icuSources/i18n/unicode/plurfmt.h

   1 /*
   2 *******************************************************************************
   3 * Copyright (C) 2007-2008, International Business Machines Corporation and
   4 * others. All Rights Reserved.
   5 *******************************************************************************
   6 *
   7
   8 * File PLURFMT.H
   9 *
  10 * Modification History:*
  11 *   Date        Name        Description
  12 *
  13 ********************************************************************************
  14 */
  15
  16 #ifndef PLURFMT
  17 #define PLURFMT
  18
  19 #include "unicode/utypes.h"
  20
  21 /**
  22  * \file
  23  * \brief C++ API: PluralFormat object
  24  */
  25
  26 #if !UCONFIG_NO_FORMATTING
  27
  28 #include "unicode/numfmt.h"
  29 #include "unicode/plurrule.h"
  30
  31 U_NAMESPACE_BEGIN
  32
  33 class Hashtable;
  34
  35 /**
  36  * <p>
  37  * <code>PluralFormat</code> supports the creation of internationalized
  38  * messages with plural inflection. It is based on <i>plural
  39  * selection</i>, i.e. the caller specifies messages for each
  40  * plural case that can appear in the users language and the
  41  * <code>PluralFormat</code> selects the appropriate message based on
  42  * the number.
  43  * </p>
  44  * <h4>The Problem of Plural Forms in Internationalized Messages</h4>
  45  * <p>
  46  * Different languages have different ways to inflect
  47  * plurals. Creating internationalized messages that include plural
  48  * forms is only feasible when the framework is able to handle plural
  49  * forms of <i>all</i> languages correctly. <code>ChoiceFormat</code>
  50  * doesn't handle this well, because it attaches a number interval to
  51  * each message and selects the message whose interval contains a
  52  * given number. This can only handle a finite number of
  53  * intervals. But in some languages, like Polish, one plural case
  54  * applies to infinitely many intervals (e.g., paucal applies to
  55  * numbers ending with 2, 3, or 4 except those ending with 12, 13, or
  56  * 14). Thus <code>ChoiceFormat</code> is not adequate.
  57  * </p><p>
  58  * <code>PluralFormat</code> deals with this by breaking the problem
  59  * into two parts:
  60  * <ul>
  61  * <li>It uses <code>PluralRules</code> that can define more complex
  62  *     conditions for a plural case than just a single interval. These plural
  63  *     rules define both what plural cases exist in a language, and to
  64  *     which numbers these cases apply.
  65  * <li>It provides predefined plural rules for many locales. Thus, the programmer
  66  *     need not worry about the plural cases of a language. On the flip side,
  67  *     the localizer does not have to specify the plural cases; he can simply
  68  *     use the predefined keywords. The whole plural formatting of messages can
  69  *     be done using localized patterns from resource bundles.
  70  * </ul>
  71  * </p>
  72  * <h4>Usage of <code>PluralFormat</code></h4>
  73  * <p>
  74  * This discussion assumes that you use <code>PluralFormat</code> with
  75  * a predefined set of plural rules. You can create one using one of
  76  * the constructors that takes a <code>locale</code> object. To
  77  * specify the message pattern, you can either pass it to the
  78  * constructor or set it explicitly using the
  79  * <code>applyPattern()</code> method. The <code>format()</code>
  80  * method takes a number object and selects the message of the
  81  * matching plural case. This message will be returned.
  82  * </p>
  83  * <h5>Patterns and Their Interpretation</h5>
  84  * <p>
  85  * The pattern text defines the message output for each plural case of the
  86  * used locale. The pattern is a sequence of
  87  * <code><i>caseKeyword</i>{<i>message</i>}</code> clauses, separated by white
  88  * space characters. Each clause assigns the message <code><i>message</i></code>
  89  * to the plural case identified by <code><i>caseKeyword</i></code>.
  90  * </p><p>
  91  * You always have to define a message text for the default plural case
  92  * "<code>other</code>" which is contained in every rule set. If the plural
  93  * rules of the <code>PluralFormat</code> object do not contain a plural case
  94  * identified by <code><i>caseKeyword</i></code>, U_DEFAULT_KEYWORD_MISSING
  95  * will be set to status.
  96  * If you do not specify a message text for a particular plural case, the
  97  * message text of the plural case "<code>other</code>" gets assigned to this
  98  * plural case. If you specify more than one message for the same plural case,
  99  * U_DUPLICATE_KEYWORD will be set to status.
 100  * <br/>
 101  * Spaces between <code><i>caseKeyword</i></code> and
 102  * <code><i>message</i></code>  will be ignored; spaces within
 103  * <code><i>message</i></code> will be preserved.
