com.ibm.icu.text.RuleBasedNumberFormat Maven / Gradle / Ivy
Show all versions of icu4j Show documentation
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
*******************************************************************************
* Copyright (C) 1996-2016, International Business Machines Corporation and
* others. All Rights Reserved.
*******************************************************************************
*/
package com.ibm.icu.text;
import java.math.BigInteger;
import java.text.FieldPosition;
import java.text.ParsePosition;
import java.util.Arrays;
import java.util.HashMap;
import java.util.Locale;
import java.util.Map;
import java.util.MissingResourceException;
import java.util.Set;
import com.ibm.icu.impl.ICUData;
import com.ibm.icu.impl.ICUDebug;
import com.ibm.icu.impl.ICUResourceBundle;
import com.ibm.icu.impl.PatternProps;
import com.ibm.icu.lang.UCharacter;
import com.ibm.icu.math.BigDecimal;
import com.ibm.icu.util.ULocale;
import com.ibm.icu.util.ULocale.Category;
import com.ibm.icu.util.UResourceBundle;
import com.ibm.icu.util.UResourceBundleIterator;
/**
* A class that formats numbers according to a set of rules. This number formatter is
* typically used for spelling out numeric values in words (e.g., 25,3476 as
* "twenty-five thousand three hundred seventy-six" or "vingt-cinq mille trois
* cents soixante-seize" or
* "funfundzwanzigtausenddreihundertsechsundsiebzig"), but can also be used for
* other complicated formatting tasks, such as formatting a number of seconds as hours,
* minutes and seconds (e.g., 3,730 as "1:02:10").
*
* The resources contain three predefined formatters for each locale: spellout, which
* spells out a value in words (123 is "one hundred twenty-three"); ordinal, which
* appends an ordinal suffix to the end of a numeral (123 is "123rd"); and
* duration, which shows a duration in seconds as hours, minutes, and seconds (123 is
* "2:03"). The client can also define more specialized RuleBasedNumberFormats
* by supplying programmer-defined rule sets.
*
* The behavior of a RuleBasedNumberFormat is specified by a textual description
* that is either passed to the constructor as a String or loaded from a resource
* bundle. In its simplest form, the description consists of a semicolon-delimited list of rules.
* Each rule has a string of output text and a value or range of values it is applicable to.
* In a typical spellout rule set, the first twenty rules are the words for the numbers from
* 0 to 19:
*
* zero; one; two; three; four; five; six; seven; eight; nine;
* ten; eleven; twelve; thirteen; fourteen; fifteen; sixteen; seventeen; eighteen; nineteen;
*
* For larger numbers, we can use the preceding set of rules to format the ones place, and
* we only have to supply the words for the multiples of 10:
*
* 20: twenty[->>];
* 30: thirty{->>];
* 40: forty[->>];
* 50: fifty[->>];
* 60: sixty[->>];
* 70: seventy[->>];
* 80: eighty[->>];
* 90: ninety[->>];
*
* In these rules, the base value is spelled out explicitly and set off from the
* rule's output text with a colon. The rules are in a sorted list, and a rule is applicable
* to all numbers from its own base value to one less than the next rule's base value. The
* ">>" token is called a substitution and tells the formatter to
* isolate the number's ones digit, format it using this same set of rules, and place the
* result at the position of the ">>" token. Text in brackets is omitted if
* the number being formatted is an even multiple of 10 (the hyphen is a literal hyphen; 24
* is "twenty-four," not "twenty four").
*
* For even larger numbers, we can actually look up several parts of the number in the
* list:
*
* 100: << hundred[ >>];
*
* The "<<" represents a new kind of substitution. The << isolates
* the hundreds digit (and any digits to its left), formats it using this same rule set, and
* places the result where the "<<" was. Notice also that the meaning of
* >> has changed: it now refers to both the tens and the ones digits. The meaning of
* both substitutions depends on the rule's base value. The base value determines the rule's divisor,
* which is the highest power of 10 that is less than or equal to the base value (the user
* can change this). To fill in the substitutions, the formatter divides the number being
* formatted by the divisor. The integral quotient is used to fill in the <<
* substitution, and the remainder is used to fill in the >> substitution. The meaning
* of the brackets changes similarly: text in brackets is omitted if the value being
* formatted is an even multiple of the rule's divisor. The rules are applied recursively, so
* if a substitution is filled in with text that includes another substitution, that
* substitution is also filled in.
*
* This rule covers values up to 999, at which point we add another rule:
*
* 1000: << thousand[ >>];
*
* Again, the meanings of the brackets and substitution tokens shift because the rule's
* base value is a higher power of 10, changing the rule's divisor. This rule can actually be
* used all the way up to 999,999. This allows us to finish out the rules as follows:
*
* 1,000,000: << million[ >>];
* 1,000,000,000: << billion[ >>];
* 1,000,000,000,000: << trillion[ >>];
* 1,000,000,000,000,000: OUT OF RANGE!;
*
* Commas, periods, and spaces can be used in the base values to improve legibility and
* are ignored by the rule parser. The last rule in the list is customarily treated as an
* "overflow rule," applying to everything from its base value on up, and often (as
* in this example) being used to print out an error message or default representation.
* Notice also that the size of the major groupings in large numbers is controlled by the
* spacing of the rules: because in English we group numbers by thousand, the higher rules
* are separated from each other by a factor of 1,000.
*
* To see how these rules actually work in practice, consider the following example:
* Formatting 25,430 with this rule set would work like this:
*
*
*
* << thousand >>
* [the rule whose base value is 1,000 is applicable to 25,340]
*
*
* twenty->> thousand >>
* [25,340 over 1,000 is 25. The rule for 20 applies.]
*
*
* twenty-five thousand >>
* [25 mod 10 is 5. The rule for 5 is "five."
*
*
* twenty-five thousand << hundred >>
* [25,340 mod 1,000 is 340. The rule for 100 applies.]
*
*
* twenty-five thousand three hundred >>
* [340 over 100 is 3. The rule for 3 is "three."]
*
*
* twenty-five thousand three hundred forty
* [340 mod 100 is 40. The rule for 40 applies. Since 40 divides
* evenly by 10, the hyphen and substitution in the brackets are omitted.]
*
*
*
* The above syntax suffices only to format positive integers. To format negative numbers,
* we add a special rule:
*
* -x: minus >>;
*
* This is called a negative-number rule, and is identified by "-x"
* where the base value would be. This rule is used to format all negative numbers. the
* >> token here means "find the number's absolute value, format it with these
* rules, and put the result here."
*
* We also add a special rule called a fraction rule for numbers with fractional
* parts:
*
* x.x: << point >>;
*
* This rule is used for all positive non-integers (negative non-integers pass through the
* negative-number rule first and then through this rule). Here, the << token refers to
* the number's integral part, and the >> to the number's fractional part. The
* fractional part is formatted as a series of single-digit numbers (e.g., 123.456 would be
* formatted as "one hundred twenty-three point four five six").
*
* To see how this rule syntax is applied to various languages, examine the resource data.
*
* There is actually much more flexibility built into the rule language than the
* description above shows. A formatter may own multiple rule sets, which can be selected by
* the caller, and which can use each other to fill in their substitutions. Substitutions can
* also be filled in with digits, using a DecimalFormat object. There is syntax that can be
* used to alter a rule's divisor in various ways. And there is provision for much more
* flexible fraction handling. A complete description of the rule syntax follows:
*
*
*
* The description of a RuleBasedNumberFormat's behavior consists of one or more rule
* sets. Each rule set consists of a name, a colon, and a list of rules. A rule
* set name must begin with a % sign. Rule sets with names that begin with a single % sign
* are public: the caller can specify that they be used to format and parse numbers.
* Rule sets with names that begin with %% are private: they exist only for the use
* of other rule sets. If a formatter only has one rule set, the name may be omitted.
*
* The user can also specify a special "rule set" named %%lenient-parse.
* The body of %%lenient-parse isn't a set of number-formatting rules, but a RuleBasedCollator
* description which is used to define equivalences for lenient parsing. For more information
* on the syntax, see RuleBasedCollator. For more information on lenient parsing,
* see setLenientParse(). Note: symbols that have syntactic meaning
* in collation rules, such as '&', have no particular meaning when appearing outside
* of the lenient-parse rule set.
*
* The body of a rule set consists of an ordered, semicolon-delimited list of rules.
* Internally, every rule has a base value, a divisor, rule text, and zero, one, or two substitutions.
