ICU 53.1  53.1
 All Data Structures Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Groups Pages
Macros | Enumerations | Functions
unorm.h File Reference

C API: Unicode Normalization. More...

#include "unicode/utypes.h"
#include "unicode/uiter.h"
#include "unicode/unorm2.h"

Go to the source code of this file.

Macros

#define UNORM_COMPARE_NORM_OPTIONS_SHIFT   20
 Lowest-order bit number of unorm_compare() options bits corresponding to normalization options bits. More...
 
#define UNORM_INPUT_IS_FCD   0x20000
 Option bit for unorm_compare: Both input strings are assumed to fulfill FCD conditions. More...
 
#define U_COMPARE_IGNORE_CASE   0x10000
 Option bit for unorm_compare: Perform case-insensitive comparison. More...
 
#define U_COMPARE_CODE_POINT_ORDER   0x8000
 Option bit for u_strCaseCompare, u_strcasecmp, unorm_compare, etc: Compare strings in code point order instead of code unit order. More...
 

Enumerations

enum  UNormalizationMode {
  UNORM_NONE = 1, UNORM_NFD = 2, UNORM_NFKD = 3, UNORM_NFC = 4,
  UNORM_DEFAULT = UNORM_NFC, UNORM_NFKC =5, UNORM_FCD = 6, UNORM_MODE_COUNT
}
 Constants for normalization modes. More...
 
enum  { UNORM_UNICODE_3_2 =0x20 }
 Constants for options flags for normalization. More...
 

Functions

int32_t unorm_normalize (const UChar *source, int32_t sourceLength, UNormalizationMode mode, int32_t options, UChar *result, int32_t resultLength, UErrorCode *status)
 Normalize a string. More...
 
UNormalizationCheckResult unorm_quickCheck (const UChar *source, int32_t sourcelength, UNormalizationMode mode, UErrorCode *status)
 Performing quick check on a string, to quickly determine if the string is in a particular normalization format. More...
 
UNormalizationCheckResult unorm_quickCheckWithOptions (const UChar *src, int32_t srcLength, UNormalizationMode mode, int32_t options, UErrorCode *pErrorCode)
 Performing quick check on a string; same as unorm_quickCheck but takes an extra options parameter like most normalization functions. More...
 
UBool unorm_isNormalized (const UChar *src, int32_t srcLength, UNormalizationMode mode, UErrorCode *pErrorCode)
 Test if a string is in a given normalization form. More...
 
UBool unorm_isNormalizedWithOptions (const UChar *src, int32_t srcLength, UNormalizationMode mode, int32_t options, UErrorCode *pErrorCode)
 Test if a string is in a given normalization form; same as unorm_isNormalized but takes an extra options parameter like most normalization functions. More...
 
int32_t unorm_next (UCharIterator *src, UChar *dest, int32_t destCapacity, UNormalizationMode mode, int32_t options, UBool doNormalize, UBool *pNeededToNormalize, UErrorCode *pErrorCode)
 Iterative normalization forward. More...
 
int32_t unorm_previous (UCharIterator *src, UChar *dest, int32_t destCapacity, UNormalizationMode mode, int32_t options, UBool doNormalize, UBool *pNeededToNormalize, UErrorCode *pErrorCode)
 Iterative normalization backward. More...
 
int32_t unorm_concatenate (const UChar *left, int32_t leftLength, const UChar *right, int32_t rightLength, UChar *dest, int32_t destCapacity, UNormalizationMode mode, int32_t options, UErrorCode *pErrorCode)
 Concatenate normalized strings, making sure that the result is normalized as well. More...
 
int32_t unorm_compare (const UChar *s1, int32_t length1, const UChar *s2, int32_t length2, uint32_t options, UErrorCode *pErrorCode)
 Compare two strings for canonical equivalence. More...
 

Detailed Description

C API: Unicode Normalization.

Unicode normalization API

Note: This API has been replaced by the unorm2.h API and is only available for backward compatibility. The functions here simply delegate to the unorm2.h functions, for example unorm2_getInstance() and unorm2_normalize(). There is one exception: The new API does not provide a replacement for unorm_compare().

unorm_normalize transforms Unicode text into an equivalent composed or decomposed form, allowing for easier sorting and searching of text. unorm_normalize supports the standard normalization forms described in Unicode Standard Annex #15: Unicode Normalization Forms.

