icuSources/common/unicode/normlzr.h

   1 // © 2016 and later: Unicode, Inc. and others.
   2 // License & terms of use: http://www.unicode.org/copyright.html
   3 /*
   4  ********************************************************************
   5  * COPYRIGHT:
   6  * Copyright (c) 1996-2015, International Business Machines Corporation and
   7  * others. All Rights Reserved.
   8  ********************************************************************
   9  */
  10
  11 #ifndef NORMLZR_H
  12 #define NORMLZR_H
  13
  14 #include "unicode/utypes.h"
  15
  16 /**
  17  * \file
  18  * \brief C++ API: Unicode Normalization
  19  */
  20
  21 #if !UCONFIG_NO_NORMALIZATION
  22
  23 #include "unicode/chariter.h"
  24 #include "unicode/normalizer2.h"
  25 #include "unicode/unistr.h"
  26 #include "unicode/unorm.h"
  27 #include "unicode/uobject.h"
  28
  29 #if U_SHOW_CPLUSPLUS_API
  30 U_NAMESPACE_BEGIN
  31 /**
  32  * Old Unicode normalization API.
  33  *
  34  * This API has been replaced by the Normalizer2 class and is only available
  35  * for backward compatibility. This class simply delegates to the Normalizer2 class.
  36  * There is one exception: The new API does not provide a replacement for Normalizer::compare().
  37  *
  38  * The Normalizer class supports the standard normalization forms described in
  39  * <a href="http://www.unicode.org/unicode/reports/tr15/" target="unicode">
  40  * Unicode Standard Annex #15: Unicode Normalization Forms</a>.
  41  *
  42  * The Normalizer class consists of two parts:
  43  * - static functions that normalize strings or test if strings are normalized
  44  * - a Normalizer object is an iterator that takes any kind of text and
  45  *   provides iteration over its normalized form
  46  *
  47  * The Normalizer class is not suitable for subclassing.
  48  *
  49  * For basic information about normalization forms and details about the C API
  50  * please see the documentation in unorm.h.
  51  *
  52  * The iterator API with the Normalizer constructors and the non-static functions
  53  * use a CharacterIterator as input. It is possible to pass a string which
  54  * is then internally wrapped in a CharacterIterator.
  55  * The input text is not normalized all at once, but incrementally where needed
  56  * (providing efficient random access).
  57  * This allows to pass in a large text but spend only a small amount of time
  58  * normalizing a small part of that text.
  59  * However, if the entire text is normalized, then the iterator will be
  60  * slower than normalizing the entire text at once and iterating over the result.
  61  * A possible use of the Normalizer iterator is also to report an index into the
  62  * original text that is close to where the normalized characters come from.
  63  *
  64  * <em>Important:</em> The iterator API was cleaned up significantly for ICU 2.0.
  65  * The earlier implementation reported the getIndex() inconsistently,
  66  * and previous() could not be used after setIndex(), next(), first(), and current().
  67  *
  68  * Normalizer allows to start normalizing from anywhere in the input text by
  69  * calling setIndexOnly(), first(), or last().
  70  * Without calling any of these, the iterator will start at the beginning of the text.
  71  *
  72  * At any time, next() returns the next normalized code point (UChar32),
  73  * with post-increment semantics (like CharacterIterator::next32PostInc()).
  74  * previous() returns the previous normalized code point (UChar32),
  75  * with pre-decrement semantics (like CharacterIterator::previous32()).
  76  *
  77  * current() returns the current code point
  78  * (respectively the one at the newly set index) without moving
  79  * the getIndex(). Note that if the text at the current position
  80  * needs to be normalized, then these functions will do that.
  81  * (This is why current() is not const.)
  82  * It is more efficient to call setIndexOnly() instead, which does not
  83  * normalize.
  84  *
  85  * getIndex() always refers to the position in the input text where the normalized
  86  * code points are returned from. It does not always change with each returned
  87  * code point.
  88  * The code point that is returned from any of the functions
  89  * corresponds to text at or after getIndex(), according to the
  90  * function's iteration semantics (post-increment or pre-decrement).
