/*
 * Copyright (c) 2012, 2017, Oracle and/or its affiliates. All rights reserved.
 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
 *
 * This code is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License version 2 only, as
 * published by the Free Software Foundation.  Oracle designates this
 * particular file as subject to the "Classpath" exception as provided
 * by Oracle in the LICENSE file that accompanied this code.
 *
 * This code is distributed in the hope that it will be useful, but WITHOUT
 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
 * version 2 for more details (a copy is included in the LICENSE file that
 * accompanied this code).
 *
 * You should have received a copy of the GNU General Public License version
 * 2 along with this work; if not, write to the Free Software Foundation,
 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
 *
 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
 * or visit www.oracle.com if you need additional information or have any
 * questions.
 */

/*
 * This file is available under and governed by the GNU General Public
 * License version 2 only, as published by the Free Software Foundation.
 * However, the following notice accompanied the original version of this
 * file:
 *
 * Copyright (c) 2008-2012, Stephen Colebourne & Michael Nascimento Santos
 *
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 *  * Redistributions of source code must retain the above copyright notice,
 *    this list of conditions and the following disclaimer.
 *
 *  * Redistributions in binary form must reproduce the above copyright notice,
 *    this list of conditions and the following disclaimer in the documentation
 *    and/or other materials provided with the distribution.
 *
 *  * Neither the name of JSR-310 nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
 * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
 * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */
package java.time.format;

import static java.time.temporal.ChronoField.DAY_OF_MONTH;
import static java.time.temporal.ChronoField.DAY_OF_WEEK;
import static java.time.temporal.ChronoField.DAY_OF_YEAR;
import static java.time.temporal.ChronoField.HOUR_OF_DAY;
import static java.time.temporal.ChronoField.MINUTE_OF_HOUR;
import static java.time.temporal.ChronoField.MONTH_OF_YEAR;
import static java.time.temporal.ChronoField.NANO_OF_SECOND;
import static java.time.temporal.ChronoField.SECOND_OF_MINUTE;
import static java.time.temporal.ChronoField.YEAR;

import java.io.IOException;
import java.text.FieldPosition;
import java.text.Format;
import java.text.ParseException;
import java.text.ParsePosition;
import java.time.DateTimeException;
import java.time.Period;
import java.time.ZoneId;
import java.time.ZoneOffset;
import java.time.chrono.ChronoLocalDateTime;
import java.time.chrono.Chronology;
import java.time.chrono.IsoChronology;
import java.time.format.DateTimeFormatterBuilder.CompositePrinterParser;
import java.time.temporal.ChronoField;
import java.time.temporal.IsoFields;
import java.time.temporal.TemporalAccessor;
import java.time.temporal.TemporalField;
import java.time.temporal.TemporalQuery;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashMap;
import java.util.HashSet;
import java.util.Locale;
import java.util.Map;
import java.util.Objects;
import java.util.Set;