* These parameters are controlled by the description syntax, which consists of a rule
* descriptor, a colon, and a rule body.
*
* A rule descriptor can take one of the following forms (text in italics is the
* name of a token):
*
*
*
* bv:
* bv specifies the rule's base value. bv is a decimal
* number expressed using ASCII digits. bv may contain spaces, period, and commas,
* which are ignored. The rule's divisor is the highest power of 10 less than or equal to
* the base value.
*
*
* bv/rad:
* bv specifies the rule's base value. The rule's divisor is the
* highest power of rad less than or equal to the base value.
*
*
* bv>:
* bv specifies the rule's base value. To calculate the divisor,
* let the radix be 10, and the exponent be the highest exponent of the radix that yields a
* result less than or equal to the base value. Every > character after the base value
* decreases the exponent by 1. If the exponent is positive or 0, the divisor is the radix
* raised to the power of the exponent; otherwise, the divisor is 1.
*
*
* bv/rad>:
* bv specifies the rule's base value. To calculate the divisor,
* let the radix be rad, and the exponent be the highest exponent of the radix that
* yields a result less than or equal to the base value. Every > character after the radix
* decreases the exponent by 1. If the exponent is positive or 0, the divisor is the radix
* raised to the power of the exponent; otherwise, the divisor is 1.
*
*
* -x:
* The rule is a negative-number rule.
*
*
* x.x:
* The rule is an improper fraction rule. If the full stop in
* the middle of the rule name is replaced with the decimal point
* that is used in the language or DecimalFormatSymbols, then that rule will
* have precedence when formatting and parsing this rule. For example, some
* languages use the comma, and can thus be written as x,x instead. For example,
* you can use "x.x: << point >>;x,x: << comma >>;" to
* handle the decimal point that matches the language's natural spelling of
* the punctuation of either the full stop or comma.
*
*
* 0.x:
* The rule is a proper fraction rule. If the full stop in
* the middle of the rule name is replaced with the decimal point
* that is used in the language or DecimalFormatSymbols, then that rule will
* have precedence when formatting and parsing this rule. For example, some
* languages use the comma, and can thus be written as 0,x instead. For example,
* you can use "0.x: point >>;0,x: comma >>;" to
* handle the decimal point that matches the language's natural spelling of
* the punctuation of either the full stop or comma
*
*
* x.0:
* The rule is a default rule. If the full stop in
* the middle of the rule name is replaced with the decimal point
* that is used in the language or DecimalFormatSymbols, then that rule will
* have precedence when formatting and parsing this rule. For example, some
* languages use the comma, and can thus be written as x,0 instead. For example,
* you can use "x.0: << point;x,0: << comma;" to
* handle the decimal point that matches the language's natural spelling of
* the punctuation of either the full stop or comma
*
*
* Inf:
* The rule for infinity.
*
*
* NaN:
* The rule for an IEEE 754 NaN (not a number).
*
*
* nothing
* If the rule's rule descriptor is left out, the base value is one plus the
* preceding rule's base value (or zero if this is the first rule in the list) in a normal
* rule set. In a fraction rule set, the base value is the same as the preceding rule's
* base value.
*
*
*
* A rule set may be either a regular rule set or a fraction rule set, depending
* on whether it is used to format a number's integral part (or the whole number) or a
* number's fractional part. Using a rule set to format a rule's fractional part makes it a
* fraction rule set.
*
* Which rule is used to format a number is defined according to one of the following
* algorithms: If the rule set is a regular rule set, do the following:
*
*
* - If the rule set includes a default rule (and the number was passed in as a double),
* use the default rule. (If the number being formatted was passed in as a long,
* the default rule is ignored.)
* - If the number is negative, use the negative-number rule.
* - If the number has a fractional part and is greater than 1, use the improper fraction
* rule.
* - If the number has a fractional part and is between 0 and 1, use the proper fraction
* rule.
* - Binary-search the rule list for the rule with the highest base value less than or equal
* to the number. If that rule has two substitutions, its base value is not an even multiple
* of its divisor, and the number is an even multiple of the rule's divisor, use the
* rule that precedes it in the rule list. Otherwise, use the rule itself.
*
*
* If the rule set is a fraction rule set, do the following:
*
*
* - Ignore negative-number and fraction rules.
* - For each rule in the list, multiply the number being formatted (which will always be
* between 0 and 1) by the rule's base value. Keep track of the distance between the result
* the nearest integer.
* - Use the rule that produced the result closest to zero in the above calculation. In the
* event of a tie or a direct hit, use the first matching rule encountered. (The idea here is
* to try each rule's base value as a possible denominator of a fraction. Whichever
* denominator produces the fraction closest in value to the number being formatted wins.) If
* the rule following the matching rule has the same base value, use it if the numerator of
* the fraction is anything other than 1; if the numerator is 1, use the original matching
* rule. (This is to allow singular and plural forms of the rule text without a lot of extra
* hassle.)
*
*
* A rule's body consists of a string of characters terminated by a semicolon. The rule
* may include zero, one, or two substitution tokens, and a range of text in
* brackets. The brackets denote optional text (and may also include one or both
* substitutions). The exact meanings of the substitution tokens, and under what conditions
* optional text is omitted, depend on the syntax of the substitution token and the context.
* The rest of the text in a rule body is literal text that is output when the rule matches
* the number being formatted.
*
* A substitution token begins and ends with a token character. The token
* character and the context together specify a mathematical operation to be performed on the
* number being formatted. An optional substitution descriptor specifies how the
* value resulting from that operation is used to fill in the substitution. The position of
* the substitution token in the rule body specifies the location of the resultant text in
* the original rule text.
*
* The meanings of the substitution token characters are as follows:
*
*
*
* >>
* in normal rule
* Divide the number by the rule's divisor and format the remainder
*
*
* in negative-number rule
* Find the absolute value of the number and format the result
*
*
* in fraction or default rule
* Isolate the number's fractional part and format it.
*
*
* in rule in fraction rule set
* Not allowed.
*
*
* >>>
* in normal rule
* Divide the number by the rule's divisor and format the remainder,
* but bypass the normal rule-selection process and just use the
* rule that precedes this one in this rule list.
*
*
* in all other rules
* Not allowed.
*
*
* <<
* in normal rule
* Divide the number by the rule's divisor and format the quotient
*
*
* in negative-number rule
* Not allowed.
*
*
* in fraction or default rule
* Isolate the number's integral part and format it.
*
*
* in rule in fraction rule set
* Multiply the number by the rule's base value and format the result.
*
*
* ==
* in all rule sets
* Format the number unchanged
*
*
* []
* in normal rule
* Omit the optional text if the number is an even multiple of the rule's divisor
*
*
* in negative-number rule
* Not allowed.
*
*
* in improper-fraction rule
* Omit the optional text if the number is between 0 and 1 (same as specifying both an
* x.x rule and a 0.x rule)
*
*
* in default rule
* Omit the optional text if the number is an integer (same as specifying both an x.x
* rule and an x.0 rule)
*
*
* in proper-fraction rule
* Not allowed.
*
*
* in rule in fraction rule set
* Omit the optional text if multiplying the number by the rule's base value yields 1.
*
*
* $(cardinal,plural syntax)$
* in all rule sets
* This provides the ability to choose a word based on the number divided by the radix to the power of the
* exponent of the base value for the specified locale, which is normally equivalent to the << value.
* This uses the cardinal plural rules from PluralFormat. All strings used in the plural format are treated
* as the same base value for parsing.
*
*
* $(ordinal,plural syntax)$
* in all rule sets
* This provides the ability to choose a word based on the number divided by the radix to the power of the
* exponent of the base value for the specified locale, which is normally equivalent to the << value.
* This uses the ordinal plural rules from PluralFormat. All strings used in the plural format are treated
* as the same base value for parsing.
*
*
*
* The substitution descriptor (i.e., the text between the token characters) may take one
* of three forms:
*
*
*
* a rule set name
* Perform the mathematical operation on the number, and format the result using the
* named rule set.
*
*
* a DecimalFormat pattern
* Perform the mathematical operation on the number, and format the result using a
* DecimalFormat with the specified pattern. The pattern must begin with 0 or #.
*
*
* nothing
* Perform the mathematical operation on the number, and format the result using the rule
* set containing the current rule, except:
* - You can't have an empty substitution descriptor with a == substitution.
* - If you omit the substitution descriptor in a >> substitution in a fraction rule,
* format the result one digit at a time using the rule set containing the current rule.
