icuSources/i18n/unicode/selfmt.h

   1 /********************************************************************
   2  * COPYRIGHT:
   3  * Copyright (c) 1997-2011, International Business Machines Corporation and
   4  * others. All Rights Reserved.
   5  * Copyright (C) 2010 , Yahoo! Inc.
   6  ********************************************************************
   7  *
   8  * File SELFMT.H
   9  *
  10  * Modification History:
  11  *
  12  *   Date        Name        Description
  13  *   11/11/09    kirtig      Finished first cut of implementation.
  14  ********************************************************************/
  15
  16 #ifndef SELFMT
  17 #define SELFMT
  18
  19 #include "unicode/messagepattern.h"
  20 #include "unicode/numfmt.h"
  21 #include "unicode/utypes.h"
  22
  23 /**
  24  * \file
  25  * \brief C++ API: SelectFormat object
  26  */
  27
  28 #if !UCONFIG_NO_FORMATTING
  29
  30 U_NAMESPACE_BEGIN
  31
  32 class MessageFormat;
  33
  34 /**
  35   * <p><code>SelectFormat</code> supports the creation of  internationalized
  36   * messages by selecting phrases based on keywords. The pattern  specifies
  37   * how to map keywords to phrases and provides a default phrase. The
  38   * object provided to the format method is a string that's matched
  39   * against the keywords. If there is a match, the corresponding phrase
  40   * is selected; otherwise, the default phrase is used.</p>
  41   *
  42   * <h4>Using <code>SelectFormat</code> for Gender Agreement</h4>
  43   *
  44   * <p>Note: Typically, select formatting is done via <code>MessageFormat</code>
  45   * with a <code>select</code> argument type,
  46   * rather than using a stand-alone <code>SelectFormat</code>.</p>
  47   *
  48   * <p>The main use case for the select format is gender based  inflection.
  49   * When names or nouns are inserted into sentences, their gender can  affect pronouns,
  50   * verb forms, articles, and adjectives. Special care needs to be
  51   * taken for the case where the gender cannot be determined.
  52   * The impact varies between languages:</p>
  53   * \htmlonly
  54   * <ul>
  55   * <li>English has three genders, and unknown gender is handled as a  special
  56   * case. Names use the gender of the named person (if known), nouns  referring
  57   * to people use natural gender, and inanimate objects are usually  neutral.
  58   * The gender only affects pronouns: "he", "she", "it", "they".
  59   *
  60   * <li>German differs from English in that the gender of nouns is  rather
  61   * arbitrary, even for nouns referring to people ("M&#x00E4;dchen", girl, is  neutral).
  62   * The gender affects pronouns ("er", "sie", "es"), articles ("der",  "die",
  63   * "das"), and adjective forms ("guter Mann", "gute Frau", "gutes  M&#x00E4;dchen").
  64   *
  65   * <li>French has only two genders; as in German the gender of nouns
  66   * is rather arbitrary - for sun and moon, the genders
  67   * are the opposite of those in German. The gender affects
  68   * pronouns ("il", "elle"), articles ("le", "la"),
  69   * adjective forms ("bon", "bonne"), and sometimes
  70   * verb forms ("all&#x00E9;", "all&#x00E9;e").
  71   *
  72   * <li>Polish distinguishes five genders (or noun classes),
  73   * human masculine, animate non-human masculine, inanimate masculine,
  74   * feminine, and neuter.
  75   * </ul>
  76   * \endhtmlonly
  77   * <p>Some other languages have noun classes that are not related to  gender,
  78   * but similar in grammatical use.
  79   * Some African languages have around 20 noun classes.</p>
  80   *
  81   * <p><b>Note:</b>For the gender of a <i>person</i> in a given sentence,
  82   * we usually need to distinguish only between female, male and other/unknown.</p>
  83   *
  84   * <p>To enable localizers to create sentence patterns that take their
  85   * language's gender dependencies into consideration, software has to  provide
  86   * information about the gender associated with a noun or name to
  87   * <code>MessageFormat</code>.
