/* * \x20@(#)Pattern.java 1.111 05/01/04 * * Copyright 2005 Sun Microsystems, Inc. All rights reserved. * SUN PROPRIETARY/CONFIDENTIAL. Use is subject to license terms. */ package java.util.regex; import java.security.AccessController; import java.security.PrivilegedAction; import java.text.CharacterIterator; import sun.text.Normalizer; import java.util.ArrayList; import java.util.HashMap; /** * A compiled representation of a regular expression. * *

A regular expression, specified as a string, must first be compiled into * an instance of this class. The resulting pattern can then be used to create * a {@link Matcher} object that can match arbitrary {@link * java.lang.CharSequence character sequences} against the regular * expression. All of the state involved in performing a match resides in the * matcher, so many matchers can share the same pattern. * *

A typical invocation sequence is thus * *

 * Pattern p = Pattern.{@link #compile compile}("a*b");
 * Matcher m = p.{@link #matcher matcher}("aaaaab");
 * boolean b = m.{@link Matcher#matches matches}();


 *
 *  A {@link #matches matches} method is defined by this class as a
 * convenience for when a regular expression is used just once.  This method
 * compiles an expression and matches an input sequence against it in a single
 * invocation.  The statement
 *
 * 
 * boolean b = Pattern.matches("a*b", "aaaaab");
 *
 * is equivalent to the three statements above, though for repeated matches it
 * is less efficient since it does not allow the compiled pattern to be reused.
 *
 *  Instances of this class are immutable and are safe for use by multiple
 * concurrent threads.  Instances of the {@link Matcher} class are not safe for
 * such use.
 *
 *
 * 
 * 
 Summary of regular-expression constructs 

 *
 * 
 *
 * 
 * 
 * 
 * 
 *
 * 
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 *
 * 
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 *
 * 
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 *
 * 
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 *
 * 
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 *
 * 
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 *
 * 
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 *
 * 
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 *
 * 
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 *
 * 
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 *
 * 
 * 
 *
 * 
 *     
 *
 * 
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 *     
 *
 * 
 * 
 *
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 *
 * Construct Matches
 
Characters
x The character x
\\ The backslash character
\0n The character with octal value 0n
 *         (0 <= n <= 7)
\0nn The character with octal value 0nn
 *         (0 <= n <= 7)
\0mnn The character with octal value 0mnn
 *         (0 <= m <= 3,
 *         0 <= n <= 7)
\xhh The character with hexadecimal value 0xhh
\uhhhh The character with hexadecimal value 0xhhhh
\t The tab character ('\u0009')
\n The newline (line feed) character ('\u000A')
\r The carriage-return character ('\u000D')
\f The form-feed character ('\u000C')
\a The alert (bell) character ('\u0007')
\e The escape character ('\u001B')
\cx The control character corresponding to x
 
Character classes
[abc] a, b, or c (simple class)
[^abc] Any character except a, b, or c (negation)
[a-zA-Z] a through z
 *         or A through Z, inclusive (range)
[a-d[m-p]] a through d,
 *      or m through p: [a-dm-p] (union)
[a-z&&[def]] d, e, or f (intersection)
[a-z&&[^bc]] a through z,
 *         except for b and c: [ad-z] (subtraction)
[a-z&&[^m-p]] a through z,
 *          and not m through p: [a-lq-z](subtraction)
 
Predefined character classes
. Any character (may or may not match line terminators)
\d A digit: [0-9]
\D A non-digit: [^0-9]
\s A whitespace character: [ \t\n\x0B\f\r]
\S A non-whitespace character: [^\s]
\w A word character: [a-zA-Z_0-9]
\W A non-word character: [^\w]
 
POSIX character classes (US-ASCII only)
\p{Lower} A lower-case alphabetic character: [a-z]
\p{Upper} An upper-case alphabetic character:[A-Z]
\p{ASCII} All ASCII:[\x00-\x7F]
\p{Alpha} An alphabetic character:[\p{Lower}\p{Upper}]
\p{Digit} A decimal digit: [0-9]
\p{Alnum} An alphanumeric character:[\p{Alpha}\p{Digit}]
\p{Punct} Punctuation: One of !"#$%&'()*+,-./:;<=>?@[\]^_`{|}~
\p{Graph} A visible character: [\p{Alnum}\p{Punct}]
\p{Print} A printable character: [\p{Graph}\x20]
\p{Blank} A space or a tab: [ \t]
\p{Cntrl} A control character: [\x00-\x1F\x7F]
\p{XDigit} A hexadecimal digit: [0-9a-fA-F]
\p{Space} A whitespace character: [ \t\n\x0B\f\r]
 
java.lang.Character classes (simple java character type)
\p{javaLowerCase} Equivalent to java.lang.Character.isLowerCase()
\p{javaUpperCase} Equivalent to java.lang.Character.isUpperCase()
\p{javaWhitespace} Equivalent to java.lang.Character.isWhitespace()
\p{javaMirrored} Equivalent to java.lang.Character.isMirrored()
 
Classes for Unicode blocks and categories
\p{InGreek} A character in the Greek block (simple block)
\p{Lu} An uppercase letter (simple category)
\p{Sc} A currency symbol
\P{InGreek} Any character except one in the Greek block (negation)
[\p{L}&&[^\p{Lu}]]  Any letter except an uppercase letter (subtraction)
 
Boundary matchers
^ The beginning of a line
$ The end of a line
\b A word boundary
\B A non-word boundary
\A The beginning of the input
\G The end of the previous match
\Z The end of the input but for the final
 *         terminator, if any
\z The end of the input
 
Greedy quantifiers
X? X, once or not at all
X* X, zero or more times
X+ X, one or more times
X{n} X, exactly n times
X{n,} X, at least n times
X{n,m} X, at least n but not more than m times
 
Reluctant quantifiers
X?? X, once or not at all
X*? X, zero or more times
X+? X, one or more times
X{n}? X, exactly n times
X{n,}? X, at least n times
X{n,m}? X, at least n but not more than m times
 
Possessive quantifiers
X?+ X, once or not at all
X*+ X, zero or more times
X++ X, one or more times
X{n}+ X, exactly n times
X{n,}+ X, at least n times
X{n,m}+ X, at least n but not more than m times
 
Logical operators
XY X followed by Y
X|Y Either X or Y
(X) X, as a capturing group
 
Back references
\n Whatever the n^th
 *     capturing group matched
 
Quotation
\ Nothing, but quotes the following character
\Q Nothing, but quotes all characters until \E
\E Nothing, but ends quoting started by \Q
 
Special constructs (non-capturing)
(?:X) X, as a non-capturing group
(?idmsux-idmsux)  Nothing, but turns match flags on - off
(?idmsux-idmsux:X)   X, as a non-capturing group with the
 *         given flags on - off
(?=X) X, via zero-width positive lookahead
(?!X) X, via zero-width negative lookahead
(?<=X) X, via zero-width positive lookbehind
(?<!X) X, via zero-width negative lookbehind
(?>X) X, as an independent, non-capturing group
 *
 * 
 *
 *
 * 
 *  Backslashes, escapes, and quoting 
 *
 *  The backslash character ('\') serves to introduce escaped
 * constructs, as defined in the table above, as well as to quote characters
 * that otherwise would be interpreted as unescaped constructs.  Thus the
 * expression \\ matches a single backslash and \{ matches a
 * left brace.
 *
 * 
 It is an error to use a backslash prior to any alphabetic character that
 * does not denote an escaped construct; these are reserved for future
 * extensions to the regular-expression language.  A backslash may be used
 * prior to a non-alphabetic character regardless of whether that character is
 * part of an unescaped construct.
 *
 *  Backslashes within string literals in Java source code are interpreted
 * as required by the Java Language
 * Specification as either Unicode
 * escapes or other character
 * escapes.  It is therefore necessary to double backslashes in string
 * literals that represent regular expressions to protect them from
 * interpretation by the Java bytecode compiler.  The string literal
 * "\b", for example, matches a single backspace character when
 * interpreted as a regular expression, while "\\b" matches a
 * word boundary.  The string literal "\(hello\)" is illegal
 * and leads to a compile-time error; in order to match the string
 * (hello) the string literal "\\(hello\\)"
 * must be used.
 *
 * 
 * 
 Character Classes 

 *
 *     Character classes may appear within other character classes, and
 *    may be composed by the union operator (implicit) and the intersection
 *    operator (&&).
 *    The union operator denotes a class that contains every character that is
 *    in at least one of its operand classes.  The intersection operator
 *    denotes a class that contains every character that is in both of its
 *    operand classes.
 *
 *    
 The precedence of character-class operators is as follows, from
 *    highest to lowest:
 *
 *    

 *      
 *	  
 *	  
 *     
 *	  
 *	  
 *     
 *	  
 *	  
 *      
 *	  
 *	  
 *      
 *	  
 *	  
 *    1     Literal escape     \x
2     Grouping [...]
3     Range a-z
4     Union [a-e][i-u]
5     Intersection [a-z&&[aeiou]]
 *
 *     Note that a different set of metacharacters are in effect inside
 *    a character class than outside a character class. For instance, the
 *    regular expression . loses its special meaning inside a
 *    character class, while the expression - becomes a range
 *    forming metacharacter.
 *
 * 
 * 
 Line terminators 

 *
 *  A line terminator is a one- or two-character sequence that marks
 * the end of a line of the input character sequence.  The following are
 * recognized as line terminators:
 *
 * 

 *
 *    A newline (line feed) character ('\n'),
 *
 *   
 A carriage-return character followed immediately by a newline
 *   character ("\r\n"),
 *
 *   
 A standalone carriage-return character ('\r'),
 *
 *   
 A next-line character ('\u0085'),
 *
 *   
 A line-separator character ('\u2028'), or
 *
 *   
 A paragraph-separator character ('\u2029).
 *
 * 
 * If {@link #UNIX_LINES} mode is activated, then the only line terminators
 * recognized are newline characters.
 *
 * 
 The regular expression . matches any character except a line
 * terminator unless the {@link #DOTALL} flag is specified.
 *
 *  By default, the regular expressions ^ and $ ignore
 * line terminators and only match at the beginning and the end, respectively,
 * of the entire input sequence. If {@link #MULTILINE} mode is activated then
 * ^ matches at the beginning of input and after any line terminator
 * except at the end of input. When in {@link #MULTILINE} mode $
 * matches just before a line terminator or the end of the input sequence.
 *
 * 
 * 
 Groups and capturing 

 *
 *  Capturing groups are numbered by counting their opening parentheses from
 * left to right.  In the expression ((A)(B(C))), for example, there
 * are four such groups: 
 *
 * 
 * 
 *     
 * 
 *     
 * 
 *     
 * 
 *     
 * 1     ((A)(B(C)))
2     (A)
3     (B(C))
4     (C)
 *
 *  Group zero always stands for the entire expression.
 *
 * 
 Capturing groups are so named because, during a match, each subsequence
 * of the input sequence that matches such a group is saved.  The captured
 * subsequence may be used later in the expression, via a back reference, and
 * may also be retrieved from the matcher once the match operation is complete.
 *
 * 
 The captured input associated with a group is always the subsequence
 * that the group most recently matched.  If a group is evaluated a second time
 * because of quantification then its previously-captured value, if any, will
 * be retained if the second evaluation fails.  Matching the string
 * "aba" against the expression (a(b)?)+, for example, leaves
 * group two set to "b".  All captured input is discarded at the
 * beginning of each match.
 *
 * 
 Groups beginning with (? are pure, non-capturing groups
 * that do not capture text and do not count towards the group total.
 *
 *
 * 
 Unicode support 
 *
 *  This class is in conformance with Level 1 of Unicode Technical
 * Standard #18: Unicode Regular Expression Guidelines, plus RL2.1
 * Canonical Equivalents.
 *
 * 
 Unicode escape sequences such as \u2014 in Java source code
 * are processed as described in \u00A73.3
 * of the Java Language Specification.  Such escape sequences are also
 * implemented directly by the regular-expression parser so that Unicode
 * escapes can be used in expressions that are read from files or from the
 * keyboard.  Thus the strings "\u2014" and "\\u2014",
 * while not equal, compile into the same pattern, which matches the character
 * with hexadecimal value 0x2014.
 *
 *  
Unicode blocks and categories are written with the
 * \p and \P constructs as in
 * Perl. \p{prop} matches if the input has the
 * property prop, while \P{prop} does not match if
 * the input has that property.  Blocks are specified with the prefix
 * In, as in InMongolian.  Categories may be specified with
 * the optional prefix Is: Both \p{L} and \p{IsL}
 * denote the category of Unicode letters.  Blocks and categories can be used
 * both inside and outside of a character class.
 *
 * 
 The supported categories are those of 
 * 
 * The Unicode Standard in the version specified by the 
 * {@link java.lang.Character Character} class. The category names are those
 * defined in the Standard, both normative and informative.
 * The block names supported by Pattern are the valid block names
 * accepted and defined by 
 * {@link java.lang.Character.UnicodeBlock#forName(String) UnicodeBlock.forName}.
 *
 *  
Categories that behave like the java.lang.Character
 * boolean ismethodname methods (except for the deprecated ones) are
 * available through the same \p{prop} syntax where
 * the specified property has the name javamethodname.
 *
 * 
 Comparison to Perl 5 

 *
 * The Pattern engine performs traditional NFA-based matching
 * with ordered alternation as occurs in Perl 5.
 *
 * 
 Perl constructs not supported by this class: 
 *
 * 
 *
 *     The conditional constructs (?{X}) and
 *    (?(condition)X|Y),
 *    
 *
 *     The embedded code constructs (?{code})
 *    and (??{code}),
 *
 *     The embedded comment syntax (?#comment), and 
 *
 *     The preprocessing operations \l \u,
 *    \L, and \U.  
 *
 * 
 *
 *  Constructs supported by this class but not by Perl: 
 *
 * 
 *
 *     Possessive quantifiers, which greedily match as much as they can
 *    and do not back off, even when doing so would allow the overall match to
 *    succeed.  
 *
 *     Character-class union and intersection as described
 *    above.
 *
 * 
 *
 *  Notable differences from Perl: 
 *
 * 
 *
 *     In Perl, \1 through \9 are always interpreted
 *    as back references; a backslash-escaped number greater than 9 is
 *    treated as a back reference if at least that many subexpressions exist,
 *    otherwise it is interpreted, if possible, as an octal escape.  In this
 *    class octal escapes must always begin with a zero. In this class,
 *    \1 through \9 are always interpreted as back
 *    references, and a larger number is accepted as a back reference if at
 *    least that many subexpressions exist at that point in the regular
 *    expression, otherwise the parser will drop digits until the number is
 *    smaller or equal to the existing number of groups or it is one digit.
 *    
 *
 *     Perl uses the g flag to request a match that resumes
 *    where the last match left off.  This functionality is provided implicitly
 *    by the {@link Matcher} class: Repeated invocations of the {@link
 *    Matcher#find find} method will resume where the last match left off,
 *    unless the matcher is reset.  
 *
 *     In Perl, embedded flags at the top level of an expression affect
 *    the whole expression.  In this class, embedded flags always take effect
 *    at the point at which they appear, whether they are at the top level or
 *    within a group; in the latter case, flags are restored at the end of the
 *    group just as in Perl.  
 *
 *     Perl is forgiving about malformed matching constructs, as in the
 *    expression *a, as well as dangling brackets, as in the
 *    expression abc], and treats them as literals.  This
 *    class also accepts dangling brackets but is strict about dangling
 *    metacharacters like +, ? and *, and will throw a
 *    {@link PatternSyntaxException} if it encounters them. 
 *
 * 
 *
 *
 *  For a more precise description of the behavior of regular expression
 * constructs, please see 
 * Mastering Regular Expressions, 2nd Edition, Jeffrey E. F. Friedl,
 * O'Reilly and Associates, 2002.
 * 
 *
 * @see java.lang.String#split(String, int)
 * @see java.lang.String#split(String)
 *
 * @author      Mike McCloskey
 * @author      Mark Reinhold
 * @author	JSR-51 Expert Group
 * @version 	1.111, 05/01/04
 * @since       1.4
 * @spec	JSR-51
 */

public final class Pattern
    implements java.io.Serializable
{

    /**
     * Regular expression modifier values.  Instead of being passed as
     * arguments, they can also be passed as inline modifiers.
     * For example, the following statements have the same effect.
     *      * RegExp r1 = RegExp.compile("abc", Pattern.I|Pattern.M);
     * RegExp r2 = RegExp.compile("(?im)abc", 0);
     * 
     *
     * The flags are duplicated so that the familiar Perl match flag
     * names are available.
     */

    /**
     * Enables Unix lines mode.
     *
     *  In this mode, only the '\n' line terminator is recognized
     * in the behavior of ., ^, and $.
     *
     * 
 Unix lines mode can also be enabled via the embedded flag
     * expression (?d).
     */
    public static final int UNIX_LINES = 0x01;