  91  *
  92  * next() returns a code point from at or after the getIndex()
  93  * from before the next() call. After the next() call, the getIndex()
  94  * might have moved to where the next code point will be returned from
  95  * (from a next() or current() call).
  96  * This is semantically equivalent to array access with array[index++]
  97  * (post-increment semantics).
  98  *
  99  * previous() returns a code point from at or after the getIndex()
 100  * from after the previous() call.
 101  * This is semantically equivalent to array access with array[--index]
 102  * (pre-decrement semantics).
 103  *
 104  * Internally, the Normalizer iterator normalizes a small piece of text
 105  * starting at the getIndex() and ending at a following "safe" index.
 106  * The normalized results is stored in an internal string buffer, and
 107  * the code points are iterated from there.
 108  * With multiple iteration calls, this is repeated until the next piece
 109  * of text needs to be normalized, and the getIndex() needs to be moved.
 110  *
 111  * The following "safe" index, the internal buffer, and the secondary
 112  * iteration index into that buffer are not exposed on the API.
 113  * This also means that it is currently not practical to return to
 114  * a particular, arbitrary position in the text because one would need to
 115  * know, and be able to set, in addition to the getIndex(), at least also the
 116  * current index into the internal buffer.
 117  * It is currently only possible to observe when getIndex() changes
 118  * (with careful consideration of the iteration semantics),
 119  * at which time the internal index will be 0.
 120  * For example, if getIndex() is different after next() than before it,
 121  * then the internal index is 0 and one can return to this getIndex()
 122  * later with setIndexOnly().
 123  *
 124  * Note: While the setIndex() and getIndex() refer to indices in the
 125  * underlying Unicode input text, the next() and previous() methods
 126  * iterate through characters in the normalized output.
 127  * This means that there is not necessarily a one-to-one correspondence
 128  * between characters returned by next() and previous() and the indices
 129  * passed to and returned from setIndex() and getIndex().
 130  * It is for this reason that Normalizer does not implement the CharacterIterator interface.
 131  *
 132  * @author Laura Werner, Mark Davis, Markus Scherer
 133  * @stable ICU 2.0
 134  */
 135 class U_COMMON_API Normalizer : public UObject {
 136 public:
 137 #ifndef U_HIDE_DEPRECATED_API
 138   /**
 139    * If DONE is returned from an iteration function that returns a code point,
 140    * then there are no more normalization results available.
 141    * @deprecated ICU 56 Use Normalizer2 instead.
 142    */
 143   enum {
 144       DONE=0xffff
 145   };
 146
 147   // Constructors
 148
 149   /**
 150    * Creates a new <code>Normalizer</code> object for iterating over the
 151    * normalized form of a given string.
 152    * <p>
 153    * @param str   The string to be normalized.  The normalization
 154    *              will start at the beginning of the string.
 155    *
 156    * @param mode  The normalization mode.
 157    * @deprecated ICU 56 Use Normalizer2 instead.
 158    */
 159   Normalizer(const UnicodeString& str, UNormalizationMode mode);
 160
 161   /**
 162    * Creates a new <code>Normalizer</code> object for iterating over the
 163    * normalized form of a given string.
 164    * <p>
 165    * @param str   The string to be normalized.  The normalization
 166    *              will start at the beginning of the string.
 167    *
 168    * @param length Length of the string, or -1 if NUL-terminated.
 169    * @param mode  The normalization mode.
 170    * @deprecated ICU 56 Use Normalizer2 instead.
 171    */
 172   Normalizer(ConstChar16Ptr str, int32_t length, UNormalizationMode mode);
 173
 174   /**
 175    * Creates a new <code>Normalizer</code> object for iterating over the
 176    * normalized form of the given text.
 177    * <p>
 178    * @param iter  The input text to be normalized.  The normalization
 179    *              will start at the beginning of the string.
 180    *
 181    * @param mode  The normalization mode.
 182    * @deprecated ICU 56 Use Normalizer2 instead.
 183    */
 184   Normalizer(const CharacterIterator& iter, UNormalizationMode mode);
 185 #endif  /* U_HIDE_DEPRECATED_API */
 186
 187   /**
 188    * Copy constructor.
 189    * @param copy The object to be copied.
 190    * @deprecated ICU 56 Use Normalizer2 instead.