* - If you omit the substitution descriptor in a << substitution in a rule in a
* fraction rule set, format the result using the default rule set for this formatter.
*
*
*
*
*
* Whitespace is ignored between a rule set name and a rule set body, between a rule
* descriptor and a rule body, or between rules. If a rule body begins with an apostrophe,
* the apostrophe is ignored, but all text after it becomes significant (this is how you can
* have a rule's rule text begin with whitespace). There is no escape function: the semicolon
* is not allowed in rule set names or in rule text, and the colon is not allowed in rule set
* names. The characters beginning a substitution token are always treated as the beginning
* of a substitution token.
*
* See the resource data and the demo program for annotated examples of real rule sets
* using these features.
*
* @author Richard Gillam
* @see NumberFormat
* @see DecimalFormat
* @see PluralFormat
* @see PluralRules
* @stable ICU 2.0
*/
public class RuleBasedNumberFormat extends NumberFormat {
//-----------------------------------------------------------------------
// constants
//-----------------------------------------------------------------------
// Generated by serialver from JDK 1.4.1_01
static final long serialVersionUID = -7664252765575395068L;
/**
* Selector code that tells the constructor to create a spellout formatter
* @stable ICU 2.0
*/
public static final int SPELLOUT = 1;
/**
* Selector code that tells the constructor to create an ordinal formatter
* @stable ICU 2.0
*/
public static final int ORDINAL = 2;
/**
* Selector code that tells the constructor to create a duration formatter
* @deprecated ICU 74 Use MeasureFormat instead.
*/
@Deprecated
public static final int DURATION = 3;
/**
* Selector code that tells the constructor to create a numbering system formatter
* @stable ICU 4.2
*/
public static final int NUMBERING_SYSTEM = 4;
//-----------------------------------------------------------------------
// data members
//-----------------------------------------------------------------------
/**
* The formatter's rule sets.
*/
private transient NFRuleSet[] ruleSets = null;
/**
* The formatter's rule names mapped to rule sets.
*/
private transient Map ruleSetsMap = null;
/**
* A pointer to the formatter's default rule set. This is always included
* in ruleSets.
*/
private transient NFRuleSet defaultRuleSet = null;
/**
* The formatter's locale. This is used to create DecimalFormatSymbols and
* Collator objects.
* @serial
*/
private ULocale locale = null;
/**
* The formatter's rounding mode.
* @serial
*/
private int roundingMode = BigDecimal.ROUND_UNNECESSARY;
/**
* Collator to be used in lenient parsing. This variable is lazy-evaluated:
* the collator is actually created the first time the client does a parse
* with lenient-parse mode turned on.
*/
private transient RbnfLenientScannerProvider scannerProvider = null;
// flag to mark whether we've previously looked for a scanner and failed
private transient boolean lookedForScanner;
/**
* The DecimalFormatSymbols object that any DecimalFormat objects this
* formatter uses should use. This variable is lazy-evaluated: it isn't
* filled in if the rule set never uses a DecimalFormat pattern.
*/
private transient DecimalFormatSymbols decimalFormatSymbols = null;
/**
* The NumberFormat used when lenient parsing numbers. This needs to reflect
* the locale. This is lazy-evaluated, like decimalFormatSymbols. It is
* here so it can be shared by different NFSubstitutions.
*/
private transient DecimalFormat decimalFormat = null;
/**
* The rule used when dealing with infinity. This is lazy-evaluated, and derived from decimalFormat.
* It is here so it can be shared by different NFRuleSets.
*/
private transient NFRule defaultInfinityRule = null;
/**
* The rule used when dealing with IEEE 754 NaN. This is lazy-evaluated, and derived from decimalFormat.
* It is here so it can be shared by different NFRuleSets.
*/
private transient NFRule defaultNaNRule = null;
/**
* Flag specifying whether lenient parse mode is on or off. Off by default.
* @serial
*/
private boolean lenientParse = false;
/**
* If the description specifies lenient-parse rules, they're stored here until
* the collator is created.
*/
private transient String lenientParseRules;
/**
* If the description specifies post-process rules, they're stored here until
* post-processing is required.
*/
private transient String postProcessRules;
/**
* Post processor lazily constructed from the postProcessRules.
*/
private transient RBNFPostProcessor postProcessor;
/**
* Localizations for rule set names.
* @serial
*/
private Map ruleSetDisplayNames;
/**
* The public rule set names;
* @serial
*/
private String[] publicRuleSetNames;
/**
* Data for handling context-based capitalization
*/
private boolean capitalizationInfoIsSet = false;
private boolean capitalizationForListOrMenu = false;
private boolean capitalizationForStandAlone = false;
private transient BreakIterator capitalizationBrkIter = null;
private static final boolean DEBUG = ICUDebug.enabled("rbnf");
//-----------------------------------------------------------------------
// constructors
//-----------------------------------------------------------------------
/**
* Creates a RuleBasedNumberFormat that behaves according to the description
* passed in. The formatter uses the default FORMAT locale.
* @param description A description of the formatter's desired behavior.
* See the class documentation for a complete explanation of the description
* syntax.
* @see Category#FORMAT
* @stable ICU 2.0
*/
public RuleBasedNumberFormat(String description) {
locale = ULocale.getDefault(Category.FORMAT);
init(description, null);
}
/**
* Creates a RuleBasedNumberFormat that behaves according to the description
* passed in. The formatter uses the default FORMAT locale.
*
* The localizations data provides information about the public
* rule sets and their localized display names for different
* locales. The first element in the list is an array of the names
* of the public rule sets. The first element in this array is
* the initial default ruleset. The remaining elements in the
* list are arrays of localizations of the names of the public
* rule sets. Each of these is one longer than the initial array,
* with the first String being the ULocale ID, and the remaining
* Strings being the localizations of the rule set names, in the
* same order as the initial array.
* @param description A description of the formatter's desired behavior.
* See the class documentation for a complete explanation of the description
* syntax.
* @param localizations a list of localizations for the rule set
* names in the description.
* @see Category#FORMAT
* @stable ICU 3.2
*/
public RuleBasedNumberFormat(String description, String[][] localizations) {
locale = ULocale.getDefault(Category.FORMAT);
init(description, localizations);
}
/**
* Creates a RuleBasedNumberFormat that behaves according to the description
* passed in. The formatter uses the specified locale to determine the
* characters to use when formatting in numerals, and to define equivalences
* for lenient parsing.
* @param description A description of the formatter's desired behavior.
* See the class documentation for a complete explanation of the description
* syntax.
* @param locale A locale, which governs which characters are used for
* formatting values in numerals, and which characters are equivalent in
* lenient parsing.
* @stable ICU 2.0
*/
public RuleBasedNumberFormat(String description, Locale locale) {
this(description, ULocale.forLocale(locale));
}
/**
* Creates a RuleBasedNumberFormat that behaves according to the description
* passed in. The formatter uses the specified locale to determine the
* characters to use when formatting in numerals, and to define equivalences
* for lenient parsing.
* @param description A description of the formatter's desired behavior.
* See the class documentation for a complete explanation of the description
* syntax.
* @param locale A locale, which governs which characters are used for
* formatting values in numerals, and which characters are equivalent in
* lenient parsing.
* @stable ICU 3.2
*/
public RuleBasedNumberFormat(String description, ULocale locale) {
this.locale = locale;
init(description, null);
}
/**
* Creates a RuleBasedNumberFormat that behaves according to the description
* passed in. The formatter uses the specified locale to determine the
* characters to use when formatting in numerals, and to define equivalences
* for lenient parsing.
*
* The localizations data provides information about the public
* rule sets and their localized display names for different
* locales. The first element in the list is an array of the names
* of the public rule sets. The first element in this array is
* the initial default ruleset. The remaining elements in the
* list are arrays of localizations of the names of the public
* rule sets. Each of these is one longer than the initial array,
* with the first String being the ULocale ID, and the remaining
* Strings being the localizations of the rule set names, in the
* same order as the initial array.
* @param description A description of the formatter's desired behavior.
* See the class documentation for a complete explanation of the description
* syntax.
* @param localizations a list of localizations for the rule set names in the description.
* @param locale A ULocale that governs which characters are used for
* formatting values in numerals, and determines which characters are equivalent in
* lenient parsing.
