2 ********************************************************************************
3 * Copyright (C) 1997-2011, International Business Machines
4 * Corporation and others. All Rights Reserved.
5 ********************************************************************************
9 * Modification History:
11 * Date Name Description
12 * 02/19/97 aliu Converted from java.
13 * 03/20/97 helena Finished first cut of implementation and got rid
14 * of nextDouble/previousDouble and replaced with
16 * 4/10/97 aliu Clean up. Modified to work on AIX.
17 * 8/6/97 nos Removed overloaded constructor, member var 'buffer'.
18 * 07/22/98 stephen Removed operator!= (implemented in Format)
19 ********************************************************************************
25 #include "unicode/utypes.h"
29 * \brief C++ API: Choice Format.
32 #if !UCONFIG_NO_FORMATTING
33 #ifndef U_HIDE_DEPRECATED_API
35 #include "unicode/fieldpos.h"
36 #include "unicode/format.h"
37 #include "unicode/messagepattern.h"
38 #include "unicode/numfmt.h"
39 #include "unicode/unistr.h"
46 * ChoiceFormat converts between ranges of numeric values and strings for those ranges.
47 * The strings must conform to the MessageFormat pattern syntax.
49 * <p><em><code>ChoiceFormat</code> is probably not what you need.
50 * Please use <code>MessageFormat</code>
51 * with <code>plural</code> arguments for proper plural selection,
52 * and <code>select</code> arguments for simple selection among a fixed set of choices!</em></p>
54 * <p>A <code>ChoiceFormat</code> splits
55 * the real number line \htmlonly<code>-∞</code> to
56 * <code>+∞</code>\endhtmlonly into two
57 * or more contiguous ranges. Each range is mapped to a
60 * <p><code>ChoiceFormat</code> was originally intended
61 * for displaying grammatically correct
62 * plurals such as "There is one file." vs. "There are 2 files."
63 * <em>However,</em> plural rules for many languages
64 * are too complex for the capabilities of ChoiceFormat,
65 * and its requirement of specifying the precise rules for each message
66 * is unmanageable for translators.</p>
68 * <p>There are two methods of defining a <code>ChoiceFormat</code>; both
69 * are equivalent. The first is by using a string pattern. This is the
70 * preferred method in most cases. The second method is through direct
71 * specification of the arrays that logically make up the
72 * <code>ChoiceFormat</code>.</p>
74 * <p>Note: Typically, choice formatting is done (if done at all) via <code>MessageFormat</code>
75 * with a <code>choice</code> argument type,
76 * rather than using a stand-alone <code>ChoiceFormat</code>.</p>
78 * <h5>Patterns and Their Interpretation</h5>
80 * <p>The pattern string defines the range boundaries and the strings for each number range.
83 * choiceStyle = number separator message ('|' number separator message)*
84 * number = normal_number | ['-'] \htmlonly∞\endhtmlonly (U+221E, infinity)
85 * normal_number = double value (unlocalized ASCII string)
86 * separator = less_than | less_than_or_equal
88 * less_than_or_equal = '#' | \htmlonly≤\endhtmlonly (U+2264)
89 * message: see {@link MessageFormat}
91 * Pattern_White_Space between syntax elements is ignored, except
92 * around each range's sub-message.</p>
94 * <p>Each numeric sub-range extends from the current range's number
95 * to the next range's number.
96 * The number itself is included in its range if a <code>less_than_or_equal</code> sign is used,
97 * and excluded from its range (and instead included in the previous range)
98 * if a <code>less_than</code> sign is used.</p>
100 * <p>When a <code>ChoiceFormat</code> is constructed from
101 * arrays of numbers, closure flags and strings,
102 * they are interpreted just like
103 * the sequence of <code>(number separator string)</code> in an equivalent pattern string.
104 * <code>closure[i]==TRUE</code> corresponds to a <code>less_than</code> separator sign.
105 * The equivalent pattern string will be constructed automatically.</p>
107 * <p>During formatting, a number is mapped to the first range
108 * where the number is not greater than the range's upper limit.
109 * That range's message string is returned. A NaN maps to the very first range.</p>
111 * <p>During parsing, a range is selected for the longest match of
112 * any range's message. That range's number is returned, ignoring the separator/closure.
