Normalizer2

public abstract class Normalizer2 extends Object

Unicode normalization functionality for standard Unicode normalization or for using custom mapping tables. All instances of this class are unmodifiable/immutable. The Normalizer2 class is not intended for public subclassing.

The primary functions are to produce a normalized string and to detect whether a string is already normalized. The most commonly used normalization forms are those defined in http://www.unicode.org/unicode/reports/tr15/ However, this API supports additional normalization forms for specialized purposes. For example, NFKC_Casefold is provided via getInstance("nfkc_cf", COMPOSE) and can be used in implementations of UTS #46.

Not only are the standard compose and decompose modes supplied, but additional modes are provided as documented in the Mode enum.

Some of the functions in this class identify normalization boundaries. At a normalization boundary, the portions of the string before it and starting from it do not interact and can be handled independently.

The spanQuickCheckYes() stops at a normalization boundary. When the goal is a normalized string, then the text before the boundary can be copied, and the remainder can be processed with normalizeSecondAndAppend().

The hasBoundaryBefore(), hasBoundaryAfter() and isInert() functions test whether a character is guaranteed to be at a normalization boundary, regardless of context. This is used for moving from one normalization boundary to the next or preceding boundary, and for performing iterative normalization.

Iterative normalization is useful when only a small portion of a longer string needs to be processed. For example, in ICU, iterative normalization is used by the NormalizationTransliterator (to avoid replacing already-normalized text) and ucol_nextSortKeyPart() (to process only the substring for which sort key bytes are computed).

The set of normalization boundaries returned by these functions may not be complete: There may be more boundaries that could be returned. Different functions may return different boundaries.

Nested Class Summary

enum Normalizer2.Mode Constants for normalization modes. 

Public Method Summary

abstract StringBuilder
append(StringBuilder first, CharSequence second)
Appends the second string to the first string (merging them at the boundary) and returns the first string.
int
composePair(int a, int b)
Performs pairwise composition of a & b and returns the composite if there is one.
int
getCombiningClass(int c)
Gets the combining class of c.
abstract String
getDecomposition(int c)
Gets the decomposition mapping of c.
static Normalizer2
getInstance(InputStream data, String name, Normalizer2.Mode mode)
Returns a Normalizer2 instance which uses the specified data file (an ICU data file if data=null, or else custom binary data) and which composes or decomposes text according to the specified mode.
static Normalizer2
getNFCInstance()
Returns a Normalizer2 instance for Unicode NFC normalization.
static Normalizer2
getNFDInstance()
Returns a Normalizer2 instance for Unicode NFD normalization.
static Normalizer2
getNFKCCasefoldInstance()
Returns a Normalizer2 instance for Unicode NFKC_Casefold normalization.
static Normalizer2
getNFKCInstance()
Returns a Normalizer2 instance for Unicode NFKC normalization.
static Normalizer2
getNFKDInstance()
Returns a Normalizer2 instance for Unicode NFKD normalization.
String
getRawDecomposition(int c)
Gets the raw decomposition mapping of c.
abstract boolean
hasBoundaryAfter(int c)
Tests if the character always has a normalization boundary after it, regardless of context.
abstract boolean
hasBoundaryBefore(int c)
Tests if the character always has a normalization boundary before it, regardless of context.
abstract boolean
isInert(int c)
Tests if the character is normalization-inert.
abstract boolean
isNormalized(CharSequence s)
Tests if the string is normalized.
String
normalize(CharSequence src)
Returns the normalized form of the source string.
abstract StringBuilder
normalize(CharSequence src, StringBuilder dest)
Writes the normalized form of the source string to the destination string (replacing its contents) and returns the destination string.
abstract Appendable
normalize(CharSequence src, Appendable dest)
Writes the normalized form of the source string to the destination Appendable and returns the destination Appendable.
abstract StringBuilder
normalizeSecondAndAppend(StringBuilder first, CharSequence second)
Appends the normalized form of the second string to the first string (merging them at the boundary) and returns the first string.
abstract Normalizer.QuickCheckResult
quickCheck(CharSequence s)
Tests if the string is normalized.
abstract int
spanQuickCheckYes(CharSequence s)
Returns the end of the normalized substring of the input string.