Characters with accents or other adornments can be encoded in several different ways in Unicode. For example, take the character A-acute. In Unicode, this can be encoded as a single character (the "composed" form):

00C1 LATIN CAPITAL LETTER A WITH ACUTE

or as two separate characters (the "decomposed" form):

0041 LATIN CAPITAL LETTER A
0301 COMBINING ACUTE ACCENT

To a user of your program, however, both of these sequences should be treated as the same "user-level" character "A with acute accent". When you are searching or comparing text, you must ensure that these two sequences are treated equivalently. In addition, you must handle characters with more than one accent. Sometimes the order of a character's combining accents is significant, while in other cases accent sequences in different orders are really equivalent.

Similarly, the string "ffi" can be encoded as three separate letters:

0066 LATIN SMALL LETTER F
0066 LATIN SMALL LETTER F
0069 LATIN SMALL LETTER I

or as the single character

FB03 LATIN SMALL LIGATURE FFI

The ffi ligature is not a distinct semantic character, and strictly speaking it shouldn't be in Unicode at all, but it was included for compatibility with existing character sets that already provided it. The Unicode standard identifies such characters by giving them "compatibility" decompositions into the corresponding semantic characters. When sorting and searching, you will often want to use these mappings.

unorm_normalize helps solve these problems by transforming text into the canonical composed and decomposed forms as shown in the first example above. In addition, you can have it perform compatibility decompositions so that you can treat compatibility characters the same as their equivalents. Finally, unorm_normalize rearranges accents into the proper canonical order, so that you do not have to worry about accent rearrangement on your own.

Form FCD, "Fast C or D", is also designed for collation. It allows to work on strings that are not necessarily normalized with an algorithm (like in collation) that works under "canonical closure", i.e., it treats precomposed characters and their decomposed equivalents the same.

It is not a normalization form because it does not provide for uniqueness of representation. Multiple strings may be canonically equivalent (their NFDs are identical) and may all conform to FCD without being identical themselves.

The form is defined such that the "raw decomposition", the recursive canonical decomposition of each character, results in a string that is canonically ordered. This means that precomposed characters are allowed for as long as their decompositions do not need canonical reordering.

Its advantage for a process like collation is that all NFD and most NFC texts - and many unnormalized texts - already conform to FCD and do not need to be normalized (NFD) for such a process. The FCD quick check will return UNORM_YES for most strings in practice.

unorm_normalize(UNORM_FCD) may be implemented with UNORM_NFD.

For more details on FCD see the collation design document: http://source.icu-project.org/repos/icu/icuhtml/trunk/design/collation/ICU_collation_design.htm

ICU collation performs either NFD or FCD normalization automatically if normalization is turned on for the collator object. Beyond collation and string search, normalized strings may be useful for string equivalence comparisons, transliteration/transcription, unique representations, etc.

The W3C generally recommends to exchange texts in NFC. Note also that most legacy character encodings use only precomposed forms and often do not encode any combining marks by themselves. For conversion to such character encodings the Unicode text needs to be normalized to NFC. For more usage examples, see the Unicode Standard Annex.

Definition in file unorm.h.

Macro Definition Documentation

#define U_COMPARE_CODE_POINT_ORDER   0x8000

Option bit for u_strCaseCompare, u_strcasecmp, unorm_compare, etc: Compare strings in code point order instead of code unit order.

Stable:
ICU 2.2

Definition at line 484 of file unorm.h.

#define U_COMPARE_IGNORE_CASE   0x10000

Option bit for unorm_compare: Perform case-insensitive comparison.

Stable:
ICU 2.2

Definition at line 475 of file unorm.h.

#define UNORM_COMPARE_NORM_OPTIONS_SHIFT   20

Lowest-order bit number of unorm_compare() options bits corresponding to normalization options bits.

The options parameter for unorm_compare() uses most bits for itself and for various comparison and folding flags. The most significant bits, however, are shifted down and passed on to the normalization implementation. (That is, from unorm_compare(..., options, ...), options>>UNORM_COMPARE_NORM_OPTIONS_SHIFT will be passed on to the internal normalization functions.)

See Also
unorm_compare
Stable:
ICU 2.6

Definition at line 185 of file unorm.h.

#define UNORM_INPUT_IS_FCD   0x20000

Option bit for unorm_compare: Both input strings are assumed to fulfill FCD conditions.

Stable:
ICU 2.2

Definition at line 468 of file unorm.h.

Enumeration Type Documentation

anonymous enum

Constants for options flags for normalization.

Use 0 for default options, including normalization according to the Unicode version that is currently supported by ICU (see u_getUnicodeVersion).