 104  * </p><p>
 105  * The message text for a particular plural case may contain other message
 106  * format patterns. <code>PluralFormat</code> preserves these so that you
 107  * can use the strings produced by <code>PluralFormat</code> with other
 108  * formatters. If you are using <code>PluralFormat</code> inside a
 109  * <code>MessageFormat</code> pattern, <code>MessageFormat</code> will
 110  * automatically evaluate the resulting format pattern.<br/>
 111  * Thus, curly braces (<code>{</code>, <code>}</code>) are <i>only</i> allowed
 112  * in message texts to define a nested format pattern.<br/>
 113  * The pound sign (<code>#</code>) will be interpreted as the number placeholder
 114  * in the message text, if it is not contained in curly braces (to preserve
 115  * <code>NumberFormat</code> patterns). <code>PluralFormat</code> will
 116  * replace each of those pound signs by the number passed to the
 117  * <code>format()</code> method. It will be formatted using a
 118  * <code>NumberFormat</code> for the <code>PluralFormat</code>'s locale. If you
 119  * need special number formatting, you have to explicitly specify a
 120  * <code>NumberFormat</code> for the <code>PluralFormat</code> to use.
 121  * </p>
 122  * Example
 123  * <pre>
 124  * UErrorCode status = U_ZERO_ERROR;
 125  * MessageFormat* msgFmt = new MessageFormat(UnicodeString("{0, plural,
 126  *   one{{0, number, C''est #,##0.0#  fichier}} other {Ce sont # fichiers}} dans la liste."),
 127  *   Locale("fr"), status);
 128  * if (U_FAILURE(status)) {
 129  *     return;
 130  * }
 131  * Formattable args1[] = {(int32_t)0};
 132  * Formattable args2[] = {(int32_t)3};
 133  * FieldPosition ignore(FieldPosition::DONT_CARE);
 134  * UnicodeString result;
 135  * msgFmt->format(args1, 1, result, ignore, status);
 136  * cout << result << endl;
 137  * result.remove();
 138  * msgFmt->format(args2, 1, result, ignore, status);
 139  * cout << result << endl;
 140  * </pre>
 141  * Produces the output:<br/>
 142  * <code>C'est 0,0 fichier dans la liste.</code><br/>
 143  * <code>Ce sont 3 fichiers dans la liste."</code>
 144  * <p>
 145  * <strong>Note:</strong><br/>
 146  *   Currently <code>PluralFormat</code>
 147  *   does not make use of quotes like <code>MessageFormat</code>.
 148  *   If you use plural format strings with <code>MessageFormat</code> and want
 149  *   to use a quote sign "<code>'</code>", you have to write "<code>''</code>".
 150  *   <code>MessageFormat</code> unquotes this pattern and  passes the unquoted
 151  *   pattern to <code>PluralFormat</code>. It's a bit trickier if you use
 152  *   nested formats that do quoting. In the example above, we wanted to insert
 153  *   "<code>'</code>" in the number format pattern. Since
 154  *   <code>NumberFormat</code> supports quotes, we had to insert
 155  *   "<code>''</code>". But since <code>MessageFormat</code> unquotes the
 156  *   pattern before it gets passed to <code>PluralFormat</code>, we have to
 157  *   double these quotes, i.e. write "<code>''''</code>".
 158  * </p>
 159  * <h4>Defining Custom Plural Rules</h4>
 160  * <p>If you need to use <code>PluralFormat</code> with custom rules, you can
 161  * create a <code>PluralRules</code> object and pass it to
 162  * <code>PluralFormat</code>'s constructor. If you also specify a locale in this
 163  * constructor, this locale will be used to format the number in the message
 164  * texts.
 165  * </p><p>
 166  * For more information about <code>PluralRules</code>, see
 167  * {@link PluralRules}.
 168  * </p>
 169  *
 170  * ported from Java
 171  * @draft ICU 4.0
 172  */
 173
 174 class U_I18N_API PluralFormat : public Format {
 175 public:
 176
 177     /**
 178      * Creates a new <code>PluralFormat</code> for the default locale.
 179      * This locale will be used to get the set of plural rules and for standard
 180      * number formatting.
 181      * @param status  output param set to success/failure code on exit, which
 182      *                must not indicate a failure before the function call.
 183      * @draft ICU 4.0
 184      */
 185     PluralFormat(UErrorCode& status);
 186
 187     /**
 188      * Creates a new <code>PluralFormat</code> for a given locale.
 189      * @param locale the <code>PluralFormat</code> will be configured with
 190      *               rules for this locale. This locale will also be used for
 191      *               standard number formatting.