    /**
     * Enables case-insensitive matching.
     *
     * 
 By default, case-insensitive matching assumes that only characters
     * in the US-ASCII charset are being matched.  Unicode-aware
     * case-insensitive matching can be enabled by specifying the {@link
     * #UNICODE_CASE} flag in conjunction with this flag.
     *
     * 
 Case-insensitive matching can also be enabled via the embedded flag
     * expression (?i).
     *
     * 
 Specifying this flag may impose a slight performance penalty.  
     */
    public static final int CASE_INSENSITIVE = 0x02;

    /**
     * Permits whitespace and comments in pattern.
     *
     *  In this mode, whitespace is ignored, and embedded comments starting
     * with # are ignored until the end of a line.
     *
     * 
 Comments mode can also be enabled via the embedded flag
     * expression (?x).
     */
    public static final int COMMENTS = 0x04;

    /**
     * Enables multiline mode.
     *
     * 
 In multiline mode the expressions ^ and $ match
     * just after or just before, respectively, a line terminator or the end of
     * the input sequence.  By default these expressions only match at the
     * beginning and the end of the entire input sequence.
     *
     * 
 Multiline mode can also be enabled via the embedded flag
     * expression (?m).  
     */
    public static final int MULTILINE = 0x08;

    /**
     * Enables literal parsing of the pattern.
     *
     *  When this flag is specified then the input string that specifies
     * the pattern is treated as a sequence of literal characters.
     * Metacharacters or escape sequences in the input sequence will be
     * given no special meaning.
     *
     * 
The flags CASE_INSENSITIVE and UNICODE_CASE retain their impact on
     * matching when used in conjunction with this flag. The other flags
     * become superfluous.
     *
     * 
 There is no embedded flag character for enabling literal parsing.
     */
    public static final int LITERAL = 0x10;

    /**
     * Enables dotall mode.
     *
     * 
 In dotall mode, the expression . matches any character,
     * including a line terminator.  By default this expression does not match
     * line terminators.
     *
     * 
 Dotall mode can also be enabled via the embedded flag
     * expression (?s).  (The s is a mnemonic for
     * "single-line" mode, which is what this is called in Perl.)  
     */
    public static final int DOTALL = 0x20;

    /**
     * Enables Unicode-aware case folding.
     *
     *  When this flag is specified then case-insensitive matching, when
     * enabled by the {@link #CASE_INSENSITIVE} flag, is done in a manner
     * consistent with the Unicode Standard.  By default, case-insensitive
     * matching assumes that only characters in the US-ASCII charset are being
     * matched.
     *
     * 
 Unicode-aware case folding can also be enabled via the embedded flag
     * expression (?u).
     *
     * 
 Specifying this flag may impose a performance penalty.  
     */
    public static final int UNICODE_CASE = 0x40;

    /**
     * Enables canonical equivalence.
     *
     *  When this flag is specified then two characters will be considered
     * to match if, and only if, their full canonical decompositions match.
     * The expression "a\u030A", for example, will match the
     * string "\u00E5" when this flag is specified.  By default,
     * matching does not take canonical equivalence into account.
     *
     * 
 There is no embedded flag character for enabling canonical
     * equivalence.
     *
     * 
 Specifying this flag may impose a performance penalty.  
     */
    public static final int CANON_EQ = 0x80;

    /* Pattern has only two serialized components: The pattern string
     * and the flags, which are all that is needed to recompile the pattern
     * when it is deserialized.
     */

    /** use serialVersionUID from Merlin b59 for interoperability */
    private static final long serialVersionUID = 5073258162644648461L;

    /**
     * The original regular-expression pattern string.
     *
     * @serial
     */
    private String pattern;

    /**
     * The original pattern flags.
     *
     * @serial
     */
    private int flags;

    /**
     * Boolean indicating this Pattern is compiled; this is necessary in order
     * to lazily compile deserialized Patterns.
     */
    private transient volatile boolean compiled = false;

    /**
     * The normalized pattern string.
     */
    private transient String normalizedPattern;

    /**
     * The starting point of state machine for the find operation.  This allows
     * a match to start anywhere in the input.
     */
    transient Node root;

    /**
     * The root of object tree for a match operation.  The pattern is matched
     * at the beginning.  This may include a find that uses BnM or a First
     * node.
     */
    transient Node matchRoot;

    /**
     * Temporary storage used by parsing pattern slice.
     */
    transient int[] buffer;

    /**
     * Temporary storage used while parsing group references.
     */
    transient GroupHead[] groupNodes;

    /**
     * Temporary null terminating char array used by pattern compiling.
     */
    private transient int[] temp;

    /**
     * The number of capturing groups in this Pattern. Used by matchers to
     * allocate storage needed to perform a match.
     */
    transient int capturingGroupCount;

    /**
     * The local variable count used by parsing tree. Used by matchers to
     * allocate storage needed to perform a match.
     */
    transient int localCount;

    /**
     * Index into the pattern string that keeps track of how much has been
     * parsed.
     */
    private transient int cursor;

    /**
     * Holds the length of the pattern string.
     */
    private transient int patternLength;

    /**
     * Compiles the given regular expression into a pattern.  
     *
     * @param  regex
     *         The expression to be compiled
     *
     * @throws  PatternSyntaxException
     *          If the expression's syntax is invalid
     */
    public static Pattern compile(String regex) {
        return new Pattern(regex, 0);
    }

    /**
     * Compiles the given regular expression into a pattern with the given
     * flags.  
     *
     * @param  regex
     *         The expression to be compiled
     *
     * @param  flags
     *         Match flags, a bit mask that may include
     *         {@link #CASE_INSENSITIVE}, {@link #MULTILINE}, {@link #DOTALL},
     *         {@link #UNICODE_CASE}, and {@link #CANON_EQ}
     *
     * @throws  IllegalArgumentException
     *          If bit values other than those corresponding to the defined
     *          match flags are set in flags
     *
     * @throws  PatternSyntaxException
     *          If the expression's syntax is invalid
     */
    public static Pattern compile(String regex, int flags) {
        return new Pattern(regex, flags);
    }

    /**
     * Returns the regular expression from which this pattern was compiled.
     * 
     *
     * @return  The source of this pattern
     */
    public String pattern() {
        return pattern;
    }

    /**
     * Returns the string representation of this pattern. This
     * is the regular expression from which this pattern was
     * compiled.
     *
     * @return  The string representation of this pattern
     * @since 1.5
     */
    public String toString() {
        return pattern;
    }

    /**
     * Creates a matcher that will match the given input against this pattern.
     * 
     *
     * @param  input
     *         The character sequence to be matched
     *
     * @return  A new matcher for this pattern
     */
    public Matcher matcher(CharSequence input) {
        synchronized(this) {
            if (!compiled)
                compile();
        }
        Matcher m = new Matcher(this, input);
        return m;
    }

    /**
     * Returns this pattern's match flags.  
     *
     * @return  The match flags specified when this pattern was compiled
     */
    public int flags() {
        return flags;
    }

    /**
     * Compiles the given regular expression and attempts to match the given
     * input against it.
     *
     *  An invocation of this convenience method of the form
     *
     * 
     * Pattern.matches(regex, input);
     *
     * behaves in exactly the same way as the expression
     *
     *      * Pattern.compile(regex).matcher(input).matches()
     *
     *  If a pattern is to be used multiple times, compiling it once and reusing
     * it will be more efficient than invoking this method each time.  
     *
     * @param  regex
     *         The expression to be compiled
     *
     * @param  input
     *         The character sequence to be matched
     *
     * @throws  PatternSyntaxException
     *          If the expression's syntax is invalid
     */
    public static boolean matches(String regex, CharSequence input) {
        Pattern p = Pattern.compile(regex);
        Matcher m = p.matcher(input);
        return m.matches();
    }

    /**
     * Splits the given input sequence around matches of this pattern.
     *
     *  The array returned by this method contains each substring of the
     * input sequence that is terminated by another subsequence that matches
     * this pattern or is terminated by the end of the input sequence.  The
     * substrings in the array are in the order in which they occur in the
     * input.  If this pattern does not match any subsequence of the input then
     * the resulting array has just one element, namely the input sequence in
     * string form.
     *
     * 
 The limit parameter controls the number of times the
     * pattern is applied and therefore affects the length of the resulting
     * array.  If the limit n is greater than zero then the pattern
     * will be applied at most n - 1 times, the array's
     * length will be no greater than n, and the array's last entry
     * will contain all input beyond the last matched delimiter.  If n
     * is non-positive then the pattern will be applied as many times as
     * possible and the array can have any length.  If n is zero then
     * the pattern will be applied as many times as possible, the array can
     * have any length, and trailing empty strings will be discarded.
     *
     * 
 The input "boo:and:foo", for example, yields the following
     * results with these parameters:
     *
     * 

     * 
     *     
     *     
     * 
     *     
     *     
     * 
     *     
     *     
     * 
     *     
     *     
     * 
     *     
     *     
     * 
     *     
     *     
     * 
     *     
     *     
     * Regex     Limit     Result    
: 2 { "boo", "and:foo" }
: 5 { "boo", "and", "foo" }
: -2 { "boo", "and", "foo" }
o 5 { "b", "", ":and:f", "", "" }
o -2 { "b", "", ":and:f", "", "" }
o 0 { "b", "", ":and:f" }
     *
     *
     * @param  input
     *         The character sequence to be split
     *
     * @param  limit
     *         The result threshold, as described above
     *
     * @return  The array of strings computed by splitting the input
     *          around matches of this pattern
     */
    public String[] split(CharSequence input, int limit) {
        int index = 0;
        boolean matchLimited = limit > 0;
        ArrayList matchList = new ArrayList();
        Matcher m = matcher(input);

        // Add segments before each match found
        while(m.find()) {
            if (!matchLimited || matchList.size() < limit - 1) {
                String match = input.subSequence(index, m.start()).toString();
                matchList.add(match);
                index = m.end();
            } else if (matchList.size() == limit - 1) { // last one
                String match = input.subSequence(index,
                                                 input.length()).toString();
                matchList.add(match);
                index = m.end();
            }
        }

        // If no match was found, return this
        if (index == 0)
            return new String[] {input.toString()};

        // Add remaining segment
        if (!matchLimited || matchList.size() < limit)
            matchList.add(input.subSequence(index, input.length()).toString());

        // Construct result
        int resultSize = matchList.size();
        if (limit == 0)
            while (resultSize > 0 && matchList.get(resultSize-1).equals(""))
                resultSize--;
        String[] result = new String[resultSize];
        return (String[])matchList.subList(0, resultSize).toArray(result);
    }

    /**
     * Splits the given input sequence around matches of this pattern.
     *
     *  This method works as if by invoking the two-argument {@link
     * #split(java.lang.CharSequence, int) split} method with the given input
     * sequence and a limit argument of zero.  Trailing empty strings are
     * therefore not included in the resulting array. 
     *
     *  The input "boo:and:foo", for example, yields the following
     * results with these expressions:
     *
     * 

     * 
     *     
     * 
     *     
     * 
     *     
     * Regex     Result
: { "boo", "and", "foo" }
o { "b", "", ":and:f" }
     *
     *
     * @param  input
     *         The character sequence to be split
     *
     * @return  The array of strings computed by splitting the input
     *          around matches of this pattern
     */
    public String[] split(CharSequence input) {
        return split(input, 0);
    }

    /**
     * Returns a literal pattern String for the specified
     * String.
     *
     * This method produces a String that can be used to
     * create a Pattern that would match the string
     * s as if it were a literal pattern. Metacharacters
     * or escape sequences in the input sequence will be given no special
     * meaning.
     *
     * @param  s The string to be literalized
     * @return  A literal string replacement
     * @since 1.5
     */
    public static String quote(String s) {
        int slashEIndex = s.indexOf("\\E");
        if (slashEIndex == -1)
            return "\\Q" + s + "\\E";

        StringBuilder sb = new StringBuilder(s.length() * 2);
        sb.append("\\Q");
        slashEIndex = 0;
        int current = 0;
        while ((slashEIndex = s.indexOf("\\E", current)) != -1) {
            sb.append(s.substring(current, slashEIndex));
            current = slashEIndex + 2;
            sb.append("\\E\\\\E\\Q");
        }
        sb.append(s.substring(current, s.length()));
        sb.append("\\E");
        return sb.toString();
    }

    /**
     * Recompile the Pattern instance from a stream.  The original pattern
     * string is read in and the object tree is recompiled from it.
     */
    private void readObject(java.io.ObjectInputStream s)
        throws java.io.IOException, ClassNotFoundException {

        // Read in all fields
	s.defaultReadObject();

        // Initialize counts
        capturingGroupCount = 1;
        localCount = 0;

        // if length > 0, the Pattern is lazily compiled
        compiled = false;
        if (pattern.length() == 0) {
            root = new Start(lastAccept);
            matchRoot = lastAccept;
            compiled = true;
        }
    }

    /**
     * This private constructor is used to create all Patterns. The pattern
     * string and match flags are all that is needed to completely describe
     * a Pattern. An empty pattern string results in an object tree with
     * only a Start node and a LastNode node.
     */
    private Pattern(String p, int f) {
        pattern = p;
        flags = f;

        // Reset group index count
        capturingGroupCount = 1;
        localCount = 0;

        if (pattern.length() > 0) {
            compile();
        } else {
            root = new Start(lastAccept);
            matchRoot = lastAccept;
        }
    }

    /**
     * The pattern is converted to normalizedD form and then a pure group
     * is constructed to match canonical equivalences of the characters.
     */
    private void normalize() {
        boolean inCharClass = false;
        int lastCodePoint = -1;

        // Convert pattern into normalizedD form
        normalizedPattern = Normalizer.decompose(pattern, false, 0);
        patternLength = normalizedPattern.length();

        // Modify pattern to match canonical equivalences
        StringBuilder newPattern = new StringBuilder(patternLength);
        for(int i=0; i= patternLength)
                        break;
                    c = normalizedPattern.codePointAt(i);
                    sequenceBuffer.appendCodePoint(c);
                }
                String ea = produceEquivalentAlternation(
                                               sequenceBuffer.toString());
                newPattern.setLength(newPattern.length()-Character.charCount(lastCodePoint));
                newPattern.append("(?:").append(ea).append(")");
            } else if (c == '[' && lastCodePoint != '\\') {
                i = normalizeCharClass(newPattern, i);
            } else {
                newPattern.appendCodePoint(c);
            }
            lastCodePoint = c;
	    i += Character.charCount(c);
        }
        normalizedPattern = newPattern.toString();
    }

    /**
     * Complete the character class being parsed and add a set
     * of alternations to it that will match the canonical equivalences
     * of the characters within the class.
     */
    private int normalizeCharClass(StringBuilder newPattern, int i) {
        StringBuilder charClass = new StringBuilder();
        StringBuilder eq = null;
        int lastCodePoint = -1;
        String result;

        i++;
        charClass.append("[");
        while(true) {
            int c = normalizedPattern.codePointAt(i);
            StringBuilder sequenceBuffer;

            if (c == ']' && lastCodePoint != '\\') {
                charClass.append((char)c);
                break;
            } else if (Character.getType(c) == Character.NON_SPACING_MARK) {
                sequenceBuffer = new StringBuilder();
                sequenceBuffer.appendCodePoint(lastCodePoint);
                while(Character.getType(c) == Character.NON_SPACING_MARK) {
                    sequenceBuffer.appendCodePoint(c);
                    i += Character.charCount(c);
                    if (i >= normalizedPattern.length())
                        break;
                    c = normalizedPattern.codePointAt(i);
                }
                String ea = produceEquivalentAlternation(
                                                  sequenceBuffer.toString());

                charClass.setLength(charClass.length()-Character.charCount(lastCodePoint));
                if (eq == null)
                    eq = new StringBuilder();
                eq.append('|');
                eq.append(ea);
            } else {
                charClass.appendCodePoint(c);
                i++;
            }
            if (i == normalizedPattern.length())
                error("Unclosed character class");
            lastCodePoint = c;
        }

        if (eq != null) {
            result = new String("(?:"+charClass.toString()+
                                eq.toString()+")");
        } else {
            result = charClass.toString();
        }

        newPattern.append(result);
        return i;
    }

    /**
     * Given a specific sequence composed of a regular character and
     * combining marks that follow it, produce the alternation that will
     * match all canonical equivalences of that sequence.
     */
    private String produceEquivalentAlternation(String source) {
	int len = countChars(source, 0, 1);
        if (source.length() == len)
	    // source has one character.
            return new String(source);

        String base = source.substring(0,len);
        String combiningMarks = source.substring(len);

        String[] perms = producePermutations(combiningMarks);
        StringBuilder result = new StringBuilder(source);