/**
 * Formatter for printing and parsing date-time objects.
 * <p>
 * This class provides the main application entry point for printing and parsing
 * and provides common implementations of {@code DateTimeFormatter}:
 * <ul>
 * <li>Using predefined constants, such as {@link #ISO_LOCAL_DATE}</li>
 * <li>Using pattern letters, such as {@code uuuu-MMM-dd}</li>
 * <li>Using localized styles, such as {@code long} or {@code medium}</li>
 * </ul>
 * <p>
 * More complex formatters are provided by
 * {@link DateTimeFormatterBuilder DateTimeFormatterBuilder}.
 *
 * <p>
 * The main date-time classes provide two methods - one for formatting,
 * {@code format(DateTimeFormatter formatter)}, and one for parsing,
 * {@code parse(CharSequence text, DateTimeFormatter formatter)}.
 * <p>For example:
 * <blockquote><pre>
 *  LocalDate date = LocalDate.now();
 *  String text = date.format(formatter);
 *  LocalDate parsedDate = LocalDate.parse(text, formatter);
 * </pre></blockquote>
 * <p>
 * In addition to the format, formatters can be created with desired Locale,
 * Chronology, ZoneId, and DecimalStyle.
 * <p>
 * The {@link #withLocale withLocale} method returns a new formatter that
 * overrides the locale. The locale affects some aspects of formatting and
 * parsing. For example, the {@link #ofLocalizedDate ofLocalizedDate} provides a
 * formatter that uses the locale specific date format.
 * <p>
 * The {@link #withChronology withChronology} method returns a new formatter
 * that overrides the chronology. If overridden, the date-time value is
 * converted to the chronology before formatting. During parsing the date-time
 * value is converted to the chronology before it is returned.
 * <p>
 * The {@link #withZone withZone} method returns a new formatter that overrides
 * the zone. If overridden, the date-time value is converted to a ZonedDateTime
 * with the requested ZoneId before formatting. During parsing the ZoneId is
 * applied before the value is returned.
 * <p>
 * The {@link #withDecimalStyle withDecimalStyle} method returns a new formatter that
 * overrides the {@link DecimalStyle}. The DecimalStyle symbols are used for
 * formatting and parsing.
 * <p>
 * Some applications may need to use the older {@link Format java.text.Format}
 * class for formatting. The {@link #toFormat()} method returns an
 * implementation of {@code java.text.Format}.
 *
 * <h3 id="predefined">Predefined Formatters</h3>
 * <table class="striped">
 * <caption>Predefined Formatters</caption>
 * <thead>
 * <tr>
 * <th style="text-align:left">Formatter</th>
 * <th style="text-align:left">Description</th>
 * <th style="text-align:left">Example</th>
 * </tr>
 * </thead>
 * <tbody>
 * <tr>
 * <td>{@link #ofLocalizedDate ofLocalizedDate(dateStyle)} </td>
 * <td> Formatter with date style from the locale </td>
 * <td> '2011-12-03'</td>
 * </tr>
 * <tr>
 * <td> {@link #ofLocalizedTime ofLocalizedTime(timeStyle)} </td>
 * <td> Formatter with time style from the locale </td>
 * <td> '10:15:30'</td>
 * </tr>
 * <tr>
 * <td> {@link #ofLocalizedDateTime ofLocalizedDateTime(dateTimeStyle)} </td>
 * <td> Formatter with a style for date and time from the locale</td>
 * <td> '3 Jun 2008 11:05:30'</td>
 * </tr>
 * <tr>
 * <td> {@link #ofLocalizedDateTime ofLocalizedDateTime(dateStyle,timeStyle)}
 * </td>
 * <td> Formatter with date and time styles from the locale </td>
 * <td> '3 Jun 2008 11:05'</td>
 * </tr>
 * <tr>
 * <td> {@link #BASIC_ISO_DATE}</td>
 * <td>Basic ISO date </td> <td>'20111203'</td>
 * </tr>
 * <tr>
 * <td> {@link #ISO_LOCAL_DATE}</td>
 * <td> ISO Local Date </td>
 * <td>'2011-12-03'</td>
 * </tr>
 * <tr>
 * <td> {@link #ISO_OFFSET_DATE}</td>
 * <td> ISO Date with offset </td>
 * <td>'2011-12-03+01:00'</td>
 * </tr>
 * <tr>
 * <td> {@link #ISO_DATE}</td>
 * <td> ISO Date with or without offset </td>
 * <td> '2011-12-03+01:00'; '2011-12-03'</td>
 * </tr>
 * <tr>
 * <td> {@link #ISO_LOCAL_TIME}</td>
 * <td> Time without offset </td>
 * <td>'10:15:30'</td>
 * </tr>
 * <tr>
 * <td> {@link #ISO_OFFSET_TIME}</td>
 * <td> Time with offset </td>
 * <td>'10:15:30+01:00'</td>
 * </tr>
 * <tr>
 * <td> {@link #ISO_TIME}</td>
 * <td> Time with or without offset </td>
 * <td>'10:15:30+01:00'; '10:15:30'</td>
 * </tr>
 * <tr>
 * <td> {@link #ISO_LOCAL_DATE_TIME}</td>
 * <td> ISO Local Date and Time </td>
 * <td>'2011-12-03T10:15:30'</td>
 * </tr>
 * <tr>
 * <td> {@link #ISO_OFFSET_DATE_TIME}</td>
 * <td> Date Time with Offset
 * </td><td>2011-12-03T10:15:30+01:00'</td>
 * </tr>
 * <tr>
 * <td> {@link #ISO_ZONED_DATE_TIME}</td>
 * <td> Zoned Date Time </td>
 * <td>'2011-12-03T10:15:30+01:00[Europe/Paris]'</td>
 * </tr>
 * <tr>
 * <td> {@link #ISO_DATE_TIME}</td>
 * <td> Date and time with ZoneId </td>
 * <td>'2011-12-03T10:15:30+01:00[Europe/Paris]'</td>
 * </tr>
 * <tr>
 * <td> {@link #ISO_ORDINAL_DATE}</td>
 * <td> Year and day of year </td>
 * <td>'2012-337'</td>
 * </tr>
 * <tr>
 * <td> {@link #ISO_WEEK_DATE}</td>
 * <td> Year and Week </td>
 * <td>2012-W48-6'</td></tr>
 * <tr>
 * <td> {@link #ISO_INSTANT}</td>
 * <td> Date and Time of an Instant </td>
 * <td>'2011-12-03T10:15:30Z' </td>
 * </tr>
 * <tr>
 * <td> {@link #RFC_1123_DATE_TIME}</td>
 * <td> RFC 1123 / RFC 822 </td>
 * <td>'Tue, 3 Jun 2008 11:05:30 GMT'</td>
 * </tr>
 * </tbody>
 * </table>
 *
 * <h3 id="patterns">Patterns for Formatting and Parsing</h3>
 * Patterns are based on a simple sequence of letters and symbols.
 * A pattern is used to create a Formatter using the
 * {@link #ofPattern(String)} and {@link #ofPattern(String, Locale)} methods.
 * For example,
 * {@code "d MMM uuuu"} will format 2011-12-03 as '3&nbsp;Dec&nbsp;2011'.
 * A formatter created from a pattern can be used as many times as necessary,
 * it is immutable and is thread-safe.
 * <p>
 * For example:
 * <blockquote><pre>
 *  LocalDate date = LocalDate.now();
 *  DateTimeFormatter formatter = DateTimeFormatter.ofPattern("yyyy MM dd");
 *  String text = date.format(formatter);
 *  LocalDate parsedDate = LocalDate.parse(text, formatter);
 * </pre></blockquote>
 * <p>
 * All letters 'A' to 'Z' and 'a' to 'z' are reserved as pattern letters. The
 * following pattern letters are defined:
 * <pre>
 *  Symbol  Meaning                     Presentation      Examples
 *  ------  -------                     ------------      -------
 *   G       era                         text              AD; Anno Domini; A
 *   u       year                        year              2004; 04
 *   y       year-of-era                 year              2004; 04
 *   D       day-of-year                 number            189
 *   M/L     month-of-year               number/text       7; 07; Jul; July; J
 *   d       day-of-month                number            10
 *   g       modified-julian-day         number            2451334
 *
 *   Q/q     quarter-of-year             number/text       3; 03; Q3; 3rd quarter
 *   Y       week-based-year             year              1996; 96
 *   w       week-of-week-based-year     number            27
 *   W       week-of-month               number            4
 *   E       day-of-week                 text              Tue; Tuesday; T
 *   e/c     localized day-of-week       number/text       2; 02; Tue; Tuesday; T
 *   F       day-of-week-in-month        number            3
 *
 *   a       am-pm-of-day                text              PM
 *   h       clock-hour-of-am-pm (1-12)  number            12
 *   K       hour-of-am-pm (0-11)        number            0
 *   k       clock-hour-of-day (1-24)    number            24
 *
 *   H       hour-of-day (0-23)          number            0
 *   m       minute-of-hour              number            30
 *   s       second-of-minute            number            55
 *   S       fraction-of-second          fraction          978
 *   A       milli-of-day                number            1234
 *   n       nano-of-second              number            987654321
 *   N       nano-of-day                 number            1234000000
 *
 *   V       time-zone ID                zone-id           America/Los_Angeles; Z; -08:30
 *   v       generic time-zone name      zone-name         Pacific Time; PT
 *   z       time-zone name              zone-name         Pacific Standard Time; PST
 *   O       localized zone-offset       offset-O          GMT+8; GMT+08:00; UTC-08:00
 *   X       zone-offset 'Z' for zero    offset-X          Z; -08; -0830; -08:30; -083015; -08:30:15
 *   x       zone-offset                 offset-x          +0000; -08; -0830; -08:30; -083015; -08:30:15
 *   Z       zone-offset                 offset-Z          +0000; -0800; -08:00
 *
 *   p       pad next                    pad modifier      1
 *
 *   '       escape for text             delimiter
 *   ''      single quote                literal           '
 *   [       optional section start
 *   ]       optional section end
 *   #       reserved for future use
 *   {       reserved for future use
 *   }       reserved for future use
 * </pre>
 * <p>
 * The count of pattern letters determines the format.
 * <p>
 * <b>Text</b>: The text style is determined based on the number of pattern
 * letters used. Less than 4 pattern letters will use the
 * {@link TextStyle#SHORT short form}. Exactly 4 pattern letters will use the
 * {@link TextStyle#FULL full form}. Exactly 5 pattern letters will use the
 * {@link TextStyle#NARROW narrow form}.
 * Pattern letters 'L', 'c', and 'q' specify the stand-alone form of the text styles.
 * <p>
 * <b>Number</b>: If the count of letters is one, then the value is output using
 * the minimum number of digits and without padding. Otherwise, the count of digits