 192      * @param status output param set to success/failure code on exit, which
 193      *               must not indicate a failure before the function call.
 194      * @draft ICU 4.0
 195      */
 196     PluralFormat(const Locale& locale, UErrorCode& status);
 197
 198     /**
 199      * Creates a new <code>PluralFormat</code> for a given set of rules.
 200      * The standard number formatting will be done using the default locale.
 201      * @param rules   defines the behavior of the <code>PluralFormat</code>
 202      *                object.
 203      * @param status  output param set to success/failure code on exit, which
 204      *                must not indicate a failure before the function call.
 205      * @draft ICU 4.0
 206      */
 207     PluralFormat(const PluralRules& rules, UErrorCode& status);
 208
 209     /**
 210      * Creates a new <code>PluralFormat</code> for a given set of rules.
 211      * The standard number formatting will be done using the given locale.
 212      * @param locale  the default number formatting will be done using this
 213      *                locale.
 214      * @param rules   defines the behavior of the <code>PluralFormat</code>
 215      *                object.
 216      * @param status  output param set to success/failure code on exit, which
 217      *                must not indicate a failure before the function call.
 218      * @draft ICU 4.0
 219      */
 220     PluralFormat(const Locale& locale, const PluralRules& rules, UErrorCode& status);
 221
 222     /**
 223      * Creates a new <code>PluralFormat</code> for a given pattern string.
 224      * The default locale will be used to get the set of plural rules and for
 225      * standard number formatting.
 226      * @param  pattern the pattern for this <code>PluralFormat</code>.
 227      *                 errors are returned to status if the pattern is invalid.
 228      * @param status   output param set to success/failure code on exit, which
 229      *                 must not indicate a failure before the function call.
 230      * @draft ICU 4.0
 231      */
 232     PluralFormat(const UnicodeString& pattern, UErrorCode& status);
 233
 234     /**
 235      * Creates a new <code>PluralFormat</code> for a given pattern string and
 236      * locale.
 237      * The locale will be used to get the set of plural rules and for
 238      * standard number formatting.
 239      * @param locale   the <code>PluralFormat</code> will be configured with
 240      *                 rules for this locale. This locale will also be used for
 241      *                 standard number formatting.
 242      * @param pattern  the pattern for this <code>PluralFormat</code>.
 243      *                 errors are returned to status if the pattern is invalid.
 244      * @param status   output param set to success/failure code on exit, which
 245      *                 must not indicate a failure before the function call.
 246      * @draft ICU 4.0
 247      */
 248     PluralFormat(const Locale& locale, const UnicodeString& pattern, UErrorCode& status);
 249
 250     /**
 251      * Creates a new <code>PluralFormat</code> for a given set of rules, a
 252      * pattern and a locale.
 253      * @param rules    defines the behavior of the <code>PluralFormat</code>
 254      *                 object.
 255      * @param pattern  the pattern for this <code>PluralFormat</code>.
 256      *                 errors are returned to status if the pattern is invalid.
 257      * @param status   output param set to success/failure code on exit, which
 258      *                 must not indicate a failure before the function call.
 259      * @draft ICU 4.0
 260      */
 261     PluralFormat(const PluralRules& rules,
 262                  const UnicodeString& pattern,
 263                  UErrorCode& status);
 264
 265     /**
 266      * Creates a new <code>PluralFormat</code> for a given set of rules, a
 267      * pattern and a locale.
 268      * @param locale  the <code>PluralFormat</code> will be configured with
 269      *                rules for this locale. This locale will also be used for
 270      *                standard number formatting.
 271      * @param rules   defines the behavior of the <code>PluralFormat</code>
 272      *                object.
 273      * @param pattern the pattern for this <code>PluralFormat</code>.
 274      *                errors are returned to status if the pattern is invalid.
 275      * @param status  output param set to success/failure code on exit, which
 276      *                must not indicate a failure before the function call.
 277      * @draft ICU 4.0
 278      */
 279     PluralFormat(const Locale& locale,
 280                  const PluralRules& rules,
 281                  const UnicodeString& pattern,
 282                  UErrorCode& status);
 283
 284     /**
 285       * copy constructor.
 286       * @draft ICU 4.0
 287       */
 288     PluralFormat(const PluralFormat& other);
 289
 290     /**
 291      * Destructor.
 292      * @draft ICU 4.0
 293      */
 294     virtual ~PluralFormat();
 295
 296     /**
 297      * Sets the pattern used by this plural format.
 298      * The method parses the pattern and creates a map of format strings
 299      * for the plural rules.
 300      * Patterns and their interpretation are specified in the class description.