        // Add combined permutations
        for(int x=0; x0)
                result.append("|"+next);
            next = composeOneStep(next);
            if (next != null)
                result.append("|"+produceEquivalentAlternation(next));
        }
        return result.toString();
    }

    /**
     * Returns an array of strings that have all the possible
     * permutations of the characters in the input string.
     * This is used to get a list of all possible orderings
     * of a set of combining marks. Note that some of the permutations
     * are invalid because of combining class collisions, and these
     * possibilities must be removed because they are not canonically
     * equivalent.
     */
    private String[] producePermutations(String input) {
        if (input.length() == countChars(input, 0, 1))
            return new String[] {input};

        if (input.length() == countChars(input, 0, 2)) {
	    int c0 = Character.codePointAt(input, 0);
	    int c1 = Character.codePointAt(input, Character.charCount(c0));
            if (getClass(c1) == getClass(c0)) {
                return new String[] {input};
            }
            String[] result = new String[2];
            result[0] = input;
            StringBuilder sb = new StringBuilder(2);
	    sb.appendCodePoint(c1);
	    sb.appendCodePoint(c0);
            result[1] = sb.toString();
            return result;
        }

        int length = 1;
	int nCodePoints = countCodePoints(input);
        for(int x=1; x=0; y--) {
                if (combClass[y] == combClass[x]) {
                    continue loop;
                }
            }
            StringBuilder sb = new StringBuilder(input);
            String otherChars = sb.delete(offset, offset+len).toString();
            String[] subResult = producePermutations(otherChars);

            String prefix = input.substring(offset, offset+len);
            for(int y=0; ynext:");
            if (node == Pattern.accept) {
                System.out.println("Accept Node");
                node = null;
            }
       }
    }

    /**
     * Used to accumulate information about a subtree of the object graph
     * so that optimizations can be applied to the subtree.
     */
    static final class TreeInfo {
        int minLength;
        int maxLength;
        boolean maxValid;
        boolean deterministic;

        TreeInfo() {
            reset();
        }
        void reset() {
            minLength = 0;
            maxLength = 0;
            maxValid = true;
            deterministic = true;
        }
    }

    /**
     * The following private methods are mainly used to improve the
     * readability of the code. In order to let the Java compiler easily
     * inline them, we should not put many assertions or error checks in them.
     */

    /**
     * Indicates whether a particular flag is set or not.
     */
    private boolean has(int f) {
        return (flags & f) > 0;
    }

    /**
     * Match next character, signal error if failed.
     */
    private void accept(int ch, String s) {
        int testChar = temp[cursor++];
        if (has(COMMENTS))
            testChar = parsePastWhitespace(testChar);
        if (ch != testChar) {
           error(s);
        }
    }

    /**
     * Mark the end of pattern with a specific character.
     */
    private void mark(int c) {
        temp[patternLength] = c;
    }

    /**
     * Peek the next character, and do not advance the cursor.
     */
    private int peek() {
        int ch = temp[cursor];
        if (has(COMMENTS))
            ch = peekPastWhitespace(ch);
        return ch;
    }

    /**
     * Read the next character, and advance the cursor by one.
     */
    private int read() {
        int ch = temp[cursor++];
        if (has(COMMENTS))
            ch = parsePastWhitespace(ch);
        return ch;
    }

    /**
     * Read the next character, and advance the cursor by one,
     * ignoring the COMMENTS setting
     */
    private int readEscaped() {
        int ch = temp[cursor++];
        return ch;
    }

    /**
     * Advance the cursor by one, and peek the next character.
     */
    private int next() {
        int ch = temp[++cursor];
        if (has(COMMENTS))
            ch = peekPastWhitespace(ch);
        return ch;
    }

    /**
     * Advance the cursor by one, and peek the next character,
     * ignoring the COMMENTS setting
     */
    private int nextEscaped() {
        int ch = temp[++cursor];
        return ch;
    }

    /**
     * If in xmode peek past whitespace and comments.
     */
    private int peekPastWhitespace(int ch) {
        while (ASCII.isSpace(ch) || ch == '#') {
            while (ASCII.isSpace(ch))
                ch = temp[++cursor];
            if (ch == '#') {
                ch = peekPastLine();
            }
        }
        return ch;
    }

    /**
     * If in xmode parse past whitespace and comments.
     */
    private int parsePastWhitespace(int ch) {
        while (ASCII.isSpace(ch) || ch == '#') {
            while (ASCII.isSpace(ch))
                ch = temp[cursor++];
            if (ch == '#')
                ch = parsePastLine();
        }
        return ch;
    }

    /**
     * xmode parse past comment to end of line.
     */
    private int parsePastLine() {
        int ch = temp[cursor++];
        while (ch != 0 && !isLineSeparator(ch))
            ch = temp[cursor++];
        return ch;
    }

    /**
     * xmode peek past comment to end of line.
     */
    private int peekPastLine() {
        int ch = temp[++cursor];
        while (ch != 0 && !isLineSeparator(ch))
            ch = temp[++cursor];
        return ch;
    }

    /**
     * Determines if character is a line separator in the current mode
     */
    private boolean isLineSeparator(int ch) {
        if (has(UNIX_LINES)) {
            return ch == '\n';
        } else {
            return (ch == '\n' ||
                    ch == '\r' ||
                    (ch|1) == '\u2029' ||
                    ch == '\u0085');
        }
    }

    /**
     * Read the character after the next one, and advance the cursor by two.
     */
    private int skip() {
        int i = cursor;
        int ch = temp[i+1];
        cursor = i + 2;
        return ch;
    }

    /**
     * Unread one next character, and retreat cursor by one.
     */
    private void unread() {
        cursor--;
    }

    /**
     * Internal method used for handling all syntax errors. The pattern is
     * displayed with a pointer to aid in locating the syntax error.
     */
    private Node error(String s) {
	throw new PatternSyntaxException(s, normalizedPattern,
					 cursor - 1);
    }

    /**
     * Determines if there is any supplementary character or unpaired
     * surrogate in the specified range.
     */
    private boolean findSupplementary(int start, int end) {
	for (int i = start; i < end; i++) {
	    if (isSupplementary(temp[i]))
		return true;
	}
	return false;
    }

    /**
     * Determines if the specified code point is a supplementary
     * character or unpaired surrogate.
     */
    private static final boolean isSupplementary(int ch) {
	return ch >= Character.MIN_SUPPLEMENTARY_CODE_POINT || isSurrogate(ch);
    }

    /**
     *  The following methods handle the main parsing. They are sorted
     *  according to their precedence order, the lowest one first.
     */

    /**
     * The expression is parsed with branch nodes added for alternations.
     * This may be called recursively to parse sub expressions that may
     * contain alternations.
     */
    private Node expr(Node end) {
        Node prev = null;
        for (;;) {
            Node node = sequence(end);
            if (prev == null) {
                prev = node;
            } else {
                prev = new Branch(prev, node);
            }
            if (peek() != '|') {
                return prev;
            }
            next();
        }
    }

    /**
     * Parsing of sequences between alternations.
     */
    private Node sequence(Node end) {
        Node head = null;
        Node tail = null;
        Node node = null;
        int i, j, ch;
    LOOP:
        for (;;) {
            ch = peek();
            switch (ch) {
            case '(':
                // Because group handles its own closure,
                // we need to treat it differently
                node = group0();
                // Check for comment or flag group
                if (node == null)
                    continue;
                if (head == null)
                    head = node;
                else
                    tail.next = node;
                // Double return: Tail was returned in root
                tail = root;
                continue;
            case '[':
                node = clazz(true);
                break;
            case '\\':
                ch = nextEscaped();
                if (ch == 'p' || ch == 'P') {
                    boolean comp = (ch == 'P');
                    boolean oneLetter = true;
                    ch = next(); // Consume { if present
                    if (ch != '{') {
                        unread();
                    } else {
                        oneLetter = false;
                    }
                    node = family(comp, oneLetter);
                } else {
                    unread();
                    node = atom();
                }
                break;
            case '^':
                next();
                if (has(MULTILINE)) {
                    if (has(UNIX_LINES))
                        node = new UnixCaret();
                    else
                        node = new Caret();
                } else {
                    node = new Begin();
                }
                break;
            case '$':
                next();
                if (has(UNIX_LINES))
                    node = new UnixDollar(has(MULTILINE));
                else
                    node = new Dollar(has(MULTILINE));
                break;
            case '.':
                next();
                if (has(DOTALL)) {
                    node = new All();
                } else {
                    if (has(UNIX_LINES))
                        node = new UnixDot();
                    else {
                        node = new Dot();
                    }
                }
                break;
            case '|':
            case ')':
                break LOOP;
            case ']': // Now interpreting dangling ] and } as literals
            case '}':
                node = atom();
                break;
            case '?':
            case '*':
            case '+':
                next();
                return error("Dangling meta character '" + ((char)ch) + "'");
            case 0:
                if (cursor >= patternLength) {
                    break LOOP;
                }
                // Fall through
            default:
                node = atom();
                break;
            }

            node = closure(node);

            if (head == null) {
                head = tail = node;
            } else {
                tail.next = node;
                tail = node;
            }
        }
        if (head == null) {
            return end;
        }
        tail.next = end;
        return head;
    }

    /**
     * Parse and add a new Single or Slice.
     */
    private Node atom() {
        int first = 0;
        int prev = -1;
	boolean hasSupplementary = false;
        int ch = peek();
        for (;;) {
            switch (ch) {
            case '*':
            case '+':
            case '?':
            case '{':
                if (first > 1) {
                    cursor = prev;    // Unwind one character
                    first--;
                }
                break;
            case '$':
            case '.':
            case '^':
            case '(':
            case '[':
            case '|':
            case ')':
                break;
            case '\\':
                ch = nextEscaped();
                if (ch == 'p' || ch == 'P') { // Property
                    if (first > 0) { // Slice is waiting; handle it first
                        unread();
                        break;
                    } else { // No slice; just return the family node
                        if (ch == 'p' || ch == 'P') {
                            boolean comp = (ch == 'P');
                            boolean oneLetter = true;
                            ch = next(); // Consume { if present
                            if (ch != '{')
                                unread();
                            else
                                oneLetter = false;
                            return family(comp, oneLetter);
                        }
                    }
                    break;
                }
                unread();
                prev = cursor;
                ch = escape(false, first == 0);
                if (ch >= 0) {
                    append(ch, first);
                    first++;
		    if (isSupplementary(ch)) {
			hasSupplementary = true;
		    }
                    ch = peek();
                    continue;
                } else if (first == 0) {
                    return root;
                }
                // Unwind meta escape sequence
                cursor = prev;
                break;
            case 0:
                if (cursor >= patternLength) {
                    break;
                }
                // Fall through
            default:
                prev = cursor;
                append(ch, first);
                first++;
		if (isSupplementary(ch)) {
		    hasSupplementary = true;
		}
                ch = next();
                continue;
            }
            break;
        }
        if (first == 1) {
            return newSingle(buffer[0]);
        } else {
            return newSlice(buffer, first, hasSupplementary);
        }
    }

    private void append(int ch, int len) {
        if (len >= buffer.length) {
            int[] tmp = new int[len+len];
            System.arraycopy(buffer, 0, tmp, 0, len);
            buffer = tmp;
        }
        buffer[len] = ch;
    }

    /**
     * Parses a backref greedily, taking as many numbers as it
     * can. The first digit is always treated as a backref, but
     * multi digit numbers are only treated as a backref if at
     * least that many backrefs exist at this point in the regex.
     */
    private Node ref(int refNum) {
        boolean done = false;
        while(!done) {
            int ch = peek();
            switch(ch) {
                case '0':
                case '1':
                case '2':
                case '3':
                case '4':
                case '5':
                case '6':
                case '7':
                case '8':
                case '9':
                    int newRefNum = (refNum * 10) + (ch - '0');
                    // Add another number if it doesn't make a group
                    // that doesn't exist
                    if (capturingGroupCount - 1 < newRefNum) {
                        done = true;
                        break;
                    }
                    refNum = newRefNum;
                    read();
                    break;
                default:
                    done = true;
                    break;
            }
        }
        if (has(CASE_INSENSITIVE) || has(UNICODE_CASE))
            return new CIBackRef(refNum);
        else
            return new BackRef(refNum);
    }

    /**
     * Parses an escape sequence to determine the actual value that needs
     * to be matched.
     * If -1 is returned and create was true a new object was added to the tree
     * to handle the escape sequence.
     * If the returned value is greater than zero, it is the value that
     * matches the escape sequence.
     */
    private int escape(boolean inclass, boolean create) {
        int ch = skip();
        switch (ch) {
            case '0':
                return o();
            case '1':
            case '2':
            case '3':
            case '4':
            case '5':
            case '6':
            case '7':
            case '8':
            case '9':
                if (inclass) break;
                if (capturingGroupCount < (ch - '0'))
                    error("No such group yet exists at this point in the pattern");
                if (create) {
                    root = ref((ch - '0'));
                }
                return -1;
            case 'A':
                if (inclass) break;
                if (create) root = new Begin();
                return -1;
            case 'B':
                if (inclass) break;
                if (create) root = new Bound(Bound.NONE);
                return -1;
            case 'C':
                break;
            case 'D':
                if (create) root = new NotCtype(ASCII.DIGIT);
                return -1;
            case 'E':
            case 'F':
                break;
            case 'G':
                if (inclass) break;
                if (create) root = new LastMatch();
                return -1;
            case 'H':
            case 'I':
            case 'J':
            case 'K':
            case 'L':
            case 'M':
            case 'N':
            case 'O':
            case 'P':
                break;
            case 'Q':
                if (create) {
                    // Disable metacharacters. We will return a slice
                    // up to the next \E
                    int i = cursor;
                    int c;
                    while ((c = readEscaped()) != 0) {
                        if (c == '\\') {
                            c = readEscaped();
                            if (c == 'E' || c == 0)
                                break;
                            else
                               unread();
                        }
                    }
                    int j = cursor-1;
                    if (c == 'E')
                        j--;
                    else
                        unread();
		    boolean hasSupplementary = false;
                    for (int x = i; x= patternLength)
                        return error("Unclosed character class");
                    break;
                case ']':
                    firstInClass = false;
                    if (prev != null) {
                        if (consume)
                            next();
                        return prev;
                    }
                    break;
                default:
                    firstInClass = false;
                    break;
            }
            node = range(bits);
            if (include) {
                if (prev == null) {
                    prev = node;
                } else {
                    if (prev != node)
                        prev = new Add(prev, node);
                }
            } else {
                if (prev == null) {
                    prev = node.dup(true);  // Complement
                } else {
                    if (prev != node)
                        prev = new Sub(prev, node);
                }
            }
            ch = peek();
        }
    }

    /**
     * Parse a single character or a character range in a character class
     * and return its representative node.
     */
    private Node range(BitClass bits) {
        int ch = peek();
        if (ch == '\\') {
            ch = nextEscaped();
            if (ch == 'p' || ch == 'P') { // A property
                boolean comp = (ch == 'P');
                boolean oneLetter = true;
                // Consume { if present
                ch = next();
                if (ch != '{')
                    unread();
                else
                    oneLetter = false;
                return family(comp, oneLetter);
            } else { // ordinary escape
                unread();
                ch = escape(true, true);
                if (ch == -1)
                    return root;
            }
        } else {
            ch = single();
        }
        if (ch >= 0) {
            if (peek() == '-') {
                int endRange = temp[cursor+1];
                if (endRange == '[') {
                    if (ch < 256)
                        return bits.add(ch, flags());
                    return newSingle(ch);
                }
                if (endRange != ']') {
                    next();
                    int m = single();
                    if (m < ch)
                        return error("Illegal character range");
                    if (has(CASE_INSENSITIVE) || has(UNICODE_CASE))
                        return new CIRange(ch, m);
                    else
                        return new Range(ch, m);
                }
            }
            if (ch < 256)
                return bits.add(ch, flags());
            return newSingle(ch);
        }
        return error("Unexpected character '"+((char)ch)+"'");
    }

    private int single() {
        int ch = peek();
        switch (ch) {
        case '\\':
            return escape(true, false);
        default:
            next();
            return ch;
        }
    }