 191    */
 192   Normalizer(const Normalizer& copy);
 193
 194   /**
 195    * Destructor
 196    * @deprecated ICU 56 Use Normalizer2 instead.
 197    */
 198   virtual ~Normalizer();
 199
 200
 201   //-------------------------------------------------------------------------
 202   // Static utility methods
 203   //-------------------------------------------------------------------------
 204
 205 #ifndef U_HIDE_DEPRECATED_API
 206   /**
 207    * Normalizes a <code>UnicodeString</code> according to the specified normalization mode.
 208    * This is a wrapper for unorm_normalize(), using UnicodeString's.
 209    *
 210    * The <code>options</code> parameter specifies which optional
 211    * <code>Normalizer</code> features are to be enabled for this operation.
 212    *
 213    * @param source    the input string to be normalized.
 214    * @param mode      the normalization mode
 215    * @param options   the optional features to be enabled (0 for no options)
 216    * @param result    The normalized string (on output).
 217    * @param status    The error code.
 218    * @deprecated ICU 56 Use Normalizer2 instead.
 219    */
 220   static void U_EXPORT2 normalize(const UnicodeString& source,
 221                         UNormalizationMode mode, int32_t options,
 222                         UnicodeString& result,
 223                         UErrorCode &status);
 224
 225   /**
 226    * Compose a <code>UnicodeString</code>.
 227    * This is equivalent to normalize() with mode UNORM_NFC or UNORM_NFKC.
 228    * This is a wrapper for unorm_normalize(), using UnicodeString's.
 229    *
 230    * The <code>options</code> parameter specifies which optional
 231    * <code>Normalizer</code> features are to be enabled for this operation.
 232    *
 233    * @param source    the string to be composed.
 234    * @param compat    Perform compatibility decomposition before composition.
 235    *                  If this argument is <code>FALSE</code>, only canonical
 236    *                  decomposition will be performed.
 237    * @param options   the optional features to be enabled (0 for no options)
 238    * @param result    The composed string (on output).
 239    * @param status    The error code.
 240    * @deprecated ICU 56 Use Normalizer2 instead.
 241    */
 242   static void U_EXPORT2 compose(const UnicodeString& source,
 243                       UBool compat, int32_t options,
 244                       UnicodeString& result,
 245                       UErrorCode &status);
 246
 247   /**
 248    * Static method to decompose a <code>UnicodeString</code>.
 249    * This is equivalent to normalize() with mode UNORM_NFD or UNORM_NFKD.
 250    * This is a wrapper for unorm_normalize(), using UnicodeString's.
 251    *
 252    * The <code>options</code> parameter specifies which optional
 253    * <code>Normalizer</code> features are to be enabled for this operation.
 254    *
 255    * @param source    the string to be decomposed.
 256    * @param compat    Perform compatibility decomposition.
 257    *                  If this argument is <code>FALSE</code>, only canonical
 258    *                  decomposition will be performed.
 259    * @param options   the optional features to be enabled (0 for no options)
 260    * @param result    The decomposed string (on output).
 261    * @param status    The error code.
 262    * @deprecated ICU 56 Use Normalizer2 instead.
 263    */
 264   static void U_EXPORT2 decompose(const UnicodeString& source,
 265                         UBool compat, int32_t options,
 266                         UnicodeString& result,
 267                         UErrorCode &status);
 268
 269   /**
 270    * Performing quick check on a string, to quickly determine if the string is
 271    * in a particular normalization format.
 272    * This is a wrapper for unorm_quickCheck(), using a UnicodeString.
 273    *
 274    * Three types of result can be returned UNORM_YES, UNORM_NO or
 275    * UNORM_MAYBE. Result UNORM_YES indicates that the argument
 276    * string is in the desired normalized format, UNORM_NO determines that
 277    * argument string is not in the desired normalized format. A
 278    * UNORM_MAYBE result indicates that a more thorough check is required,
 279    * the user may have to put the string in its normalized form and compare the
 280    * results.
 281    * @param source       string for determining if it is in a normalized format
 282    * @param mode         normalization format
 283    * @param status A reference to a UErrorCode to receive any errors
 284    * @return UNORM_YES, UNORM_NO or UNORM_MAYBE
 285    *
 286    * @see isNormalized
 287    * @deprecated ICU 56 Use Normalizer2 instead.