  88   * Two main cases can be distinguished:</p>
  89   *
  90   * <ul>
  91   * <li>For people, natural gender information should be maintained  for each person.
  92   * Keywords like "male", "female", "mixed" (for groups of people)
  93   * and "unknown" could be used.
  94   *
  95   * <li>For nouns, grammatical gender information should be maintained  for
  96   * each noun and per language, e.g., in resource bundles.
  97   * The keywords "masculine", "feminine", and "neuter" are commonly  used,
  98   * but some languages may require other keywords.
  99   * </ul>
 100   *
 101   * <p>The resulting keyword is provided to <code>MessageFormat</code>  as a
 102   * parameter separate from the name or noun it's associated with. For  example,
 103   * to generate a message such as "Jean went to Paris", three separate  arguments
 104   * would be provided: The name of the person as argument 0, the  gender of
 105   * the person as argument 1, and the name of the city as argument 2.
 106   * The sentence pattern for English, where the gender of the person has
 107   * no impact on this simple sentence, would not refer to argument 1  at all:</p>
 108   *
 109   * <pre>{0} went to {2}.</pre>
 110   *
 111   * <p><b>Note:</b> The entire sentence should be included (and partially repeated)
 112   * inside each phrase. Otherwise translators would have to be trained on how to
 113   * move bits of the sentence in and out of the select argument of a message.
 114   * (The examples below do not follow this recommendation!)</p>
 115   *
 116   * <p>The sentence pattern for French, where the gender of the person affects
 117   * the form of the participle, uses a select format based on argument 1:</p>
 118   *
 119   * \htmlonly<pre>{0} est {1, select, female {all&#x00E9;e} other {all&#x00E9;}} &#x00E0; {2}.</pre>\endhtmlonly
 120   *
 121   * <p>Patterns can be nested, so that it's possible to handle  interactions of
 122   * number and gender where necessary. For example, if the above  sentence should
 123   * allow for the names of several people to be inserted, the  following sentence
 124   * pattern can be used (with argument 0 the list of people's names,
 125   * argument 1 the number of people, argument 2 their combined gender, and
 126   * argument 3 the city name):</p>
 127   *
 128   * \htmlonly
 129   * <pre>{0} {1, plural,
 130   *                 one {est {2, select, female {all&#x00E9;e} other  {all&#x00E9;}}}
 131   *                 other {sont {2, select, female {all&#x00E9;es} other {all&#x00E9;s}}}
 132   *          }&#x00E0; {3}.</pre>
 133   * \endhtmlonly
 134   *
 135   * <h4>Patterns and Their Interpretation</h4>
 136   *
 137   * <p>The <code>SelectFormat</code> pattern string defines the phrase output
 138   * for each user-defined keyword.
 139   * The pattern is a sequence of (keyword, message) pairs.
 140   * A keyword is a "pattern identifier": [^[[:Pattern_Syntax:][:Pattern_White_Space:]]]+</p>
 141   *
 142   * <p>Each message is a MessageFormat pattern string enclosed in {curly braces}.</p>
 143   *
 144   * <p>You always have to define a phrase for the default keyword
 145   * <code>other</code>; this phrase is returned when the keyword
 146   * provided to
 147   * the <code>format</code> method matches no other keyword.
 148   * If a pattern does not provide a phrase for <code>other</code>, the  method
 149   * it's provided to returns the error  <code>U_DEFAULT_KEYWORD_MISSING</code>.
 150   * <br>
 151   * Pattern_White_Space between keywords and messages is ignored.