 * is used as the width of the output field, with the value zero-padded as necessary.
 * The following pattern letters have constraints on the count of letters.
 * Only one letter of 'c' and 'F' can be specified.
 * Up to two letters of 'd', 'H', 'h', 'K', 'k', 'm', and 's' can be specified.
 * Up to three letters of 'D' can be specified.
 * <p>
 * <b>Number/Text</b>: If the count of pattern letters is 3 or greater, use the
 * Text rules above. Otherwise use the Number rules above.
 * <p>
 * <b>Fraction</b>: Outputs the nano-of-second field as a fraction-of-second.
 * The nano-of-second value has nine digits, thus the count of pattern letters
 * is from 1 to 9. If it is less than 9, then the nano-of-second value is
 * truncated, with only the most significant digits being output.
 * <p>
 * <b>Year</b>: The count of letters determines the minimum field width below
 * which padding is used. If the count of letters is two, then a
 * {@link DateTimeFormatterBuilder#appendValueReduced reduced} two digit form is
 * used. For printing, this outputs the rightmost two digits. For parsing, this
 * will parse using the base value of 2000, resulting in a year within the range
 * 2000 to 2099 inclusive. If the count of letters is less than four (but not
 * two), then the sign is only output for negative years as per
 * {@link SignStyle#NORMAL}. Otherwise, the sign is output if the pad width is
 * exceeded, as per {@link SignStyle#EXCEEDS_PAD}.
 * <p>
 * <b>ZoneId</b>: This outputs the time-zone ID, such as 'Europe/Paris'. If the
 * count of letters is two, then the time-zone ID is output. Any other count of
 * letters throws {@code IllegalArgumentException}.
 * <p>
 * <b>Zone names</b>: This outputs the display name of the time-zone ID. If the
 * pattern letter is 'z' the output is the daylight savings aware zone name.
 * If there is insufficient information to determine whether DST applies,
 * the name ignoring daylight savings time will be used.
 * If the count of letters is one, two or three, then the short name is output.
 * If the count of letters is four, then the full name is output.
 * Five or more letters throws {@code IllegalArgumentException}.
 * <p>
 * If the pattern letter is 'v' the output provides the zone name ignoring
 * daylight savings time. If the count of letters is one, then the short name is output.
 * If the count of letters is four, then the full name is output.
 * Two, three and five or more letters throw {@code IllegalArgumentException}.
 * <p>
 * <b>Offset X and x</b>: This formats the offset based on the number of pattern
 * letters. One letter outputs just the hour, such as '+01', unless the minute
 * is non-zero in which case the minute is also output, such as '+0130'. Two
 * letters outputs the hour and minute, without a colon, such as '+0130'. Three
 * letters outputs the hour and minute, with a colon, such as '+01:30'. Four
 * letters outputs the hour and minute and optional second, without a colon,
 * such as '+013015'. Five letters outputs the hour and minute and optional
 * second, with a colon, such as '+01:30:15'. Six or more letters throws
 * {@code IllegalArgumentException}. Pattern letter 'X' (upper case) will output
 * 'Z' when the offset to be output would be zero, whereas pattern letter 'x'
 * (lower case) will output '+00', '+0000', or '+00:00'.
 * <p>
 * <b>Offset O</b>: This formats the localized offset based on the number of
 * pattern letters. One letter outputs the {@linkplain TextStyle#SHORT short}
 * form of the localized offset, which is localized offset text, such as 'GMT',
 * with hour without leading zero, optional 2-digit minute and second if
 * non-zero, and colon, for example 'GMT+8'. Four letters outputs the
 * {@linkplain TextStyle#FULL full} form, which is localized offset text,
 * such as 'GMT, with 2-digit hour and minute field, optional second field
 * if non-zero, and colon, for example 'GMT+08:00'. Any other count of letters
 * throws {@code IllegalArgumentException}.
 * <p>
 * <b>Offset Z</b>: This formats the offset based on the number of pattern
 * letters. One, two or three letters outputs the hour and minute, without a
 * colon, such as '+0130'. The output will be '+0000' when the offset is zero.
 * Four letters outputs the {@linkplain TextStyle#FULL full} form of localized
 * offset, equivalent to four letters of Offset-O. The output will be the
 * corresponding localized offset text if the offset is zero. Five
 * letters outputs the hour, minute, with optional second if non-zero, with
 * colon. It outputs 'Z' if the offset is zero.
 * Six or more letters throws {@code IllegalArgumentException}.
 * <p>
 * <b>Optional section</b>: The optional section markers work exactly like
 * calling {@link DateTimeFormatterBuilder#optionalStart()} and
 * {@link DateTimeFormatterBuilder#optionalEnd()}.
 * <p>
 * <b>Pad modifier</b>: Modifies the pattern that immediately follows to be
 * padded with spaces. The pad width is determined by the number of pattern
 * letters. This is the same as calling
 * {@link DateTimeFormatterBuilder#padNext(int)}.
 * <p>
 * For example, 'ppH' outputs the hour-of-day padded on the left with spaces to
 * a width of 2.
 * <p>
 * Any unrecognized letter is an error. Any non-letter character, other than
 * '[', ']', '{', '}', '#' and the single quote will be output directly.
 * Despite this, it is recommended to use single quotes around all characters
 * that you want to output directly to ensure that future changes do not break
 * your application.
 *
 * <h3 id="resolving">Resolving</h3>
 * Parsing is implemented as a two-phase operation.
 * First, the text is parsed using the layout defined by the formatter, producing
 * a {@code Map} of field to value, a {@code ZoneId} and a {@code Chronology}.
 * Second, the parsed data is <em>resolved</em>, by validating, combining and
 * simplifying the various fields into more useful ones.
 * <p>
 * Five parsing methods are supplied by this class.
 * Four of these perform both the parse and resolve phases.
 * The fifth method, {@link #parseUnresolved(CharSequence, ParsePosition)},
 * only performs the first phase, leaving the result unresolved.
 * As such, it is essentially a low-level operation.
 * <p>
 * The resolve phase is controlled by two parameters, set on this class.
 * <p>
 * The {@link ResolverStyle} is an enum that offers three different approaches,
 * strict, smart and lenient. The smart option is the default.
 * It can be set using {@link #withResolverStyle(ResolverStyle)}.
 * <p>
 * The {@link #withResolverFields(TemporalField...)} parameter allows the
 * set of fields that will be resolved to be filtered before resolving starts.
 * For example, if the formatter has parsed a year, month, day-of-month
 * and day-of-year, then there are two approaches to resolve a date:
 * (year + month + day-of-month) and (year + day-of-year).
 * The resolver fields allows one of the two approaches to be selected.
 * If no resolver fields are set then both approaches must result in the same date.
 * <p>
 * Resolving separate fields to form a complete date and time is a complex
 * process with behaviour distributed across a number of classes.
 * It follows these steps:
 * <ol>
 * <li>The chronology is determined.
 * The chronology of the result is either the chronology that was parsed,
 * or if no chronology was parsed, it is the chronology set on this class,
 * or if that is null, it is {@code IsoChronology}.
 * <li>The {@code ChronoField} date fields are resolved.
 * This is achieved using {@link Chronology#resolveDate(Map, ResolverStyle)}.
 * Documentation about field resolution is located in the implementation
 * of {@code Chronology}.
 * <li>The {@code ChronoField} time fields are resolved.
 * This is documented on {@link ChronoField} and is the same for all chronologies.
 * <li>Any fields that are not {@code ChronoField} are processed.
 * This is achieved using {@link TemporalField#resolve(Map, TemporalAccessor, ResolverStyle)}.
 * Documentation about field resolution is located in the implementation
 * of {@code TemporalField}.
 * <li>The {@code ChronoField} date and time fields are re-resolved.
 * This allows fields in step four to produce {@code ChronoField} values
 * and have them be processed into dates and times.
 * <li>A {@code LocalTime} is formed if there is at least an hour-of-day available.
 * This involves providing default values for minute, second and fraction of second.
 * <li>Any remaining unresolved fields are cross-checked against any
 * date and/or time that was resolved. Thus, an earlier stage would resolve
 * (year + month + day-of-month) to a date, and this stage would check that
 * day-of-week was valid for the date.
 * <li>If an {@linkplain #parsedExcessDays() excess number of days}
 * was parsed then it is added to the date if a date is available.
 * <li> If a second-based field is present, but {@code LocalTime} was not parsed,
 * then the resolver ensures that milli, micro and nano second values are
 * available to meet the contract of {@link ChronoField}.
 * These will be set to zero if missing.
 * <li>If both date and time were parsed and either an offset or zone is present,
 * the field {@link ChronoField#INSTANT_SECONDS} is created.
 * If an offset was parsed then the offset will be combined with the
 * {@code LocalDateTime} to form the instant, with any zone ignored.
 * If a {@code ZoneId} was parsed without an offset then the zone will be
 * combined with the {@code LocalDateTime} to form the instant using the rules
 * of {@link ChronoLocalDateTime#atZone(ZoneId)}.
 * </ol>
 *
 * @implSpec
 * This class is immutable and thread-safe.
 *
 * @since 1.8
 */
public final class DateTimeFormatter {