* @stable ICU 3.2
*/
public RuleBasedNumberFormat(String description, String[][] localizations, ULocale locale) {
this.locale = locale;
init(description, localizations);
}
/**
* Creates a RuleBasedNumberFormat from a predefined description. The selector
* code chooses among three possible predefined formats: spellout, ordinal,
* and duration.
* @param locale The locale for the formatter.
* @param format A selector code specifying which kind of formatter to create for that
* locale. There are three legal values: SPELLOUT, which creates a formatter that
* spells out a value in words in the desired language, ORDINAL, which attaches
* an ordinal suffix from the desired language to the end of a number (e.g. "123rd"),
* and DURATION, which formats a duration in seconds as hours, minutes, and seconds.
* @stable ICU 2.0
*/
public RuleBasedNumberFormat(Locale locale, int format) {
this(ULocale.forLocale(locale), format);
}
/**
* Creates a RuleBasedNumberFormat from a predefined description. The selector
* code chooses among three possible predefined formats: spellout, ordinal,
* and duration.
* @param locale The locale for the formatter.
* @param format A selector code specifying which kind of formatter to create for that
* locale. There are four legal values: SPELLOUT, which creates a formatter that
* spells out a value in words in the desired language, ORDINAL, which attaches
* an ordinal suffix from the desired language to the end of a number (e.g. "123rd"),
* DURATION, which formats a duration in seconds as hours, minutes, and seconds, and
* NUMBERING_SYSTEM, which is used to invoke rules for alternate numbering
* systems such as the Hebrew numbering system, or for Roman numerals, etc..
* @stable ICU 3.2
*/
public RuleBasedNumberFormat(ULocale locale, int format) {
this.locale = locale;
ICUResourceBundle bundle = (ICUResourceBundle)UResourceBundle.
getBundleInstance(ICUData.ICU_RBNF_BASE_NAME, locale);
// TODO: determine correct actual/valid locale. Note ambiguity
// here -- do actual/valid refer to pattern, DecimalFormatSymbols,
// or Collator?
ULocale uloc = bundle.getULocale();
setLocale(uloc, uloc);
StringBuilder description = new StringBuilder();
String[][] localizations = null;
try {
ICUResourceBundle rules = bundle.getWithFallback("RBNFRules/"+rulenames[format-1]);
UResourceBundleIterator it = rules.getIterator();
while (it.hasNext()) {
description.append(it.nextString());
}
}
catch (MissingResourceException e1) {
}
// We use findTopLevel() instead of get() because
// it's faster when we know that it's usually going to fail.
UResourceBundle locNamesBundle = bundle.findTopLevel(locnames[format - 1]);
if (locNamesBundle != null) {
localizations = new String[locNamesBundle.getSize()][];
for (int i = 0; i < localizations.length; ++i) {
localizations[i] = locNamesBundle.get(i).getStringArray();
}
}
// else there are no localized names. It's not that important.
init(description.toString(), localizations);
}
private static final String[] rulenames = {
"SpelloutRules", "OrdinalRules", "DurationRules", "NumberingSystemRules",
};
private static final String[] locnames = {
"SpelloutLocalizations", "OrdinalLocalizations", "DurationLocalizations", "NumberingSystemLocalizations",
};
/**
* Creates a RuleBasedNumberFormat from a predefined description. Uses the
* default FORMAT locale.
* @param format A selector code specifying which kind of formatter to create.
* There are three legal values: SPELLOUT, which creates a formatter that spells
* out a value in words in the default locale's language, ORDINAL, which attaches
* an ordinal suffix from the default locale's language to a numeral, and
* DURATION, which formats a duration in seconds as hours, minutes, and seconds always rounding down.
* or NUMBERING_SYSTEM, which is used for alternate numbering systems such as Hebrew.
* @see Category#FORMAT
* @stable ICU 2.0
*/
public RuleBasedNumberFormat(int format) {
this(ULocale.getDefault(Category.FORMAT), format);
}
//-----------------------------------------------------------------------
// boilerplate
//-----------------------------------------------------------------------
/**
* Duplicates this formatter.
* @return A RuleBasedNumberFormat that is equal to this one.
* @stable ICU 2.0
*/
@Override
public Object clone() {
return super.clone();
}
/**
* Tests two RuleBasedNumberFormats for equality.
* @param that The formatter to compare against this one.
* @return true if the two formatters have identical behavior.
* @stable ICU 2.0
*/
@Override
public boolean equals(Object that) {
// if the other object isn't a RuleBasedNumberFormat, that's
// all we need to know
// Test for capitalization info equality is adequately handled
// by the NumberFormat test for capitalizationSetting equality;
// the info here is just derived from that.
if (!(that instanceof RuleBasedNumberFormat)) {
return false;
} else {
// cast the other object's pointer to a pointer to a
// RuleBasedNumberFormat
RuleBasedNumberFormat that2 = (RuleBasedNumberFormat)that;
// compare their locales and lenient-parse modes
if (!locale.equals(that2.locale) || lenientParse != that2.lenientParse) {
return false;
}
// if that succeeds, then compare their rule set lists
if (ruleSets.length != that2.ruleSets.length) {
return false;
}
for (int i = 0; i < ruleSets.length; i++) {
if (!ruleSets[i].equals(that2.ruleSets[i])) {
return false;
}
}
return true;
}
}
/**
* {@inheritDoc}
* @stable ICU 2.0
*/
@Override
public int hashCode() {
return super.hashCode();
}
/**
* Generates a textual description of this formatter.
* @return a String containing a rule set that will produce a RuleBasedNumberFormat
* with identical behavior to this one. This won't necessarily be identical
* to the rule set description that was originally passed in, but will produce
* the same result.
* @stable ICU 2.0
*/
@Override
public String toString() {
// accumulate the descriptions of all the rule sets in a
// StringBuffer, then cast it to a String and return it
StringBuilder result = new StringBuilder();
for (NFRuleSet ruleSet : ruleSets) {
result.append(ruleSet.toString());
}
return result.toString();
}
/**
* Writes this object to a stream.
* @param out The stream to write to.
*/
private void writeObject(java.io.ObjectOutputStream out)
throws java.io.IOException {
// we just write the textual description to the stream, so we
// have an implementation-independent streaming format
out.writeUTF(this.toString());
out.writeObject(this.locale);
out.writeInt(this.roundingMode);
}
/**
* Reads this object in from a stream.
* @param in The stream to read from.
*/
private void readObject(java.io.ObjectInputStream in)
throws java.io.IOException {
// read the description in from the stream
String description = in.readUTF();
ULocale loc;
try {
loc = (ULocale) in.readObject();
} catch (Exception e) {
loc = ULocale.getDefault(Category.FORMAT);
}
try {
roundingMode = in.readInt();
} catch (Exception ignored) {
}
// build a brand-new RuleBasedNumberFormat from the description,
// then steal its substructure. This object's substructure and
// the temporary RuleBasedNumberFormat drop on the floor and
// get swept up by the garbage collector
RuleBasedNumberFormat temp = new RuleBasedNumberFormat(description, loc);
ruleSets = temp.ruleSets;
ruleSetsMap = temp.ruleSetsMap;
defaultRuleSet = temp.defaultRuleSet;
publicRuleSetNames = temp.publicRuleSetNames;
decimalFormatSymbols = temp.decimalFormatSymbols;
decimalFormat = temp.decimalFormat;
locale = temp.locale;
defaultInfinityRule = temp.defaultInfinityRule;
defaultNaNRule = temp.defaultNaNRule;
}
//-----------------------------------------------------------------------
// public API functions
//-----------------------------------------------------------------------
/**
* Returns a list of the names of all of this formatter's public rule sets.
* @return A list of the names of all of this formatter's public rule sets.
* @stable ICU 2.0
*/
public String[] getRuleSetNames() {
return publicRuleSetNames.clone();
}
/**
* Return a list of locales for which there are locale-specific display names
* for the rule sets in this formatter. If there are no localized display names, return null.