113 * Only a simple string match is performed, without parsing of arguments that
114 * might be specified in the message strings.</p>
116 * <p>Note that the first range's number is ignored in formatting
117 * but may be returned from parsing.</p>
121 * <p>Here is an example of two arrays that map the number
122 * <code>1..7</code> to the English day of the week abbreviations
123 * <code>Sun..Sat</code>. No closures array is given; this is the same as
124 * specifying all closures to be <code>FALSE</code>.</p>
126 * <pre> {1,2,3,4,5,6,7},
127 * {"Sun","Mon","Tue","Wed","Thur","Fri","Sat"}</pre>
129 * <p>Here is an example that maps the ranges [-Inf, 1), [1, 1], and (1,
130 * +Inf] to three strings. That is, the number line is split into three
131 * ranges: x < 1.0, x = 1.0, and x > 1.0.
132 * (The round parentheses in the notation above indicate an exclusive boundary,
133 * like the turned bracket in European notation: [-Inf, 1) == [-Inf, 1[ )</p>
136 * {FALSE, FALSE, TRUE},
137 * {"no files", "one file", "many files"}</pre>
139 * <p>Here is an example that shows formatting and parsing: </p>
142 * #include <unicode/choicfmt.h>
143 * #include <unicode/unistr.h>
144 * #include <iostream.h>
146 * int main(int argc, char *argv[]) {
147 * double limits[] = {1,2,3,4,5,6,7};
148 * UnicodeString monthNames[] = {
149 * "Sun","Mon","Tue","Wed","Thu","Fri","Sat"};
150 * ChoiceFormat fmt(limits, monthNames, 7);
153 * for (double x = 1.0; x <= 8.0; x += 1.0) {
154 * fmt.format(x, str);
155 * str.extract(0, str.length(), buf, 256, "");
157 * cout << x << " -> "
165 * <p><em>User subclasses are not supported.</em> While clients may write
166 * subclasses, such code will not necessarily work and will not be
167 * guaranteed to work stably from release to release.
169 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
171 class U_I18N_API ChoiceFormat
: public NumberFormat
{
174 * Constructs a new ChoiceFormat from the pattern string.
176 * @param pattern Pattern used to construct object.
177 * @param status Output param to receive success code. If the
178 * pattern cannot be parsed, set to failure code.
179 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
181 ChoiceFormat(const UnicodeString
& pattern
,
186 * Constructs a new ChoiceFormat with the given limits and message strings.
187 * All closure flags default to <code>FALSE</code>,
188 * equivalent to <code>less_than_or_equal</code> separators.
190 * Copies the limits and formats instead of adopting them.
192 * @param limits Array of limit values.
193 * @param formats Array of formats.
194 * @param count Size of 'limits' and 'formats' arrays.
195 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
197 ChoiceFormat(const double* limits
,
198 const UnicodeString
* formats
,
202 * Constructs a new ChoiceFormat with the given limits, closure flags and message strings.
204 * Copies the limits and formats instead of adopting them.
206 * @param limits Array of limit values
207 * @param closures Array of booleans specifying whether each
208 * element of 'limits' is open or closed. If FALSE, then the
209 * corresponding limit number is a member of its range.
210 * If TRUE, then the limit number belongs to the previous range it.
211 * @param formats Array of formats
212 * @param count Size of 'limits', 'closures', and 'formats' arrays
213 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
215 ChoiceFormat(const double* limits
,
216 const UBool
* closures
,
217 const UnicodeString
* formats
,
223 * @param that ChoiceFormat object to be copied from
224 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
226 ChoiceFormat(const ChoiceFormat
& that
);
229 * Assignment operator.
231 * @param that ChoiceFormat object to be copied
232 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
234 const ChoiceFormat
& operator=(const ChoiceFormat
& that
);
238 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
240 virtual ~ChoiceFormat();
243 * Clones this Format object. The caller owns the
244 * result and must delete it when done.
246 * @return a copy of this object
247 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
249 virtual Format
* clone(void) const;
252 * Returns true if the given Format objects are semantically equal.
253 * Objects of different subclasses are considered unequal.
255 * @param other ChoiceFormat object to be compared
256 * @return true if other is the same as this.
257 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
259 virtual UBool
operator==(const Format
& other
) const;
263 * @param pattern The pattern to be applied.
264 * @param status Output param set to success/failure code on
265 * exit. If the pattern is invalid, this will be
266 * set to a failure result.
267 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
269 virtual void applyPattern(const UnicodeString
& pattern
,
274 * @param pattern The pattern to be applied.
