of the Islamic calendar in existence. The first 140 * is the <em>civil</em> calendar, which uses a fixed cycle of alternating 29- 141 * and 30-day months, with a leap day added to the last month of 11 out of 142 * every 30 years. This calendar is easily calculated and thus predictable in 143 * advance, so it is used as the civil calendar in a number of Arab countries. 144 * This is the default behavior of a newly-created <code>IslamicCalendar</code> 145 * object. 146 * <p> 147 * The Islamic <em>religious</em> calendar and Saudi Arabia's <em>Umm al-Qura</em> 148 * calendar, however, are based on the <em>observation</em> of the crescent moon. 149 * It is thus affected by the position at which the 150 * observations are made, seasonal variations in the time of sunset, the 151 * eccentricities of the moon's orbit, and even the weather at the observation 152 * site. This makes it impossible to calculate in advance, and it causes the 153 * start of a month in the religious calendar to differ from the civil calendar 154 * by up to three days. 155 * <p> 156 * Using astronomical calculations for the position of the sun and moon, the 157 * moon's illumination, and other factors, it is possible to determine the start 158 * of a lunar month with a fairly high degree of certainty. However, these 159 * calculations are extremely complicated and thus slow, so most algorithms, 160 * including the one used here, are only approximations of the true astronomical 161 * calculations. At present, the approximations used in this class are fairly 162 * simplistic; they will be improved in later versions of the code. 163 * <p> 164 * Like the Islamic religious calendar, <em>Umm al-Qura</em> is also based 165 * on the sighting method of the crescent moon but is standardized by Saudi Arabia. 166 * <p> 167 * The fixed-cycle <em>civil</em> calendar is used. 168 * <p> 169 * This class should not be subclassed.</p> 170 * <p> 171 * IslamicCalendar usually should be instantiated using 172 * {@link com.ibm.icu.util.Calendar#getInstance(ULocale)} passing in a <code>ULocale</code> 173 * with the tag <code>"@calendar=islamic"</code> or <code>"@calendar=islamic-civil"</code> 174 * or <code>"@calendar=islamic-umalqura"</code>.</p> 175 * 176 * @see com.ibm.icu.util.GregorianCalendar 177 * @see com.ibm.icu.util.Calendar 178 * 179 * @author Laura Werner 180 * @author Alan Liu 181 * @stable ICU 2.8 182 */ 183 -- 184 185 # Remove @see reference to non-public class TimeUnitAmount. 186 --type:android.icu.util.TimeUnit 187 /** 188 * Measurement unit for time units. 189 * @see TimeUnit 190 * @author markdavis 191 * @stable ICU 4.0 192 */ 193 -- 194 195 # Remove references to setDefault* methods that are hidden on Android. 196 --type:android.icu.util.TimeZone 197 /** 198 * {@icuenhanced java.util.TimeZone}.{@icu _usage_} 199 * 200 * <p><code>TimeZone</code> represents a time zone offset, and also computes daylight 201 * savings. 202 * 203 * <p>Typically, you get a <code>TimeZone</code> using {@link #getDefault()} 204 * which creates a <code>TimeZone</code> based on the time zone where the program 205 * is running. For example, for a program running in Japan, <code>getDefault</code> 206 * creates a <code>TimeZone</code> object based on Japanese Standard Time. 207 * 208 * <p>You can also get a <code>TimeZone</code> using {@link #getTimeZone(String)} 209 * along with a time zone ID. For instance, the time zone ID for the 210 * U.S. Pacific Time zone is "America/Los_Angeles". So, you can get a 211 * U.S. Pacific Time <code>TimeZone</code> object with: 212 * 213 * <blockquote> 214 * <pre> 215 * TimeZone tz = TimeZone.getTimeZone("America/Los_Angeles"); 216 * </pre> 217 * </blockquote> 218 * You can use the {@link #getAvailableIDs()} method to iterate through 219 * all the supported time zone IDs, or getCanonicalID method to check 220 * if a time zone ID is supported or not. You can then choose a 221 * supported ID to get a <code>TimeZone</code>. 222 * If the time zone you want is not represented by one of the 223 * supported IDs, then you can create a custom time zone ID with 224 * the following syntax: 225 * 226 * <blockquote> 227 * <pre> 228 * GMT[+|-]hh[[:]mm] 229 * </pre> 230 * </blockquote> 231 * 232 * For example, you might specify GMT+14:00 as a custom 233 * time zone ID. The <code>TimeZone</code> that is returned 234 * when you specify a custom time zone ID uses the specified 235 * offset from GMT(=UTC) and does not observe daylight saving 236 * time. For example, you might specify GMT+14:00 as a custom 237 * time zone ID to create a TimeZone representing 14 hours ahead 238 * of GMT (with no daylight saving time). In addition, 239 * <code>getCanonicalID</code> can also be used to 240 * normalize a custom time zone ID. 241 * 242 * <p>For compatibility with JDK 1.1.x, some other three-letter time zone IDs 243 * (such as "PST", "CTT", "AST") are also supported. However, <strong>their 244 * use is deprecated</strong> because the same abbreviation is often used 245 * for multiple time zones (for example, "CST" could be U.S. "Central Standard 246 * Time" and "China Standard Time"), and the Java platform can then only 247 * recognize one of them. 248 * 249 * @see Calendar 250 * @see GregorianCalendar 251 * @see SimpleTimeZone 252 * @author Mark Davis, Deborah Goldsmith, Chen-Lieh Huang, Alan Liu 253 * @stable ICU 2.0 254 */ 255 -- 256 257 # Remove references to {get,set}DefaultTimezoneType methods that are hidden on Android. 258 --field:android.icu.util.TimeZone#TIMEZONE_ICU 259 /** 260 * <strong>[icu]</strong> A time zone implementation type indicating ICU's own TimeZone used by 261 * <code>getTimeZone</code>. 