    /**
     * Parses a Unicode character family and returns its representative node.
     * Reference to an unknown character family results in a list of supported
     * families in the error.
     */
    private Node family(boolean not, boolean singleLetter) {
        next();
        String name;

        if (singleLetter) {
	    int c = temp[cursor];
	    if (!Character.isSupplementaryCodePoint(c)) {
		name = String.valueOf((char)c);
	    } else {
		name = new String(temp, cursor, 1);
	    }
	    name = name.intern();
            read();
        } else {
            int i = cursor;
            mark('}');
            while(read() != '}') {
            }
            mark('\000');
            int j = cursor;
            if (j > patternLength)
                return error("Unclosed character family");
            if (i + 1 >= j)
                return error("Empty character family");
            name = new String(temp, i, j-i-1).intern();
        }

        if (name.startsWith("In")) {
            name = name.substring(2, name.length()).intern();
            return retrieveFamilyNode(name, not);
        }
        if (name.startsWith("Is"))
            name = name.substring(2, name.length()).intern();
        return retrieveCategoryNode(name).dup(not);
    }

    private Node retrieveFamilyNode(String name, boolean not) {
        if (name == null) {
            return familyError("", "Null character family.");
        }
        Node n = null;
        try {
            Character.UnicodeBlock block = Character.UnicodeBlock.forName(name);
            n = new UBlock(block, not);
        } catch (IllegalArgumentException iae) {
            return familyError(name, "Unknown character family {");
        }
        return n;
    }

    private Node retrieveCategoryNode(String name) {
        Node n = (Node)categoryNames.cMap.get(name);
        if (n != null)
            return n;

        return familyError(name, "Unknown character category {");
    }

    private Node familyError(String name, String type) {
        StringBuilder sb = new StringBuilder();
        sb.append(type);
        sb.append(name);
        sb.append("}");
        name = sb.toString();
        return error(name);
    }

    /**
     * Parses a group and returns the head node of a set of nodes that process
     * the group. Sometimes a double return system is used where the tail is
     * returned in root.
     */
    private Node group0() {
        boolean capturingGroup = false;
        Node head = null;
        Node tail = null;
        int save = flags;
        root = null;
        int ch = next();
        if (ch == '?') {
            ch = skip();
            switch (ch) {
            case ':':   //  (?:xxx) pure group
                head = createGroup(true);
                tail = root;
                head.next = expr(tail);
                break;
            case '=':   // (?=xxx) and (?!xxx) lookahead
            case '!':
                head = createGroup(true);
                tail = root;
                head.next = expr(tail);
                if (ch == '=') {
                    head = tail = new Pos(head);
                } else {
                    head = tail = new Neg(head);
                }
                break;
            case '>':   // (?>xxx)  independent group
                head = createGroup(true);
                tail = root;
                head.next = expr(tail);
                head = tail = new Ques(head, INDEPENDENT);
                break;
            case '<':   // (?= 0) {
            int m = read();
            if (((m-'0')|('7'-m)) >= 0) {
                int o = read();
                if ((((o-'0')|('7'-o)) >= 0) && (((n-'0')|('3'-n)) >= 0)) {
                    return (n - '0') * 64 + (m - '0') * 8 + (o - '0');
                }
                unread();
                return (n - '0') * 8 + (m - '0');
            }
            unread();
            return (n - '0');
        }
        error("Illegal octal escape sequence");
        return -1;
    }

    /**
     *  Utility method for parsing hexadecimal escape sequences.
     */
    private int x() {
        int n = read();
        if (ASCII.isHexDigit(n)) {
            int m = read();
            if (ASCII.isHexDigit(m)) {
                return ASCII.toDigit(n) * 16 + ASCII.toDigit(m);
            }
        }
        error("Illegal hexadecimal escape sequence");
        return -1;
    }

    /**
     *  Utility method for parsing unicode escape sequences.
     */
    private int u() {
        int n = 0;
        for (int i = 0; i < 4; i++) {
            int ch = read();
            if (!ASCII.isHexDigit(ch)) {
                error("Illegal Unicode escape sequence");
            }
            n = n * 16 + ASCII.toDigit(ch);
        }
        return n;
    }

    //
    // Utility methods for code point support
    //

    /**
     * Tests a surrogate value.
     */
    private static final boolean isSurrogate(int c) {
	return c >= Character.MIN_HIGH_SURROGATE && c <= Character.MAX_LOW_SURROGATE;
    }

    private static final int countChars(CharSequence seq, int index,
					int lengthInCodePoints) {
	// optimization
	if (lengthInCodePoints == 1 && !Character.isHighSurrogate(seq.charAt(index))) {
	    assert (index >= 0 && index < seq.length());
	    return 1;
	}
	int length = seq.length();
	int x = index;
	if (lengthInCodePoints >= 0) {
	    assert (index >= 0 && index < length);
	    for (int i = 0; x < length && i < lengthInCodePoints; i++) {
		if (Character.isHighSurrogate(seq.charAt(x++))) {
		    if (x < length && Character.isLowSurrogate(seq.charAt(x))) {
			x++;
		    }
		}
	    }
	    return x - index;
	}

	assert (index >= 0 && index <= length);
	if (index == 0) {
	    return 0;
	}
	int len = -lengthInCodePoints;
	for (int i = 0; x > 0 && i < len; i++) {
	    if (Character.isLowSurrogate(seq.charAt(--x))) {
		if (x > 0 && Character.isHighSurrogate(seq.charAt(x-1))) {
		    x--;
		}
	    }
	}
	return index - x;
    }

    private static final int countCodePoints(CharSequence seq) {
	int length = seq.length();
	int n = 0;
	for (int i = 0; i < length; ) {
	    n++;
	    if (Character.isHighSurrogate(seq.charAt(i++))) {
		if (i < length && Character.isLowSurrogate(seq.charAt(i))) {
		    i++;
		}
	    }
	}
	return n;
    }

    /**
     *  Creates a bit vector for matching Latin-1 values. A normal BitClass
     *  never matches values above Latin-1, and a complemented BitClass always
     *  matches values above Latin-1.
     */
    static final class BitClass extends Node {
        boolean[] bits = new boolean[256];
        boolean complementMe = false;
        BitClass(boolean not) {
            complementMe = not;
        }
        BitClass(boolean[] newBits, boolean not) {
            complementMe = not;
            bits = newBits;
        }
        Node add(int c, int f) {
            if ((f & CASE_INSENSITIVE) == 0) {
                bits[c] = true;
            } else if (ASCII.isAscii(c)) {
                bits[ASCII.toUpper(c)] = true;
                bits[ASCII.toLower(c)] = true;
            } else {
                bits[Character.toLowerCase((char)c)] = true;
                bits[Character.toUpperCase((char)c)] = true;
            }
            return this;
        }
        Node dup(boolean not) {
            return new BitClass(bits, not);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i >= matcher.to) {
                matcher.hitEnd = true;
                return false;
            }
	    int c = Character.codePointAt(seq, i);
	    boolean charMatches =
		(c > 255) ? complementMe : (bits[c] ^ complementMe);
            return charMatches && next.match(matcher, i+Character.charCount(c), seq);
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    /**
     *  Utility method for creating a single character matcher.
     */
    private Node newSingle(int ch) {
        int f = flags;
        if ((f & (CASE_INSENSITIVE|UNICODE_CASE)) == 0) {
            return new Single(ch);
        }
        if ((f & UNICODE_CASE) == 0) {
            return new SingleA(ch);
        }
        return new SingleU(ch);
    }

    /**
     *  Utility method for creating a string slice matcher.
     */
    private Node newSlice(int[] buf, int count, boolean hasSupplementary) {
        int[] tmp = new int[count];
        int i = flags;
        if ((i & (CASE_INSENSITIVE|UNICODE_CASE)) == 0) {
            for (i = 0; i < count; i++) {
                tmp[i] = buf[i];
            }
            return hasSupplementary ? new SliceS(tmp) : new Slice(tmp);
        } else if ((i & UNICODE_CASE) == 0) {
            for (i = 0; i < count; i++) {
                tmp[i] = (char)ASCII.toLower(buf[i]);
            }
            return new SliceA(tmp);
        } else {
            for (i = 0; i < count; i++) {
                int c = buf[i];
                c = Character.toUpperCase(c);
                c = Character.toLowerCase(c);
                tmp[i] = c;
            }
            return new SliceU(tmp);
        }
    }

    /**
     * The following classes are the building components of the object
     * tree that represents a compiled regular expression. The object tree
     * is made of individual elements that handle constructs in the Pattern.
     * Each type of object knows how to match its equivalent construct with
     * the match() method.
     */

    /**
     * Base class for all node classes. Subclasses should override the match()
     * method as appropriate. This class is an accepting node, so its match()
     * always returns true.
     */
    static class Node extends Object {
        Node next;
        Node() {
            next = Pattern.accept;
        }
        Node dup(boolean not) {
            if (not) {
                return new Not(this);
            } else {
                throw new RuntimeException("internal error in Node dup()");
            }
        }
        /**
         * This method implements the classic accept node.
         */
        boolean match(Matcher matcher, int i, CharSequence seq) {
            matcher.last = i;
            matcher.groups[0] = matcher.first;
            matcher.groups[1] = matcher.last;
            return true;
        }
        /**
         * This method is good for all zero length assertions.
         */
        boolean study(TreeInfo info) {
            if (next != null) {
                return next.study(info);
            } else {
                return info.deterministic;
            }
        }
    }

    static class LastNode extends Node {
        /**
         * This method implements the classic accept node with
         * the addition of a check to see if the match occurred
         * using all of the input.
         */
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (matcher.acceptMode == Matcher.ENDANCHOR && i != matcher.to)
                return false;
            matcher.last = i;
            matcher.groups[0] = matcher.first;
            matcher.groups[1] = matcher.last;
            return true;
        }
    }

    /**
     * Dummy node to assist in connecting branches.
     */
    static class Dummy extends Node {
        boolean match(Matcher matcher, int i, CharSequence seq) {
            return next.match(matcher, i, seq);
        }
    }

    /**
     * Used for REs that can start anywhere within the input string.
     * This basically tries to match repeatedly at each spot in the
     * input string, moving forward after each try. An anchored search
     * or a BnM will bypass this node completely.
     */
    static class Start extends Node {
        int minLength;
        Start(Node node) {
            this.next = node;
            TreeInfo info = new TreeInfo();
            next.study(info);
            minLength = info.minLength;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i > matcher.to - minLength) {
                matcher.hitEnd = true;
                return false;
            }
            boolean ret = false;
            int guard = matcher.to - minLength;
            for (; i <= guard; i++) {
                if (ret = next.match(matcher, i, seq))
                    break;
                if (i == guard)
                    matcher.hitEnd = true;
            }
            if (ret) {
                matcher.first = i;
                matcher.groups[0] = matcher.first;
                matcher.groups[1] = matcher.last;
            }
            return ret;
        }
        boolean study(TreeInfo info) {
            next.study(info);
            info.maxValid = false;
            info.deterministic = false;
            return false;
        }
    }

    /*
     * StartS supports supplementary characters, including unpaired surrogates.
     */
    static final class StartS extends Start {
        StartS(Node node) {
	    super(node);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i > matcher.to - minLength) {
                matcher.hitEnd = true;
                return false;
            }
            boolean ret = false;
            int guard = matcher.to - minLength;
            while (i <= guard) {
                if ((ret = next.match(matcher, i, seq)) || i == guard)
                    break;
		// Optimization to move to the next character. This is
		// faster than countChars(seq, i, 1).
		if (Character.isHighSurrogate(seq.charAt(i++))) {
		    if (i < seq.length() && Character.isLowSurrogate(seq.charAt(i))) {
			i++;
		    }
		}
                if (i == guard)
                    matcher.hitEnd = true;
            }
            if (ret) {
                matcher.first = i;
                matcher.groups[0] = matcher.first;
                matcher.groups[1] = matcher.last;
            }
            return ret;
        }
    }

    /**
     * Node to anchor at the beginning of input. This object implements the
     * match for a \A sequence, and the caret anchor will use this if not in
     * multiline mode.
     */
    static final class Begin extends Node {
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int fromIndex = (matcher.anchoringBounds) ?
                matcher.from : 0;
            if (i == fromIndex && next.match(matcher, i, seq)) {
                matcher.first = i;
                matcher.groups[0] = i;
                matcher.groups[1] = matcher.last;
                return true;
            } else {
                return false;
            }
        }
    }

    /**
     * Node to anchor at the end of input. This is the absolute end, so this
     * should not match at the last newline before the end as $ will.
     */
    static final class End extends Node {
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int endIndex = (matcher.anchoringBounds) ?
                matcher.to : matcher.getTextLength();
            if (i == endIndex) {
                matcher.hitEnd = true;
                return next.match(matcher, i, seq);
            }
            return false;
        }
    }

    /**
     * Node to anchor at the beginning of a line. This is essentially the
     * object to match for the multiline ^.
     */
    static final class Caret extends Node {
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int startIndex = matcher.from;
            int endIndex = matcher.to;
            if (!matcher.anchoringBounds) {
                startIndex = 0;
                endIndex = matcher.getTextLength();
            }
            // Perl does not match ^ at end of input even after newline
            if (i == endIndex) {
                matcher.hitEnd = true;
                return false;
            }
            if (i > startIndex) {
                char ch = seq.charAt(i-1);
                if (ch != '\n' && ch != '\r'
                    && (ch|1) != '\u2029'
                    && ch != '\u0085' ) {
                    return false;
                }
                // Should treat /r/n as one newline
                if (ch == '\r' && seq.charAt(i) == '\n')
                    return false;
            }
            return next.match(matcher, i, seq);
        }
    }

    /**
     * Node to anchor at the beginning of a line when in unixdot mode.
     */
    static final class UnixCaret extends Node {
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int startIndex = matcher.from;
            int endIndex = matcher.to;
            if (!matcher.anchoringBounds) {
                startIndex = 0;
                endIndex = matcher.getTextLength();
            }
            // Perl does not match ^ at end of input even after newline
            if (i == endIndex) {
                matcher.hitEnd = true;
                return false;
            }
            if (i > startIndex) {
                char ch = seq.charAt(i-1);
                if (ch != '\n') {
                    return false;
                }
            }
            return next.match(matcher, i, seq);
        }
    }

    /**
     * Node to match the location where the last match ended.
     * This is used for the \G construct.
     */
    static final class LastMatch extends Node {
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i != matcher.oldLast)
                return false;
            return next.match(matcher, i, seq);
        }
    }

    /**
     * Node to anchor at the end of a line or the end of input based on the
     * multiline mode.
     *
     * When not in multiline mode, the $ can only match at the very end
     * of the input, unless the input ends in a line terminator in which
     * it matches right before the last line terminator.
     *
     * Note that \r\n is considered an atomic line terminator.
     * 
     * Like ^ the $ operator matches at a position, it does not match the
     * line terminators themselves.
     */
    static final class Dollar extends Node {
        boolean multiline;
        Dollar(boolean mul) {
            multiline = mul;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int endIndex = (matcher.anchoringBounds) ?
                matcher.to : matcher.getTextLength();
            if (!multiline) {
                if (i < endIndex - 2)
                    return false;
                if (i == endIndex - 2) {
                    char ch = seq.charAt(i);
                    if (ch != '\r')
                        return false;
                    ch = seq.charAt(i + 1);
                    if (ch != '\n')
                        return false;
                }
            }
            // Matches before any line terminator; also matches at the
            // end of input
            // Before line terminator:
            // If multiline, we match here no matter what
            // If not multiline, fall through so that the end
            // is marked as hit; this must be a /r/n or a /n
            // at the very end so the end was hit; more input
            // could make this not match here
            if (i < endIndex) {
                char ch = seq.charAt(i);
                 if (ch == '\n') {
                     // No match between \r\n
                     if (i > 0 && seq.charAt(i-1) == '\r')
                         return false;
                     if (multiline)
                         return next.match(matcher, i, seq);
                 } else if (ch == '\r' || ch == '\u0085' ||
                            (ch|1) == '\u2029') {
                     if (multiline)
                         return next.match(matcher, i, seq);
                 } else { // No line terminator, no match
                     return false;
                 }
            }
            // Matched at current end so hit end
            matcher.hitEnd = true;
            // If a $ matches because of end of input, then more input
            // could cause it to fail!
            matcher.requireEnd = true;
            return next.match(matcher, i, seq);
        }
        boolean study(TreeInfo info) {
            next.study(info);
            return info.deterministic;
        }
    }

    /**
     * Node to anchor at the end of a line or the end of input based on the
     * multiline mode when in unix lines mode.
     */
    static final class UnixDollar extends Node {
        boolean multiline;
        UnixDollar(boolean mul) {
            multiline = mul;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int endIndex = (matcher.anchoringBounds) ?
                matcher.to : matcher.getTextLength();
            if (i < endIndex) {
                char ch = seq.charAt(i);
                if (ch == '\n') {
                    // If not multiline, then only possible to
                    // match at very end or one before end
                    if (multiline == false && i != endIndex - 1)
                        return false;
                    // If multiline return next.match without setting
                    // matcher.hitEnd
                    if (multiline)
                        return next.match(matcher, i, seq);
                } else {
                    return false;
                }
            }
            // Matching because at the end or 1 before the end; 
            // more input could change this so set hitEnd
            matcher.hitEnd = true;
            // If a $ matches because of end of input, then more input
            // could cause it to fail!
            matcher.requireEnd = true;
            return next.match(matcher, i, seq);
        }
        boolean study(TreeInfo info) {
            next.study(info);
            return info.deterministic;
        }
    }