    /**
     * The printer and/or parser to use, not null.
     */
    private final CompositePrinterParser printerParser;
    /**
     * The locale to use for formatting, not null.
     */
    private final Locale locale;
    /**
     * The symbols to use for formatting, not null.
     */
    private final DecimalStyle decimalStyle;
    /**
     * The resolver style to use, not null.
     */
    private final ResolverStyle resolverStyle;
    /**
     * The fields to use in resolving, null for all fields.
     */
    private final Set<TemporalField> resolverFields;
    /**
     * The chronology to use for formatting, null for no override.
     */
    private final Chronology chrono;
    /**
     * The zone to use for formatting, null for no override.
     */
    private final ZoneId zone;

    //-----------------------------------------------------------------------
    /**
     * Creates a formatter using the specified pattern.
     * <p>
     * This method will create a formatter based on a simple
     * <a href="#patterns">pattern of letters and symbols</a>
     * as described in the class documentation.
     * For example, {@code d MMM uuuu} will format 2011-12-03 as '3 Dec 2011'.
     * <p>
     * The formatter will use the {@link Locale#getDefault(Locale.Category) default FORMAT locale}.
     * This can be changed using {@link DateTimeFormatter#withLocale(Locale)} on the returned formatter
     * Alternatively use the {@link #ofPattern(String, Locale)} variant of this method.
     * <p>
     * The returned formatter has no override chronology or zone.
     * It uses {@link ResolverStyle#SMART SMART} resolver style.
     *
     * @param pattern  the pattern to use, not null
     * @return the formatter based on the pattern, not null
     * @throws IllegalArgumentException if the pattern is invalid
     * @see DateTimeFormatterBuilder#appendPattern(String)
     */
    public static DateTimeFormatter ofPattern(String pattern) {
        return new DateTimeFormatterBuilder().appendPattern(pattern).toFormatter();
    }

    /**
     * Creates a formatter using the specified pattern and locale.
     * <p>
     * This method will create a formatter based on a simple
     * <a href="#patterns">pattern of letters and symbols</a>
     * as described in the class documentation.
     * For example, {@code d MMM uuuu} will format 2011-12-03 as '3 Dec 2011'.
     * <p>
     * The formatter will use the specified locale.
     * This can be changed using {@link DateTimeFormatter#withLocale(Locale)} on the returned formatter
     * <p>
     * The returned formatter has no override chronology or zone.
     * It uses {@link ResolverStyle#SMART SMART} resolver style.
     *
     * @param pattern  the pattern to use, not null
     * @param locale  the locale to use, not null
     * @return the formatter based on the pattern, not null
     * @throws IllegalArgumentException if the pattern is invalid
     * @see DateTimeFormatterBuilder#appendPattern(String)
     */
    public static DateTimeFormatter ofPattern(String pattern, Locale locale) {
        return new DateTimeFormatterBuilder().appendPattern(pattern).toFormatter(locale);
    }

    //-----------------------------------------------------------------------
    /**
     * Returns a locale specific date format for the ISO chronology.
     * <p>
     * This returns a formatter that will format or parse a date.
     * The exact format pattern used varies by locale.
     * <p>
     * The locale is determined from the formatter. The formatter returned directly by
     * this method will use the {@link Locale#getDefault(Locale.Category) default FORMAT locale}.
     * The locale can be controlled using {@link DateTimeFormatter#withLocale(Locale) withLocale(Locale)}
     * on the result of this method.
     * <p>
     * Note that the localized pattern is looked up lazily.
     * This {@code DateTimeFormatter} holds the style required and the locale,
     * looking up the pattern required on demand.
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#SMART SMART} resolver style.
     *
     * @param dateStyle  the formatter style to obtain, not null
     * @return the date formatter, not null
     */
    public static DateTimeFormatter ofLocalizedDate(FormatStyle dateStyle) {
        Objects.requireNonNull(dateStyle, "dateStyle");
        return new DateTimeFormatterBuilder().appendLocalized(dateStyle, null)
                .toFormatter(ResolverStyle.SMART, IsoChronology.INSTANCE);
    }

    /**
     * Returns a locale specific time format for the ISO chronology.
     * <p>
     * This returns a formatter that will format or parse a time.
     * The exact format pattern used varies by locale.
     * <p>
     * The locale is determined from the formatter. The formatter returned directly by
     * this method will use the {@link Locale#getDefault(Locale.Category) default FORMAT locale}.
     * The locale can be controlled using {@link DateTimeFormatter#withLocale(Locale) withLocale(Locale)}
     * on the result of this method.
     * <p>
     * Note that the localized pattern is looked up lazily.
     * This {@code DateTimeFormatter} holds the style required and the locale,
     * looking up the pattern required on demand.
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#SMART SMART} resolver style.
     * The {@code FULL} and {@code LONG} styles typically require a time-zone.
     * When formatting using these styles, a {@code ZoneId} must be available,
     * either by using {@code ZonedDateTime} or {@link DateTimeFormatter#withZone}.
     *
     * @param timeStyle  the formatter style to obtain, not null
     * @return the time formatter, not null
     */
    public static DateTimeFormatter ofLocalizedTime(FormatStyle timeStyle) {
        Objects.requireNonNull(timeStyle, "timeStyle");
        return new DateTimeFormatterBuilder().appendLocalized(null, timeStyle)
                .toFormatter(ResolverStyle.SMART, IsoChronology.INSTANCE);
    }

    /**
     * Returns a locale specific date-time formatter for the ISO chronology.
     * <p>
     * This returns a formatter that will format or parse a date-time.
     * The exact format pattern used varies by locale.
     * <p>
     * The locale is determined from the formatter. The formatter returned directly by
     * this method will use the {@link Locale#getDefault(Locale.Category) default FORMAT locale}.
     * The locale can be controlled using {@link DateTimeFormatter#withLocale(Locale) withLocale(Locale)}
     * on the result of this method.
     * <p>
     * Note that the localized pattern is looked up lazily.
     * This {@code DateTimeFormatter} holds the style required and the locale,
     * looking up the pattern required on demand.
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#SMART SMART} resolver style.
     * The {@code FULL} and {@code LONG} styles typically require a time-zone.
     * When formatting using these styles, a {@code ZoneId} must be available,
     * either by using {@code ZonedDateTime} or {@link DateTimeFormatter#withZone}.
     *
     * @param dateTimeStyle  the formatter style to obtain, not null
     * @return the date-time formatter, not null
     */
    public static DateTimeFormatter ofLocalizedDateTime(FormatStyle dateTimeStyle) {
        Objects.requireNonNull(dateTimeStyle, "dateTimeStyle");
        return new DateTimeFormatterBuilder().appendLocalized(dateTimeStyle, dateTimeStyle)
                .toFormatter(ResolverStyle.SMART, IsoChronology.INSTANCE);
    }