* @return an array of the ULocales for which there is rule set display name information
* @stable ICU 3.2
*/
public ULocale[] getRuleSetDisplayNameLocales() {
if (ruleSetDisplayNames != null) {
Set s = ruleSetDisplayNames.keySet();
String[] locales = s.toArray(new String[s.size()]);
Arrays.sort(locales, String.CASE_INSENSITIVE_ORDER);
ULocale[] result = new ULocale[locales.length];
for (int i = 0; i < locales.length; ++i) {
result[i] = new ULocale(locales[i]);
}
return result;
}
return null;
}
private String[] getNameListForLocale(ULocale loc) {
if (loc != null && ruleSetDisplayNames != null) {
String[] localeNames = { loc.getBaseName(), ULocale.getDefault(Category.DISPLAY).getBaseName() };
for (String lname : localeNames) {
while (lname.length() > 0) {
String[] names = ruleSetDisplayNames.get(lname);
if (names != null) {
return names;
}
lname = ULocale.getFallback(lname);
}
}
}
return null;
}
/**
* Return the rule set display names for the provided locale. These are in the same order
* as those returned by getRuleSetNames. The locale is matched against the locales for
* which there is display name data, using normal fallback rules. If no locale matches,
* the default display names are returned. (These are the internal rule set names minus
* the leading '%'.)
* @return an array of the locales that have display name information
* @see #getRuleSetNames
* @stable ICU 3.2
*/
public String[] getRuleSetDisplayNames(ULocale loc) {
String[] names = getNameListForLocale(loc);
if (names != null) {
return names.clone();
}
names = getRuleSetNames();
for (int i = 0; i < names.length; ++i) {
names[i] = names[i].substring(1);
}
return names;
}
/**
* Return the rule set display names for the current default DISPLAY locale.
* @return an array of the display names
* @see #getRuleSetDisplayNames(ULocale)
* @see Category#DISPLAY
* @stable ICU 3.2
*/
public String[] getRuleSetDisplayNames() {
return getRuleSetDisplayNames(ULocale.getDefault(Category.DISPLAY));
}
/**
* Return the rule set display name for the provided rule set and locale.
* The locale is matched against the locales for which there is display name data, using
* normal fallback rules. If no locale matches, the default display name is returned.
* @return the display name for the rule set
* @see #getRuleSetDisplayNames
* @throws IllegalArgumentException if ruleSetName is not a valid rule set name for this format
* @stable ICU 3.2
*/
public String getRuleSetDisplayName(String ruleSetName, ULocale loc) {
String[] rsnames = publicRuleSetNames;
for (int ix = 0; ix < rsnames.length; ++ix) {
if (rsnames[ix].equals(ruleSetName)) {
String[] names = getNameListForLocale(loc);
if (names != null) {
return names[ix];
}
return rsnames[ix].substring(1);
}
}
throw new IllegalArgumentException("unrecognized rule set name: " + ruleSetName);
}
/**
* Return the rule set display name for the provided rule set in the current default DISPLAY locale.
* @return the display name for the rule set
* @see #getRuleSetDisplayName(String,ULocale)
* @see Category#DISPLAY
* @stable ICU 3.2
*/
public String getRuleSetDisplayName(String ruleSetName) {
return getRuleSetDisplayName(ruleSetName, ULocale.getDefault(Category.DISPLAY));
}
/**
* Formats the specified number according to the specified rule set.
* @param number The number to format.
* @param ruleSet The name of the rule set to format the number with.
* This must be the name of a valid public rule set for this formatter.
* @return A textual representation of the number.
* @stable ICU 2.0
*/
public String format(double number, String ruleSet) throws IllegalArgumentException {
if (ruleSet.startsWith("%%")) {
throw new IllegalArgumentException("Can't use internal rule set");
}
return adjustForContext(format(number, findRuleSet(ruleSet)));
}
/**
* Formats the specified number according to the specified rule set.
* (If the specified rule set specifies a default ["x.0"] rule, this function
* ignores it. Convert the number to a double first if you ned it.) This
* function preserves all the precision in the long-- it doesn't convert it
* to a double.
* @param number The number to format.
* @param ruleSet The name of the rule set to format the number with.
* This must be the name of a valid public rule set for this formatter.
* @return A textual representation of the number.
* @stable ICU 2.0
*/
public String format(long number, String ruleSet) throws IllegalArgumentException {
if (ruleSet.startsWith("%%")) {
throw new IllegalArgumentException("Can't use internal rule set");
}
return adjustForContext(format(number, findRuleSet(ruleSet)));
}
/**
* Formats the specified number using the formatter's default rule set.
* (The default rule set is the last public rule set defined in the description.)
* @param number The number to format.
* @param toAppendTo A StringBuffer that the result should be appended to.
* @param ignore This function doesn't examine or update the field position.
* @return toAppendTo
* @stable ICU 2.0
*/
@Override
public StringBuffer format(double number,
StringBuffer toAppendTo,
FieldPosition ignore) {
// this is one of the inherited format() methods. Since it doesn't
// have a way to select the rule set to use, it just uses the
// default one
// Note, the BigInteger/BigDecimal methods below currently go through this.
if (toAppendTo.length() == 0) {
toAppendTo.append(adjustForContext(format(number, defaultRuleSet)));
} else {
// appending to other text, don't capitalize
toAppendTo.append(format(number, defaultRuleSet));
}
return toAppendTo;
}
/**
* Formats the specified number using the formatter's default rule set.
* (The default rule set is the last public rule set defined in the description.)
* (If the specified rule set specifies a default ["x.0"] rule, this function
* ignores it. Convert the number to a double first if you ned it.) This
* function preserves all the precision in the long-- it doesn't convert it
* to a double.
* @param number The number to format.
* @param toAppendTo A StringBuffer that the result should be appended to.
* @param ignore This function doesn't examine or update the field position.
* @return toAppendTo
* @stable ICU 2.0
*/
@Override
public StringBuffer format(long number,
StringBuffer toAppendTo,
FieldPosition ignore) {
// this is one of the inherited format() methods. Since it doesn't
// have a way to select the rule set to use, it just uses the
// default one
if (toAppendTo.length() == 0) {
toAppendTo.append(adjustForContext(format(number, defaultRuleSet)));
} else {
// appending to other text, don't capitalize
toAppendTo.append(format(number, defaultRuleSet));
}
return toAppendTo;
}
/**
* NEW
* Implement com.ibm.icu.text.NumberFormat:
* Format a BigInteger.
* @stable ICU 2.0
*/
@Override
public StringBuffer format(BigInteger number,
StringBuffer toAppendTo,
FieldPosition pos) {
return format(new com.ibm.icu.math.BigDecimal(number), toAppendTo, pos);
}
/**
* NEW
* Implement com.ibm.icu.text.NumberFormat:
* Format a BigDecimal.
* @stable ICU 2.0
*/
@Override
public StringBuffer format(java.math.BigDecimal number,
StringBuffer toAppendTo,
FieldPosition pos) {
return format(new com.ibm.icu.math.BigDecimal(number), toAppendTo, pos);
}
private static final com.ibm.icu.math.BigDecimal MAX_VALUE = com.ibm.icu.math.BigDecimal.valueOf(Long.MAX_VALUE);
private static final com.ibm.icu.math.BigDecimal MIN_VALUE = com.ibm.icu.math.BigDecimal.valueOf(Long.MIN_VALUE);
/**
* NEW
* Implement com.ibm.icu.text.NumberFormat:
* Format a BigDecimal.
* @stable ICU 2.0
*/
@Override
public StringBuffer format(com.ibm.icu.math.BigDecimal number,
StringBuffer toAppendTo,
FieldPosition pos) {
if (MIN_VALUE.compareTo(number) > 0 || MAX_VALUE.compareTo(number) < 0) {
// We're outside of our normal range that this framework can handle.
// The DecimalFormat will provide more accurate results.
return getDecimalFormat().format(number, toAppendTo, pos);
}
if (number.scale() == 0) {
return format(number.longValue(), toAppendTo, pos);
}
return format(number.doubleValue(), toAppendTo, pos);
}
/**
* Parses the specified string, beginning at the specified position, according
* to this formatter's rules. This will match the string against all of the
* formatter's public rule sets and return the value corresponding to the longest
* parseable substring. This function's behavior is affected by the lenient
* parse mode.
* @param text The string to parse
* @param parsePosition On entry, contains the position of the first character
* in "text" to examine. On exit, has been updated to contain the position
* of the first character in "text" that wasn't consumed by the parse.
* @return The number that corresponds to the parsed text. This will be an
* instance of either Long or Double, depending on whether the result has a
* fractional part.