Stable:
ICU 2.6
Enumerator
UNORM_UNICODE_3_2 

Options bit set value to select Unicode 3.2 normalization (except NormalizationCorrections).

At most one Unicode version can be selected at a time.

Stable:
ICU 2.6

Definition at line 160 of file unorm.h.

Constants for normalization modes.

Stable:
ICU 2.0
Enumerator
UNORM_NONE 

No decomposition/composition.

Stable:
ICU 2.0
UNORM_NFD 

Canonical decomposition.

Stable:
ICU 2.0
UNORM_NFKD 

Compatibility decomposition.

Stable:
ICU 2.0
UNORM_NFC 

Canonical decomposition followed by canonical composition.

Stable:
ICU 2.0
UNORM_DEFAULT 

Default normalization.

Stable:
ICU 2.0
UNORM_NFKC 

Compatibility decomposition followed by canonical composition.

Stable:
ICU 2.0
UNORM_FCD 

"Fast C or D" form.

Stable:
ICU 2.0
UNORM_MODE_COUNT 

One more than the highest normalization mode constant.

Stable:
ICU 2.0

Definition at line 133 of file unorm.h.

Function Documentation

int32_t unorm_compare ( const UChar s1,
int32_t  length1,
const UChar s2,
int32_t  length2,
uint32_t  options,
UErrorCode pErrorCode 
)

Compare two strings for canonical equivalence.

Further options include case-insensitive comparison and code point order (as opposed to code unit order).

Canonical equivalence between two strings is defined as their normalized forms (NFD or NFC) being identical. This function compares strings incrementally instead of normalizing (and optionally case-folding) both strings entirely, improving performance significantly.

Bulk normalization is only necessary if the strings do not fulfill the FCD conditions. Only in this case, and only if the strings are relatively long, is memory allocated temporarily. For FCD strings and short non-FCD strings there is no memory allocation.

Semantically, this is equivalent to strcmp[CodePointOrder](NFD(foldCase(NFD(s1))), NFD(foldCase(NFD(s2)))) where code point order and foldCase are all optional.

UAX 21 2.5 Caseless Matching specifies that for a canonical caseless match the case folding must be performed first, then the normalization.

Parameters
s1First source string.
length1Length of first source string, or -1 if NUL-terminated.
s2Second source string.
length2Length of second source string, or -1 if NUL-terminated.
optionsA bit set of options:
  • U_FOLD_CASE_DEFAULT or 0 is used for default options: Case-sensitive comparison in code unit order, and the input strings are quick-checked for FCD.
  • UNORM_INPUT_IS_FCD Set if the caller knows that both s1 and s2 fulfill the FCD conditions. If not set, the function will quickCheck for FCD and normalize if necessary.
  • U_COMPARE_CODE_POINT_ORDER Set to choose code point order instead of code unit order (see u_strCompare for details).
  • U_COMPARE_IGNORE_CASE Set to compare strings case-insensitively using case folding, instead of case-sensitively. If set, then the following case folding options are used.
  • Options as used with case-insensitive comparisons, currently:
  • U_FOLD_CASE_EXCLUDE_SPECIAL_I (see u_strCaseCompare for details)
  • regular normalization options shifted left by UNORM_COMPARE_NORM_OPTIONS_SHIFT
pErrorCodeICU error code in/out parameter. Must fulfill U_SUCCESS before the function call.
Returns
<0 or 0 or >0 as usual for string comparisons
See Also
unorm_normalize
UNORM_FCD
u_strCompare
u_strCaseCompare
Stable:
ICU 2.2

Referenced by icu::Normalizer::compare().

int32_t unorm_concatenate ( const UChar left,
int32_t  leftLength,
const UChar right,
int32_t  rightLength,
UChar dest,
int32_t  destCapacity,
UNormalizationMode  mode,
int32_t  options,
UErrorCode pErrorCode 
)

Concatenate normalized strings, making sure that the result is normalized as well.

If both the left and the right strings are in the normalization form according to "mode/options", then the result will be

dest=normalize(left+right, mode, options)

With the input strings already being normalized, this function will use unorm_next() and unorm_previous() to find the adjacent end pieces of the input strings. Only the concatenation of these end pieces will be normalized and then concatenated with the remaining parts of the input strings.

It is allowed to have dest==left to avoid copying the entire left string.