 288    */
 289   static inline UNormalizationCheckResult
 290   quickCheck(const UnicodeString &source, UNormalizationMode mode, UErrorCode &status);
 291
 292   /**
 293    * Performing quick check on a string; same as the other version of quickCheck
 294    * but takes an extra options parameter like most normalization functions.
 295    *
 296    * @param source       string for determining if it is in a normalized format
 297    * @param mode         normalization format
 298    * @param options      the optional features to be enabled (0 for no options)
 299    * @param status A reference to a UErrorCode to receive any errors
 300    * @return UNORM_YES, UNORM_NO or UNORM_MAYBE
 301    *
 302    * @see isNormalized
 303    * @deprecated ICU 56 Use Normalizer2 instead.
 304    */
 305   static UNormalizationCheckResult
 306   quickCheck(const UnicodeString &source, UNormalizationMode mode, int32_t options, UErrorCode &status);
 307
 308   /**
 309    * Test if a string is in a given normalization form.
 310    * This is semantically equivalent to source.equals(normalize(source, mode)) .
 311    *
 312    * Unlike unorm_quickCheck(), this function returns a definitive result,
 313    * never a "maybe".
 314    * For NFD, NFKD, and FCD, both functions work exactly the same.
 315    * For NFC and NFKC where quickCheck may return "maybe", this function will
 316    * perform further tests to arrive at a TRUE/FALSE result.
 317    *
 318    * @param src        String that is to be tested if it is in a normalization format.
 319    * @param mode       Which normalization form to test for.
 320    * @param errorCode  ICU error code in/out parameter.
 321    *                   Must fulfill U_SUCCESS before the function call.
 322    * @return Boolean value indicating whether the source string is in the
 323    *         "mode" normalization form.
 324    *
 325    * @see quickCheck
 326    * @deprecated ICU 56 Use Normalizer2 instead.
 327    */
 328   static inline UBool
 329   isNormalized(const UnicodeString &src, UNormalizationMode mode, UErrorCode &errorCode);
 330
 331   /**
 332    * Test if a string is in a given normalization form; same as the other version of isNormalized
 333    * but takes an extra options parameter like most normalization functions.
 334    *
 335    * @param src        String that is to be tested if it is in a normalization format.
 336    * @param mode       Which normalization form to test for.
 337    * @param options      the optional features to be enabled (0 for no options)
 338    * @param errorCode  ICU error code in/out parameter.
 339    *                   Must fulfill U_SUCCESS before the function call.
 340    * @return Boolean value indicating whether the source string is in the
 341    *         "mode" normalization form.
 342    *
 343    * @see quickCheck
 344    * @deprecated ICU 56 Use Normalizer2 instead.
 345    */
 346   static UBool
 347   isNormalized(const UnicodeString &src, UNormalizationMode mode, int32_t options, UErrorCode &errorCode);
 348
 349   /**
 350    * Concatenate normalized strings, making sure that the result is normalized as well.
 351    *
 352    * If both the left and the right strings are in
 353    * the normalization form according to "mode/options",
 354    * then the result will be
 355    *
 356    * \code
 357    *     dest=normalize(left+right, mode, options)
 358    * \endcode
 359    *
 360    * For details see unorm_concatenate in unorm.h.
 361    *
 362    * @param left Left source string.
 363    * @param right Right source string.
 364    * @param result The output string.
 365    * @param mode The normalization mode.
 366    * @param options A bit set of normalization options.
 367    * @param errorCode ICU error code in/out parameter.
 368    *                   Must fulfill U_SUCCESS before the function call.
 369    * @return result
 370    *
 371    * @see unorm_concatenate
 372    * @see normalize
 373    * @see unorm_next
 374    * @see unorm_previous
 375    *
 376    * @deprecated ICU 56 Use Normalizer2 instead.
 377    */
 378   static UnicodeString &
 379   U_EXPORT2 concatenate(const UnicodeString &left, const UnicodeString &right,
 380               UnicodeString &result,
 381               UNormalizationMode mode, int32_t options,
 382               UErrorCode &errorCode);
 383 #endif  /* U_HIDE_DEPRECATED_API */
 384
 385   /**
 386    * Compare two strings for canonical equivalence.