 301      *
 302      * @param pattern the pattern for this plural format
 303      *                errors are returned to status if the pattern is invalid.
 304      * @param status  output param set to success/failure code on exit, which
 305      *                must not indicate a failure before the function call.
 306      * @draft ICU 4.0
 307      */
 308     void applyPattern(const UnicodeString& pattern, UErrorCode& status);
 309
 310     /**
 311      * Formats a plural message for a given number.
 312      *
 313      * @param number  a number for which the plural message should be formatted
 314      *                for. If no pattern has been applied to this
 315      *                <code>PluralFormat</code> object yet, the formatted number
 316      *                will be returned.
 317      * @param status  output param set to success/failure code on exit, which
 318      *                must not indicate a failure before the function call.
 319      * @return        the string containing the formatted plural message.
 320      * @draft ICU 4.0
 321      */
 322     UnicodeString format(int32_t number, UErrorCode& status) const;
 323
 324     /**
 325      * Formats a plural message for a given number.
 326      *
 327      * @param number  a number for which the plural message should be formatted
 328      *                for. If no pattern has been applied to this
 329      *                <code>PluralFormat</code> object yet, the formatted number
 330      *                will be returned.
 331      * @param status  output param set to success/failure code on exit, which
 332      *                must not indicate a failure before the function call.
 333      * @return        the string containing the formatted plural message.
 334      * @draft ICU 4.0
 335      */
 336     UnicodeString format(double number, UErrorCode& status) const;
 337
 338     /**
 339      * Formats a plural message for a given number.
 340      *
 341      * @param number   a number for which the plural message should be formatted
 342      *                 for. If no pattern has been applied to this
 343      *                 <code>PluralFormat</code> object yet, the formatted number
 344      *                 will be returned.
 345      * @param appendTo output parameter to receive result.
 346      *                 result is appended to existing contents.
 347      * @param pos      On input: an alignment field, if desired.
 348      *                 On output: the offsets of the alignment field.
 349      * @param status   output param set to success/failure code on exit, which
 350      *                 must not indicate a failure before the function call.
 351      * @return         the string containing the formatted plural message.
 352      * @draft ICU 4.0
 353      */
 354     UnicodeString& format(int32_t number,
 355                           UnicodeString& appendTo,
 356                           FieldPosition& pos,
 357                           UErrorCode& status) const;
 358
 359     /**
 360      * Formats a plural message for a given number.
 361      *
 362      * @param number   a number for which the plural message should be formatted
 363      *                 for. If no pattern has been applied to this
 364      *                 <code>PluralFormat</code> object yet, the formatted number
 365      *                 will be returned.
 366      * @param appendTo output parameter to receive result.
 367      *                 result is appended to existing contents.
 368      * @param pos      On input: an alignment field, if desired.
 369      *                 On output: the offsets of the alignment field.
 370      * @param status   output param set to success/failure code on exit, which
 371      *                 must not indicate a failure before the function call.
 372      * @return         the string containing the formatted plural message.
 373      * @draft ICU 4.0
 374      */
 375     UnicodeString& format(double number,
 376                           UnicodeString& appendTo,
 377                           FieldPosition& pos,
 378                           UErrorCode& status) const;
 379
 380     /**
 381      * Sets the locale used by this <code>PluraFormat</code> object.
 382      * Note: Calling this method resets this <code>PluraFormat</code> object,
 383      *     i.e., a pattern that was applied previously will be removed,
 384      *     and the NumberFormat is set to the default number format for
 385      *     the locale.  The resulting format behaves the same as one
 386      *     constructed from {@link #PluralFormat(locale)}.
 387      * @param locale  the <code>locale</code> to use to configure the formatter.
 388      * @param status  output param set to success/failure code on exit, which
 389      *                must not indicate a failure before the function call.
 390      * @draft ICU 4.0
 391      */
 392     void setLocale(const Locale& locale, UErrorCode& status);
 393
 394     /**
 395       * Sets the number format used by this formatter.  You only need to
 396       * call this if you want a different number format than the default
 397       * formatter for the locale.
 398       * @param format  the number format to use.
 399       * @param status  output param set to success/failure code on exit, which
 400       *                must not indicate a failure before the function call.
 401       * @draft ICU 4.0
 402       */
 403     void setNumberFormat(const NumberFormat* format, UErrorCode& status);
 404
 405     /**
 406        * Assignment operator
 407        *
 408        * @param other    the PluralFormat object to copy from.
 409        * @draft ICU 4.0
 410        */
 411     PluralFormat& operator=(const PluralFormat& other);
 412
 413     /**
 414       * Return true if another object is semantically equal to this one.