Parameters
leftLeft source string, may be same as dest.
leftLengthLength of left source string, or -1 if NUL-terminated.
rightRight source string. Must not be the same as dest, nor overlap.
rightLengthLength of right source string, or -1 if NUL-terminated.
destThe output buffer; can be NULL if destCapacity==0 for pure preflighting.
destCapacityThe number of UChars that fit into dest.
modeThe normalization mode.
optionsThe normalization options, ORed together (0 for no options).
pErrorCodeICU error code in/out parameter. Must fulfill U_SUCCESS before the function call.
Returns
Length of output (number of UChars) when successful or buffer overflow.
See Also
unorm_normalize
unorm_next
unorm_previous
Stable:
ICU 2.1
UBool unorm_isNormalized ( const UChar src,
int32_t  srcLength,
UNormalizationMode  mode,
UErrorCode pErrorCode 
)

Test if a string is in a given normalization form.

This is semantically equivalent to source.equals(normalize(source, mode)) .

Unlike unorm_quickCheck(), this function returns a definitive result, never a "maybe". For NFD, NFKD, and FCD, both functions work exactly the same. For NFC and NFKC where quickCheck may return "maybe", this function will perform further tests to arrive at a TRUE/FALSE result.

Parameters
srcString that is to be tested if it is in a normalization format.
srcLengthLength of source to test, or -1 if NUL-terminated.
modeWhich normalization form to test for.
pErrorCodeICU error code in/out parameter. Must fulfill U_SUCCESS before the function call.
Returns
Boolean value indicating whether the source string is in the "mode" normalization form.
See Also
unorm_quickCheck
Stable:
ICU 2.2
UBool unorm_isNormalizedWithOptions ( const UChar src,
int32_t  srcLength,
UNormalizationMode  mode,
int32_t  options,
UErrorCode pErrorCode 
)

Test if a string is in a given normalization form; same as unorm_isNormalized but takes an extra options parameter like most normalization functions.

Parameters
srcString that is to be tested if it is in a normalization format.
srcLengthLength of source to test, or -1 if NUL-terminated.
modeWhich normalization form to test for.
optionsThe normalization options, ORed together (0 for no options).
pErrorCodeICU error code in/out parameter. Must fulfill U_SUCCESS before the function call.
Returns
Boolean value indicating whether the source string is in the "mode/options" normalization form.
See Also
unorm_quickCheck
unorm_isNormalized
Stable:
ICU 2.6
int32_t unorm_next ( UCharIterator src,
UChar dest,
int32_t  destCapacity,
UNormalizationMode  mode,
int32_t  options,
UBool  doNormalize,
UBool pNeededToNormalize,
UErrorCode pErrorCode 
)

Iterative normalization forward.

This function (together with unorm_previous) is somewhat similar to the C++ Normalizer class (see its non-static functions).

Iterative normalization is useful when only a small portion of a longer string/text needs to be processed.

For example, the likelihood may be high that processing the first 10% of some text will be sufficient to find certain data. Another example: When one wants to concatenate two normalized strings and get a normalized result, it is much more efficient to normalize just a small part of the result around the concatenation place instead of re-normalizing everything.

The input text is an instance of the C character iteration API UCharIterator. It may wrap around a simple string, a CharacterIterator, a Replaceable, or any other kind of text object.

If a buffer overflow occurs, then the caller needs to reset the iterator to the old index and call the function again with a larger buffer - if the caller cares for the actual output. Regardless of the output buffer, the iterator will always be moved to the next normalization boundary.

This function (like unorm_previous) serves two purposes:

1) To find the next boundary so that the normalization of the part of the text from the current position to that boundary does not affect and is not affected by the part of the text beyond that boundary.

2) To normalize the text up to the boundary.

The second step is optional, per the doNormalize parameter. It is omitted for operations like string concatenation, where the two adjacent string ends need to be normalized together. In such a case, the output buffer will just contain a copy of the text up to the boundary.

pNeededToNormalize is an output-only parameter. Its output value is only defined if normalization was requested (doNormalize) and successful (especially, no buffer overflow). It is useful for operations like a normalizing transliterator, where one would not want to replace a piece of text if it is not modified.

If doNormalize==TRUE and pNeededToNormalize!=NULL then *pNeeded... is set TRUE if the normalization was necessary.

If doNormalize==FALSE then *pNeededToNormalize will be set to FALSE.

If the buffer overflows, then *pNeededToNormalize will be undefined; essentially, whenever U_FAILURE is true (like in buffer overflows), this result will be undefined.