 387    * Further options include case-insensitive comparison and
 388    * code point order (as opposed to code unit order).
 389    *
 390    * Canonical equivalence between two strings is defined as their normalized
 391    * forms (NFD or NFC) being identical.
 392    * This function compares strings incrementally instead of normalizing
 393    * (and optionally case-folding) both strings entirely,
 394    * improving performance significantly.
 395    *
 396    * Bulk normalization is only necessary if the strings do not fulfill the FCD
 397    * conditions. Only in this case, and only if the strings are relatively long,
 398    * is memory allocated temporarily.
 399    * For FCD strings and short non-FCD strings there is no memory allocation.
 400    *
 401    * Semantically, this is equivalent to
 402    *   strcmp[CodePointOrder](NFD(foldCase(s1)), NFD(foldCase(s2)))
 403    * where code point order and foldCase are all optional.
 404    *
 405    * UAX 21 2.5 Caseless Matching specifies that for a canonical caseless match
 406    * the case folding must be performed first, then the normalization.
 407    *
 408    * @param s1 First source string.
 409    * @param s2 Second source string.
 410    *
 411    * @param options A bit set of options:
 412    *   - U_FOLD_CASE_DEFAULT or 0 is used for default options:
 413    *     Case-sensitive comparison in code unit order, and the input strings
 414    *     are quick-checked for FCD.
 415    *
 416    *   - UNORM_INPUT_IS_FCD
 417    *     Set if the caller knows that both s1 and s2 fulfill the FCD conditions.
 418    *     If not set, the function will quickCheck for FCD
 419    *     and normalize if necessary.
 420    *
 421    *   - U_COMPARE_CODE_POINT_ORDER
 422    *     Set to choose code point order instead of code unit order
 423    *     (see u_strCompare for details).
 424    *
 425    *   - U_COMPARE_IGNORE_CASE
 426    *     Set to compare strings case-insensitively using case folding,
 427    *     instead of case-sensitively.
 428    *     If set, then the following case folding options are used.
 429    *
 430    *   - Options as used with case-insensitive comparisons, currently:
 431    *
 432    *   - U_FOLD_CASE_EXCLUDE_SPECIAL_I
 433    *    (see u_strCaseCompare for details)
 434    *
 435    *   - regular normalization options shifted left by UNORM_COMPARE_NORM_OPTIONS_SHIFT
 436    *
 437    * @param errorCode ICU error code in/out parameter.
 438    *                  Must fulfill U_SUCCESS before the function call.
 439    * @return <0 or 0 or >0 as usual for string comparisons
 440    *
 441    * @see unorm_compare
 442    * @see normalize
 443    * @see UNORM_FCD
 444    * @see u_strCompare
 445    * @see u_strCaseCompare
 446    *
 447    * @stable ICU 2.2
 448    */
 449   static inline int32_t
 450   compare(const UnicodeString &s1, const UnicodeString &s2,
 451           uint32_t options,
 452           UErrorCode &errorCode);
 453
 454 #ifndef U_HIDE_DEPRECATED_API
 455   //-------------------------------------------------------------------------
 456   // Iteration API
 457   //-------------------------------------------------------------------------
 458
 459   /**
 460    * Return the current character in the normalized text.
 461    * current() may need to normalize some text at getIndex().
 462    * The getIndex() is not changed.
 463    *
 464    * @return the current normalized code point
 465    * @deprecated ICU 56 Use Normalizer2 instead.
 466    */
 467   UChar32              current(void);
 468
 469   /**
 470    * Return the first character in the normalized text.
 471    * This is equivalent to setIndexOnly(startIndex()) followed by next().
 472    * (Post-increment semantics.)
 473    *
 474    * @return the first normalized code point
 475    * @deprecated ICU 56 Use Normalizer2 instead.
 476    */
 477   UChar32              first(void);
 478
 479   /**
 480    * Return the last character in the normalized text.
 481    * This is equivalent to setIndexOnly(endIndex()) followed by previous().
 482    * (Pre-decrement semantics.)