275 * @param parseError Struct to receive information on position
276 * of error if an error is encountered
277 * @param status Output param set to success/failure code on
278 * exit. If the pattern is invalid, this will be
279 * set to a failure result.
280 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
282 virtual void applyPattern(const UnicodeString
& pattern
,
283 UParseError
& parseError
,
288 * @param pattern Output param which will receive the pattern
289 * Previous contents are deleted.
290 * @return A reference to 'pattern'
291 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
293 virtual UnicodeString
& toPattern(UnicodeString
&pattern
) const;
296 * Sets the choices to be used in formatting.
297 * For details see the constructor with the same parameter list.
299 * @param limitsToCopy Contains the top value that you want
300 * parsed with that format,and should be in
301 * ascending sorted order. When formatting X,
302 * the choice will be the i, where limit[i]
303 * <= X < limit[i+1].
304 * @param formatsToCopy The format strings you want to use for each limit.
305 * @param count The size of the above arrays.
306 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
308 virtual void setChoices(const double* limitsToCopy
,
309 const UnicodeString
* formatsToCopy
,
313 * Sets the choices to be used in formatting.
314 * For details see the constructor with the same parameter list.
316 * @param limits Array of limits
317 * @param closures Array of limit booleans
318 * @param formats Array of format string
319 * @param count The size of the above arrays
320 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
322 virtual void setChoices(const double* limits
,
323 const UBool
* closures
,
324 const UnicodeString
* formats
,
328 * Returns NULL and 0.
329 * Before ICU 4.8, this used to return the choice limits array.
331 * @param count Will be set to 0.
333 * @deprecated ICU 4.8 Use the MessagePattern class to analyze a ChoiceFormat pattern.
335 virtual const double* getLimits(int32_t& count
) const;
338 * Returns NULL and 0.
339 * Before ICU 4.8, this used to return the limit booleans array.
341 * @param count Will be set to 0.
343 * @deprecated ICU 4.8 Use the MessagePattern class to analyze a ChoiceFormat pattern.
345 virtual const UBool
* getClosures(int32_t& count
) const;
348 * Returns NULL and 0.
349 * Before ICU 4.8, this used to return the array of choice strings.
351 * @param count Will be set to 0.
353 * @deprecated ICU 4.8 Use the MessagePattern class to analyze a ChoiceFormat pattern.
355 virtual const UnicodeString
* getFormats(int32_t& count
) const;
358 using NumberFormat::format
;
361 * Formats a double number using this object's choices.
363 * @param number The value to be formatted.
364 * @param appendTo Output parameter to receive result.
365 * Result is appended to existing contents.
366 * @param pos On input: an alignment field, if desired.
367 * On output: the offsets of the alignment field.
368 * @return Reference to 'appendTo' parameter.
369 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
371 virtual UnicodeString
& format(double number
,
372 UnicodeString
& appendTo
,
373 FieldPosition
& pos
) const;
375 * Formats an int32_t number using this object's choices.
377 * @param number The value to be formatted.
378 * @param appendTo Output parameter to receive result.
379 * Result is appended to existing contents.
380 * @param pos On input: an alignment field, if desired.
381 * On output: the offsets of the alignment field.
382 * @return Reference to 'appendTo' parameter.
383 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
385 virtual UnicodeString
& format(int32_t number
,
386 UnicodeString
& appendTo
,
387 FieldPosition
& pos
) const;
390 * Formats an int64_t number using this object's choices.
392 * @param number The value to be formatted.
393 * @param appendTo Output parameter to receive result.
394 * Result is appended to existing contents.
395 * @param pos On input: an alignment field, if desired.
396 * On output: the offsets of the alignment field.
397 * @return Reference to 'appendTo' parameter.
398 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
400 virtual UnicodeString
& format(int64_t number
,
401 UnicodeString
& appendTo
,
402 FieldPosition
& pos
) const;
405 * Formats an array of objects using this object's choices.
407 * @param objs The array of objects to be formatted.
408 * @param cnt The size of objs.
409 * @param appendTo Output parameter to receive result.
410 * Result is appended to existing contents.
411 * @param pos On input: an alignment field, if desired.
412 * On output: the offsets of the alignment field.
413 * @param success Output param set to success/failure code on
415 * @return Reference to 'appendTo' parameter.
416 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
418 virtual UnicodeString
& format(const Formattable
* objs
,
420 UnicodeString
& appendTo
,
422 UErrorCode
& success
) const;
424 * Formats an object using this object's choices.