 152   * Pattern_White_Space within a message is preserved and output.</p>
 153   *
 154   * <p><pre>Example:
 155   * \htmlonly
 156   *
 157   * UErrorCode status = U_ZERO_ERROR;
 158   * MessageFormat *msgFmt = new MessageFormat(UnicodeString("{0} est  {1, select, female {all&#x00E9;e} other {all&#x00E9;}} &#x00E0; Paris."), Locale("fr"),  status);
 159   * if (U_FAILURE(status)) {
 160   *       return;
 161   * }
 162   * FieldPosition ignore(FieldPosition::DONT_CARE);
 163   * UnicodeString result;
 164   *
 165   * char* str1= "Kirti,female";
 166   * Formattable args1[] = {"Kirti","female"};
 167   * msgFmt->format(args1, 2, result, ignore, status);
 168   * cout << "Input is " << str1 << " and result is: " << result << endl;
 169   * delete msgFmt;
 170   *
 171   * \endhtmlonly
 172   * </pre>
 173   * </p>
 174   *
 175   * Produces the output:<br>
 176   * \htmlonly
 177   * <code>Kirti est all&#x00E9;e &#x00E0; Paris.</code>
 178   * \endhtmlonly
 179   *
 180   * @stable ICU 4.4
 181   */
 182
 183 class U_I18N_API SelectFormat : public Format {
 184 public:
 185
 186     /**
 187      * Creates a new <code>SelectFormat</code> for a given pattern string.
 188      * @param  pattern the pattern for this <code>SelectFormat</code>.
 189      *                 errors are returned to status if the pattern is invalid.
 190      * @param status   output param set to success/failure code on exit, which
 191      *                 must not indicate a failure before the function call.
 192      * @stable ICU 4.4
 193      */
 194     SelectFormat(const UnicodeString& pattern, UErrorCode& status);
 195
 196     /**
 197      * copy constructor.
 198      * @stable ICU 4.4
 199      */
 200     SelectFormat(const SelectFormat& other);
 201
 202     /**
 203      * Destructor.
 204      * @stable ICU 4.4
 205      */
 206     virtual ~SelectFormat();
 207
 208     /**
 209      * Sets the pattern used by this select format.
 210      * for the keyword rules.
 211      * Patterns and their interpretation are specified in the class description.
 212      *
 213      * @param pattern the pattern for this select format
 214      *                errors are returned to status if the pattern is invalid.
 215      * @param status  output param set to success/failure code on exit, which
 216      *                must not indicate a failure before the function call.
 217      * @stable ICU 4.4
 218      */
 219     void applyPattern(const UnicodeString& pattern, UErrorCode& status);
 220
 221
 222     using Format::format;
 223
 224     /**
 225      * Selects the phrase for  the given keyword
 226      *
 227      * @param keyword  The keyword that is used to select an alternative.
 228      * @param appendTo output parameter to receive result.
 229      *                 result is appended to existing contents.
 230      * @param pos      On input: an alignment field, if desired.
 231      *                 On output: the offsets of the alignment field.
 232      * @param status  output param set to success/failure code on exit, which
 233      *                 must not indicate a failure before the function call.
 234      * @return         Reference to 'appendTo' parameter.
 235      * @stable ICU 4.4
 236      */
 237     UnicodeString& format(const UnicodeString& keyword,
 238                             UnicodeString& appendTo,
 239                             FieldPosition& pos,
 240                             UErrorCode& status) const;
 241
 242     /**
 243      * Assignment operator
 244      *
 245      * @param other    the SelectFormat object to copy from.
 246      * @stable ICU 4.4
 247      */
 248     SelectFormat& operator=(const SelectFormat& other);
 249
 250     /**
 251      * Return true if another object is semantically equal to this one.
 252      *
 253      * @param other    the SelectFormat object to be compared with.
 254      * @return         true if other is semantically equal to this.
 255      * @stable ICU 4.4
 256      */
 257     virtual UBool operator==(const Format& other) const;
 258
 259     /**
 260      * Return true if another object is semantically unequal to this one.
 261      *
 262      * @param other    the SelectFormat object to be compared with.
 263      * @return         true if other is semantically unequal to this.