 483    *
 484    * @return the last normalized code point
 485    * @deprecated ICU 56 Use Normalizer2 instead.
 486    */
 487   UChar32              last(void);
 488
 489   /**
 490    * Return the next character in the normalized text.
 491    * (Post-increment semantics.)
 492    * If the end of the text has already been reached, DONE is returned.
 493    * The DONE value could be confused with a U+FFFF non-character code point
 494    * in the text. If this is possible, you can test getIndex()<endIndex()
 495    * before calling next(), or (getIndex()<endIndex() || last()!=DONE)
 496    * after calling next(). (Calling last() will change the iterator state!)
 497    *
 498    * The C API unorm_next() is more efficient and does not have this ambiguity.
 499    *
 500    * @return the next normalized code point
 501    * @deprecated ICU 56 Use Normalizer2 instead.
 502    */
 503   UChar32              next(void);
 504
 505   /**
 506    * Return the previous character in the normalized text and decrement.
 507    * (Pre-decrement semantics.)
 508    * If the beginning of the text has already been reached, DONE is returned.
 509    * The DONE value could be confused with a U+FFFF non-character code point
 510    * in the text. If this is possible, you can test
 511    * (getIndex()>startIndex() || first()!=DONE). (Calling first() will change
 512    * the iterator state!)
 513    *
 514    * The C API unorm_previous() is more efficient and does not have this ambiguity.
 515    *
 516    * @return the previous normalized code point
 517    * @deprecated ICU 56 Use Normalizer2 instead.
 518    */
 519   UChar32              previous(void);
 520
 521   /**
 522    * Set the iteration position in the input text that is being normalized,
 523    * without any immediate normalization.
 524    * After setIndexOnly(), getIndex() will return the same index that is
 525    * specified here.
 526    *
 527    * @param index the desired index in the input text.
 528    * @deprecated ICU 56 Use Normalizer2 instead.
 529    */
 530   void                 setIndexOnly(int32_t index);
 531
 532   /**
 533    * Reset the index to the beginning of the text.
 534    * This is equivalent to setIndexOnly(startIndex)).
 535    * @deprecated ICU 56 Use Normalizer2 instead.
 536    */
 537   void                reset(void);
 538
 539   /**
 540    * Retrieve the current iteration position in the input text that is
 541    * being normalized.
 542    *
 543    * A following call to next() will return a normalized code point from
 544    * the input text at or after this index.
 545    *
 546    * After a call to previous(), getIndex() will point at or before the
 547    * position in the input text where the normalized code point
 548    * was returned from with previous().
 549    *
 550    * @return the current index in the input text
 551    * @deprecated ICU 56 Use Normalizer2 instead.
 552    */
 553   int32_t            getIndex(void) const;
 554
 555   /**
 556    * Retrieve the index of the start of the input text. This is the begin index
 557    * of the <code>CharacterIterator</code> or the start (i.e. index 0) of the string
 558    * over which this <code>Normalizer</code> is iterating.
 559    *
 560    * @return the smallest index in the input text where the Normalizer operates
 561    * @deprecated ICU 56 Use Normalizer2 instead.
 562    */
 563   int32_t            startIndex(void) const;
 564
 565   /**
 566    * Retrieve the index of the end of the input text. This is the end index
 567    * of the <code>CharacterIterator</code> or the length of the string
 568    * over which this <code>Normalizer</code> is iterating.
 569    * This end index is exclusive, i.e., the Normalizer operates only on characters
 570    * before this index.
 571    *
 572    * @return the first index in the input text where the Normalizer does not operate
 573    * @deprecated ICU 56 Use Normalizer2 instead.
 574    */
 575   int32_t            endIndex(void) const;
 576
 577   /**
 578    * Returns TRUE when both iterators refer to the same character in the same
 579    * input text.
 580    *
 581    * @param that a Normalizer object to compare this one to
 582    * @return comparison result
 583    * @deprecated ICU 56 Use Normalizer2 instead.
 584    */
 585   UBool        operator==(const Normalizer& that) const;
 586
 587   /**
 588    * Returns FALSE when both iterators refer to the same character in the same
 589    * input text.
 590    *
 591    * @param that a Normalizer object to compare this one to
 592    * @return comparison result
 593    * @deprecated ICU 56 Use Normalizer2 instead.