Inherited Method Summary

Public Methods

public abstract StringBuilder append (StringBuilder first, CharSequence second)

Appends the second string to the first string (merging them at the boundary) and returns the first string. The result is normalized if both the strings were normalized. The first and second strings must be different objects.

Parameters
first string, should be normalized
second string, should be normalized
Returns
  • first

public int composePair (int a, int b)

Performs pairwise composition of a & b and returns the composite if there is one.

Returns a composite code point c only if c has a two-way mapping to a+b. In standard Unicode normalization, this means that c has a canonical decomposition to a+b and c does not have the Full_Composition_Exclusion property.

This function is independent of the mode of the Normalizer2. The default implementation returns a negative value.

Parameters
a A (normalization starter) code point.
b Another code point.
Returns
  • The non-negative composite code point if there is one; otherwise a negative value.

public int getCombiningClass (int c)

Gets the combining class of c. The default implementation returns 0 but all standard implementations return the Unicode Canonical_Combining_Class value.

Parameters
c code point
Returns
  • c's combining class

public abstract String getDecomposition (int c)

Gets the decomposition mapping of c. Roughly equivalent to normalizing the String form of c on a DECOMPOSE Normalizer2 instance, but much faster, and except that this function returns null if c does not have a decomposition mapping in this instance's data. This function is independent of the mode of the Normalizer2.

Parameters
c code point
Returns
  • c's decomposition mapping, if any; otherwise null

public static Normalizer2 getInstance (InputStream data, String name, Normalizer2.Mode mode)

Returns a Normalizer2 instance which uses the specified data file (an ICU data file if data=null, or else custom binary data) and which composes or decomposes text according to the specified mode. Returns an unmodifiable singleton instance.

  • Use data=null for data files that are part of ICU's own data.
  • Use name="nfc" and COMPOSE/DECOMPOSE for Unicode standard NFC/NFD.
  • Use name="nfkc" and COMPOSE/DECOMPOSE for Unicode standard NFKC/NFKD.
  • Use name="nfkc_cf" and COMPOSE for Unicode standard NFKC_CF=NFKC_Casefold.
If data!=null, then the binary data is read once and cached using the provided name as the key. If you know or expect the data to be cached already, you can use data!=null for non-ICU data as well.

Any IOException is wrapped into a ICUUncheckedIOException.

Parameters
data the binary, big-endian normalization (.nrm file) data, or null for ICU data
name "nfc" or "nfkc" or "nfkc_cf" or name of custom data file
mode normalization mode (compose or decompose etc.)
Returns
  • the requested Normalizer2, if successful

public static Normalizer2 getNFCInstance ()

Returns a Normalizer2 instance for Unicode NFC normalization. Same as getInstance(null, "nfc", Mode.COMPOSE). Returns an unmodifiable singleton instance.

Returns
  • the requested Normalizer2, if successful

public static Normalizer2 getNFDInstance ()

Returns a Normalizer2 instance for Unicode NFD normalization. Same as getInstance(null, "nfc", Mode.DECOMPOSE). Returns an unmodifiable singleton instance.

Returns
  • the requested Normalizer2, if successful

public static Normalizer2 getNFKCCasefoldInstance ()

Returns a Normalizer2 instance for Unicode NFKC_Casefold normalization. Same as getInstance(null, "nfkc_cf", Mode.COMPOSE). Returns an unmodifiable singleton instance.

Returns
  • the requested Normalizer2, if successful

public static Normalizer2 getNFKCInstance ()

Returns a Normalizer2 instance for Unicode NFKC normalization. Same as getInstance(null, "nfkc", Mode.COMPOSE). Returns an unmodifiable singleton instance.

Returns
  • the requested Normalizer2, if successful

public static Normalizer2 getNFKDInstance ()

Returns a Normalizer2 instance for Unicode NFKD normalization. Same as getInstance(null, "nfkc", Mode.DECOMPOSE). Returns an unmodifiable singleton instance.

Returns
  • the requested Normalizer2, if successful

public String getRawDecomposition (int c)

Gets the raw decomposition mapping of c.

This is similar to the getDecomposition() method but returns the raw decomposition mapping as specified in UnicodeData.txt or (for custom data) in the mapping files processed by the gennorm2 tool. By contrast, getDecomposition() returns the processed, recursively-decomposed version of this mapping.