    /**
     * Node class for a single character value.
     */
    static final class Single extends Node {
        int ch;
	int len;
        Single(int n) {
            ch = n;
	    len = Character.charCount(ch);
        }
        Node dup(boolean not) {
            if (not)
                return new NotSingle(ch);
            else
                return new Single(ch);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i >= matcher.to) {
                matcher.hitEnd = true;
		return false;
            }
	    int c = Character.codePointAt(seq, i);
            return (c == ch
                && next.match(matcher, i+len, seq));
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    /**
     * Node class to match any character except a single char value.
     */
    static final class NotSingle extends Node {
        int ch;
        NotSingle(int n) {
            ch = n;
        }
        Node dup(boolean not) {
            if (not)
                return new Single(ch);
            else
                return new NotSingle(ch);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i >= matcher.to) {
                matcher.hitEnd = true;
		return false;
            }
	    int c = Character.codePointAt(seq, i);
            return (c != ch
                && next.match(matcher, i+Character.charCount(c), seq));
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    /**
     * Case independent ASCII value.
     */
    static final class SingleA extends Node {
        int ch;
        SingleA(int n) {
            ch = ASCII.toLower(n);
        }
        Node dup(boolean not) {
            if (not)
                return new NotSingleA(ch);
            else
                return new SingleA(ch);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                int c = seq.charAt(i);
                if (c == ch || ASCII.toLower(c) == ch) {
                    return next.match(matcher, i+1, seq);
                }
            }
            matcher.hitEnd = true;
            return false;
        }

        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    static final class NotSingleA extends Node {
        int ch;
        NotSingleA(int n) {
            ch = ASCII.toLower(n);
        }
        Node dup(boolean not) {
            if (not)
                return new SingleA(ch);
            else
                return new NotSingleA(ch);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                int c = Character.codePointAt(seq, i);
                if (c != ch && ASCII.toLower(c) != ch) {
                    return next.match(matcher, i+Character.charCount(c), seq);
                }
            }
            matcher.hitEnd = true;
            return false;
        }

        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    /**
     * Case independent unicode (including supplementary characters) value.
     */
    static final class SingleU extends Node {
        int ch;
	int len;
        SingleU(int c) {
            ch = Character.toLowerCase(Character.toUpperCase(c));
	    len = Character.charCount(ch);
        }
        Node dup(boolean not) {
            if (not)
                return new NotSingleU(ch);
            else
                return new SingleU(ch);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                int c = Character.codePointAt(seq, i);
                if (c == ch)
                    return next.match(matcher, i+len, seq);
                int cc = Character.toUpperCase(c);
                cc = Character.toLowerCase(cc);
                if (cc == ch)
                    return next.match(matcher, i+Character.charCount(c), seq);
            }
            matcher.hitEnd = true;
            return false;
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    /**
     * Case independent unicode value.
     */
    static final class NotSingleU extends Node {
        int ch;
        NotSingleU(int c) {
            ch = Character.toLowerCase(Character.toUpperCase((char)c));
        }
        Node dup(boolean not) {
            if (not)
                return new SingleU(ch);
            else
                return new NotSingleU(ch);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                int c = Character.codePointAt(seq, i);
                if (c == ch)
                    return false;
                int cc = Character.toUpperCase(c);
                cc = Character.toLowerCase(cc);
                if (cc != ch)
                    return next.match(matcher, i+Character.charCount(c), seq);
            }
            matcher.hitEnd = true;
            return false;
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    /**
     * Node class that matches a Unicode category.
     */
    static final class Category extends Node {
        int atype;
        Category(int type) {
            atype = type;
        }
        Node dup(boolean not) {
            return new Category(not ? ~atype : atype);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i >= matcher.to) {
                matcher.hitEnd = true;
		return false;
            }
	    int c = Character.codePointAt(seq, i);
	    return (atype & (1 << Character.getType(c))) != 0
                && next.match(matcher, i+Character.charCount(c), seq);
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    /**
     * Node class that matches a POSIX type.
     */
    static final class Ctype extends Node {
        int ctype;
        Ctype(int type) {
            ctype = type;
        }
        Node dup(boolean not) {
            if (not) {
                return new NotCtype(ctype);
            } else {
                return new Ctype(ctype);
            }
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i >= matcher.to) {
                matcher.hitEnd = true;
		return false;
            }
	    int c = Character.codePointAt(seq, i);
	    return (c < 128 && ASCII.isType(c, ctype)
                && next.match(matcher, i+1, seq));
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    static final class NotCtype extends Node {
        int ctype;
        NotCtype(int type) {
            ctype = type;
        }
        Node dup(boolean not) {
            if (not) {
                return new Ctype(ctype);
            } else {
                return new NotCtype(ctype);
            }
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i >= matcher.to) {
                matcher.hitEnd = true;
		return false;
            }
	    int c = Character.codePointAt(seq, i);
	    return ((c >= 128 || !ASCII.isType(c, ctype))
                && next.match(matcher, i+Character.charCount(c), seq));
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    static abstract class JavaTypeClass extends Node {
        JavaTypeClass() {
        }
        Node dup(boolean not) {
            Node duplicate = null;
            try {
                duplicate = (Node)this.getClass().newInstance();
            } catch (InstantiationException ie) {
                throw new Error("Cannot instantiate node");
            } catch (IllegalAccessException iae) {
                throw new Error("Cannot instantiate node");
            }
            if (not)
                return new Not(duplicate);
            else
                return duplicate;
        }
	abstract boolean isProperty(int ch);
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i >= matcher.to) {
                matcher.hitEnd = true;
		return false;
            }
	    int c = Character.codePointAt(seq, i);
	    return (isProperty(c)
                && next.match(matcher, i+Character.charCount(c), seq));
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    static final class JavaLowerCase extends JavaTypeClass {
        JavaLowerCase() {
        }
	boolean isProperty(int ch) {
	    return Character.isLowerCase(ch);
	}
    }

    static final class JavaUpperCase extends JavaTypeClass {
        JavaUpperCase() {
        }
	boolean isProperty(int ch) {
	    return Character.isUpperCase(ch);
        }
    }

    static final class JavaTitleCase extends JavaTypeClass {
        JavaTitleCase() {
        }
	boolean isProperty(int ch) {
	    return Character.isTitleCase(ch);
	}
    }

    static final class JavaDigit extends JavaTypeClass {
        JavaDigit() {
        }
	boolean isProperty(int ch) {
	    return Character.isDigit(ch);
        }
    }

    static final class JavaDefined extends JavaTypeClass {
        JavaDefined() {
        }
	boolean isProperty(int ch) {
	    return Character.isDefined(ch);
        }
    }

    static final class JavaLetter extends JavaTypeClass {
        JavaLetter() {
        }
	boolean isProperty(int ch) {
	    return Character.isLetter(ch);
        }
    }

    static final class JavaLetterOrDigit extends JavaTypeClass {
        JavaLetterOrDigit() {
        }
	boolean isProperty(int ch) {
	    return Character.isLetterOrDigit(ch);
        }
    }

    static final class JavaJavaIdentifierStart extends JavaTypeClass {
        JavaJavaIdentifierStart() {
        }
	boolean isProperty(int ch) {
	    return Character.isJavaIdentifierStart(ch);
        }
    }

    static final class JavaJavaIdentifierPart extends JavaTypeClass {
        JavaJavaIdentifierPart() {
        }
	boolean isProperty(int ch) {
	    return Character.isJavaIdentifierPart(ch);
        }
    }

    static final class JavaUnicodeIdentifierStart extends JavaTypeClass {
        JavaUnicodeIdentifierStart() {
        }
	boolean isProperty(int ch) {
	    return Character.isUnicodeIdentifierStart(ch);
        }
    }

    static final class JavaUnicodeIdentifierPart extends JavaTypeClass {
        JavaUnicodeIdentifierPart() {
        }
	boolean isProperty(int ch) {
	    return Character.isUnicodeIdentifierPart(ch);
        }
    }

    static final class JavaIdentifierIgnorable extends JavaTypeClass {
        JavaIdentifierIgnorable() {
        }
	boolean isProperty(int ch) {
	    return Character.isIdentifierIgnorable(ch);
        }
    }

    static final class JavaSpaceChar extends JavaTypeClass {
        JavaSpaceChar() {
        }
	boolean isProperty(int ch) {
	    return Character.isSpaceChar(ch);
        }
    }

    static final class JavaWhitespace extends JavaTypeClass {
        JavaWhitespace() {
        }
	boolean isProperty(int ch) {
	    return Character.isWhitespace(ch);
        }
    }

    static final class JavaISOControl extends JavaTypeClass {
        JavaISOControl() {
        }
	boolean isProperty(int ch) {
	    return Character.isISOControl(ch);
        }
    }

    static final class JavaMirrored extends JavaTypeClass {
        JavaMirrored() {
        }
	boolean isProperty(int ch) {
	    return Character.isMirrored(ch);
        }
    }

    static final class Specials extends Node {
        Specials() {
        }
        Node dup(boolean not) {
            if (not)
                return new Not(this);
            else
                return new Specials();
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                int ch = seq.charAt(i);
                return (((ch-0xFFF0) | (0xFFFD-ch)) >= 0 || ch == 0xFEFF)
                    && next.match(matcher, i+1, seq);
            }
            matcher.hitEnd = true;
            return false;
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    static final class Not extends Node {
        Node atom;
        Not(Node atom) {
            this.atom = atom;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            return !atom.match(matcher, i, seq)
		&& next.match(matcher, i+countChars(seq, i, 1), seq);
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    /**
     * Node class for a case sensitive sequence of literal characters.
     */
    static class Slice extends Node {
        int[] buffer;
        Slice(int[] buf) {
            buffer = buf;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int[] buf = buffer;
            int len = buf.length;

            // Unfortunately we must now void this opto
            // in order to properly report hitEnd...
            //if (i + len > matcher.to) {
            //    matcher.hitEnd = true;
            //    return false;
            //}

            for (int j=0; j= matcher.to) {
                    matcher.hitEnd = true;
                    return false;
                }
                if (buf[j] != seq.charAt(i+j))
                    return false;
            }

            return next.match(matcher, i+len, seq);
        }
        boolean study(TreeInfo info) {
            info.minLength += buffer.length;
            info.maxLength += buffer.length;
            return next.study(info);
        }
    }

    /**
     * Node class for a case insensitive sequence of literal characters.
     */
    static final class SliceA extends Node {
        int[] buffer;
        SliceA(int[] buf) {
            buffer = buf;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int[] buf = buffer;
            int len = buf.length;
            for (int j=0; j= matcher.to) {
                    matcher.hitEnd = true;
                    return false;
                }
                int c = ASCII.toLower(seq.charAt(i+j));
                if (buf[j] != c)
                    return false;
            }
            return next.match(matcher, i+len, seq);
        }
        boolean study(TreeInfo info) {
            info.minLength += buffer.length;
            info.maxLength += buffer.length;
            return next.study(info);
        }
    }

    /**
     * Node class for a case insensitive sequence of literal characters.
     * Uses unicode case folding.
     */
    static final class SliceU extends Node {
        int[] buffer;
        SliceU(int[] buf) {
            buffer = buf;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int[] buf = buffer;
            int x = i;
            for (int j = 0; j < buf.length; j++) {
                if (x >= matcher.to) {
                    matcher.hitEnd = true;
                    return false;
                }
                int c = Character.codePointAt(seq, x);
                int cc = Character.toUpperCase(c);
                cc = Character.toLowerCase(cc);
                if (buf[j] != cc) {
                    return false;
                }
		x += Character.charCount(c);
		if (x > matcher.to) {
                    matcher.hitEnd = true;
		    return false;
		}
            }
            return next.match(matcher, x, seq);
        }
        boolean study(TreeInfo info) {
            info.minLength += buffer.length;
            info.maxLength += buffer.length;
            return next.study(info);
        }
    }

    /**
     * Node class for a case sensitive sequence of literal characters
     * including supplementary characters.
     */
    static final class SliceS extends Slice {
        SliceS(int[] buf) {
            super(buf);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
	    int[] buf = buffer;
	    int x = i;
	    for (int j = 0; j < buf.length; j++) {
                if (x >= matcher.to) {
                    matcher.hitEnd = true;
                    return false;
                }
		int c = Character.codePointAt(seq, x);
		if (buf[j] != c)
		    return false;
		x += Character.charCount(c);
		if (x > matcher.to) {
                    matcher.hitEnd = true;
		    return false;
                }
	    }

            return next.match(matcher, x, seq);
        }
    }

    /**
     * Node class for matching characters within an explicit value range.
     */
    static class Range extends Node {
        int lower, upper;
        Range() {
        }
        Range(int n) {
            lower = n >>> 16;
            upper = n & 0xFFFF;
        }
	Range(int lower, int upper) {
	    this.lower = lower;
	    this.upper = upper;
	}
        Node dup(boolean not) {
            if (not)
                return new NotRange(lower, upper);
            else
                return new Range(lower, upper);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                int ch = Character.codePointAt(seq, i);
                return ((ch-lower)|(upper-ch)) >= 0
                    && next.match(matcher, i+Character.charCount(ch), seq);
            }
            matcher.hitEnd = true;
            return false;
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    /**
     * Node class for matching characters within an explicit value range
     * in a case insensitive manner.
     */
    static final class CIRange extends Range {
        CIRange(int n) {
            lower = n >>> 16;
            upper = n & 0xFFFF;
        }
        CIRange(int lower, int upper) {
	    super(lower, upper);
        }
        Node dup(boolean not) {
            if (not)
                return new CINotRange(lower, upper);
            else
                return new CIRange(lower, upper);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                int ch = Character.codePointAt(seq, i);
                boolean m = (((ch-lower)|(upper-ch)) >= 0);
                if (!m) {
                    int cc = Character.toUpperCase(ch);
                    m = (((cc-lower)|(upper-cc)) >= 0);
                    if (!m) {
                        cc = Character.toLowerCase(cc);
                        m = (((cc-lower)|(upper-cc)) >= 0);
                    }
                }
                return (m && next.match(matcher, i+Character.charCount(ch), seq));
            }
            matcher.hitEnd = true;
            return false;
        }
    }

    static class NotRange extends Node {
        int lower, upper;
        NotRange() {
        }
        NotRange(int n) {
            lower = n >>> 16;
            upper = n & 0xFFFF;
        }
        NotRange(int lower, int upper) {
            this.lower = lower;
	    this.upper = upper;
	}
        Node dup(boolean not) {
            if (not) {
                return new Range(lower, upper);
            } else {
                return new NotRange(lower, upper);
            }
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                int ch = Character.codePointAt(seq, i);
                return ((ch-lower)|(upper-ch)) < 0
                    && next.match(matcher, i+Character.charCount(ch), seq);
            }
            matcher.hitEnd = true;
            return false;
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    static class CINotRange extends NotRange {
        int lower, upper;
        CINotRange(int n) {
            lower = n >>> 16;
            upper = n & 0xFFFF;
        }
        CINotRange(int lower, int upper) {
            this.lower = lower;
            this.upper = upper;
	}
        Node dup(boolean not) {
            if (not) {
                return new CIRange(lower, upper);
            } else {
                return new CINotRange(lower, upper);
            }
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                int ch = Character.codePointAt(seq, i);
                boolean m = (((ch-lower)|(upper-ch)) < 0);
                if (m) {
                    int cc = Character.toUpperCase(ch);
                    m = (((cc-lower)|(upper-cc)) < 0);
                    if (m) {
                        cc = Character.toLowerCase(cc);
                        m = (((cc-lower)|(upper-cc)) < 0);
                    }
                }

                return (m && next.match(matcher, i+Character.charCount(ch), seq));
            }
            matcher.hitEnd = true;
            return false;
        }
    }

    /**
     * Implements the Unicode category ALL and the dot metacharacter when
     * in dotall mode.
     */
    static final class All extends Node {
        All() {
            super();
        }
        Node dup(boolean not) {
            if (not) {
                return new Single(-1);
            } else {
                return new All();
            }
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                return next.match(matcher, i+countChars(seq, i, 1), seq);
            }
            matcher.hitEnd = true;
            return false;            
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    /**
     * Node class for the dot metacharacter when dotall is not enabled.
     */
    static final class Dot extends Node {
        Dot() {
            super();
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                int ch = Character.codePointAt(seq, i);
                return (ch != '\n' && ch != '\r'
                    && (ch|1) != '\u2029'
                    && ch != '\u0085'
                    && next.match(matcher, i+Character.charCount(ch), seq));
            }
            matcher.hitEnd = true;
            return false;
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    /**
     * Node class for the dot metacharacter when dotall is not enabled
     * but UNIX_LINES is enabled.
     */
    static final class UnixDot extends Node {
        UnixDot() {
            super();
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                int ch = Character.codePointAt(seq, i);
                return (ch != '\n' && next.match(matcher, i+Character.charCount(ch), seq));
            }
            matcher.hitEnd = true;
            return false;
        }
        boolean study(TreeInfo info) {
            info.minLength++;
            info.maxLength++;
            return next.study(info);
        }
    }