Parameters
srcThe input text in the form of a C character iterator.
destThe output buffer; can be NULL if destCapacity==0 for pure preflighting.
destCapacityThe number of UChars that fit into dest.
modeThe normalization mode.
optionsThe normalization options, ORed together (0 for no options).
doNormalizeIndicates if the source text up to the next boundary is to be normalized (TRUE) or just copied (FALSE).
pNeededToNormalizeOutput flag indicating if the normalization resulted in different text from the input. Not defined if an error occurs including buffer overflow. Always FALSE if !doNormalize.
pErrorCodeICU error code in/out parameter. Must fulfill U_SUCCESS before the function call.
Returns
Length of output (number of UChars) when successful or buffer overflow.
See Also
unorm_previous
unorm_normalize
Stable:
ICU 2.1
int32_t unorm_normalize ( const UChar source,
int32_t  sourceLength,
UNormalizationMode  mode,
int32_t  options,
UChar result,
int32_t  resultLength,
UErrorCode status 
)

Normalize a string.

The string will be normalized according the specified normalization mode and options. The source and result buffers must not be the same, nor overlap.

Parameters
sourceThe string to normalize.
sourceLengthThe length of source, or -1 if NUL-terminated.
modeThe normalization mode; one of UNORM_NONE, UNORM_NFD, UNORM_NFC, UNORM_NFKC, UNORM_NFKD, UNORM_DEFAULT.
optionsThe normalization options, ORed together (0 for no options).
resultA pointer to a buffer to receive the result string. The result string is NUL-terminated if possible.
resultLengthThe maximum size of result.
statusA pointer to a UErrorCode to receive any errors.
Returns
The total buffer size needed; if greater than resultLength, the output was truncated, and the error code is set to U_BUFFER_OVERFLOW_ERROR.
Stable:
ICU 2.0
int32_t unorm_previous ( UCharIterator src,
UChar dest,
int32_t  destCapacity,
UNormalizationMode  mode,
int32_t  options,
UBool  doNormalize,
UBool pNeededToNormalize,
UErrorCode pErrorCode 
)

Iterative normalization backward.

This function (together with unorm_next) is somewhat similar to the C++ Normalizer class (see its non-static functions). For all details see unorm_next.

Parameters
srcThe input text in the form of a C character iterator.
destThe output buffer; can be NULL if destCapacity==0 for pure preflighting.
destCapacityThe number of UChars that fit into dest.
modeThe normalization mode.
optionsThe normalization options, ORed together (0 for no options).
doNormalizeIndicates if the source text up to the next boundary is to be normalized (TRUE) or just copied (FALSE).
pNeededToNormalizeOutput flag indicating if the normalization resulted in different text from the input. Not defined if an error occurs including buffer overflow. Always FALSE if !doNormalize.
pErrorCodeICU error code in/out parameter. Must fulfill U_SUCCESS before the function call.
Returns
Length of output (number of UChars) when successful or buffer overflow.
See Also
unorm_next
unorm_normalize
Stable:
ICU 2.1
UNormalizationCheckResult unorm_quickCheck ( const UChar source,
int32_t  sourcelength,
UNormalizationMode  mode,
UErrorCode status 
)

Performing quick check on a string, to quickly determine if the string is in a particular normalization format.

Three types of result can be returned UNORM_YES, UNORM_NO or UNORM_MAYBE. Result UNORM_YES indicates that the argument string is in the desired normalized format, UNORM_NO determines that argument string is not in the desired normalized format. A UNORM_MAYBE result indicates that a more thorough check is required, the user may have to put the string in its normalized form and compare the results.

Parameters
sourcestring for determining if it is in a normalized format
sourcelengthlength of source to test, or -1 if NUL-terminated
modewhich normalization form to test for
statusa pointer to a UErrorCode to receive any errors
Returns
UNORM_YES, UNORM_NO or UNORM_MAYBE
See Also
unorm_isNormalized
Stable:
ICU 2.0
UNormalizationCheckResult unorm_quickCheckWithOptions ( const UChar src,
int32_t  srcLength,
UNormalizationMode  mode,
int32_t  options,
UErrorCode pErrorCode 
)

Performing quick check on a string; same as unorm_quickCheck but takes an extra options parameter like most normalization functions.

Parameters
srcString that is to be tested if it is in a normalization format.
srcLengthLength of source to test, or -1 if NUL-terminated.
modeWhich normalization form to test for.
optionsThe normalization options, ORed together (0 for no options).
pErrorCodeICU error code in/out parameter. Must fulfill U_SUCCESS before the function call.
Returns
UNORM_YES, UNORM_NO or UNORM_MAYBE
See Also
unorm_quickCheck
unorm_isNormalized
Stable:
ICU 2.6