* @see #setLenientParseMode
* @stable ICU 2.0
*/
@Override
public Number parse(String text, ParsePosition parsePosition) {
// parsePosition tells us where to start parsing. We copy the
// text in the string from here to the end inro a new string,
// and create a new ParsePosition and result variable to use
// for the duration of the parse operation
String workingText = text.substring(parsePosition.getIndex());
ParsePosition workingPos = new ParsePosition(0);
Number tempResult = null;
// keep track of the largest number of characters consumed in
// the various trials, and the result that corresponds to it
Number result = NFRule.ZERO;
ParsePosition highWaterMark = new ParsePosition(workingPos.getIndex());
// iterate over the public rule sets (beginning with the default one)
// and try parsing the text with each of them. Keep track of which
// one consumes the most characters: that's the one that determines
// the result we return
for (int i = ruleSets.length - 1; i >= 0; i--) {
// skip private or unparseable rule sets
if (!ruleSets[i].isPublic() || !ruleSets[i].isParseable()) {
continue;
}
// try parsing the string with the rule set. If it gets past the
// high-water mark, update the high-water mark and the result
tempResult = ruleSets[i].parse(workingText, workingPos, Double.MAX_VALUE, 0);
if (workingPos.getIndex() > highWaterMark.getIndex()) {
result = tempResult;
highWaterMark.setIndex(workingPos.getIndex());
}
// commented out because this API on ParsePosition doesn't exist in 1.1.x
// if (workingPos.getErrorIndex() > highWaterMark.getErrorIndex()) {
// highWaterMark.setErrorIndex(workingPos.getErrorIndex());
// }
// if we manage to use up all the characters in the string,
// we don't have to try any more rule sets
if (highWaterMark.getIndex() == workingText.length()) {
break;
}
// otherwise, reset our internal parse position to the
// beginning and try again with the next rule set
workingPos.setIndex(0);
}
// add the high water mark to our original parse position and
// return the result
parsePosition.setIndex(parsePosition.getIndex() + highWaterMark.getIndex());
// commented out because this API on ParsePosition doesn't exist in 1.1.x
// if (highWaterMark.getIndex() == 0) {
// parsePosition.setErrorIndex(parsePosition.getIndex() + highWaterMark.getErrorIndex());
// }
return result;
}
/**
* Turns lenient parse mode on and off.
*
* When in lenient parse mode, the formatter uses an RbnfLenientScanner
* for parsing the text. Lenient parsing is only in effect if a scanner
* is set. If a provider is not set, and this is used for parsing,
* a default scanner RbnfLenientScannerProviderImpl will be set if
* it is available on the classpath. Otherwise this will have no effect.
*
* @param enabled If true, turns lenient-parse mode on; if false, turns it off.
* @see RbnfLenientScanner
* @see RbnfLenientScannerProvider
* @stable ICU 2.0
*/
public void setLenientParseMode(boolean enabled) {
lenientParse = enabled;
}
/**
* Returns true if lenient-parse mode is turned on. Lenient parsing is off
* by default.
* @return true if lenient-parse mode is turned on.
* @see #setLenientParseMode
* @stable ICU 2.0
*/
public boolean lenientParseEnabled() {
return lenientParse;
}
/**
* Sets the provider for the lenient scanner. If this has not been set,
* {@link #setLenientParseMode}
* has no effect. This is necessary to decouple collation from format code.
* @param scannerProvider the provider
* @see #setLenientParseMode
* @see #getLenientScannerProvider
* @stable ICU 4.4
*/
public void setLenientScannerProvider(RbnfLenientScannerProvider scannerProvider) {
this.scannerProvider = scannerProvider;
}
/**
* Returns the lenient scanner provider. If none was set, and lenient parse is
* enabled, this will attempt to instantiate a default scanner, setting it if
* it was successful. Otherwise this returns false.
*
* @see #setLenientScannerProvider
* @stable ICU 4.4
*/
public RbnfLenientScannerProvider getLenientScannerProvider() {
// there's a potential race condition if two threads try to set/get the scanner at
// the same time, but you get what you get, and you shouldn't be using this from
// multiple threads anyway.
if (scannerProvider == null && lenientParse && !lookedForScanner) {
try {
lookedForScanner = true;
Class cls = Class.forName("com.ibm.icu.impl.text.RbnfScannerProviderImpl");
RbnfLenientScannerProvider provider = (RbnfLenientScannerProvider)cls.newInstance();
setLenientScannerProvider(provider);
}
catch (Exception e) {
// any failure, we just ignore and return null
}
}
return scannerProvider;
}
/**
* Override the default rule set to use. If ruleSetName is null, reset
* to the initial default rule set.
* @param ruleSetName the name of the rule set, or null to reset the initial default.
* @throws IllegalArgumentException if ruleSetName is not the name of a public ruleset.
* @stable ICU 2.0
*/
public void setDefaultRuleSet(String ruleSetName) {
if (ruleSetName == null) {
if (publicRuleSetNames.length > 0) {
defaultRuleSet = findRuleSet(publicRuleSetNames[0]);
} else {
defaultRuleSet = null;
int n = ruleSets.length;
while (--n >= 0) {
String currentName = ruleSets[n].getName();
if (currentName.equals("%spellout-numbering") ||
currentName.equals("%digits-ordinal") ||
currentName.equals("%duration")) {
defaultRuleSet = ruleSets[n];
return;
}
}
n = ruleSets.length;
while (--n >= 0) {
if (ruleSets[n].isPublic()) {
defaultRuleSet = ruleSets[n];
break;
}
}
}
} else if (ruleSetName.startsWith("%%")) {
throw new IllegalArgumentException("cannot use private rule set: " + ruleSetName);
} else {
defaultRuleSet = findRuleSet(ruleSetName);
}
}
/**
* Return the name of the current default rule set.
* @return the name of the current default rule set, if it is public, else the empty string.
* @stable ICU 3.0
*/
public String getDefaultRuleSetName() {
if (defaultRuleSet != null && defaultRuleSet.isPublic()) {
return defaultRuleSet.getName();
}
return "";
}
/**
* Sets the decimal format symbols used by this formatter. The formatter uses a copy of the
* provided symbols.
*
* @param newSymbols desired DecimalFormatSymbols
* @see DecimalFormatSymbols
* @stable ICU 49
*/
public void setDecimalFormatSymbols(DecimalFormatSymbols newSymbols) {
if (newSymbols != null) {
decimalFormatSymbols = (DecimalFormatSymbols) newSymbols.clone();
if (decimalFormat != null) {
decimalFormat.setDecimalFormatSymbols(decimalFormatSymbols);
}
if (defaultInfinityRule != null) {
defaultInfinityRule = null;
getDefaultInfinityRule(); // Reset with the new DecimalFormatSymbols
}
if (defaultNaNRule != null) {
defaultNaNRule = null;
getDefaultNaNRule(); // Reset with the new DecimalFormatSymbols
}
// Apply the new decimalFormatSymbols by reparsing the rulesets
for (NFRuleSet ruleSet : ruleSets) {
ruleSet.setDecimalFormatSymbols(decimalFormatSymbols);
}
}
}
/**
* {@icu} Set a particular DisplayContext value in the formatter,
* such as CAPITALIZATION_FOR_STANDALONE. Note: For getContext, see
* NumberFormat.
*
* @param context The DisplayContext value to set.
* @stable ICU 53
*/
// Here we override the NumberFormat implementation in order to
// lazily initialize relevant items
@Override
public void setContext(DisplayContext context) {
super.setContext(context);
if (!capitalizationInfoIsSet &&
(context==DisplayContext.CAPITALIZATION_FOR_UI_LIST_OR_MENU || context==DisplayContext.CAPITALIZATION_FOR_STANDALONE)) {
initCapitalizationContextInfo(locale);
capitalizationInfoIsSet = true;
}
if (capitalizationBrkIter == null && (context==DisplayContext.CAPITALIZATION_FOR_BEGINNING_OF_SENTENCE ||
(context==DisplayContext.CAPITALIZATION_FOR_UI_LIST_OR_MENU && capitalizationForListOrMenu) ||
(context==DisplayContext.CAPITALIZATION_FOR_STANDALONE && capitalizationForStandAlone) )) {
capitalizationBrkIter = BreakIterator.getSentenceInstance(locale);
}
}
/**
* Returns the rounding mode.
*
* @return A rounding mode, between BigDecimal.ROUND_UP and
* BigDecimal.ROUND_UNNECESSARY.
* @see #setRoundingMode
* @see java.math.BigDecimal
* @stable ICU 56
*/
@Override
public int getRoundingMode() {
return roundingMode;
}
/**
* Sets the rounding mode. This has no effect unless the rounding increment is greater
* than zero.
*
* @param roundingMode A rounding mode, between BigDecimal.ROUND_UP and
* BigDecimal.ROUND_UNNECESSARY.
