package client.net.sf.saxon.ce.expr.number;
import client.net.sf.saxon.ce.lib.Numberer;
import client.net.sf.saxon.ce.tree.util.UTF16CharacterSet;
import client.net.sf.saxon.ce.tree.util.FastStringBuffer;
import client.net.sf.saxon.ce.value.StringValue;
/**
* Class AbstractNumberer is a base implementation of Numberer that provides language-independent
* default numbering
* This supports the xsl:number element.
* Methods and data are declared as protected, and static is avoided, to allow easy subclassing.
*
* @author Michael H. Kay
*/
public abstract class AbstractNumberer implements Numberer {
private String country;
public static final int UPPER_CASE = 0;
public static final int LOWER_CASE = 1;
public static final int TITLE_CASE = 2;
/**
* Set the country used by this numberer (currenly used only for names of timezones)
*/
public void setCountry(String country) {
this.country = country;
}
/**
* Get the country used by this numberer.
*
* @return the country used by this numberer, or null if no country has been set
*/
public String getCountry() {
return country;
}
/**
* Format a number into a string. This method is provided for backwards compatibility. It merely
* calls the other format method after constructing a RegularGroupFormatter. The method is final;
* localization subclasses should implement the method
* {@link #format(long, String, NumericGroupFormatter, String, String)} rather than this method.
*
* @param number The number to be formatted
* @param picture The format token. This is a single component of the format attribute
* of xsl:number, e.g. "1", "01", "i", or "a"
* @param groupSize number of digits per group (0 implies no grouping)
* @param groupSeparator string to appear between groups of digits
* @param letterValue The letter-value specified to xsl:number: "alphabetic" or
* "traditional". Can also be an empty string or null.
* @param ordinal The value of the ordinal attribute specified to xsl:number
* The value "yes" indicates that ordinal numbers should be used; "" or null indicates
* that cardinal numbers
* @return the formatted number. Note that no errors are reported; if the request
* is invalid, the number is formatted as if the string() function were used.
*/
public final String format(long number,
String picture,
int groupSize,
String groupSeparator,
String letterValue,
String ordinal) {
return format(number, picture, new RegularGroupFormatter(groupSize, groupSeparator), letterValue, ordinal);
}
/**
* Format a number into a string
*
* @param number The number to be formatted
* @param picture The format token. This is a single component of the format attribute
* of xsl:number, e.g. "1", "01", "i", or "a"
* @param numGroupFormatter object contains separators to appear between groups of digits
* @param letterValue The letter-value specified to xsl:number: "alphabetic" or
* "traditional". Can also be an empty string or null.
* @param ordinal The value of the ordinal attribute specified to xsl:number
* The value "yes" indicates that ordinal numbers should be used; "" or null indicates
* that cardinal numbers
* @return the formatted number. Note that no errors are reported; if the request
* is invalid, the number is formatted as if the string() function were used.
*/
public String format(long number,
String picture,
NumericGroupFormatter numGroupFormatter,
String letterValue,
String ordinal) {
if (number < 0) {
return "" + number;
}
if (picture == null || picture.length() == 0) {
return "" + number;
}
int pictureLength = StringValue.getStringLength(picture);
FastStringBuffer sb = new FastStringBuffer(FastStringBuffer.TINY);
int formchar = picture.charAt(0);
if (UTF16CharacterSet.isHighSurrogate(formchar)) {
formchar = UTF16CharacterSet.combinePair((char) formchar, picture.charAt(1));
}
switch (formchar) {
case '0':
case '1':