    /**
     * The 0 or 1 quantifier. This one class implements all three types.
     */
    static final class Ques extends Node {
        Node atom;
        int type;
        Ques(Node node, int type) {
            this.atom = node;
            this.type = type;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            switch (type) {
            case GREEDY:
                return (atom.match(matcher, i, seq) && next.match(matcher, matcher.last, seq))
                    || next.match(matcher, i, seq);
            case LAZY:
                return next.match(matcher, i, seq)
                    || (atom.match(matcher, i, seq) && next.match(matcher, matcher.last, seq));
            case POSSESSIVE:
                if (atom.match(matcher, i, seq)) i = matcher.last;
                return next.match(matcher, i, seq);
            default:
                return atom.match(matcher, i, seq) && next.match(matcher, matcher.last, seq);
            }
        }
        boolean study(TreeInfo info) {
            if (type != INDEPENDENT) {
                int minL = info.minLength;
                atom.study(info);
                info.minLength = minL;
                info.deterministic = false;
                return next.study(info);
            } else {
                atom.study(info);
                return next.study(info);
            }
        }
    }

    /**
     * Handles the curly-brace style repetition with a specified minimum and
     * maximum occurrences. The * quantifier is handled as a special case.
     * This class handles the three types.
     */
    static final class Curly extends Node {
        Node atom;
        int type;
        int cmin;
        int cmax;

        Curly(Node node, int cmin, int cmax, int type) {
            this.atom = node;
            this.type = type;
            this.cmin = cmin;
            this.cmax = cmax;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int j;
            for (j = 0; j < cmin; j++) {
                if (atom.match(matcher, i, seq)) {
                    i = matcher.last;
                    continue;
                }
                return false;
            }
            if (type == GREEDY)
                return match0(matcher, i, j, seq);
            else if (type == LAZY)
                return match1(matcher, i, j, seq);
            else
                return match2(matcher, i, j, seq);
        }
        // Greedy match.
        // i is the index to start matching at
        // j is the number of atoms that have matched
        boolean match0(Matcher matcher, int i, int j, CharSequence seq) {
            if (j >= cmax) {
                // We have matched the maximum... continue with the rest of
                // the regular expression
                return next.match(matcher, i, seq);
            }
            int backLimit = j;
            while (atom.match(matcher, i, seq)) {
                // k is the length of this match
                int k = matcher.last - i;
                if (k == 0) // Zero length match
                    break;
                // Move up index and number matched
                i = matcher.last;
                j++;
                // We are greedy so match as many as we can
                while (j < cmax) {
                    if (!atom.match(matcher, i, seq))
                        break;
                    if (i + k != matcher.last) {
                        if (match0(matcher, matcher.last, j+1, seq))
                            return true;
                        break;
                    }
                    i += k;
                    j++;
                }
                // Handle backing off if match fails
                while (j >= backLimit) {
                   if (next.match(matcher, i, seq))
                        return true;
                    i -= k;
                    j--;
                }
                return false;
            }
            return next.match(matcher, i, seq);
        }
        // Reluctant match. At this point, the minimum has been satisfied.
        // i is the index to start matching at
        // j is the number of atoms that have matched
        boolean match1(Matcher matcher, int i, int j, CharSequence seq) {
            for (;;) {
                // Try finishing match without consuming any more
                if (next.match(matcher, i, seq))
                    return true;
                // At the maximum, no match found
                if (j >= cmax)
                    return false;
                // Okay, must try one more atom
                if (!atom.match(matcher, i, seq))
                    return false;
                // If we haven't moved forward then must break out
                if (i == matcher.last)
                    return false;
                // Move up index and number matched
                i = matcher.last;
                j++;
            }
        }
        boolean match2(Matcher matcher, int i, int j, CharSequence seq) {
            for (; j < cmax; j++) {
                if (!atom.match(matcher, i, seq))
                    break;
                if (i == matcher.last)
                    break;
                i = matcher.last;
            }
            return next.match(matcher, i, seq);
        }
        boolean study(TreeInfo info) {
            // Save original info
            int minL = info.minLength;
            int maxL = info.maxLength;
            boolean maxV = info.maxValid;
            boolean detm = info.deterministic;
            info.reset();

            atom.study(info);

            int temp = info.minLength * cmin + minL;
            if (temp < minL) {
                temp = 0xFFFFFFF; // arbitrary large number
            }
            info.minLength = temp;

            if (maxV & info.maxValid) {
                temp = info.maxLength * cmax + maxL;
                info.maxLength = temp;
                if (temp < maxL) {
                    info.maxValid = false;
                }
            } else {
                info.maxValid = false;
            }

            if (info.deterministic && cmin == cmax)
                info.deterministic = detm;
            else
                info.deterministic = false;

            return next.study(info);
        }
    }

    /**
     * Handles the curly-brace style repetition with a specified minimum and
     * maximum occurrences in deterministic cases. This is an iterative
     * optimization over the Prolog and Loop system which would handle this
     * in a recursive way. The * quantifier is handled as a special case.
     * If capture is true then this class saves group settings and ensures
     * that groups are unset when backing off of a group match.
     */
    static final class GroupCurly extends Node {
        Node atom;
        int type;
        int cmin;
        int cmax;
        int localIndex;
        int groupIndex;
        boolean capture;

        GroupCurly(Node node, int cmin, int cmax, int type, int local,
                   int group, boolean capture) {
            this.atom = node;
            this.type = type;
            this.cmin = cmin;
            this.cmax = cmax;
            this.localIndex = local;
            this.groupIndex = group;  
            this.capture = capture;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int[] groups = matcher.groups;
            int[] locals = matcher.locals;
            int save0 = locals[localIndex];
            int save1 = 0;
            int save2 = 0;

            if (capture) {
                save1 = groups[groupIndex];
                save2 = groups[groupIndex+1];
            }

            // Notify GroupTail there is no need to setup group info
            // because it will be set here
            locals[localIndex] = -1;

            boolean ret = true;
            for (int j = 0; j < cmin; j++) {
                if (atom.match(matcher, i, seq)) {
                    if (capture) {
                        groups[groupIndex] = i;
                        groups[groupIndex+1] = matcher.last;
                    }
                    i = matcher.last;
                } else {
                    ret = false;
                    break;
                }
            }
            if (!ret) {
                locals[localIndex] = save0;
                if (capture) {
                    groups[groupIndex] = save1;
                    groups[groupIndex+1] = save2;
                }
            } else if (type == GREEDY) {
                ret = match0(matcher, i, cmin, seq);
            } else if (type == LAZY) {
                ret = match1(matcher, i, cmin, seq);
            } else {
                ret = match2(matcher, i, cmin, seq);
            }
            return ret;
        }
        // Aggressive group match
        boolean match0(Matcher matcher, int i, int j, CharSequence seq) {
            int[] groups = matcher.groups;
            int save0 = 0;
            int save1 = 0;
            if (capture) {
                save0 = groups[groupIndex];
                save1 = groups[groupIndex+1];
            }
            for (;;) {
                if (j >= cmax)
                    break;
                if (!atom.match(matcher, i, seq))
                    break;
                int k = matcher.last - i;
                if (k <= 0) {
                    if (capture) {
                        groups[groupIndex] = i;
                        groups[groupIndex+1] = i + k;
                    }
                    i = i + k;
                    break;
                }
                for (;;) {
                    if (capture) {
                        groups[groupIndex] = i;
                        groups[groupIndex+1] = i + k;
                    }
                    i = i + k;
                    if (++j >= cmax)
                        break;
                    if (!atom.match(matcher, i, seq))
                        break;
                    if (i + k != matcher.last) {
                        if (match0(matcher, i, j, seq))
                            return true;
                        break;
                    }
                }
                while (j > cmin) {
                    if (next.match(matcher, i, seq)) {
                        if (capture) {
                            groups[groupIndex+1] = i;
                            groups[groupIndex] = i - k;
                        }
                        i = i - k;
                        return true;
                    }
                    // backing off
                    if (capture) {
                        groups[groupIndex+1] = i;
                        groups[groupIndex] = i - k;
                    }
                    i = i - k;
                    j--;
                }
                break;
            }
            if (capture) {
                groups[groupIndex] = save0;
                groups[groupIndex+1] = save1;
            }
            return next.match(matcher, i, seq);
        }
        // Reluctant matching
        boolean match1(Matcher matcher, int i, int j, CharSequence seq) {
            for (;;) {
                if (next.match(matcher, i, seq))
                    return true;
                if (j >= cmax)
                    return false;
                if (!atom.match(matcher, i, seq))
                    return false;
                if (i == matcher.last)
                    return false;
                if (capture) {
                    matcher.groups[groupIndex] = i;
                    matcher.groups[groupIndex+1] = matcher.last;
                }
                i = matcher.last;
                j++;
            }
        }
        // Possessive matching
        boolean match2(Matcher matcher, int i, int j, CharSequence seq) {
            for (; j < cmax; j++) {
                if (!atom.match(matcher, i, seq)) {
                    break;
                }
                if (capture) {
                    matcher.groups[groupIndex] = i;
                    matcher.groups[groupIndex+1] = matcher.last;
                }
                if (i == matcher.last) {
                    break;
                }
                i = matcher.last;
            }
            return next.match(matcher, i, seq);
        }
        boolean study(TreeInfo info) {
            // Save original info
            int minL = info.minLength;
            int maxL = info.maxLength;
            boolean maxV = info.maxValid;
            boolean detm = info.deterministic;
            info.reset();

            atom.study(info);

            int temp = info.minLength * cmin + minL;
            if (temp < minL) {
                temp = 0xFFFFFFF; // Arbitrary large number
            }
            info.minLength = temp;

            if (maxV & info.maxValid) {
                temp = info.maxLength * cmax + maxL;
                info.maxLength = temp;
                if (temp < maxL) {
                    info.maxValid = false;
                }
            } else {
                info.maxValid = false;
            }

            if (info.deterministic && cmin == cmax) {
                info.deterministic = detm;
            } else {
                info.deterministic = false;
            }

            return next.study(info);
        }
    }

    /**
     * Handles the branching of alternations. Note this is also used for
     * the ? quantifier to branch between the case where it matches once
     * and where it does not occur.
     */
    static final class Branch extends Node {
        Node prev;
        Branch(Node lhs, Node rhs) {
            this.prev = lhs;
            this.next = rhs;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            return (prev.match(matcher, i, seq) || next.match(matcher, i, seq));
        }
        boolean study(TreeInfo info) {
            int minL = info.minLength;
            int maxL = info.maxLength;
            boolean maxV = info.maxValid;
            info.reset();
            prev.study(info);

            int minL2 = info.minLength;
            int maxL2 = info.maxLength;
            boolean maxV2 = info.maxValid;
            info.reset();
            next.study(info);

            info.minLength = minL + Math.min(minL2, info.minLength);
            info.maxLength = maxL + Math.max(maxL2, info.maxLength);
            info.maxValid = (maxV & maxV2 & info.maxValid);
            info.deterministic = false;
            return false;
        }
    }

    /**
     * The GroupHead saves the location where the group begins in the locals
     * and restores them when the match is done.
     *
     * The matchRef is used when a reference to this group is accessed later
     * in the expression. The locals will have a negative value in them to
     * indicate that we do not want to unset the group if the reference
     * doesn't match.
     */
    static final class GroupHead extends Node {
        int localIndex;
        GroupHead(int localCount) {
            localIndex = localCount;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int save = matcher.locals[localIndex];
            matcher.locals[localIndex] = i;
            boolean ret = next.match(matcher, i, seq);
            matcher.locals[localIndex] = save;
            return ret;
        }
        boolean matchRef(Matcher matcher, int i, CharSequence seq) {
            int save = matcher.locals[localIndex];
            matcher.locals[localIndex] = ~i; // HACK
            boolean ret = next.match(matcher, i, seq);
            matcher.locals[localIndex] = save;
            return ret;
        }
    }

    /**
     * Recursive reference to a group in the regular expression. It calls
     * matchRef because if the reference fails to match we would not unset
     * the group.
     */
    static final class GroupRef extends Node {
        GroupHead head;
        GroupRef(GroupHead head) {
            this.head = head;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            return head.matchRef(matcher, i, seq)
                && next.match(matcher, matcher.last, seq);
        }
        boolean study(TreeInfo info) {
            info.maxValid = false;
            info.deterministic = false;
            return next.study(info);
        }
    }

    /**
     * The GroupTail handles the setting of group beginning and ending
     * locations when groups are successfully matched. It must also be able to
     * unset groups that have to be backed off of.
     *
     * The GroupTail node is also used when a previous group is referenced,
     * and in that case no group information needs to be set.
     */
    static final class GroupTail extends Node {
        int localIndex;
        int groupIndex;
        GroupTail(int localCount, int groupCount) {
            localIndex = localCount;
            groupIndex = groupCount + groupCount;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int tmp = matcher.locals[localIndex];
            if (tmp >= 0) { // This is the normal group case.
                // Save the group so we can unset it if it
                // backs off of a match.
                int groupStart = matcher.groups[groupIndex];
                int groupEnd = matcher.groups[groupIndex+1];

                matcher.groups[groupIndex] = tmp;
                matcher.groups[groupIndex+1] = i;
                if (next.match(matcher, i, seq)) {
                    return true;
                }
                matcher.groups[groupIndex] = groupStart;
                matcher.groups[groupIndex+1] = groupEnd;
                return false;
            } else {
                // This is a group reference case. We don't need to save any
                // group info because it isn't really a group.
                matcher.last = i;
                return true;
            }
        }
    }

    /**
     * This sets up a loop to handle a recursive quantifier structure.
     */
    static final class Prolog extends Node {
        Loop loop;
        Prolog(Loop loop) {
            this.loop = loop;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            return loop.matchInit(matcher, i, seq);
        }
        boolean study(TreeInfo info) {
            return loop.study(info);
        }
    }

    /**
     * Handles the repetition count for a greedy Curly. The matchInit
     * is called from the Prolog to save the index of where the group
     * beginning is stored. A zero length group check occurs in the
     * normal match but is skipped in the matchInit.
     */
    static class Loop extends Node {
        Node body;
        int countIndex; // local count index in matcher locals
        int beginIndex; // group beginning index
        int cmin, cmax;
        Loop(int countIndex, int beginIndex) {
            this.countIndex = countIndex;
            this.beginIndex = beginIndex;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            // Avoid infinite loop in zero-length case.
            if (i > matcher.locals[beginIndex]) {
                int count = matcher.locals[countIndex];

                // This block is for before we reach the minimum
                // iterations required for the loop to match
                if (count < cmin) {
                    matcher.locals[countIndex] = count + 1;
                    boolean b = body.match(matcher, i, seq);
                    // If match failed we must backtrack, so
                    // the loop count should NOT be incremented
                    if (!b)
                        matcher.locals[countIndex] = count;
                    // Return success or failure since we are under
                    // minimum
                    return b;
                }
                // This block is for after we have the minimum
                // iterations required for the loop to match
                if (count < cmax) {
                    matcher.locals[countIndex] = count + 1;
                    boolean b = body.match(matcher, i, seq);
                    // If match failed we must backtrack, so
                    // the loop count should NOT be incremented
                    if (!b)
                        matcher.locals[countIndex] = count;
                    else
                        return true;
                }
            }
            return next.match(matcher, i, seq);
        }
        boolean matchInit(Matcher matcher, int i, CharSequence seq) {
            int save = matcher.locals[countIndex];
            boolean ret = false;
            if (0 < cmin) {
                matcher.locals[countIndex] = 1;
                ret = body.match(matcher, i, seq);
            } else if (0 < cmax) {
                matcher.locals[countIndex] = 1;
                ret = body.match(matcher, i, seq);
                if (ret == false)
                    ret = next.match(matcher, i, seq);
            } else {
                ret = next.match(matcher, i, seq);
            }
            matcher.locals[countIndex] = save;
            return ret;
        }
        boolean study(TreeInfo info) {
            info.maxValid = false;
            info.deterministic = false;
            return false;
        }
    }