    /**
     * Returns a locale specific date and time format for the ISO chronology.
     * <p>
     * This returns a formatter that will format or parse a date-time.
     * The exact format pattern used varies by locale.
     * <p>
     * The locale is determined from the formatter. The formatter returned directly by
     * this method will use the {@link Locale#getDefault() default FORMAT locale}.
     * The locale can be controlled using {@link DateTimeFormatter#withLocale(Locale) withLocale(Locale)}
     * on the result of this method.
     * <p>
     * Note that the localized pattern is looked up lazily.
     * This {@code DateTimeFormatter} holds the style required and the locale,
     * looking up the pattern required on demand.
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#SMART SMART} resolver style.
     * The {@code FULL} and {@code LONG} styles typically require a time-zone.
     * When formatting using these styles, a {@code ZoneId} must be available,
     * either by using {@code ZonedDateTime} or {@link DateTimeFormatter#withZone}.
     *
     * @param dateStyle  the date formatter style to obtain, not null
     * @param timeStyle  the time formatter style to obtain, not null
     * @return the date, time or date-time formatter, not null
     */
    public static DateTimeFormatter ofLocalizedDateTime(FormatStyle dateStyle, FormatStyle timeStyle) {
        Objects.requireNonNull(dateStyle, "dateStyle");
        Objects.requireNonNull(timeStyle, "timeStyle");
        return new DateTimeFormatterBuilder().appendLocalized(dateStyle, timeStyle)
                .toFormatter(ResolverStyle.SMART, IsoChronology.INSTANCE);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO date formatter that formats or parses a date without an
     * offset, such as '2011-12-03'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 extended local date format.
     * The format consists of:
     * <ul>
     * <li>Four digits or more for the {@link ChronoField#YEAR year}.
     * Years in the range 0000 to 9999 will be pre-padded by zero to ensure four digits.
     * Years outside that range will have a prefixed positive or negative symbol.
     * <li>A dash
     * <li>Two digits for the {@link ChronoField#MONTH_OF_YEAR month-of-year}.
     *  This is pre-padded by zero to ensure two digits.
     * <li>A dash
     * <li>Two digits for the {@link ChronoField#DAY_OF_MONTH day-of-month}.
     *  This is pre-padded by zero to ensure two digits.
     * </ul>
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_LOCAL_DATE;
    static {
        ISO_LOCAL_DATE = new DateTimeFormatterBuilder()
                .appendValue(YEAR, 4, 10, SignStyle.EXCEEDS_PAD)
                .appendLiteral('-')
                .appendValue(MONTH_OF_YEAR, 2)
                .appendLiteral('-')
                .appendValue(DAY_OF_MONTH, 2)
                .toFormatter(ResolverStyle.STRICT, IsoChronology.INSTANCE);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO date formatter that formats or parses a date with an
     * offset, such as '2011-12-03+01:00'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 extended offset date format.
     * The format consists of:
     * <ul>
     * <li>The {@link #ISO_LOCAL_DATE}
     * <li>The {@link ZoneOffset#getId() offset ID}. If the offset has seconds then
     *  they will be handled even though this is not part of the ISO-8601 standard.
     *  Parsing is case insensitive.
     * </ul>
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_OFFSET_DATE;
    static {
        ISO_OFFSET_DATE = new DateTimeFormatterBuilder()
                .parseCaseInsensitive()
                .append(ISO_LOCAL_DATE)
                .appendOffsetId()
                .toFormatter(ResolverStyle.STRICT, IsoChronology.INSTANCE);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO date formatter that formats or parses a date with the
     * offset if available, such as '2011-12-03' or '2011-12-03+01:00'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 extended date format.
     * The format consists of:
     * <ul>
     * <li>The {@link #ISO_LOCAL_DATE}
     * <li>If the offset is not available then the format is complete.
     * <li>The {@link ZoneOffset#getId() offset ID}. If the offset has seconds then
     *  they will be handled even though this is not part of the ISO-8601 standard.
     *  Parsing is case insensitive.
     * </ul>
     * <p>
     * As this formatter has an optional element, it may be necessary to parse using
     * {@link DateTimeFormatter#parseBest}.
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_DATE;
    static {
        ISO_DATE = new DateTimeFormatterBuilder()
                .parseCaseInsensitive()
                .append(ISO_LOCAL_DATE)
                .optionalStart()
                .appendOffsetId()
                .toFormatter(ResolverStyle.STRICT, IsoChronology.INSTANCE);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO time formatter that formats or parses a time without an
     * offset, such as '10:15' or '10:15:30'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 extended local time format.
     * The format consists of:
     * <ul>
     * <li>Two digits for the {@link ChronoField#HOUR_OF_DAY hour-of-day}.
     *  This is pre-padded by zero to ensure two digits.
     * <li>A colon
     * <li>Two digits for the {@link ChronoField#MINUTE_OF_HOUR minute-of-hour}.
     *  This is pre-padded by zero to ensure two digits.
     * <li>If the second-of-minute is not available then the format is complete.
     * <li>A colon
     * <li>Two digits for the {@link ChronoField#SECOND_OF_MINUTE second-of-minute}.
     *  This is pre-padded by zero to ensure two digits.
     * <li>If the nano-of-second is zero or not available then the format is complete.
     * <li>A decimal point
     * <li>One to nine digits for the {@link ChronoField#NANO_OF_SECOND nano-of-second}.
     *  As many digits will be output as required.
     * </ul>
     * <p>
     * The returned formatter has no override chronology or zone.
     * It uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_LOCAL_TIME;
    static {
        ISO_LOCAL_TIME = new DateTimeFormatterBuilder()
                .appendValue(HOUR_OF_DAY, 2)
                .appendLiteral(':')
                .appendValue(MINUTE_OF_HOUR, 2)
                .optionalStart()
                .appendLiteral(':')
                .appendValue(SECOND_OF_MINUTE, 2)
                .optionalStart()
                .appendFraction(NANO_OF_SECOND, 0, 9, true)
                .toFormatter(ResolverStyle.STRICT, null);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO time formatter that formats or parses a time with an
     * offset, such as '10:15+01:00' or '10:15:30+01:00'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 extended offset time format.
     * The format consists of:
     * <ul>
     * <li>The {@link #ISO_LOCAL_TIME}
     * <li>The {@link ZoneOffset#getId() offset ID}. If the offset has seconds then
     *  they will be handled even though this is not part of the ISO-8601 standard.
     *  Parsing is case insensitive.
     * </ul>
     * <p>
     * The returned formatter has no override chronology or zone.
     * It uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_OFFSET_TIME;
    static {
        ISO_OFFSET_TIME = new DateTimeFormatterBuilder()
                .parseCaseInsensitive()
                .append(ISO_LOCAL_TIME)
                .appendOffsetId()
                .toFormatter(ResolverStyle.STRICT, null);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO time formatter that formats or parses a time, with the
     * offset if available, such as '10:15', '10:15:30' or '10:15:30+01:00'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 extended offset time format.
     * The format consists of:
     * <ul>
     * <li>The {@link #ISO_LOCAL_TIME}
     * <li>If the offset is not available then the format is complete.
     * <li>The {@link ZoneOffset#getId() offset ID}. If the offset has seconds then
     *  they will be handled even though this is not part of the ISO-8601 standard.
     *  Parsing is case insensitive.
     * </ul>
     * <p>
     * As this formatter has an optional element, it may be necessary to parse using
     * {@link DateTimeFormatter#parseBest}.
     * <p>
     * The returned formatter has no override chronology or zone.
     * It uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_TIME;
    static {
        ISO_TIME = new DateTimeFormatterBuilder()
                .parseCaseInsensitive()
                .append(ISO_LOCAL_TIME)
                .optionalStart()
                .appendOffsetId()
                .toFormatter(ResolverStyle.STRICT, null);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO date-time formatter that formats or parses a date-time without
     * an offset, such as '2011-12-03T10:15:30'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 extended offset date-time format.
     * The format consists of:
     * <ul>
     * <li>The {@link #ISO_LOCAL_DATE}
     * <li>The letter 'T'. Parsing is case insensitive.
     * <li>The {@link #ISO_LOCAL_TIME}
     * </ul>
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_LOCAL_DATE_TIME;
    static {
        ISO_LOCAL_DATE_TIME = new DateTimeFormatterBuilder()
                .parseCaseInsensitive()
                .append(ISO_LOCAL_DATE)
                .appendLiteral('T')
                .append(ISO_LOCAL_TIME)
                .toFormatter(ResolverStyle.STRICT, IsoChronology.INSTANCE);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO date-time formatter that formats or parses a date-time with an
     * offset, such as '2011-12-03T10:15:30+01:00'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 extended offset date-time format.
     * The format consists of:
     * <ul>
     * <li>The {@link #ISO_LOCAL_DATE_TIME}
     * <li>The {@link ZoneOffset#getId() offset ID}. If the offset has seconds then
     *  they will be handled even though this is not part of the ISO-8601 standard.
     *  The offset parsing is lenient, which allows the minutes and seconds to be optional.
     *  Parsing is case insensitive.
     * </ul>
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_OFFSET_DATE_TIME;
    static {
        ISO_OFFSET_DATE_TIME = new DateTimeFormatterBuilder()
                .parseCaseInsensitive()
                .append(ISO_LOCAL_DATE_TIME)
                .parseLenient()
                .appendOffsetId()
                .parseStrict()
                .toFormatter(ResolverStyle.STRICT, IsoChronology.INSTANCE);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO-like date-time formatter that formats or parses a date-time with
     * offset and zone, such as '2011-12-03T10:15:30+01:00[Europe/Paris]'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * a format that extends the ISO-8601 extended offset date-time format
     * to add the time-zone.
     * The section in square brackets is not part of the ISO-8601 standard.
     * The format consists of:
     * <ul>
     * <li>The {@link #ISO_OFFSET_DATE_TIME}
     * <li>If the zone ID is not available or is a {@code ZoneOffset} then the format is complete.
     * <li>An open square bracket '['.
     * <li>The {@link ZoneId#getId() zone ID}. This is not part of the ISO-8601 standard.
     *  Parsing is case sensitive.
     * <li>A close square bracket ']'.
     * </ul>
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_ZONED_DATE_TIME;
    static {
        ISO_ZONED_DATE_TIME = new DateTimeFormatterBuilder()
                .append(ISO_OFFSET_DATE_TIME)
                .optionalStart()
                .appendLiteral('[')
                .parseCaseSensitive()
                .appendZoneRegionId()
                .appendLiteral(']')
                .toFormatter(ResolverStyle.STRICT, IsoChronology.INSTANCE);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO-like date-time formatter that formats or parses a date-time with
     * the offset and zone if available, such as '2011-12-03T10:15:30',
     * '2011-12-03T10:15:30+01:00' or '2011-12-03T10:15:30+01:00[Europe/Paris]'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 extended local or offset date-time format, as well as the
     * extended non-ISO form specifying the time-zone.
     * The format consists of:
     * <ul>
     * <li>The {@link #ISO_LOCAL_DATE_TIME}
     * <li>If the offset is not available to format or parse then the format is complete.
     * <li>The {@link ZoneOffset#getId() offset ID}. If the offset has seconds then
     *  they will be handled even though this is not part of the ISO-8601 standard.
     * <li>If the zone ID is not available or is a {@code ZoneOffset} then the format is complete.
     * <li>An open square bracket '['.
     * <li>The {@link ZoneId#getId() zone ID}. This is not part of the ISO-8601 standard.
     *  Parsing is case sensitive.
     * <li>A close square bracket ']'.
     * </ul>
     * <p>
     * As this formatter has an optional element, it may be necessary to parse using
     * {@link DateTimeFormatter#parseBest}.
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_DATE_TIME;
    static {
        ISO_DATE_TIME = new DateTimeFormatterBuilder()
                .append(ISO_LOCAL_DATE_TIME)
                .optionalStart()
                .appendOffsetId()
                .optionalStart()
                .appendLiteral('[')
                .parseCaseSensitive()
                .appendZoneRegionId()
                .appendLiteral(']')
                .toFormatter(ResolverStyle.STRICT, IsoChronology.INSTANCE);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO date formatter that formats or parses the ordinal date
     * without an offset, such as '2012-337'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 extended ordinal date format.
     * The format consists of:
     * <ul>
     * <li>Four digits or more for the {@link ChronoField#YEAR year}.
     * Years in the range 0000 to 9999 will be pre-padded by zero to ensure four digits.
     * Years outside that range will have a prefixed positive or negative symbol.
     * <li>A dash
     * <li>Three digits for the {@link ChronoField#DAY_OF_YEAR day-of-year}.
     *  This is pre-padded by zero to ensure three digits.
     * <li>If the offset is not available to format or parse then the format is complete.
     * <li>The {@link ZoneOffset#getId() offset ID}. If the offset has seconds then
     *  they will be handled even though this is not part of the ISO-8601 standard.
     *  Parsing is case insensitive.
     * </ul>
     * <p>
     * As this formatter has an optional element, it may be necessary to parse using
     * {@link DateTimeFormatter#parseBest}.
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_ORDINAL_DATE;
    static {
        ISO_ORDINAL_DATE = new DateTimeFormatterBuilder()
                .parseCaseInsensitive()
                .appendValue(YEAR, 4, 10, SignStyle.EXCEEDS_PAD)
                .appendLiteral('-')
                .appendValue(DAY_OF_YEAR, 3)
                .optionalStart()
                .appendOffsetId()
                .toFormatter(ResolverStyle.STRICT, IsoChronology.INSTANCE);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO date formatter that formats or parses the week-based date
     * without an offset, such as '2012-W48-6'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 extended week-based date format.
     * The format consists of:
     * <ul>
     * <li>Four digits or more for the {@link IsoFields#WEEK_BASED_YEAR week-based-year}.
     * Years in the range 0000 to 9999 will be pre-padded by zero to ensure four digits.
     * Years outside that range will have a prefixed positive or negative symbol.
     * <li>A dash
     * <li>The letter 'W'. Parsing is case insensitive.
     * <li>Two digits for the {@link IsoFields#WEEK_OF_WEEK_BASED_YEAR week-of-week-based-year}.
     *  This is pre-padded by zero to ensure three digits.
     * <li>A dash
     * <li>One digit for the {@link ChronoField#DAY_OF_WEEK day-of-week}.
     *  The value run from Monday (1) to Sunday (7).
     * <li>If the offset is not available to format or parse then the format is complete.
     * <li>The {@link ZoneOffset#getId() offset ID}. If the offset has seconds then
     *  they will be handled even though this is not part of the ISO-8601 standard.
     *  Parsing is case insensitive.
     * </ul>
     * <p>
     * As this formatter has an optional element, it may be necessary to parse using
     * {@link DateTimeFormatter#parseBest}.
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_WEEK_DATE;
    static {
        ISO_WEEK_DATE = new DateTimeFormatterBuilder()
                .parseCaseInsensitive()
                .appendValue(IsoFields.WEEK_BASED_YEAR, 4, 10, SignStyle.EXCEEDS_PAD)
                .appendLiteral("-W")
                .appendValue(IsoFields.WEEK_OF_WEEK_BASED_YEAR, 2)
                .appendLiteral('-')
                .appendValue(DAY_OF_WEEK, 1)
                .optionalStart()
                .appendOffsetId()
                .toFormatter(ResolverStyle.STRICT, IsoChronology.INSTANCE);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO instant formatter that formats or parses an instant in UTC,
     * such as '2011-12-03T10:15:30Z'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 instant format.
     * When formatting, the second-of-minute is always output.
     * The nano-of-second outputs zero, three, six or nine digits as necessary.
     * When parsing, time to at least the seconds field is required.
     * Fractional seconds from zero to nine are parsed.
     * The localized decimal style is not used.
     * <p>
     * This is a special case formatter intended to allow a human readable form
     * of an {@link java.time.Instant}. The {@code Instant} class is designed to
     * only represent a point in time and internally stores a value in nanoseconds
     * from a fixed epoch of 1970-01-01Z. As such, an {@code Instant} cannot be
     * formatted as a date or time without providing some form of time-zone.
     * This formatter allows the {@code Instant} to be formatted, by providing
     * a suitable conversion using {@code ZoneOffset.UTC}.
     * <p>
     * The format consists of:
     * <ul>
     * <li>The {@link #ISO_OFFSET_DATE_TIME} where the instant is converted from
     *  {@link ChronoField#INSTANT_SECONDS} and {@link ChronoField#NANO_OF_SECOND}
     *  using the {@code UTC} offset. Parsing is case insensitive.
     * </ul>
     * <p>
     * The returned formatter has no override chronology or zone.
     * It uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter ISO_INSTANT;
    static {
        ISO_INSTANT = new DateTimeFormatterBuilder()
                .parseCaseInsensitive()
                .appendInstant()
                .toFormatter(ResolverStyle.STRICT, null);
    }