262 */ 263 -- 264 265 # Remove references to {get,set}DefaultTimezoneType methods that are hidden on Android. 266 --field:android.icu.util.TimeZone#TIMEZONE_JDK 267 /** 268 * <strong>[icu]</strong> A time zone implementation type indicating the {@link java.util.TimeZone} 269 * used by <code>getTimeZone</code>. 270 */ 271 -- 272 273 # Remove ICU class docs that refer to provisional methods. 274 --type:android.icu.util.ULocale 275 /** 276 * {@icuenhanced java.util.Locale}.{@icu _usage_} 277 * 278 * A class analogous to {@link java.util.Locale} that provides additional 279 * support for ICU protocol. In ICU 3.0 this class is enhanced to support 280 * RFC 3066 language identifiers. 281 * 282 * <p>Many classes and services in ICU follow a factory idiom, in 283 * which a factory method or object responds to a client request with 284 * an object. The request includes a locale (the <i>requested</i> 285 * locale), and the returned object is constructed using data for that 286 * locale. The system may lack data for the requested locale, in 287 * which case the locale fallback mechanism will be invoked until a 288 * populated locale is found (the <i>valid</i> locale). Furthermore, 289 * even when a populated locale is found (the <i>valid</i> locale), 290 * further fallback may be required to reach a locale containing the 291 * specific data required by the service (the <i>actual</i> locale). 292 * 293 * <p>ULocale performs <b>'normalization'</b> and <b>'canonicalization'</b> of locale ids. 294 * Normalization 'cleans up' ICU locale ids as follows: 295 * <ul> 296 * <li>language, script, country, variant, and keywords are properly cased<br> 297 * (lower, title, upper, upper, and lower case respectively)</li> 298 * <li>hyphens used as separators are converted to underscores</li> 299 * <li>three-letter language and country ids are converted to two-letter 300 * equivalents where available</li> 301 * <li>surrounding spaces are removed from keywords and values</li> 302 * <li>if there are multiple keywords, they are put in sorted order</li> 303 * </ul> 304 * Canonicalization additionally performs the following: 305 * <ul> 306 * <li>POSIX ids are converted to ICU format IDs</li> 307 * <li>'grandfathered' 3066 ids are converted to ICU standard form</li> 308 * <li>'PREEURO' and 'EURO' variants are converted to currency keyword form, 309 * with the currency 310 * id appropriate to the country of the locale (for PREEURO) or EUR (for EURO). 311 * </ul> 312 * All ULocale constructors automatically normalize the locale id. To handle 313 * POSIX ids, <code>canonicalize</code> can be called to convert the id 314 * to canonical form, or the <code>canonicalInstance</code> factory method 315 * can be called. 316 * 317 * <p>Note: The <i>actual</i> locale is returned correctly, but the <i>valid</i> 318 * locale is not, in most cases. 319 * 320 * @see java.util.Locale 321 * @author weiv 322 * @author Alan Liu 323 * @author Ram Viswanadha 324 * @stable ICU 2.8 325 */ 326 -- 327 328 # Remove the reference to ULocale.setDefault() and remove system properties information. 329 --method:android.icu.util.ULocale#getDefault() 330 /** 331 * Returns the current default ULocale. 332 * <p> 333 * The default ULocale is synchronized to the default Java Locale. This method checks 334 * the current default Java Locale and returns an equivalent ULocale. 335 * 336 * @return the default ULocale. 337 * @stable ICU 2.8 338 */ 339 -- 340 341 # Removal of sentence containing link to class that is not exposed in Android API 342 --type:android.icu.text.UnicodeFilter 343 /** 344 * <code>UnicodeFilter</code> defines a protocol for selecting a 345 * subset of the full range (U+0000 to U+FFFF) of Unicode characters. 346 * @stable ICU 2.0 347 */ 348 -- 349 350 # Document that collation rules are omitted. 351 --method:android.icu.text.RuleBasedCollator#getRules() 352 /** 353 * Gets the collation tailoring rules for this RuleBasedCollator. 354 * Equivalent to String getRules(false). 355 * 356 * <p>On Android, the returned string will be empty unless this instance was 357 * constructed using {@link #RuleBasedCollator(String)}. 358 * 359 * @return the collation tailoring rules 360 * @see #getRules(boolean) 361 * @stable ICU 2.8 362 */ 363 -- 364 365 # Deprecate method. 366 --method:android.icu.util.Calendar#computeZoneOffset(long,int) 367 /** 368 * This method can assume EXTENDED_YEAR has been set. 369 * @param millis milliseconds of the date fields (local midnight millis) 370 * @param millisInDay milliseconds of the time fields; may be out 371 * or range. 372 * @return total zone offset (raw + DST) for the given moment 373 * @stable ICU 2.0 374 * @deprecated This method suffers from a potential integer overflow and may be removed or 375 * changed in a future release. See <a href="http://bugs.icu-project.org/trac/ticket/11632"> 376 * ICU ticket #11632</a> for details. 377 */ 378 -- 379 380 # Deprecate method. 381 --method:android.icu.util.Calendar#computeMillisInDay() 382 /** 383 * Compute the milliseconds in the day from the fields. This is a 384 * value from 0 to 23:59:59.999 inclusive, unless fields are out of 385 * range, in which case it can be an arbitrary value. This value 386 * reflects local zone wall time. 387 * @stable ICU 2.0 388 * @deprecated This method suffers from a potential integer overflow and may be removed or 389 * changed in a future release. See <a href="http://bugs.icu-project.org/trac/ticket/11632"> 390 * ICU ticket #11632</a> for details. 391 */ 392 -- 393