    /**
     * Handles the repetition count for a reluctant Curly. The matchInit
     * is called from the Prolog to save the index of where the group
     * beginning is stored. A zero length group check occurs in the
     * normal match but is skipped in the matchInit.
     */
    static final class LazyLoop extends Loop {
        LazyLoop(int countIndex, int beginIndex) {
            super(countIndex, beginIndex);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            // Check for zero length group
            if (i > matcher.locals[beginIndex]) {
                int count = matcher.locals[countIndex];
                if (count < cmin) {
                    matcher.locals[countIndex] = count + 1;
                    boolean result = body.match(matcher, i, seq);
                    // If match failed we must backtrack, so
                    // the loop count should NOT be incremented
                    if (!result)
                        matcher.locals[countIndex] = count;
                    return result;
                }
                if (next.match(matcher, i, seq))
                    return true;
                if (count < cmax) {
                    matcher.locals[countIndex] = count + 1;
                    boolean result = body.match(matcher, i, seq);
                    // If match failed we must backtrack, so
                    // the loop count should NOT be incremented
                    if (!result)
                        matcher.locals[countIndex] = count;
                    return result;
                }
                return false;
            }
            return next.match(matcher, i, seq);
        }
        boolean matchInit(Matcher matcher, int i, CharSequence seq) {
            int save = matcher.locals[countIndex];
            boolean ret = false;
            if (0 < cmin) {
                matcher.locals[countIndex] = 1;
                ret = body.match(matcher, i, seq);
            } else if (next.match(matcher, i, seq)) {
                ret = true;
            } else if (0 < cmax) {
                matcher.locals[countIndex] = 1;
                ret = body.match(matcher, i, seq);
            }
            matcher.locals[countIndex] = save;
            return ret;
        }
        boolean study(TreeInfo info) {
            info.maxValid = false;
            info.deterministic = false;
            return false;
        }
    }

    /**
     * Refers to a group in the regular expression. Attempts to match
     * whatever the group referred to last matched.
     */
    static class BackRef extends Node {
        int groupIndex;
        BackRef(int groupCount) {
            super();
            groupIndex = groupCount + groupCount;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int j = matcher.groups[groupIndex];
            int k = matcher.groups[groupIndex+1];

            int groupSize = k - j;

            // If the referenced group didn't match, neither can this
            if (j < 0)
                return false;

            // If there isn't enough input left no match
            if (i + groupSize > matcher.to) {
                matcher.hitEnd = true;
                return false;
            }

            // Check each new char to make sure it matches what the group
            // referenced matched last time around
            for (int index=0; index matcher.to) {
                matcher.hitEnd = true;
                return false;
            }

            // Check each new char to make sure it matches what the group
            // referenced matched last time around
	    int x = i;
            for (int index=0; index matcher.to) {
                    matcher.hitEnd = true;
                    return false;
                }
                if (atom.match(matcher, i, seq)) {
                    return next.match(matcher, matcher.last, seq);
                }
                i += countChars(seq, i, 1);
                matcher.first++;
            }
        }
        boolean study(TreeInfo info) {
            atom.study(info);
            info.maxValid = false;
            info.deterministic = false;
            return next.study(info);
        }
    }

    static final class Conditional extends Node {
        Node cond, yes, not;
        Conditional(Node cond, Node yes, Node not) {
            this.cond = cond;
            this.yes = yes;
            this.not = not;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (cond.match(matcher, i, seq)) {
                return yes.match(matcher, i, seq);
            } else {
                return not.match(matcher, i, seq);
            }
        }
        boolean study(TreeInfo info) {
            int minL = info.minLength;
            int maxL = info.maxLength;
            boolean maxV = info.maxValid;
            info.reset();
            yes.study(info);

            int minL2 = info.minLength;
            int maxL2 = info.maxLength;
            boolean maxV2 = info.maxValid;
            info.reset();
            not.study(info);

            info.minLength = minL + Math.min(minL2, info.minLength);
            info.maxLength = maxL + Math.max(maxL2, info.maxLength);
            info.maxValid = (maxV & maxV2 & info.maxValid);
            info.deterministic = false;
            return next.study(info);
        }
    }

    /**
     * Zero width positive lookahead.
     */
    static final class Pos extends Node {
        Node cond;
        Pos(Node cond) {
            this.cond = cond;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int savedTo = matcher.to;
            boolean conditionMatched = false;

            // Relax transparent region boundaries for lookahead
            if (matcher.transparentBounds)
                matcher.to = matcher.getTextLength();
            try {
                conditionMatched = cond.match(matcher, i, seq);
            } finally {
                // Reinstate region boundaries
                matcher.to = savedTo;
            }
            return conditionMatched && next.match(matcher, i, seq);
        }
    }

    /**
     * Zero width negative lookahead.
     */
    static final class Neg extends Node {
        Node cond;
        Neg(Node cond) {
            this.cond = cond;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int savedTo = matcher.to;
            boolean conditionMatched = false;
            
            // Relax transparent region boundaries for lookahead
            if (matcher.transparentBounds)
                matcher.to = matcher.getTextLength();
            try {
                if (i < matcher.to) {
                    conditionMatched = !cond.match(matcher, i, seq);
                } else {
                    // If a negative lookahead succeeds then more input
                    // could cause it to fail!
                    matcher.requireEnd = true;
                    conditionMatched = !cond.match(matcher, i, seq);
                }
            } finally {
                // Reinstate region boundaries
                matcher.to = savedTo;
            }
            return conditionMatched && next.match(matcher, i, seq);
        }
    }

    /**
     * Zero width positive lookbehind.
     */
    static class Behind extends Node {
        Node cond;
        int rmax, rmin;
        Behind(Node cond, int rmax, int rmin) {
            this.cond = cond;
            this.rmax = rmax;
            this.rmin = rmin;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int savedFrom = matcher.from;
            boolean conditionMatched = false;
            int startIndex = (!matcher.transparentBounds) ?
                             matcher.from : 0;
            int from = Math.max(i - rmax, startIndex);
            for (int j = i - rmin; j >= from; j--) {

                // Relax transparent region boundaries for lookbehind
                if (matcher.transparentBounds)
                    matcher.from = 0;
                try {
                    conditionMatched = 
                        (cond.match(matcher, j, seq) && matcher.last == i);
                } finally { // Reinstate region boundaries
                    matcher.from = savedFrom;
                }
                
                if (conditionMatched)
                    return next.match(matcher, i, seq);
            }
            return false;
        }
    }

    /**
     * Zero width positive lookbehind, including supplementary
     * characters or unpaired surrogates.
     */
    static final class BehindS extends Behind {
        BehindS(Node cond, int rmax, int rmin) {
            super(cond, rmax, rmin);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
	    int rmaxChars = countChars(seq, i, -rmax);
	    int rminChars = countChars(seq, i, -rmin);
            int savedFrom = matcher.from;
            int startIndex = (!matcher.transparentBounds) ?
                             matcher.from : 0;
            boolean conditionMatched = false;
            int from = Math.max(i - rmaxChars, startIndex);
            for (int j = i - rminChars; 
                 j >= from; 
                 j -= j>from ? countChars(seq, j, -1) : 1) {
                // Relax transparent region boundaries for lookbehind
                if (matcher.transparentBounds)
                    matcher.from = 0;
                try {
                    conditionMatched = 
                        (cond.match(matcher, j, seq) && matcher.last == i);
                } finally { // Reinstate region boundaries
                    matcher.from = savedFrom;
                }
                if (conditionMatched)
                    return next.match(matcher, i, seq);
            }
            return false;
        }
    }

    /**
     * Zero width negative lookbehind.
     */
    static class NotBehind extends Node {
        Node cond;
        int rmax, rmin;
        NotBehind(Node cond, int rmax, int rmin) {
            this.cond = cond;
            this.rmax = rmax;
            this.rmin = rmin;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int savedFrom = matcher.from;
            boolean conditionMatched = false;
            int startIndex = (!matcher.transparentBounds) ?
                             matcher.from : 0;
            int from = Math.max(i - rmax, startIndex);
            for (int j = i - rmin; j >= from; j--) {

                // Relax transparent region boundaries for lookbehind
                if (matcher.transparentBounds)
                    matcher.from = 0;
                try {
                    conditionMatched = 
                        (cond.match(matcher, j, seq) && matcher.last == i);
                } finally { // Reinstate region boundaries
                    matcher.from = savedFrom;
                }
                if (conditionMatched)
                    return false;
            }
            return next.match(matcher, i, seq);
        }
    }

    /**
     * Zero width negative lookbehind, including supplementary
     * characters or unpaired surrogates.
     */
    static final class NotBehindS extends NotBehind {
        NotBehindS(Node cond, int rmax, int rmin) {
            super(cond, rmax, rmin);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
	    int rmaxChars = countChars(seq, i, -rmax);
	    int rminChars = countChars(seq, i, -rmin);
            int savedFrom = matcher.from;
            boolean conditionMatched = false;
            int startIndex = (!matcher.transparentBounds) ?
                             matcher.from : 0;
            int from = Math.max(i - rmaxChars, startIndex);
            for (int j = i - rminChars;
                 j >= from;
                 j -= j>from ? countChars(seq, j, -1) : 1) {
                // Relax transparent region boundaries for lookbehind
                if (matcher.transparentBounds)
                    matcher.from = 0;
                try {
                    conditionMatched = 
                        (cond.match(matcher, j, seq) && matcher.last == i);
                } finally { // Reinstate region boundaries
                    matcher.from = savedFrom;
                }
                if (conditionMatched)
                    return false;
            }
            return next.match(matcher, i, seq);
        }
    }

    /**
     * An object added to the tree when a character class has an additional
     * range added to it.
     */
    static class Add extends Node {
        Node lhs, rhs;
        Add(Node lhs, Node rhs) {
            this.lhs = lhs;
            this.rhs = rhs;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                return ((lhs.match(matcher, i, seq) || 
                         rhs.match(matcher, i, seq)) &&
                        next.match(matcher, matcher.last, seq));
            }
            matcher.hitEnd = true;
            return false;
        }
        boolean study(TreeInfo info) {
            boolean maxV = info.maxValid;
            boolean detm = info.deterministic;

            int minL = info.minLength;
            int maxL = info.maxLength;
            lhs.study(info);

            int minL2 = info.minLength;
            int maxL2 = info.maxLength;

            info.minLength = minL;
            info.maxLength = maxL;
            rhs.study(info);

            info.minLength = Math.min(minL2, info.minLength);
            info.maxLength = Math.max(maxL2, info.maxLength);
            info.maxValid = maxV;
            info.deterministic = detm;

            return next.study(info);
        }
    }

    /**
     * An object added to the tree when a character class has another
     * nested class in it.
     */
    static class Both extends Node {
        Node lhs, rhs;
        Both(Node lhs, Node rhs) {
            this.lhs = lhs;
            this.rhs = rhs;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                return ((lhs.match(matcher, i, seq) && 
                         rhs.match(matcher, i, seq)) &&
                        next.match(matcher, matcher.last, seq));
            }
            matcher.hitEnd = true;
            return false;
        }
        boolean study(TreeInfo info) {
            boolean maxV = info.maxValid;
            boolean detm = info.deterministic;

            int minL = info.minLength;
            int maxL = info.maxLength;
            lhs.study(info);

            int minL2 = info.minLength;
            int maxL2 = info.maxLength;

            info.minLength = minL;
            info.maxLength = maxL;
            rhs.study(info);

            info.minLength = Math.min(minL2, info.minLength);
            info.maxLength = Math.max(maxL2, info.maxLength);
            info.maxValid = maxV;
            info.deterministic = detm;

            return next.study(info);
        }
    }

    /**
     * An object added to the tree when a character class has a range
     * or single subtracted from it.
     */
    static final class Sub extends Add  {
        Sub(Node lhs, Node rhs) {
            super(lhs, rhs);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            if (i < matcher.to) {
                return !rhs.match(matcher, i, seq)
                    && lhs.match(matcher, i, seq)
                    && next.match(matcher, matcher.last, seq);
            }
            matcher.hitEnd = true;
            return false;
        }
        boolean study(TreeInfo info) {
            lhs.study(info);
            return next.study(info);
        }
    }

    /**
     * Handles word boundaries. Includes a field to allow this one class to
     * deal with the different types of word boundaries we can match. The word
     * characters include underscores, letters, and digits. Non spacing marks
     * can are also part of a word if they have a base character, otherwise
     * they are ignored for purposes of finding word boundaries.
     */
    static final class Bound extends Node {
        static int LEFT = 0x1;
        static int RIGHT= 0x2;
        static int BOTH = 0x3;
        static int NONE = 0x4;
        int type;
        Bound(int n) {
            type = n;
        }
        int check(Matcher matcher, int i, CharSequence seq) {
            int ch;
            boolean left = false;
            int startIndex = matcher.from;
            int endIndex = matcher.to;
            if (matcher.transparentBounds) {
                startIndex = 0;
                endIndex = matcher.getTextLength();
            }
            if (i > startIndex) {
                ch = Character.codePointBefore(seq, i);
                left = (ch == '_' || Character.isLetterOrDigit(ch) ||
                    ((Character.getType(ch) == Character.NON_SPACING_MARK)
                     && hasBaseCharacter(matcher, i-1, seq)));
            }
            boolean right = false;
            if (i < endIndex) {
                ch = Character.codePointAt(seq, i);
                right = (ch == '_' || Character.isLetterOrDigit(ch) ||
                    ((Character.getType(ch) == Character.NON_SPACING_MARK)
                     && hasBaseCharacter(matcher, i, seq)));
            } else {
                // Tried to access char past the end
                matcher.hitEnd = true;
                // The addition of another char could wreck a boundary
                matcher.requireEnd = true;
            }
            return ((left ^ right) ? (right ? LEFT : RIGHT) : NONE);
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            return (check(matcher, i, seq) & type) > 0
                && next.match(matcher, i, seq);
        }
    }

    /**
     * Non spacing marks only count as word characters in bounds calculations 
     * if they have a base character.
     */
    private static boolean hasBaseCharacter(Matcher matcher, int i, 
                                            CharSequence seq)
    {
        int start = (!matcher.transparentBounds) ?
            matcher.from : 0;
        for (int x=i; x >= start; x--) {
            int ch = Character.codePointAt(seq, x);
            if (Character.isLetterOrDigit(ch))
                return true;
            if (Character.getType(ch) == Character.NON_SPACING_MARK)
                continue;
            return false;
        }
        return false;
    }

    /**
     * Attempts to match a slice in the input using the Boyer-Moore string
     * matching algorithm. The algorithm is based on the idea that the
     * pattern can be shifted farther ahead in the search text if it is
     * matched right to left.
     * 
     * The pattern is compared to the input one character at a time, from
     * the rightmost character in the pattern to the left. If the characters
     * all match the pattern has been found. If a character does not match,
     * the pattern is shifted right a distance that is the maximum of two
     * functions, the bad character shift and the good suffix shift. This
     * shift moves the attempted match position through the input more
     * quickly than a naive one position at a time check.
     * 

     * The bad character shift is based on the character from the text that
     * did not match. If the character does not appear in the pattern, the
     * pattern can be shifted completely beyond the bad character. If the
     * character does occur in the pattern, the pattern can be shifted to
     * line the pattern up with the next occurrence of that character.
     * 

     * The good suffix shift is based on the idea that some subset on the right
     * side of the pattern has matched. When a bad character is found, the
     * pattern can be shifted right by the pattern length if the subset does
     * not occur again in pattern, or by the amount of distance to the
     * next occurrence of the subset in the pattern.
     *
     * Boyer-Moore search methods adapted from code by Amy Yu.
     */
    static class BnM extends Node {
        int[] buffer;
        int[] lastOcc;
        int[] optoSft;

        /**
         * Pre calculates arrays needed to generate the bad character
         * shift and the good suffix shift. Only the last seven bits
         * are used to see if chars match; This keeps the tables small
         * and covers the heavily used ASCII range, but occasionally
         * results in an aliased match for the bad character shift.
         */
        static Node optimize(Node node) {
            if (!(node instanceof Slice)) {
                return node;
            }

            int[] src = ((Slice) node).buffer;
            int patternLength = src.length;
            // The BM algorithm requires a bit of overhead;
            // If the pattern is short don't use it, since
            // a shift larger than the pattern length cannot
            // be used anyway.
            if (patternLength < 4) {
                return node;
            }
            int i, j, k;
            int[] lastOcc = new int[128];
            int[] optoSft = new int[patternLength];
            // Precalculate part of the bad character shift
            // It is a table for where in the pattern each
            // lower 7-bit value occurs
            for (i = 0; i < patternLength; i++) {
                lastOcc[src[i]&0x7F] = i + 1;
            }
            // Precalculate the good suffix shift
            // i is the shift amount being considered
NEXT:       for (i = patternLength; i > 0; i--) {
                // j is the beginning index of suffix being considered
                for (j = patternLength - 1; j >= i; j--) {
                    // Testing for good suffix
                    if (src[j] == src[j-i]) {
                        // src[j..len] is a good suffix
                        optoSft[j-1] = i;
                    } else {
                        // No match. The array has already been
                        // filled up with correct values before.
                        continue NEXT;
                    }
                }
                // This fills up the remaining of optoSft
                // any suffix can not have larger shift amount
                // then its sub-suffix. Why???
                while (j > 0) {
                    optoSft[--j] = i;
                }
            }
            // Set the guard value because of unicode compression
            optoSft[patternLength-1] = 1;
	    if (node instanceof SliceS)
		return new BnMS(src, lastOcc, optoSft, node.next);
            return new BnM(src, lastOcc, optoSft, node.next);
        }
        BnM(int[] src, int[] lastOcc, int[] optoSft, Node next) {
            this.buffer = src;
            this.lastOcc = lastOcc;
            this.optoSft = optoSft;
            this.next = next;
        }
        boolean match(Matcher matcher, int i, CharSequence seq) {
            int[] src = buffer;
            int patternLength = src.length;
            int last = matcher.to - patternLength;