427 * @param obj The object to be formatted.
428 * @param appendTo Output parameter to receive result.
429 * Result is appended to existing contents.
430 * @param pos On input: an alignment field, if desired.
431 * On output: the offsets of the alignment field.
432 * @param status Output param set to success/failure code on
434 * @return Reference to 'appendTo' parameter.
435 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
437 virtual UnicodeString
& format(const Formattable
& obj
,
438 UnicodeString
& appendTo
,
440 UErrorCode
& status
) const;
443 * Redeclared NumberFormat method.
445 * @param obj The object to be formatted.
446 * @param appendTo Output parameter to receive result.
447 * Result is appended to existing contents.
448 * @param status Output param set to success/failure code on
450 * @return Reference to 'appendTo' parameter.
451 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
453 UnicodeString
& format(const Formattable
& obj
,
454 UnicodeString
& appendTo
,
455 UErrorCode
& status
) const;
458 * Redeclared NumberFormat method.
459 * Formats a double number. These methods call the NumberFormat
460 * pure virtual format() methods with the default FieldPosition.
462 * @param number The value to be formatted.
463 * @param appendTo Output parameter to receive result.
464 * Result is appended to existing contents.
465 * @return Reference to 'appendTo' parameter.
466 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
468 UnicodeString
& format( double number
,
469 UnicodeString
& appendTo
) const;
472 * Redeclared NumberFormat method.
473 * Formats an int32_t number. These methods call the NumberFormat
474 * pure virtual format() methods with the default FieldPosition.
476 * @param number The value to be formatted.
477 * @param appendTo Output parameter to receive result.
478 * Result is appended to existing contents.
479 * @return Reference to 'appendTo' parameter.
480 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
482 UnicodeString
& format( int32_t number
,
483 UnicodeString
& appendTo
) const;
486 * Looks for the longest match of any message string on the input text and,
487 * if there is a match, sets the result object to the corresponding range's number.
489 * If no string matches, then the parsePosition is unchanged.
491 * @param text The text to be parsed.
492 * @param result Formattable to be set to the parse result.
493 * If parse fails, return contents are undefined.
494 * @param parsePosition The position to start parsing at on input.
495 * On output, moved to after the last successfully
496 * parse character. On parse failure, does not change.
497 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
499 virtual void parse(const UnicodeString
& text
,
501 ParsePosition
& parsePosition
) const;
504 * Looks for the longest match of any message string on the input text and,
505 * if there is a match, sets the result object to the corresponding range's number.
507 * If no string matches, then the UErrorCode is set to U_INVALID_FORMAT_ERROR.
509 * @param text The text to be parsed.
510 * @param result Formattable to be set to the parse result.
511 * If parse fails, return contents are undefined.
512 * @param status Output param with the formatted string.
513 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
515 virtual void parse(const UnicodeString
& text
,
517 UErrorCode
& status
) const;
520 * Returns a unique class ID POLYMORPHICALLY. Part of ICU's "poor man's RTTI".
522 * @return The class ID for this object. All objects of a
523 * given class have the same class ID. Objects of
524 * other classes have different class IDs.
525 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
527 virtual UClassID
getDynamicClassID(void) const;
530 * Returns the class ID for this class. This is useful only for
531 * comparing to a return value from getDynamicClassID(). For example:
533 * . Base* polymorphic_pointer = createPolymorphicObject();
534 * . if (polymorphic_pointer->getDynamicClassID() ==
535 * . Derived::getStaticClassID()) ...
537 * @return The class ID for all objects of this class.
538 * @deprecated ICU 49 Use MessageFormat instead, with plural and select arguments.
540 static UClassID U_EXPORT2
getStaticClassID(void);
544 * Converts a double value to a string.
545 * @param value the double number to be converted.
546 * @param string the result string.
547 * @return the converted string.
549 static UnicodeString
& dtos(double value
, UnicodeString
& string
);
551 ChoiceFormat(); // default constructor not implemented
554 * Construct a new ChoiceFormat with the limits and the corresponding formats
555 * based on the pattern.
557 * @param newPattern Pattern used to construct object.
558 * @param parseError Struct to receive information on position
559 * of error if an error is encountered.
560 * @param status Output param to receive success code. If the
561 * pattern cannot be parsed, set to failure code.