    //-----------------------------------------------------------------------
    /**
     * The ISO date formatter that formats or parses a date without an
     * offset, such as '20111203'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * the ISO-8601 basic local date format.
     * The format consists of:
     * <ul>
     * <li>Four digits for the {@link ChronoField#YEAR year}.
     *  Only years in the range 0000 to 9999 are supported.
     * <li>Two digits for the {@link ChronoField#MONTH_OF_YEAR month-of-year}.
     *  This is pre-padded by zero to ensure two digits.
     * <li>Two digits for the {@link ChronoField#DAY_OF_MONTH day-of-month}.
     *  This is pre-padded by zero to ensure two digits.
     * <li>If the offset is not available to format or parse then the format is complete.
     * <li>The {@link ZoneOffset#getId() offset ID} without colons. If the offset has
     *  seconds then they will be handled even though this is not part of the ISO-8601 standard.
     *  The offset parsing is lenient, which allows the minutes and seconds to be optional.
     *  Parsing is case insensitive.
     * </ul>
     * <p>
     * As this formatter has an optional element, it may be necessary to parse using
     * {@link DateTimeFormatter#parseBest}.
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#STRICT STRICT} resolver style.
     */
    public static final DateTimeFormatter BASIC_ISO_DATE;
    static {
        BASIC_ISO_DATE = new DateTimeFormatterBuilder()
                .parseCaseInsensitive()
                .appendValue(YEAR, 4)
                .appendValue(MONTH_OF_YEAR, 2)
                .appendValue(DAY_OF_MONTH, 2)
                .optionalStart()
                .parseLenient()
                .appendOffset("+HHMMss", "Z")
                .parseStrict()
                .toFormatter(ResolverStyle.STRICT, IsoChronology.INSTANCE);
    }