sb.append(toRadical(number, westernDigits, pictureLength, numGroupFormatter));
if (ordinal != null && ordinal.length() > 0) {
sb.append(ordinalSuffix(ordinal, number));
}
break;
case 'A':
if (number == 0) {
return "0";
}
return toAlphaSequence(number, latinUpper);
case 'a':
if (number == 0) {
return "0";
}
return toAlphaSequence(number, latinLower);
case 'w':
case 'W':
int wordCase;
if (picture.equals("W")) {
wordCase = UPPER_CASE;
} else if (picture.equals("w")) {
wordCase = LOWER_CASE;
} else {
// includes cases like "ww" or "Wz". The action here is conformant, but it's not clear what's best
wordCase = TITLE_CASE;
}
if (ordinal != null && ordinal.length() > 0) {
return toOrdinalWords(ordinal, number, wordCase);
} else {
return toWords(number, wordCase);
}
case 'i':
if (number == 0) {
return "0";
}
if (letterValue == null || letterValue.length() == 0 ||
letterValue.equals("traditional")) {
return toRoman(number);
} else {
alphaDefault(number, 'i', sb);
}
break;
case 'I':
if (number == 0) {
return "0";
}
if (letterValue == null || letterValue.length() == 0 ||
letterValue.equals("traditional")) {
return toRoman(number).toUpperCase();
} else {
alphaDefault(number, 'I', sb);
}
break;
case '\u2460':
// circled digits
if (number == 0 || number > 20) {
return "" + number;
}
return "" + (char) (0x2460 + number - 1);
case '\u2474':
// parenthesized digits
if (number == 0 || number > 20) {
return "" + number;
}
return "" + (char) (0x2474 + number - 1);
case '\u2488':
// digit full stop
if (number == 0 || number > 20) {
return "" + number;
}
return "" + (char) (0x2488 + number - 1);
case '\u0391':
if (number == 0) {
return "0";
}
return toAlphaSequence(number, greekUpper);
case '\u03b1':
if (number == 0) {
return "0";
}
return toAlphaSequence(number, greekLower);
case '\u0410':
if (number == 0) {
return "0";
}
return toAlphaSequence(number, cyrillicUpper);
case '\u0430':
if (number == 0) {
return "0";
}
return toAlphaSequence(number, cyrillicLower);
case '\u05d0':
if (number == 0) {
return "0";
}
return toAlphaSequence(number, hebrew);
case '\u3042':
if (number == 0) {
return "0";
}
return toAlphaSequence(number, hiraganaA);
case '\u30a2':
if (number == 0) {
return "0";
}
return toAlphaSequence(number, katakanaA);
case '\u3044':
if (number == 0) {
return "0";
}
return toAlphaSequence(number, hiraganaI);
case '\u30a4':
if (number == 0) {
return "0";
}
return toAlphaSequence(number, katakanaI);
case '\u4e00':
return toJapanese(number);
default:
int digitValue = Alphanumeric.getDigitValue(formchar);
if (digitValue >= 0) {
int zero = formchar - digitValue;
int[] digits = new int[10];
for (int z = 0; z <= 9; z++) {
digits[z] = zero + z;
}
return toRadical(number, digits, pictureLength, numGroupFormatter);
} else {
if (number == 0) {
return "0";
}
if (formchar < '\u1100') {
alphaDefault(number, (char) formchar, sb);
} else {
// fallback to western numbering
return toRadical(number, westernDigits, pictureLength, numGroupFormatter);
}
break;
}
}
return sb.toString();
}
/**
* Construct the ordinal suffix for a number, for example "st", "nd", "rd". The default
* (language-neutral) implementation returns a zero-length string
*
* @param ordinalParam the value of the ordinal attribute (used in non-English
* language implementations)
* @param number the number being formatted
* @return the ordinal suffix to be appended to the formatted number
*/
protected String ordinalSuffix(String ordinalParam, long number) {
return "";
}
protected static final int[] westernDigits = {'0', '1', '2', '3', '4', '5', '6', '7', '8', '9'};
protected static final String latinUpper =
"ABCDEFGHIJKLMNOPQRSTUVWXYZ";
protected static final String latinLower =
"abcdefghijklmnopqrstuvwxyz";
protected static final String greekUpper =
"\u0391\u0392\u0393\u0394\u0395\u0396\u0397\u0398\u0399\u039a" +
"\u039b\u039c\u039d\u039e\u039f\u03a0\u03a1\u03a2\u03a3\u03a4" +