 264      * @stable ICU 4.4
 265      */
 266     virtual UBool operator!=(const Format& other) const;
 267
 268     /**
 269      * Clones this Format object polymorphically.  The caller owns the
 270      * result and should delete it when done.
 271      * @stable ICU 4.4
 272      */
 273     virtual Format* clone(void) const;
 274
 275     /**
 276      * Format an object to produce a string.
 277      * This method handles keyword strings.
 278      * If the Formattable object is not a <code>UnicodeString</code>,
 279      * then it returns a failing UErrorCode.
 280      *
 281      * @param obj       A keyword string that is used to select an alternative.
 282      * @param appendTo  output parameter to receive result.
 283      *                  Result is appended to existing contents.
 284      * @param pos       On input: an alignment field, if desired.
 285      *                  On output: the offsets of the alignment field.
 286      * @param status    output param filled with success/failure status.
 287      * @return          Reference to 'appendTo' parameter.
 288      * @stable ICU 4.4
 289      */
 290     UnicodeString& format(const Formattable& obj,
 291                          UnicodeString& appendTo,
 292                          FieldPosition& pos,
 293                          UErrorCode& status) const;
 294
 295     /**
 296      * Returns the pattern from applyPattern() or constructor.
 297      *
 298      * @param  appendTo  output parameter to receive result.
 299      *                  Result is appended to existing contents.
 300      * @return the UnicodeString with inserted pattern.
 301      * @stable ICU 4.4
 302      */
 303     UnicodeString& toPattern(UnicodeString& appendTo);
 304
 305     /**
 306      * This method is not yet supported by <code>SelectFormat</code>.
 307      * <P>
 308      * Before calling, set parse_pos.index to the offset you want to start
 309      * parsing at in the source. After calling, parse_pos.index is the end of
 310      * the text you parsed. If error occurs, index is unchanged.
 311      * <P>
 312      * When parsing, leading whitespace is discarded (with a successful parse),
 313      * while trailing whitespace is left as is.
 314      * <P>
 315      * See Format::parseObject() for more.
 316      *
 317      * @param source     The string to be parsed into an object.
 318      * @param result     Formattable to be set to the parse result.
 319      *     If parse fails, return contents are undefined.
 320      * @param parse_pos The position to start parsing at. Upon return
 321      *     this param is set to the position after the
 322      *     last character successfully parsed. If the
 323      *     source is not parsed successfully, this param
 324      *     will remain unchanged.
 325      * @stable ICU 4.4
 326      */
 327     virtual void parseObject(const UnicodeString& source,
 328                             Formattable& result,
 329                             ParsePosition& parse_pos) const;
 330
 331     /**
 332      * ICU "poor man's RTTI", returns a UClassID for this class.
 333      * @stable ICU 4.4
 334      */
 335     static UClassID U_EXPORT2 getStaticClassID(void);
 336
 337     /**
 338      * ICU "poor man's RTTI", returns a UClassID for the actual class.
 339      * @stable ICU 4.4
 340      */
 341     virtual UClassID getDynamicClassID() const;
 342
 343 private:
 344     friend class MessageFormat;
 345
 346     SelectFormat();   // default constructor not implemented.
 347
 348     /**
 349      * Finds the SelectFormat sub-message for the given keyword, or the "other" sub-message.
 350      * @param pattern A MessagePattern.
 351      * @param partIndex the index of the first SelectFormat argument style part.
 352      * @param keyword a keyword to be matched to one of the SelectFormat argument's keywords.
 353      * @param ec Error code.
 354      * @return the sub-message start part index.
 355      */
 356     static int32_t findSubMessage(const MessagePattern& pattern, int32_t partIndex,
 357                                   const UnicodeString& keyword, UErrorCode& ec);
 358
 359     MessagePattern msgPattern;
 360 };
 361
 362 U_NAMESPACE_END
 363
 364 #endif /* #if !UCONFIG_NO_FORMATTING */
 365
 366 #endif // _SELFMT
 367 //eof