 415       *
 416       * @param other    the PluralFormat object to be compared with.
 417       * @return         true if other is semantically equal to this.
 418       * @draft ICU 4.0
 419       */
 420     virtual UBool operator==(const Format& other) const;
 421
 422     /**
 423      * Return true if another object is semantically unequal to this one.
 424      *
 425      * @param other    the PluralFormat object to be compared with.
 426      * @return         true if other is semantically unequal to this.
 427      * @draft ICU 4.0
 428      */
 429     virtual UBool operator!=(const Format& other) const;
 430
 431     /**
 432      * Clones this Format object polymorphically.  The caller owns the
 433      * result and should delete it when done.
 434      * @draft ICU 4.0
 435      */
 436     virtual Format* clone(void) const;
 437
 438     /**
 439     * Redeclared Format method.
 440     *
 441     * @param obj       The object to be formatted into a string.
 442     * @param appendTo  output parameter to receive result.
 443     *                  Result is appended to existing contents.
 444     * @param pos       On input: an alignment field, if desired.
 445     *                  On output: the offsets of the alignment field.
 446     * @param status    output param filled with success/failure status.
 447     * @return          Reference to 'appendTo' parameter.
 448     * @draft ICU 4.0
 449     */
 450    UnicodeString& format(const Formattable& obj,
 451                          UnicodeString& appendTo,
 452                          FieldPosition& pos,
 453                          UErrorCode& status) const;
 454
 455    /**
 456     * Returns the pattern from applyPattern() or constructor().
 457     *
 458     * @param  appendTo  output parameter to receive result.
 459      *                  Result is appended to existing contents.
 460     * @return the UnicodeString with inserted pattern.
 461     * @draft ICU 4.0
 462     */
 463    UnicodeString& toPattern(UnicodeString& appendTo);
 464
 465    /**
 466     * This method is not yet supported by <code>PluralFormat</code>.
 467     * <P>
 468     * Before calling, set parse_pos.index to the offset you want to start
 469     * parsing at in the source. After calling, parse_pos.index is the end of
 470     * the text you parsed. If error occurs, index is unchanged.
 471     * <P>
 472     * When parsing, leading whitespace is discarded (with a successful parse),
 473     * while trailing whitespace is left as is.
 474     * <P>
 475     * See Format::parseObject() for more.
 476     *
 477     * @param source    The string to be parsed into an object.
 478     * @param result    Formattable to be set to the parse result.
 479     *                  If parse fails, return contents are undefined.
 480     * @param parse_pos The position to start parsing at. Upon return
 481     *                  this param is set to the position after the
 482     *                  last character successfully parsed. If the
 483     *                  source is not parsed successfully, this param
 484     *                  will remain unchanged.
 485     * @draft ICU 4.0
 486     */
 487    virtual void parseObject(const UnicodeString& source,
 488                             Formattable& result,
 489                             ParsePosition& parse_pos) const;
 490
 491     /**
 492      * ICU "poor man's RTTI", returns a UClassID for this class.
 493      *
 494      * @draft ICU 4.0
 495      *
 496      */
 497     static UClassID U_EXPORT2 getStaticClassID(void);
 498
 499     /**
 500      * ICU "poor man's RTTI", returns a UClassID for the actual class.
 501      *
 502      * @draft ICU 4.0
 503      */
 504      virtual UClassID getDynamicClassID() const;
 505
 506 private:
 507     typedef enum fmtToken {
 508         none,
 509         tLetter,
 510         tNumber,
 511         tSpace,
 512         tNumberSign,
 513         tLeftBrace,
 514         tRightBrace
 515     }fmtToken;
 516
 517     Locale  locale;
 518     PluralRules* pluralRules;
 519     UnicodeString pattern;
 520     Hashtable  *fParsedValuesHash;
 521     NumberFormat*  numberFormat;
 522     NumberFormat*  replacedNumberFormat;
 523
 524     PluralFormat();   // default constructor not implemented
 525     void init(const PluralRules* rules, const Locale& curlocale, UErrorCode& status);
 526     UBool inRange(UChar ch, fmtToken& type);
 527     UBool checkSufficientDefinition();
 528     void parsingFailure();
 529     UnicodeString insertFormattedNumber(double number,
 530                                         UnicodeString& message,
 531                                         UnicodeString& appendTo,
 532                                         FieldPosition& pos) const;
 533     void copyHashtable(Hashtable *other, UErrorCode& status);
 534 };
 535
 536 U_NAMESPACE_END
 537
 538 #endif /* #if !UCONFIG_NO_FORMATTING */
 539
 540 #endif // _PLURFMT
 541 //eof