    //-----------------------------------------------------------------------
    /**
     * The RFC-1123 date-time formatter, such as 'Tue, 3 Jun 2008 11:05:30 GMT'.
     * <p>
     * This returns an immutable formatter capable of formatting and parsing
     * most of the RFC-1123 format.
     * RFC-1123 updates RFC-822 changing the year from two digits to four.
     * This implementation requires a four digit year.
     * This implementation also does not handle North American or military zone
     * names, only 'GMT' and offset amounts.
     * <p>
     * The format consists of:
     * <ul>
     * <li>If the day-of-week is not available to format or parse then jump to day-of-month.
     * <li>Three letter {@link ChronoField#DAY_OF_WEEK day-of-week} in English.
     * <li>A comma
     * <li>A space
     * <li>One or two digits for the {@link ChronoField#DAY_OF_MONTH day-of-month}.
     * <li>A space
     * <li>Three letter {@link ChronoField#MONTH_OF_YEAR month-of-year} in English.
     * <li>A space
     * <li>Four digits for the {@link ChronoField#YEAR year}.
     *  Only years in the range 0000 to 9999 are supported.
     * <li>A space
     * <li>Two digits for the {@link ChronoField#HOUR_OF_DAY hour-of-day}.
     *  This is pre-padded by zero to ensure two digits.
     * <li>A colon
     * <li>Two digits for the {@link ChronoField#MINUTE_OF_HOUR minute-of-hour}.
     *  This is pre-padded by zero to ensure two digits.
     * <li>If the second-of-minute is not available then jump to the next space.
     * <li>A colon
     * <li>Two digits for the {@link ChronoField#SECOND_OF_MINUTE second-of-minute}.
     *  This is pre-padded by zero to ensure two digits.
     * <li>A space
     * <li>The {@link ZoneOffset#getId() offset ID} without colons or seconds.
     *  An offset of zero uses "GMT". North American zone names and military zone names are not handled.
     * </ul>
     * <p>
     * Parsing is case insensitive.
     * <p>
     * The returned formatter has a chronology of ISO set to ensure dates in
     * other calendar systems are correctly converted.
     * It has no override zone and uses the {@link ResolverStyle#SMART SMART} resolver style.
     */
    public static final DateTimeFormatter RFC_1123_DATE_TIME;
    static {
        // manually code maps to ensure correct data always used
        // (locale data can be changed by application code)
        Map<Long, String> dow = new HashMap<>();
        dow.put(1L, "Mon");
        dow.put(2L, "Tue");
        dow.put(3L, "Wed");
        dow.put(4L, "Thu");
        dow.put(5L, "Fri");
        dow.put(6L, "Sat");
        dow.put(7L, "Sun");
        Map<Long, String> moy = new HashMap<>();
        moy.put(1L, "Jan");
        moy.put(2L, "Feb");
        moy.put(3L, "Mar");
        moy.put(4L, "Apr");
        moy.put(5L, "May");
        moy.put(6L, "Jun");
        moy.put(7L, "Jul");
        moy.put(8L, "Aug");
        moy.put(9L, "Sep");
        moy.put(10L, "Oct");
        moy.put(11L, "Nov");
        moy.put(12L, "Dec");
        RFC_1123_DATE_TIME = new DateTimeFormatterBuilder()
                .parseCaseInsensitive()
                .parseLenient()
                .optionalStart()
                .appendText(DAY_OF_WEEK, dow)
                .appendLiteral(", ")
                .optionalEnd()
                .appendValue(DAY_OF_MONTH, 1, 2, SignStyle.NOT_NEGATIVE)
                .appendLiteral(' ')
                .appendText(MONTH_OF_YEAR, moy)
                .appendLiteral(' ')
                .appendValue(YEAR, 4)  // 2 digit year not handled
                .appendLiteral(' ')
                .appendValue(HOUR_OF_DAY, 2)
                .appendLiteral(':')
                .appendValue(MINUTE_OF_HOUR, 2)
                .optionalStart()
                .appendLiteral(':')
                .appendValue(SECOND_OF_MINUTE, 2)
                .optionalEnd()
                .appendLiteral(' ')
                .appendOffset("+HHMM", "GMT")  // should handle UT/Z/EST/EDT/CST/CDT/MST/MDT/PST/MDT
                .toFormatter(ResolverStyle.SMART, IsoChronology.INSTANCE);
    }

    //-----------------------------------------------------------------------
    /**
     * A query that provides access to the excess days that were parsed.
     * <p>
     * This returns a singleton {@linkplain TemporalQuery query} that provides
     * access to additional information from the parse. The query always returns
     * a non-null period, with a zero period returned instead of null.
     * <p>
     * There are two situations where this query may return a non-zero period.
     * <ul>
     * <li>If the {@code ResolverStyle} is {@code LENIENT} and a time is parsed
     *  without a date, then the complete result of the parse consists of a
     *  {@code LocalTime} and an excess {@code Period} in days.
     *
     * <li>If the {@code ResolverStyle} is {@code SMART} and a time is parsed
     *  without a date where the time is 24:00:00, then the complete result of
     *  the parse consists of a {@code LocalTime} of 00:00:00 and an excess
     *  {@code Period} of one day.
     * </ul>
     * <p>
     * In both cases, if a complete {@code ChronoLocalDateTime} or {@code Instant}
     * is parsed, then the excess days are added to the date part.
     * As a result, this query will return a zero period.
     * <p>
     * The {@code SMART} behaviour handles the common "end of day" 24:00 value.
     * Processing in {@code LENIENT} mode also produces the same result:
     * <pre>
     *  Text to parse        Parsed object                         Excess days
     *  "2012-12-03T00:00"   LocalDateTime.of(2012, 12, 3, 0, 0)   ZERO
     *  "2012-12-03T24:00"   LocalDateTime.of(2012, 12, 4, 0, 0)   ZERO
     *  "00:00"              LocalTime.of(0, 0)                    ZERO
     *  "24:00"              LocalTime.of(0, 0)                    Period.ofDays(1)
     * </pre>
     * The query can be used as follows:
     * <pre>
     *  TemporalAccessor parsed = formatter.parse(str);
     *  LocalTime time = parsed.query(LocalTime::from);
     *  Period extraDays = parsed.query(DateTimeFormatter.parsedExcessDays());
     * </pre>
     * @return a query that provides access to the excess days that were parsed
     */
    public static final TemporalQuery<Period> parsedExcessDays() {
        return PARSED_EXCESS_DAYS;
    }
    private static final TemporalQuery<Period> PARSED_EXCESS_DAYS = t -> {
        if (t instanceof Parsed) {
            return ((Parsed) t).excessDays;
        } else {
            return Period.ZERO;
        }
    };