"\u03a5\u03a6\u03a7\u03a8\u03a9";
protected static final String greekLower =
"\u03b1\u03b2\u03b3\u03b4\u03b5\u03b6\u03b7\u03b8\u03b9\u03ba" +
"\u03bb\u03bc\u03bd\u03be\u03bf\u03c0\u03c1\u03c2\u03c3\u03c4" +
"\u03c5\u03c6\u03c7\u03c8\u03c9";
// Cyrillic information from Dmitry Kirsanov [dmitry@kirsanov.com]
// (based on his personal knowledge of Russian texts, not any authoritative source)
protected static final String cyrillicUpper =
"\u0410\u0411\u0412\u0413\u0414\u0415\u0416\u0417\u0418" +
"\u041a\u041b\u041c\u041d\u041e\u041f\u0420\u0421\u0421\u0423" +
"\u0424\u0425\u0426\u0427\u0428\u0429\u042b\u042d\u042e\u042f";
protected static final String cyrillicLower =
"\u0430\u0431\u0432\u0433\u0434\u0435\u0436\u0437\u0438" +
"\u043a\u043b\u043c\u043d\u043e\u043f\u0440\u0441\u0441\u0443" +
"\u0444\u0445\u0446\u0447\u0448\u0449\u044b\u044d\u044e\u044f";
protected static final String hebrew =
"\u05d0\u05d1\u05d2\u05d3\u05d4\u05d5\u05d6\u05d7\u05d8\u05d9\u05db\u05dc" +
"\u05de\u05e0\u05e1\u05e2\u05e4\u05e6\u05e7\u05e8\u05e9\u05ea";
// The following Japanese sequences were supplied by
// MURAKAMI Shinyu [murakami@nadita.com]
protected static final String hiraganaA =
"\u3042\u3044\u3046\u3048\u304a\u304b\u304d\u304f\u3051\u3053" +
"\u3055\u3057\u3059\u305b\u305d\u305f\u3061\u3064\u3066\u3068" +
"\u306a\u306b\u306c\u306d\u306e\u306f\u3072\u3075\u3078\u307b" +
"\u307e\u307f\u3080\u3081\u3082\u3084\u3086\u3088\u3089\u308a" +
"\u308b\u308c\u308d\u308f\u3092\u3093";
protected static final String katakanaA =
"\u30a2\u30a4\u30a6\u30a8\u30aa\u30ab\u30ad\u30af\u30b1\u30b3" +
"\u30b5\u30b7\u30b9\u30bb\u30bd\u30bf\u30c1\u30c4\u30c6\u30c8" +
"\u30ca\u30cb\u30cc\u30cd\u30ce\u30cf\u30d2\u30d5\u30d8\u30db" +
"\u30de\u30df\u30e0\u30e1\u30e2\u30e4\u30e6\u30e8\u30e9\u30ea" +
"\u30eb\u30ec\u30ed\u30ef\u30f2\u30f3";
protected static final String hiraganaI =
"\u3044\u308d\u306f\u306b\u307b\u3078\u3068\u3061\u308a\u306c" +
"\u308b\u3092\u308f\u304b\u3088\u305f\u308c\u305d\u3064\u306d" +
"\u306a\u3089\u3080\u3046\u3090\u306e\u304a\u304f\u3084\u307e" +
"\u3051\u3075\u3053\u3048\u3066\u3042\u3055\u304d\u3086\u3081" +
"\u307f\u3057\u3091\u3072\u3082\u305b\u3059";
protected static final String katakanaI =
"\u30a4\u30ed\u30cf\u30cb\u30db\u30d8\u30c8\u30c1\u30ea\u30cc" +
"\u30eb\u30f2\u30ef\u30ab\u30e8\u30bf\u30ec\u30bd\u30c4\u30cd" +
"\u30ca\u30e9\u30e0\u30a6\u30f0\u30ce\u30aa\u30af\u30e4\u30de" +
"\u30b1\u30d5\u30b3\u30a8\u30c6\u30a2\u30b5\u30ad\u30e6\u30e1" +
"\u30df\u30b7\u30f1\u30d2\u30e2\u30bb\u30b9";
/**
* Default processing with an alphabetic format token: use the contiguous
* range of Unicode letters starting with that token.
*
* @param number the number to be formatted
* @param formchar the format character, for example 'A' for the numbering sequence A,B,C
* @param sb buffer to hold the result of the formatting
*/
protected void alphaDefault(long number, char formchar, FastStringBuffer sb) {
int min = (int) formchar;
int max = (int) formchar;
// use the contiguous range of letters starting with the specified one
while (Character.isLetterOrDigit((char) (max + 1))) {
max++;
}
sb.append(toAlpha(number, min, max));
}
/**
* Format the number as an alphabetic label using the alphabet consisting
* of consecutive Unicode characters from min to max
*
* @param number the number to be formatted
* @param min the start of the Unicode codepoint range
* @param max the end of the Unicode codepoint range
* @return the formatted number
*/
protected String toAlpha(long number, int min, int max) {
if (number <= 0) {
return "" + number;
}
int range = max - min + 1;
char last = (char) (((number - 1) % range) + min);
if (number > range) {
return toAlpha((number - 1) / range, min, max) + last;
} else {
return "" + last;
}
}
/**
* Convert the number into an alphabetic label using a given alphabet.