 594    */
 595   inline UBool        operator!=(const Normalizer& that) const;
 596
 597   /**
 598    * Returns a pointer to a new Normalizer that is a clone of this one.
 599    * The caller is responsible for deleting the new clone.
 600    * @return a pointer to a new Normalizer
 601    * @deprecated ICU 56 Use Normalizer2 instead.
 602    */
 603   Normalizer*        clone(void) const;
 604
 605   /**
 606    * Generates a hash code for this iterator.
 607    *
 608    * @return the hash code
 609    * @deprecated ICU 56 Use Normalizer2 instead.
 610    */
 611   int32_t                hashCode(void) const;
 612
 613   //-------------------------------------------------------------------------
 614   // Property access methods
 615   //-------------------------------------------------------------------------
 616
 617   /**
 618    * Set the normalization mode for this object.
 619    * <p>
 620    * <b>Note:</b>If the normalization mode is changed while iterating
 621    * over a string, calls to {@link #next() } and {@link #previous() } may
 622    * return previously buffers characters in the old normalization mode
 623    * until the iteration is able to re-sync at the next base character.
 624    * It is safest to call {@link #setIndexOnly }, {@link #reset() },
 625    * {@link #setText }, {@link #first() },
 626    * {@link #last() }, etc. after calling <code>setMode</code>.
 627    * <p>
 628    * @param newMode the new mode for this <code>Normalizer</code>.
 629    * @see #getUMode
 630    * @deprecated ICU 56 Use Normalizer2 instead.
 631    */
 632   void setMode(UNormalizationMode newMode);
 633
 634   /**
 635    * Return the normalization mode for this object.
 636    *
 637    * This is an unusual name because there used to be a getMode() that
 638    * returned a different type.
 639    *
 640    * @return the mode for this <code>Normalizer</code>
 641    * @see #setMode
 642    * @deprecated ICU 56 Use Normalizer2 instead.
 643    */
 644   UNormalizationMode getUMode(void) const;
 645
 646   /**
 647    * Set options that affect this <code>Normalizer</code>'s operation.
 648    * Options do not change the basic composition or decomposition operation
 649    * that is being performed, but they control whether
 650    * certain optional portions of the operation are done.
 651    * Currently the only available option is obsolete.
 652    *
 653    * It is possible to specify multiple options that are all turned on or off.
 654    *
 655    * @param   option  the option(s) whose value is/are to be set.
 656    * @param   value   the new setting for the option.  Use <code>TRUE</code> to
 657    *                  turn the option(s) on and <code>FALSE</code> to turn it/them off.
 658    *
 659    * @see #getOption
 660    * @deprecated ICU 56 Use Normalizer2 instead.
 661    */
 662   void setOption(int32_t option,
 663          UBool value);
 664
 665   /**
 666    * Determine whether an option is turned on or off.
 667    * If multiple options are specified, then the result is TRUE if any
 668    * of them are set.
 669    * <p>
 670    * @param option the option(s) that are to be checked
 671    * @return TRUE if any of the option(s) are set
 672    * @see #setOption
 673    * @deprecated ICU 56 Use Normalizer2 instead.
 674    */
 675   UBool getOption(int32_t option) const;
 676
 677   /**
 678    * Set the input text over which this <code>Normalizer</code> will iterate.
 679    * The iteration position is set to the beginning.
 680    *
 681    * @param newText a string that replaces the current input text
 682    * @param status a UErrorCode
 683    * @deprecated ICU 56 Use Normalizer2 instead.
 684    */
 685   void setText(const UnicodeString& newText,
 686            UErrorCode &status);
 687
 688   /**
 689    * Set the input text over which this <code>Normalizer</code> will iterate.
 690    * The iteration position is set to the beginning.
 691    *
 692    * @param newText a CharacterIterator object that replaces the current input text
 693    * @param status a UErrorCode
 694    * @deprecated ICU 56 Use Normalizer2 instead.
 695    */
 696   void setText(const CharacterIterator& newText,
 697            UErrorCode &status);
 698
 699   /**
 700    * Set the input text over which this <code>Normalizer</code> will iterate.