When used on a standard NFKC Normalizer2 instance, getRawDecomposition() returns the Unicode Decomposition_Mapping (dm) property.

When used on a standard NFC Normalizer2 instance, it returns the Decomposition_Mapping only if the Decomposition_Type (dt) is Canonical (Can); in this case, the result contains either one or two code points (=1..4 Java chars).

This function is independent of the mode of the Normalizer2. The default implementation returns null.

Parameters
c code point
Returns
  • c's raw decomposition mapping, if any; otherwise null

public abstract boolean hasBoundaryAfter (int c)

Tests if the character always has a normalization boundary after it, regardless of context. If true, then the character does not normalization-interact with following characters. In other words, a string containing this character can be normalized by processing portions up to this character and after this character independently. This is used for iterative normalization. See the class documentation for details.

Note that this operation may be significantly slower than hasBoundaryBefore().

Parameters
c character to test
Returns
  • true if c has a normalization boundary after it

public abstract boolean hasBoundaryBefore (int c)

Tests if the character always has a normalization boundary before it, regardless of context. If true, then the character does not normalization-interact with preceding characters. In other words, a string containing this character can be normalized by processing portions before this character and starting from this character independently. This is used for iterative normalization. See the class documentation for details.

Parameters
c character to test
Returns
  • true if c has a normalization boundary before it

public abstract boolean isInert (int c)

Tests if the character is normalization-inert. If true, then the character does not change, nor normalization-interact with preceding or following characters. In other words, a string containing this character can be normalized by processing portions before this character and after this character independently. This is used for iterative normalization. See the class documentation for details.

Note that this operation may be significantly slower than hasBoundaryBefore().

Parameters
c character to test
Returns
  • true if c is normalization-inert

public abstract boolean isNormalized (CharSequence s)

Tests if the string is normalized. Internally, in cases where the quickCheck() method would return "maybe" (which is only possible for the two COMPOSE modes) this method resolves to "yes" or "no" to provide a definitive result, at the cost of doing more work in those cases.

Parameters
s input string
Returns
  • true if s is normalized

public String normalize (CharSequence src)

Returns the normalized form of the source string.

Parameters
src source string
Returns
  • normalized src

public abstract StringBuilder normalize (CharSequence src, StringBuilder dest)

Writes the normalized form of the source string to the destination string (replacing its contents) and returns the destination string. The source and destination strings must be different objects.

Parameters
src source string
dest destination string; its contents is replaced with normalized src
Returns
  • dest

public abstract Appendable normalize (CharSequence src, Appendable dest)

Writes the normalized form of the source string to the destination Appendable and returns the destination Appendable. The source and destination strings must be different objects.

Any IOException is wrapped into a ICUUncheckedIOException.

Parameters
src source string
dest destination Appendable; gets normalized src appended
Returns
  • dest

public abstract StringBuilder normalizeSecondAndAppend (StringBuilder first, CharSequence second)

Appends the normalized form of the second string to the first string (merging them at the boundary) and returns the first string. The result is normalized if the first string was normalized. The first and second strings must be different objects.

Parameters
first string, should be normalized
second string, will be normalized
Returns
  • first

public abstract Normalizer.QuickCheckResult quickCheck (CharSequence s)

Tests if the string is normalized. For the two COMPOSE modes, the result could be "maybe" in cases that would take a little more work to resolve definitively. Use spanQuickCheckYes() and normalizeSecondAndAppend() for a faster combination of quick check + normalization, to avoid re-checking the "yes" prefix.

Parameters
s input string
Returns
  • the quick check result

public abstract int spanQuickCheckYes (CharSequence s)

Returns the end of the normalized substring of the input string. In other words, with end=spanQuickCheckYes(s); the substring s.subSequence(0, end) will pass the quick check with a "yes" result.

The returned end index is usually one or more characters before the "no" or "maybe" character: The end index is at a normalization boundary. (See the class documentation for more about normalization boundaries.)

When the goal is a normalized string and most input strings are expected to be normalized already, then call this method, and if it returns a prefix shorter than the input string, copy that prefix and use normalizeSecondAndAppend() for the remainder.

Parameters
s input string
Returns
  • "yes" span end index