* For example, if the alphabet is "xyz" the sequence is x, y, z, xx, xy, xz, ....
*
* @param number the number to be formatted
* @param alphabet a string containing the characters to be used, for example "abc...xyz"
* @return the formatted number
*/
protected String toAlphaSequence(long number, String alphabet) {
if (number <= 0) {
return "" + number;
}
int range = alphabet.length();
char last = alphabet.charAt((int) ((number - 1) % range));
if (number > range) {
return toAlphaSequence((number - 1) / range, alphabet) + last;
} else {
return "" + last;
}
}
/**
* Convert the number into a decimal or other representation using the given set of
* digits.
* For example, if the digits are "01" the sequence is 1, 10, 11, 100, 101, 110, 111, ...
* More commonly, the digits will be "0123456789", giving the usual decimal numbering.
*
* @param number the number to be formatted
* @param digits the codepoints to be used for the digits
* @param pictureLength the length of the picture that is significant: for example "3" if the
* picture is "001"
* @param numGroupFormatter an object that encapsulates the rules for inserting grouping separators
* @return the formatted number
*/
private String toRadical(long number, int[] digits, int pictureLength,
NumericGroupFormatter numGroupFormatter) {
FastStringBuffer temp = new FastStringBuffer(FastStringBuffer.TINY);
int base = digits.length;
FastStringBuffer s = new FastStringBuffer(FastStringBuffer.TINY);
long n = number;
int count = 0;
while (n > 0) {
int digit = digits[(int) (n % base)];
s.prependWideChar(digit);
count++;
n = n / base;
}
for (int i = 0; i < (pictureLength - count); i++) {
temp.appendWideChar(digits[0]);
}
temp.append(s);
if (numGroupFormatter == null) {
return temp.toString();
}
return numGroupFormatter.format(temp);
}
/**
* Generate a Roman numeral (in lower case)
*
* @param n the number to be formatted
* @return the Roman numeral representation of the number in lower case
*/
public static String toRoman(long n) {
if (n <= 0 || n > 9999) {
return "" + n;
}
return romanThousands[(int) n / 1000] +
romanHundreds[((int) n / 100) % 10] +
romanTens[((int) n / 10) % 10] +
romanUnits[(int) n % 10];
}
// Roman numbers beyond 4000 use overlining and other conventions which we won't
// attempt to reproduce. We'll go high enough to handle present-day Gregorian years.
private static String[] romanThousands =
{"", "m", "mm", "mmm", "mmmm", "mmmmm", "mmmmmm", "mmmmmmm", "mmmmmmmm", "mmmmmmmmm"};
private static String[] romanHundreds =
{"", "c", "cc", "ccc", "cd", "d", "dc", "dcc", "dccc", "cm"};
private static String[] romanTens =
{"", "x", "xx", "xxx", "xl", "l", "lx", "lxx", "lxxx", "xc"};
private static String[] romanUnits =
{"", "i", "ii", "iii", "iv", "v", "vi", "vii", "viii", "ix"};
/**
* Format the number in Japanese.
* @param number the number to be formatted: formatted in Western decimal style unless in the range 1 to 9999
* @return the Japanese Kanji representation of the number if in the range 1-9999
*/
public String toJapanese(long number) {
FastStringBuffer fsb = new FastStringBuffer(FastStringBuffer.TINY);
if (number == 0) {
fsb.appendWideChar(0x3007);
} else if (number <= 9999) {
toJapanese((int)number, fsb, false);
} else {
fsb.append("" + number);
}
return fsb.toString();
}
/**
* Format the number in Japanese.
* @param nr the number to be formatted: must be in the range 1 to 9999 (or 0 on a recursive call)
* @param fsb buffer to receive the formatted number
* @return the Japanese Kanji representation of the number in lower case
*/
private static void toJapanese(int nr, FastStringBuffer fsb, boolean isInitial) {
if (nr == 0) {
// no action (not used at top level)
} else if (nr <= 9) {
if (!(nr == 1 && isInitial)) {
fsb.appendWideChar(kanjiDigits[nr]);
}
} else if (nr == 10) {
fsb.appendWideChar(0x5341);
} else if (nr <= 99) {
toJapanese(nr / 10, fsb, true);
fsb.appendWideChar(0x5341);
toJapanese(nr % 10, fsb, false);
} else if (nr <= 999) {
toJapanese(nr / 100, fsb, true);
fsb.appendWideChar(0x767e);
toJapanese(nr % 100, fsb, false);
} else if (nr <= 9999) {
toJapanese(nr / 1000, fsb, true);
fsb.appendWideChar(0x5343);
toJapanese(nr % 1000, fsb, false);
}
}
private static final int[] kanjiDigits =
{0x3007, 0x4e00, 0x4e8c, 0x4e09, 0x56db, 0x4e94, 0x516d, 0x4e03, 0x516b, 0x4e5d};
/**
* Show the number as words in title case. (We choose title case because
* the result can then be converted algorithmically to lower case or upper case).