* @exception IllegalArgumentException if roundingMode is unrecognized.
* @see #getRoundingMode
* @see java.math.BigDecimal
* @stable ICU 56
*/
@Override
public void setRoundingMode(int roundingMode) {
if (roundingMode < BigDecimal.ROUND_UP || roundingMode > BigDecimal.ROUND_UNNECESSARY) {
throw new IllegalArgumentException("Invalid rounding mode: " + roundingMode);
}
this.roundingMode = roundingMode;
}
//-----------------------------------------------------------------------
// package-internal API
//-----------------------------------------------------------------------
/**
* Returns a reference to the formatter's default rule set. The default
* rule set is the last public rule set in the description, or the one
* most recently set by setDefaultRuleSet.
* @return The formatter's default rule set.
*/
NFRuleSet getDefaultRuleSet() {
return defaultRuleSet;
}
/**
* Returns the scanner to use for lenient parsing. The scanner is
* provided by the provider.
* @return The collator to use for lenient parsing, or null if lenient parsing
* is turned off.
*/
RbnfLenientScanner getLenientScanner() {
if (lenientParse) {
RbnfLenientScannerProvider provider = getLenientScannerProvider();
if (provider != null) {
return provider.get(locale, lenientParseRules);
}
}
return null;
}
/**
* Returns the DecimalFormatSymbols object that should be used by all DecimalFormat
* instances owned by this formatter. This object is lazily created: this function
* creates it the first time it's called.
* @return The DecimalFormatSymbols object that should be used by all DecimalFormat
* instances owned by this formatter.
*/
DecimalFormatSymbols getDecimalFormatSymbols() {
// lazy-evaluate the DecimalFormatSymbols object. This object
// is shared by all DecimalFormat instances belonging to this
// formatter
if (decimalFormatSymbols == null) {
decimalFormatSymbols = new DecimalFormatSymbols(locale);
}
return decimalFormatSymbols;
}
DecimalFormat getDecimalFormat() {
if (decimalFormat == null) {
// Don't use NumberFormat.getInstance, which can cause a recursive call
String pattern = getPattern(locale, NUMBERSTYLE);
decimalFormat = new DecimalFormat(pattern, getDecimalFormatSymbols());
}
return decimalFormat;
}
PluralFormat createPluralFormat(PluralRules.PluralType pluralType, String pattern) {
return new PluralFormat(locale, pluralType, pattern, getDecimalFormat());
}
/**
* Returns the default rule for infinity. This object is lazily created: this function
* creates it the first time it's called.
*/
NFRule getDefaultInfinityRule() {
if (defaultInfinityRule == null) {
defaultInfinityRule = new NFRule(this, "Inf: " + getDecimalFormatSymbols().getInfinity());
}
return defaultInfinityRule;
}
/**
* Returns the default rule for NaN. This object is lazily created: this function
* creates it the first time it's called.
*/
NFRule getDefaultNaNRule() {
if (defaultNaNRule == null) {
defaultNaNRule = new NFRule(this, "NaN: " + getDecimalFormatSymbols().getNaN());
}
return defaultNaNRule;
}
//-----------------------------------------------------------------------
// construction implementation
//-----------------------------------------------------------------------
/**
* This extracts the special information from the rule sets before the
* main parsing starts. Extra whitespace must have already been removed
* from the description. If found, the special information is removed from the
* description and returned, otherwise the description is unchanged and null
* is returned. Note: the trailing semicolon at the end of the special
* rules is stripped.
* @param description the rbnf description with extra whitespace removed
* @param specialName the name of the special rule text to extract
* @return the special rule text, or null if the rule was not found
*/
private String extractSpecial(StringBuilder description, String specialName) {
String result = null;
int lp = description.indexOf(specialName);
if (lp != -1) {
// we've got to make sure we're not in the middle of a rule
// (where specialName would actually get treated as
// rule text)
if (lp == 0 || description.charAt(lp - 1) == ';') {
// locate the beginning and end of the actual special
// rules (there may be whitespace between the name and
// the first token in the description)
int lpEnd = description.indexOf(";%", lp);
if (lpEnd == -1) {
lpEnd = description.length() - 1; // later we add 1 back to get the '%'
}
int lpStart = lp + specialName.length();
while (lpStart < lpEnd &&
PatternProps.isWhiteSpace(description.charAt(lpStart))) {
++lpStart;
}
// copy out the special rules
result = description.substring(lpStart, lpEnd);
// remove the special rule from the description
description.delete(lp, lpEnd+1); // delete the semicolon but not the '%'
}
}
return result;
}
/**
* This function parses the description and uses it to build all of
* internal data structures that the formatter uses to do formatting
* @param description The description of the formatter's desired behavior.
* This is either passed in by the caller or loaded out of a resource
* by one of the constructors, and is in the description format specified
* in the class docs.
*/
private void init(String description, String[][] localizations) {
initLocalizations(localizations);
// start by stripping the trailing whitespace from all the rules
// (this is all the whitespace following each semicolon in the
// description). This allows us to look for rule-set boundaries
// by searching for ";%" without having to worry about whitespace
// between the ; and the %
StringBuilder descBuf = stripWhitespace(description);
// check to see if there's a set of lenient-parse rules. If there
// is, pull them out into our temporary holding place for them,
// and delete them from the description before the real description-
// parsing code sees them
lenientParseRules = extractSpecial(descBuf, "%%lenient-parse:");
postProcessRules = extractSpecial(descBuf, "%%post-process:");
// pre-flight parsing the description and count the number of
// rule sets (";%" marks the end of one rule set and the beginning
// of the next)
int numRuleSets = 1;
int p = 0;
while ((p = descBuf.indexOf(";%", p)) != -1) {
++numRuleSets;
p += 2; // Skip the length of ";%"
}
// our rule list is an array of the appropriate size
ruleSets = new NFRuleSet[numRuleSets];
ruleSetsMap = new HashMap<>(numRuleSets * 2 + 1);
defaultRuleSet = null;
// Used to count the number of public rule sets
// Public rule sets have names that begin with % instead of %%.
int publicRuleSetCount = 0;
// divide up the descriptions into individual rule-set descriptions
// and store them in a temporary array. At each step, we also
// new up a rule set, but all this does is initialize its name
// and remove it from its description. We can't actually parse
// the rest of the descriptions and finish initializing everything
// because we have to know the names and locations of all the rule
// sets before we can actually set everything up
String[] ruleSetDescriptions = new String[numRuleSets];
int curRuleSet = 0;
int start = 0;
while (curRuleSet < ruleSets.length) {
p = descBuf.indexOf(";%", start);
if (p < 0) {
p = descBuf.length() - 1;
}
ruleSetDescriptions[curRuleSet] = descBuf.substring(start, p + 1);
NFRuleSet ruleSet = new NFRuleSet(this, ruleSetDescriptions, curRuleSet);
ruleSets[curRuleSet] = ruleSet;
String currentName = ruleSet.getName();
ruleSetsMap.put(currentName, ruleSet);
if (!currentName.startsWith("%%")) {
++publicRuleSetCount;
if (defaultRuleSet == null
&& currentName.equals("%spellout-numbering")
|| currentName.equals("%digits-ordinal")
|| currentName.equals("%duration"))
{
defaultRuleSet = ruleSet;
}
}
++curRuleSet;
start = p + 1;
}
// now we can take note of the formatter's default rule set, which
// is the last public rule set in the description (it's the last
// rather than the first so that a user can create a new formatter
// from an existing formatter and change its default behavior just
// by appending more rule sets to the end)
// {dlf} Initialization of a fraction rule set requires the default rule
// set to be known. For purposes of initialization, this is always the
// last public rule set, no matter what the localization data says.
// Set the default ruleset to the last public ruleset, unless one of the predefined
// ruleset names %spellout-numbering, %digits-ordinal, or %duration is found
if (defaultRuleSet == null) {
for (int i = ruleSets.length - 1; i >= 0; --i) {
if (!ruleSets[i].getName().startsWith("%%")) {
defaultRuleSet = ruleSets[i];
break;
}
}
}
if (defaultRuleSet == null) {
defaultRuleSet = ruleSets[ruleSets.length - 1];
}
// finally, we can go back through the temporary descriptions
// list and finish setting up the substructure
for (int i = 0; i < ruleSets.length; i++) {
ruleSets[i].parseRules(ruleSetDescriptions[i]);
}
// Now that the rules are initialized, the 'real' default rule
// set can be adjusted by the localization data.