            // Loop over all possible match positions in text
NEXT:       while (i <= last) {
                // Loop over pattern from right to left
                for (int j = patternLength - 1; j >= 0; j--) {
                    int ch = seq.charAt(i+j);
                    if (ch != src[j]) {
                        // Shift search to the right by the maximum of the
                        // bad character shift and the good suffix shift
                        i += Math.max(j + 1 - lastOcc[ch&0x7F], optoSft[j]);
                        continue NEXT;
                    }
                }
                // Entire pattern matched starting at i
                matcher.first = i;
                boolean ret = next.match(matcher, i + patternLength, seq);
                if (ret) {
                    matcher.first = i;
                    matcher.groups[0] = matcher.first;
                    matcher.groups[1] = matcher.last;
                    return true;
                }
                i++;
            }
            // BnM is only used as the leading node in the unanchored case,
            // and it replaced its Start() which always searches to the end
            // if it doesn't find what it's looking for, so hitEnd is true.
            matcher.hitEnd = true;
            return false;
        }
        boolean study(TreeInfo info) {
            info.minLength += buffer.length;
            info.maxValid = false;
            return next.study(info);
        }
    }

    /**
     * Supplementary support version of BnM(). Unpaired surrogates are
     * also handled by this class.
     */
    static final class BnMS extends BnM {
	int lengthInChars;

	BnMS(int[] src, int[] lastOcc, int[] optoSft, Node next) {
	    super(src, lastOcc, optoSft, next);
	    for (int x = 0; x < buffer.length; x++) {
		lengthInChars += Character.charCount(buffer[x]);
	    }
	}
	boolean match(Matcher matcher, int i, CharSequence seq) {
            int[] src = buffer;
            int patternLength = src.length;
	    int last = matcher.to - lengthInChars;

            // Loop over all possible match positions in text
NEXT:       while (i <= last) {
                // Loop over pattern from right to left
		int ch;    
                for (int j = countChars(seq, i, patternLength), x = patternLength - 1;
		     j > 0; j -= Character.charCount(ch), x--) {
		    ch = Character.codePointBefore(seq, i+j);
                    if (ch != src[x]) {
                        // Shift search to the right by the maximum of the
                        // bad character shift and the good suffix shift
                        int n = Math.max(x + 1 - lastOcc[ch&0x7F], optoSft[x]);
			i += countChars(seq, i, n);
                        continue NEXT;
                    }
                }
                // Entire pattern matched starting at i
                matcher.first = i;
                boolean ret = next.match(matcher, i + lengthInChars, seq);
                if (ret) {
                    matcher.first = i;
                    matcher.groups[0] = matcher.first;
                    matcher.groups[1] = matcher.last;
                    return true;
                }
		i += countChars(seq, i, 1);
            }
            matcher.hitEnd = true;
            return false;
        }
    }

     /**
      * Node class for matching characters in a Unicode block
      */
     static final class UBlock extends Node {
         Character.UnicodeBlock block;
         boolean complementMe = false;

         UBlock() {
         }

         UBlock(Character.UnicodeBlock block, boolean not) {
             this.block = block;
             this.complementMe = not;
         }

         Node dup(boolean not) {
             if (not)
                 return new UBlock(block, !complementMe);
             else
                 return new UBlock(block, complementMe);
         }

         boolean match(Matcher matcher, int i, CharSequence seq) {
             if (complementMe)
                 return notMatch(matcher, i, seq);
             if (i < matcher.to) {
                 int ch = Character.codePointAt(seq, i);
                 return (block == Character.UnicodeBlock.of(ch) && 
                         (next.match(matcher, i+Character.charCount(ch), seq)));
             }
             matcher.hitEnd = true;
             return false;
         }

         boolean notMatch(Matcher matcher, int i, CharSequence seq) {
             if (i < matcher.to) {
                 int ch = Character.codePointAt(seq, i);
                 return (block != Character.UnicodeBlock.of(ch) && 
                         (next.match(matcher, i+Character.charCount(ch), seq)));
             }
             matcher.hitEnd = true;
             return false;
         }

         boolean study(TreeInfo info) {
             info.minLength++;
             info.maxLength++;
             return next.study(info);
         }

     }
 
 

///////////////////////////////////////////////////////////////////////////////
///////////////////////////////////////////////////////////////////////////////

    /**
     *  This must be the very first initializer.
     */
    static Node accept = new Node();

    static Node lastAccept = new LastNode();

    static class categoryNames {

        static HashMap cMap = new HashMap();

        static {
            cMap.put("Cn", new Category(1<<0));        // UNASSIGNED
            cMap.put("Lu", new Category(1<<1));        // UPPERCASE_LETTER
            cMap.put("Ll", new Category(1<<2));        // LOWERCASE_LETTER
            cMap.put("Lt", new Category(1<<3));        // TITLECASE_LETTER
            cMap.put("Lm", new Category(1<<4));        // MODIFIER_LETTER
            cMap.put("Lo", new Category(1<<5));        // OTHER_LETTER
            cMap.put("Mn", new Category(1<<6));        // NON_SPACING_MARK
            cMap.put("Me", new Category(1<<7));        // ENCLOSING_MARK
            cMap.put("Mc", new Category(1<<8));        // COMBINING_SPACING_MARK
            cMap.put("Nd", new Category(1<<9));        // DECIMAL_DIGIT_NUMBER
            cMap.put("Nl", new Category(1<<10));       // LETTER_NUMBER
            cMap.put("No", new Category(1<<11));       // OTHER_NUMBER
            cMap.put("Zs", new Category(1<<12));       // SPACE_SEPARATOR
            cMap.put("Zl", new Category(1<<13));       // LINE_SEPARATOR
            cMap.put("Zp", new Category(1<<14));       // PARAGRAPH_SEPARATOR
            cMap.put("Cc", new Category(1<<15));       // CNTRL
            cMap.put("Cf", new Category(1<<16));       // FORMAT
            cMap.put("Co", new Category(1<<18));       // PRIVATE USE
            cMap.put("Cs", new Category(1<<19));       // SURROGATE
            cMap.put("Pd", new Category(1<<20));       // DASH_PUNCTUATION
            cMap.put("Ps", new Category(1<<21));       // START_PUNCTUATION
            cMap.put("Pe", new Category(1<<22));       // END_PUNCTUATION
            cMap.put("Pc", new Category(1<<23));       // CONNECTOR_PUNCTUATION
            cMap.put("Po", new Category(1<<24));       // OTHER_PUNCTUATION
            cMap.put("Sm", new Category(1<<25));       // MATH_SYMBOL
            cMap.put("Sc", new Category(1<<26));       // CURRENCY_SYMBOL
            cMap.put("Sk", new Category(1<<27));       // MODIFIER_SYMBOL
            cMap.put("So", new Category(1<<28));       // OTHER_SYMBOL
            cMap.put("L", new Category(0x0000003E));   // LETTER
            cMap.put("M", new Category(0x000001C0));   // MARK
            cMap.put("N", new Category(0x00000E00));   // NUMBER
            cMap.put("Z", new Category(0x00007000));   // SEPARATOR
            cMap.put("C", new Category(0x000D8000));   // CONTROL
            cMap.put("P", new Category(0x01F00000));   // PUNCTUATION
            cMap.put("S", new Category(0x1E000000));   // SYMBOL
            cMap.put("LD", new Category(0x0000023E));   // LETTER_OR_DIGIT
            cMap.put("L1", new Range(0x000000FF));      // Latin-1
            cMap.put("all", new All());                  // ALL
            cMap.put("ASCII", new Range(0x0000007F));    // ASCII
            cMap.put("Alnum", new Ctype(ASCII.ALNUM));    // Alphanumeric characters
            cMap.put("Alpha", new Ctype(ASCII.ALPHA));    // Alphabetic characters
            cMap.put("Blank", new Ctype(ASCII.BLANK));    // Space and tab characters
            cMap.put("Cntrl", new Ctype(ASCII.CNTRL));    // Control characters
            cMap.put("Digit", new Range(('0'<<16)|'9'));  // Numeric characters
            cMap.put("Graph", new Ctype(ASCII.GRAPH));    // printable and visible
            cMap.put("Lower", new Range(('a'<<16)|'z'));  // Lower-case alphabetic
            cMap.put("Print", new Range(0x0020007E));     // Printable characters
            cMap.put("Punct", new Ctype(ASCII.PUNCT));    // Punctuation characters
            cMap.put("Space", new Ctype(ASCII.SPACE));    // Space characters
            cMap.put("Upper", new Range(('A'<<16)|'Z'));  // Upper-case alphabetic
            cMap.put("XDigit", new Ctype(ASCII.XDIGIT));  // hexadecimal digits
            cMap.put("javaLowerCase", new JavaLowerCase());
            cMap.put("javaUpperCase", new JavaUpperCase());
            cMap.put("javaTitleCase", new JavaTitleCase());
            cMap.put("javaDigit", new JavaDigit());
            cMap.put("javaDefined", new JavaDefined());
            cMap.put("javaLetter", new JavaLetter());
            cMap.put("javaLetterOrDigit", new JavaLetterOrDigit());
            cMap.put("javaJavaIdentifierStart", new JavaJavaIdentifierStart());
            cMap.put("javaJavaIdentifierPart", new JavaJavaIdentifierPart());
            cMap.put("javaUnicodeIdentifierStart", new JavaUnicodeIdentifierStart());
            cMap.put("javaUnicodeIdentifierPart", new JavaUnicodeIdentifierPart());
            cMap.put("javaIdentifierIgnorable", new JavaIdentifierIgnorable());
            cMap.put("javaSpaceChar", new JavaSpaceChar());
            cMap.put("javaWhitespace", new JavaWhitespace());
            cMap.put("javaISOControl", new JavaISOControl());
            cMap.put("javaMirrored", new JavaMirrored());
        }
    }
}

Construct	Matches

Characters
x	The character x
`\\`	The backslash character
`\0`n	The character with octal value `0`n * (0 `<=` n `<=` 7)
`\0`nn	The character with octal value `0`nn * (0 `<=` n `<=` 7)
`\0`mnn	The character with octal value `0`mnn * (0 `<=` m `<=` 3, * 0 `<=` n `<=` 7)
`\x`hh	The character with hexadecimal value `0x`hh
`\u`hhhh	The character with hexadecimal value `0x`hhhh
`\t`	The tab character (`'\u0009'`)
`\n`	The newline (line feed) character (`'\u000A'`)
`\r`	The carriage-return character (`'\u000D'`)
`\f`	The form-feed character (`'\u000C'`)
`\a`	The alert (bell) character (`'\u0007'`)
`\e`	The escape character (`'\u001B'`)
`\c`x	The control character corresponding to x

Character classes
`[abc]`	`a`, `b`, or `c` (simple class)
`[^abc]`	Any character except `a`, `b`, or `c` (negation)
`[a-zA-Z]`	`a` through `z` * or `A` through `Z`, inclusive (range)
`[a-d[m-p]]`	`a` through `d`, * or `m` through `p`: `[a-dm-p]` (union)
`[a-z&&[def]]`	`d`, `e`, or `f` (intersection)
`[a-z&&[^bc]]`	`a` through `z`, * except for `b` and `c`: `[ad-z]` (subtraction)
`[a-z&&[^m-p]]`	`a` through `z`, * and not `m` through `p`: `[a-lq-z]`(subtraction)

Predefined character classes
`.`	Any character (may or may not match line terminators)
`\d`	A digit: `[0-9]`
`\D`	A non-digit: `[^0-9]`
`\s`	A whitespace character: `[ \t\n\x0B\f\r]`
`\S`	A non-whitespace character: `[^\s]`
`\w`	A word character: `[a-zA-Z_0-9]`
`\W`	A non-word character: `[^\w]`

POSIX character classes (US-ASCII only)
`\p{Lower}`	A lower-case alphabetic character: `[a-z]`
`\p{Upper}`	An upper-case alphabetic character:`[A-Z]`
`\p{ASCII}`	All ASCII:`[\x00-\x7F]`
`\p{Alpha}`	An alphabetic character:`[\p{Lower}\p{Upper}]`
`\p{Digit}`	A decimal digit: `[0-9]`
`\p{Alnum}`	An alphanumeric character:`[\p{Alpha}\p{Digit}]`
`\p{Punct}`	Punctuation: One of !"#$%&'()*+,-./:;<=>?@[\]^_`{\|}~
`\p{Graph}`	A visible character: `[\p{Alnum}\p{Punct}]`
`\p{Print}`	A printable character: `[\p{Graph}\x20]`
`\p{Blank}`	A space or a tab: `[ \t]`
`\p{Cntrl}`	A control character: `[\x00-\x1F\x7F]`
`\p{XDigit}`	A hexadecimal digit: `[0-9a-fA-F]`
`\p{Space}`	A whitespace character: `[ \t\n\x0B\f\r]`

java.lang.Character classes (simple java character type)
`\p{javaLowerCase}`	Equivalent to java.lang.Character.isLowerCase()
`\p{javaUpperCase}`	Equivalent to java.lang.Character.isUpperCase()
`\p{javaWhitespace}`	Equivalent to java.lang.Character.isWhitespace()
`\p{javaMirrored}`	Equivalent to java.lang.Character.isMirrored()

Classes for Unicode blocks and categories
`\p{InGreek}`	A character in the Greek block (simple block)
`\p{Lu}`	An uppercase letter (simple category)
`\p{Sc}`	A currency symbol
`\P{InGreek}`	Any character except one in the Greek block (negation)
`[\p{L}&&[^\p{Lu}]]`	Any letter except an uppercase letter (subtraction)

Boundary matchers
`^`	The beginning of a line
`$`	The end of a line
`\b`	A word boundary
`\B`	A non-word boundary
`\A`	The beginning of the input
`\G`	The end of the previous match
`\Z`	The end of the input but for the final * terminator, if any
`\z`	The end of the input

Greedy quantifiers
X`?`	X, once or not at all
X`*`	X, zero or more times
X`+`	X, one or more times
X`{`n`}`	X, exactly n times
X`{`n`,}`	X, at least n times
X`{`n`,`m`}`	X, at least n but not more than m times

Reluctant quantifiers
X`??`	X, once or not at all
X`*?`	X, zero or more times
X`+?`	X, one or more times
X`{`n`}?`	X, exactly n times
X`{`n`,}?`	X, at least n times
X`{`n`,`m`}?`	X, at least n but not more than m times

Possessive quantifiers
X`?+`	X, once or not at all
X`*+`	X, zero or more times
X`++`	X, one or more times
X`{`n`}+`	X, exactly n times
X`{`n`,}+`	X, at least n times
X`{`n`,`m`}+`	X, at least n but not more than m times

Logical operators
XY	X followed by Y
X`\|`Y	Either X or Y
`(`X`)`	X, as a capturing group

Back references
`\`n	Whatever the n^th * capturing group matched

Quotation
`\`	Nothing, but quotes the following character
`\Q`	Nothing, but quotes all characters until `\E`
`\E`	Nothing, but ends quoting started by `\Q`

Special constructs (non-capturing)
`(?:`X`)`	X, as a non-capturing group
`(?idmsux-idmsux)`	Nothing, but turns match flags on - off
`(?idmsux-idmsux:`X`)`	X, as a non-capturing group with the * given flags on - off
`(?=`X`)`	X, via zero-width positive lookahead
`(?!`X`)`	X, via zero-width negative lookahead
`(?<=`X`)`	X, via zero-width positive lookbehind
`(?<!`X`)`	X, via zero-width negative lookbehind
`(?>`X`)`	X, as an independent, non-capturing group

1	`((A)(B(C)))`
2	`(A)`
3	`(B(C))`
4	`(C)`

Regex	Limit	Result
:	2	`{ "boo", "and:foo" }`
:	5	`{ "boo", "and", "foo" }`
:	-2	`{ "boo", "and", "foo" }`
o	5	`{ "b", "", ":and:f", "", "" }`
o	-2	`{ "b", "", ":and:f", "", "" }`
o	0	`{ "b", "", ":and:f" }`

Regex	Result
:	`{ "boo", "and", "foo" }`
o	`{ "b", "", ":and:f" }`

1	Literal escape	`\x`
2	Grouping	`[...]`
3	Range	`a-z`
4	Union	`[a-e][i-u]`
5	Intersection	`[a-z&&[aeiou]]`