*
* @param number the number to be formatted
* @return the number formatted as English words
*/
public abstract String toWords(long number);
/**
* Format a number as English words with specified case options
*
* @param number the number to be formatted
* @param wordCase the required case for example {@link #UPPER_CASE},
* {@link #LOWER_CASE}, {@link #TITLE_CASE}
* @return the formatted number
*/
public String toWords(long number, int wordCase) {
String s;
if (number == 0) {
s = "Zero";
} else {
s = toWords(number);
}
if (wordCase == UPPER_CASE) {
return s.toUpperCase();
} else if (wordCase == LOWER_CASE) {
return s.toLowerCase();
} else {
return s;
}
}
/**
* Show an ordinal number as English words in a requested case (for example, Twentyfirst)
*
* @param ordinalParam the value of the "ordinal" attribute as supplied by the user
* @param number the number to be formatted
* @param wordCase the required case for example {@link #UPPER_CASE},
* {@link #LOWER_CASE}, {@link #TITLE_CASE}
* @return the formatted number
*/
public abstract String toOrdinalWords(String ordinalParam, long number, int wordCase);
/**
* Get a month name or abbreviation
*
* @param month The month number (1=January, 12=December)
* @param minWidth The minimum number of characters
* @param maxWidth The maximum number of characters
*/
public abstract String monthName(int month, int minWidth, int maxWidth);
/**
* Get a day name or abbreviation
*
* @param day The day of the week (1=Monday, 7=Sunday)
* @param minWidth The minimum number of characters
* @param maxWidth The maximum number of characters
*/
public abstract String dayName(int day, int minWidth, int maxWidth);
/**
* Get an am/pm indicator. Default implementation works for English, on the basis that some
* other languages might like to copy this (most non-English countries don't actually use the
* 12-hour clock, so it's irrelevant).
*
* @param minutes the minutes within the day
* @param minWidth minimum width of output
* @param maxWidth maximum width of output
* @return the AM or PM indicator
*/
public String halfDayName(int minutes, int minWidth, int maxWidth) {
String s;
if (minutes == 0 && maxWidth >= 8) {
s = "Midnight";
} else if (minutes < 12 * 60) {
switch (maxWidth) {
case 1:
s = "A";
break;
case 2:
case 3:
s = "Am";
break;
default:
s = "A.M.";
}
} else if (minutes == 12 * 60 && maxWidth >= 8) {
s = "Noon";
} else {
switch (maxWidth) {
case 1:
s = "P";
break;
case 2:
case 3:
s = "Pm";
break;
default:
s = "P.M.";
}
}
return s;
}
/**
* Get an ordinal suffix for a particular component of a date/time.
*
* @param component the component specifier from a format-dateTime picture, for
* example "M" for the month or "D" for the day.
* @return a string that is acceptable in the ordinal attribute of xsl:number
* to achieve the required ordinal representation. For example, "-e" for the day component
* in German, to have the day represented as "dritte August".
*/
public String getOrdinalSuffixForDateTime(String component) {
return "yes";
}
/**
* Get the name for an era (e.g. "BC" or "AD")
*
* @param year the proleptic gregorian year, using "0" for the year before 1AD
*/
public String getEraName(int year) {
return (year > 0 ? "AD" : "BC");
}
/**
* Get the name of a calendar
*
* @param code The code representing the calendar as in the XSLT 2.0 spec, e.g. AD for the Gregorian calendar
*/
public String getCalendarName(String code) {
if (code.equals("AD")) {
return "Gregorian";
} else {
return code;
}
}
}
// This Source Code Form is subject to the terms of the Mozilla Public License, v. 2.0.
// If a copy of the MPL was not distributed with this file, You can obtain one at http://mozilla.org/MPL/2.0/.
// This Source Code Form is “Incompatible With Secondary Licenses”, as defined by the Mozilla Public License, v. 2.0.