 701    * The iteration position is set to the beginning.
 702    *
 703    * @param newText a string that replaces the current input text
 704    * @param length the length of the string, or -1 if NUL-terminated
 705    * @param status a UErrorCode
 706    * @deprecated ICU 56 Use Normalizer2 instead.
 707    */
 708   void setText(ConstChar16Ptr newText,
 709                     int32_t length,
 710             UErrorCode &status);
 711   /**
 712    * Copies the input text into the UnicodeString argument.
 713    *
 714    * @param result Receives a copy of the text under iteration.
 715    * @deprecated ICU 56 Use Normalizer2 instead.
 716    */
 717   void            getText(UnicodeString&  result);
 718
 719   /**
 720    * ICU "poor man's RTTI", returns a UClassID for this class.
 721    * @returns a UClassID for this class.
 722    * @deprecated ICU 56 Use Normalizer2 instead.
 723    */
 724   static UClassID U_EXPORT2 getStaticClassID();
 725 #endif  /* U_HIDE_DEPRECATED_API */
 726
 727   /**
 728    * ICU "poor man's RTTI", returns a UClassID for the actual class.
 729    * @return a UClassID for the actual class.
 730    * @deprecated ICU 56 Use Normalizer2 instead.
 731    */
 732   virtual UClassID getDynamicClassID() const;
 733
 734 private:
 735   //-------------------------------------------------------------------------
 736   // Private functions
 737   //-------------------------------------------------------------------------
 738
 739   Normalizer(); // default constructor not implemented
 740   Normalizer &operator=(const Normalizer &that); // assignment operator not implemented
 741
 742   // Private utility methods for iteration
 743   // For documentation, see the source code
 744   UBool nextNormalize();
 745   UBool previousNormalize();
 746
 747   void    init();
 748   void    clearBuffer(void);
 749
 750   //-------------------------------------------------------------------------
 751   // Private data
 752   //-------------------------------------------------------------------------
 753
 754   FilteredNormalizer2*fFilteredNorm2;  // owned if not NULL
 755   const Normalizer2  *fNorm2;  // not owned; may be equal to fFilteredNorm2
 756   UNormalizationMode  fUMode;  // deprecated
 757   int32_t             fOptions;
 758
 759   // The input text and our position in it
 760   CharacterIterator  *text;
 761
 762   // The normalization buffer is the result of normalization
 763   // of the source in [currentIndex..nextIndex[ .
 764   int32_t         currentIndex, nextIndex;
 765
 766   // A buffer for holding intermediate results
 767   UnicodeString       buffer;
 768   int32_t         bufferPos;
 769 };
 770
 771 //-------------------------------------------------------------------------
 772 // Inline implementations
 773 //-------------------------------------------------------------------------
 774
 775 #ifndef U_HIDE_DEPRECATED_API
 776 inline UBool
 777 Normalizer::operator!= (const Normalizer& other) const
 778 { return ! operator==(other); }
 779
 780 inline UNormalizationCheckResult
 781 Normalizer::quickCheck(const UnicodeString& source,
 782                        UNormalizationMode mode,
 783                        UErrorCode &status) {
 784     return quickCheck(source, mode, 0, status);
 785 }
 786
 787 inline UBool
 788 Normalizer::isNormalized(const UnicodeString& source,
 789                          UNormalizationMode mode,
 790                          UErrorCode &status) {
 791     return isNormalized(source, mode, 0, status);
 792 }
 793 #endif  /* U_HIDE_DEPRECATED_API */
 794
 795 inline int32_t
 796 Normalizer::compare(const UnicodeString &s1, const UnicodeString &s2,
 797                     uint32_t options,
 798                     UErrorCode &errorCode) {
 799   // all argument checking is done in unorm_compare
 800   return unorm_compare(toUCharPtr(s1.getBuffer()), s1.length(),
 801                        toUCharPtr(s2.getBuffer()), s2.length(),
 802                        options,
 803                        &errorCode);
 804 }
 805
 806 U_NAMESPACE_END
 807 #endif // U_SHOW_CPLUSPLUS_API
 808
 809 #endif /* #if !UCONFIG_NO_NORMALIZATION */
 810
 811 #endif // NORMLZR_H