563 ChoiceFormat(const UnicodeString
& newPattern
,
564 UParseError
& parseError
,
567 friend class MessageFormat
;
569 virtual void setChoices(const double* limits
,
570 const UBool
* closures
,
571 const UnicodeString
* formats
,
573 UErrorCode
&errorCode
);
576 * Finds the ChoiceFormat sub-message for the given number.
577 * @param pattern A MessagePattern.
578 * @param partIndex the index of the first ChoiceFormat argument style part.
579 * @param number a number to be mapped to one of the ChoiceFormat argument's intervals
580 * @return the sub-message start part index.
582 static int32_t findSubMessage(const MessagePattern
&pattern
, int32_t partIndex
, double number
);
584 static double parseArgument(
585 const MessagePattern
&pattern
, int32_t partIndex
,
586 const UnicodeString
&source
, ParsePosition
&pos
);
589 * Matches the pattern string from the end of the partIndex to
590 * the beginning of the limitPartIndex,
591 * including all syntax except SKIP_SYNTAX,
592 * against the source string starting at sourceOffset.
593 * If they match, returns the length of the source string match.
594 * Otherwise returns -1.
596 static int32_t matchStringUntilLimitPart(
597 const MessagePattern
&pattern
, int32_t partIndex
, int32_t limitPartIndex
,
598 const UnicodeString
&source
, int32_t sourceOffset
);
601 * Some of the ChoiceFormat constructors do not have a UErrorCode paramater.
602 * We need _some_ way to provide one for the MessagePattern constructor.
603 * Alternatively, the MessagePattern could be a pointer field, but that is
606 UErrorCode constructorErrorCode
;
609 * The MessagePattern which contains the parsed structure of the pattern string.
611 * Starting with ICU 4.8, the MessagePattern contains a sequence of
612 * numeric/selector/message parts corresponding to the parsed pattern.
613 * For details see the MessagePattern class API docs.
615 MessagePattern msgPattern
;
618 * Docs & fields from before ICU 4.8, before MessagePattern was used.
619 * Commented out, and left only for explanation of semantics.
621 * Each ChoiceFormat divides the range -Inf..+Inf into fCount
622 * intervals. The intervals are:
624 * 0: fChoiceLimits[0]..fChoiceLimits[1]
625 * 1: fChoiceLimits[1]..fChoiceLimits[2]
627 * fCount-2: fChoiceLimits[fCount-2]..fChoiceLimits[fCount-1]
628 * fCount-1: fChoiceLimits[fCount-1]..+Inf
630 * Interval 0 is special; during formatting (mapping numbers to
631 * strings), it also contains all numbers less than
632 * fChoiceLimits[0], as well as NaN values.
634 * Interval i maps to and from string fChoiceFormats[i]. When
635 * parsing (mapping strings to numbers), then intervals map to
636 * their lower limit, that is, interval i maps to fChoiceLimit[i].
638 * The intervals may be closed, half open, or open. This affects
639 * formatting but does not affect parsing. Interval i is affected
640 * by fClosures[i] and fClosures[i+1]. If fClosures[i]
641 * is FALSE, then the value fChoiceLimits[i] is in interval i.
642 * That is, intervals i and i are:
644 * i-1: ... x < fChoiceLimits[i]
645 * i: fChoiceLimits[i] <= x ...
647 * If fClosures[i] is TRUE, then the value fChoiceLimits[i] is
648 * in interval i-1. That is, intervals i-1 and i are:
650 * i-1: ... x <= fChoiceLimits[i]
651 * i: fChoiceLimits[i] < x ...
653 * Because of the nature of interval 0, fClosures[0] has no
656 // double* fChoiceLimits;
658 // UnicodeString* fChoiceFormats;
662 inline UnicodeString
&
663 ChoiceFormat::format(const Formattable
& obj
,
664 UnicodeString
& appendTo
,
665 UErrorCode
& status
) const {
666 // Don't use Format:: - use immediate base class only,
667 // in case immediate base modifies behavior later.
668 return NumberFormat::format(obj
, appendTo
, status
);
671 inline UnicodeString
&
672 ChoiceFormat::format(double number
,
673 UnicodeString
& appendTo
) const {
674 return NumberFormat::format(number
, appendTo
);
677 inline UnicodeString
&
678 ChoiceFormat::format(int32_t number
,
679 UnicodeString
& appendTo
) const {
680 return NumberFormat::format(number
, appendTo
);
684 #endif // U_HIDE_DEPRECATED_API
685 #endif /* #if !UCONFIG_NO_FORMATTING */