// prepare an array of the proper size and copy the names into it
String[] publicRuleSetTemp = new String[publicRuleSetCount];
publicRuleSetCount = 0;
for (int i = ruleSets.length - 1; i >= 0; i--) {
if (!ruleSets[i].getName().startsWith("%%")) {
publicRuleSetTemp[publicRuleSetCount++] = ruleSets[i].getName();
}
}
if (publicRuleSetNames != null) {
// confirm the names, if any aren't in the rules, that's an error
// it is ok if the rules contain public rule sets that are not in this list
loop: for (int i = 0; i < publicRuleSetNames.length; ++i) {
String name = publicRuleSetNames[i];
for (int j = 0; j < publicRuleSetTemp.length; ++j) {
if (name.equals(publicRuleSetTemp[j])) {
continue loop;
}
}
throw new IllegalArgumentException("did not find public rule set: " + name);
}
defaultRuleSet = findRuleSet(publicRuleSetNames[0]); // might be different
} else {
publicRuleSetNames = publicRuleSetTemp;
}
}
/**
* Take the localizations array and create a Map from the locale strings to
* the localization arrays.
*/
private void initLocalizations(String[][] localizations) {
if (localizations != null) {
publicRuleSetNames = localizations[0].clone();
Map m = new HashMap<>();
for (int i = 1; i < localizations.length; ++i) {
String[] data = localizations[i];
String loc = data[0];
String[] names = new String[data.length-1];
if (names.length != publicRuleSetNames.length) {
throw new IllegalArgumentException("public name length: " + publicRuleSetNames.length +
" != localized names[" + i + "] length: " + names.length);
}
System.arraycopy(data, 1, names, 0, names.length);
m.put(loc, names);
}
if (!m.isEmpty()) {
ruleSetDisplayNames = m;
}
}
}
/**
* Set capitalizationForListOrMenu, capitalizationForStandAlone
*/
private void initCapitalizationContextInfo(ULocale theLocale) {
ICUResourceBundle rb = (ICUResourceBundle) UResourceBundle.getBundleInstance(ICUData.ICU_BASE_NAME, theLocale);
try {
ICUResourceBundle rdb = rb.getWithFallback("contextTransforms/number-spellout");
int[] intVector = rdb.getIntVector();
if (intVector.length >= 2) {
capitalizationForListOrMenu = (intVector[0] != 0);
capitalizationForStandAlone = (intVector[1] != 0);
}
} catch (MissingResourceException e) {
// use default
}
}
/**
* This function is used by init() to strip whitespace between rules (i.e.,
* after semicolons).
* @param description The formatter description
* @return The description with all the whitespace that follows semicolons
* taken out.
*/
private StringBuilder stripWhitespace(String description) {
// since we don't have a method that deletes characters (why?!!)
// create a new StringBuffer to copy the text into
StringBuilder result = new StringBuilder();
int descriptionLength = description.length();
// iterate through the characters...
int start = 0;
while (start < descriptionLength) {
// seek to the first non-whitespace character...
while (start < descriptionLength
&& PatternProps.isWhiteSpace(description.charAt(start)))
{
++start;
}
//if the first non-whitespace character is semicolon, skip it and continue
if (start < descriptionLength && description.charAt(start) == ';') {
start += 1;
continue;
}
// locate the next semicolon in the text and copy the text from
// our current position up to that semicolon into the result
int p = description.indexOf(';', start);
if (p == -1) {
// or if we don't find a semicolon, just copy the rest of
// the string into the result
result.append(description.substring(start));
break;
}
else if (p < descriptionLength) {
result.append(description.substring(start, p + 1));
start = p + 1;
}
else {
// when we get here, we've seeked off the end of the string, and
// we terminate the loop (we continue until *start* is -1 rather
// than until *p* is -1, because otherwise we'd miss the last
// rule in the description)
break;
}
}
return result;
}
//-----------------------------------------------------------------------
// formatting implementation
//-----------------------------------------------------------------------
/**
* Bottleneck through which all the public format() methods
* that take a double pass. By the time we get here, we know
* which rule set we're using to do the formatting.
* @param number The number to format
* @param ruleSet The rule set to use to format the number
* @return The text that resulted from formatting the number
*/
private String format(double number, NFRuleSet ruleSet) {
// all API format() routines that take a double vector through
// here. Create an empty string buffer where the result will
// be built, and pass it to the rule set (along with an insertion
// position of 0 and the number being formatted) to the rule set
// for formatting
StringBuilder result = new StringBuilder();
if (getRoundingMode() != BigDecimal.ROUND_UNNECESSARY && !Double.isNaN(number) && !Double.isInfinite(number)) {
// We convert to a string because BigDecimal insists on excessive precision.
number = new BigDecimal(Double.toString(number)).setScale(getMaximumFractionDigits(), roundingMode).doubleValue();
}
ruleSet.format(number, result, 0, 0);
postProcess(result, ruleSet);
return result.toString();
}
/**
* Bottleneck through which all the public format() methods
* that take a long pass. By the time we get here, we know
* which rule set we're using to do the formatting.
* @param number The number to format
* @param ruleSet The rule set to use to format the number
* @return The text that resulted from formatting the number
*/
private String format(long number, NFRuleSet ruleSet) {
// all API format() routines that take a double vector through
// here. We have these two identical functions-- one taking a
// double and one taking a long-- the couple digits of precision
// that long has but double doesn't (both types are 8 bytes long,
// but double has to borrow some of the mantissa bits to hold
// the exponent).
// Create an empty string buffer where the result will
// be built, and pass it to the rule set (along with an insertion
// position of 0 and the number being formatted) to the rule set
// for formatting
StringBuilder result = new StringBuilder();
if (number == Long.MIN_VALUE) {
// We can't handle this value right now. Provide an accurate default value.
result.append(getDecimalFormat().format(Long.MIN_VALUE));
}
else {
ruleSet.format(number, result, 0, 0);
}
postProcess(result, ruleSet);
return result.toString();
}
/**
* Post-process the rules if we have a post-processor.
*/
private void postProcess(StringBuilder result, NFRuleSet ruleSet) {
if (postProcessRules != null) {
if (postProcessor == null) {
int ix = postProcessRules.indexOf(";");
if (ix == -1) {
ix = postProcessRules.length();
}
String ppClassName = postProcessRules.substring(0, ix).trim();
try {
Class cls = Class.forName(ppClassName);
postProcessor = (RBNFPostProcessor)cls.newInstance();
postProcessor.init(this, postProcessRules);
}
catch (Exception e) {
// if debug, print it out
if (DEBUG) System.out.println("could not locate " + ppClassName + ", error " +
e.getClass().getName() + ", " + e.getMessage());
postProcessor = null;
postProcessRules = null; // don't try again
return;
}
}
postProcessor.process(result, ruleSet);
}
}
/**
* Adjust capitalization of formatted result for display context
*/
private String adjustForContext(String result) {
DisplayContext capitalization = getContext(DisplayContext.Type.CAPITALIZATION);
if (capitalization != DisplayContext.CAPITALIZATION_NONE && result != null && result.length() > 0
&& UCharacter.isLowerCase(result.codePointAt(0)))
{
if ( capitalization==DisplayContext.CAPITALIZATION_FOR_BEGINNING_OF_SENTENCE ||
(capitalization == DisplayContext.CAPITALIZATION_FOR_UI_LIST_OR_MENU && capitalizationForListOrMenu) ||
(capitalization == DisplayContext.CAPITALIZATION_FOR_STANDALONE && capitalizationForStandAlone) ) {
if (capitalizationBrkIter == null) {
// should only happen when deserializing, etc.
capitalizationBrkIter = BreakIterator.getSentenceInstance(locale);
}
return UCharacter.toTitleCase(locale, result, capitalizationBrkIter,
UCharacter.TITLECASE_NO_LOWERCASE | UCharacter.TITLECASE_NO_BREAK_ADJUSTMENT);
}
}
return result;
}
/**
* Returns the named rule set. Throws an IllegalArgumentException
* if this formatter doesn't have a rule set with that name.
* @param name The name of the desired rule set
* @return The rule set with that name
*/
NFRuleSet findRuleSet(String name) throws IllegalArgumentException {
NFRuleSet result = ruleSetsMap.get(name);
if (result == null) {
throw new IllegalArgumentException("No rule set named " + name);
}
return result;
}
}