    /**
     * A query that provides access to whether a leap-second was parsed.
     * <p>
     * This returns a singleton {@linkplain TemporalQuery query} that provides
     * access to additional information from the parse. The query always returns
     * a non-null boolean, true if parsing saw a leap-second, false if not.
     * <p>
     * Instant parsing handles the special "leap second" time of '23:59:60'.
     * Leap seconds occur at '23:59:60' in the UTC time-zone, but at other
     * local times in different time-zones. To avoid this potential ambiguity,
     * the handling of leap-seconds is limited to
     * {@link DateTimeFormatterBuilder#appendInstant()}, as that method
     * always parses the instant with the UTC zone offset.
     * <p>
     * If the time '23:59:60' is received, then a simple conversion is applied,
     * replacing the second-of-minute of 60 with 59. This query can be used
     * on the parse result to determine if the leap-second adjustment was made.
     * The query will return {@code true} if it did adjust to remove the
     * leap-second, and {@code false} if not. Note that applying a leap-second
     * smoothing mechanism, such as UTC-SLS, is the responsibility of the
     * application, as follows:
     * <pre>
     *  TemporalAccessor parsed = formatter.parse(str);
     *  Instant instant = parsed.query(Instant::from);
     *  if (parsed.query(DateTimeFormatter.parsedLeapSecond())) {
     *    // validate leap-second is correct and apply correct smoothing
     *  }
     * </pre>
     * @return a query that provides access to whether a leap-second was parsed
     */
    public static final TemporalQuery<Boolean> parsedLeapSecond() {
        return PARSED_LEAP_SECOND;
    }
    private static final TemporalQuery<Boolean> PARSED_LEAP_SECOND = t -> {
        if (t instanceof Parsed) {
            return ((Parsed) t).leapSecond;
        } else {
            return Boolean.FALSE;
        }
    };

    //-----------------------------------------------------------------------
    /**
     * Constructor.
     *
     * @param printerParser  the printer/parser to use, not null
     * @param locale  the locale to use, not null
     * @param decimalStyle  the DecimalStyle to use, not null
     * @param resolverStyle  the resolver style to use, not null
     * @param resolverFields  the fields to use during resolving, null for all fields
     * @param chrono  the chronology to use, null for no override
     * @param zone  the zone to use, null for no override
     */
    DateTimeFormatter(CompositePrinterParser printerParser,
            Locale locale, DecimalStyle decimalStyle,
            ResolverStyle resolverStyle, Set<TemporalField> resolverFields,
            Chronology chrono, ZoneId zone) {
        this.printerParser = Objects.requireNonNull(printerParser, "printerParser");
        this.resolverFields = resolverFields;
        this.locale = Objects.requireNonNull(locale, "locale");
        this.decimalStyle = Objects.requireNonNull(decimalStyle, "decimalStyle");
        this.resolverStyle = Objects.requireNonNull(resolverStyle, "resolverStyle");
        this.chrono = chrono;
        this.zone = zone;
    }

    //-----------------------------------------------------------------------
    /**
     * Gets the locale to be used during formatting.
     * <p>
     * This is used to lookup any part of the formatter needing specific
     * localization, such as the text or localized pattern.
     *
     * @return the locale of this formatter, not null
     */
    public Locale getLocale() {
        return locale;
    }

    /**
     * Returns a copy of this formatter with a new locale.
     * <p>
     * This is used to lookup any part of the formatter needing specific
     * localization, such as the text or localized pattern.
     * <p>
     * This instance is immutable and unaffected by this method call.
     *
     * @param locale  the new locale, not null
     * @return a formatter based on this formatter with the requested locale, not null
     */
    public DateTimeFormatter withLocale(Locale locale) {
        if (this.locale.equals(locale)) {
            return this;
        }
        return new DateTimeFormatter(printerParser, locale, decimalStyle, resolverStyle, resolverFields, chrono, zone);
    }

    //-----------------------------------------------------------------------
    /**
     * Gets the DecimalStyle to be used during formatting.
     *
     * @return the locale of this formatter, not null
     */
    public DecimalStyle getDecimalStyle() {
        return decimalStyle;
    }

    /**
     * Returns a copy of this formatter with a new DecimalStyle.
     * <p>
     * This instance is immutable and unaffected by this method call.
     *
     * @param decimalStyle  the new DecimalStyle, not null
     * @return a formatter based on this formatter with the requested DecimalStyle, not null
     */
    public DateTimeFormatter withDecimalStyle(DecimalStyle decimalStyle) {
        if (this.decimalStyle.equals(decimalStyle)) {
            return this;
        }
        return new DateTimeFormatter(printerParser, locale, decimalStyle, resolverStyle, resolverFields, chrono, zone);
    }

    //-----------------------------------------------------------------------
    /**
     * Gets the overriding chronology to be used during formatting.
     * <p>
     * This returns the override chronology, used to convert dates.
     * By default, a formatter has no override chronology, returning null.
     * See {@link #withChronology(Chronology)} for more details on overriding.
     *
     * @return the override chronology of this formatter, null if no override
     */
    public Chronology getChronology() {
        return chrono;
    }

    /**
     * Returns a copy of this formatter with a new override chronology.
     * <p>
     * This returns a formatter with similar state to this formatter but
     * with the override chronology set.
     * By default, a formatter has no override chronology, returning null.
     * <p>
     * If an override is added, then any date that is formatted or parsed will be affected.
     * <p>
     * When formatting, if the temporal object contains a date, then it will
     * be converted to a date in the override chronology.
     * Whether the temporal contains a date is determined by querying the
     * {@link ChronoField#EPOCH_DAY EPOCH_DAY} field.
     * Any time or zone will be retained unaltered unless overridden.
     * <p>
     * If the temporal object does not contain a date, but does contain one
     * or more {@code ChronoField} date fields, then a {@code DateTimeException}
     * is thrown. In all other cases, the override chronology is added to the temporal,
     * replacing any previous chronology, but without changing the date/time.
     * <p>
     * When parsing, there are two distinct cases to consider.
     * If a chronology has been parsed directly from the text, perhaps because
     * {@link DateTimeFormatterBuilder#appendChronologyId()} was used, then
     * this override chronology has no effect.
     * If no zone has been parsed, then this override chronology will be used
     * to interpret the {@code ChronoField} values into a date according to the
     * date resolving rules of the chronology.
     * <p>
     * This instance is immutable and unaffected by this method call.
     *
     * @param chrono  the new chronology, null if no override
     * @return a formatter based on this formatter with the requested override chronology, not null
     */
    public DateTimeFormatter withChronology(Chronology chrono) {
        if (Objects.equals(this.chrono, chrono)) {
            return this;
        }
        return new DateTimeFormatter(printerParser, locale, decimalStyle, resolverStyle, resolverFields, chrono, zone);
    }

    //-----------------------------------------------------------------------
    /**
     * Gets the overriding zone to be used during formatting.
     * <p>
     * This returns the override zone, used to convert instants.
     * By default, a formatter has no override zone, returning null.
     * See {@link #withZone(ZoneId)} for more details on overriding.
     *
     * @return the override zone of this formatter, null if no override
     */
    public ZoneId getZone() {
        return zone;
    }

    /**
     * Returns a copy of this formatter with a new override zone.
     * <p>
     * This returns a formatter with similar state to this formatter but
     * with the override zone set.
     * By default, a formatter has no override zone, returning null.
     * <p>
     * If an override is added, then any instant that is formatted or parsed will be affected.
     * <p>
     * When formatting, if the temporal object contains an instant, then it will
     * be converted to a zoned date-time using the override zone.
     * Whether the temporal is an instant is determined by querying the
     * {@link ChronoField#INSTANT_SECONDS INSTANT_SECONDS} field.
     * If the input has a chronology then it will be retained unless overridden.
     * If the input does not have a chronology, such as {@code Instant}, then
     * the ISO chronology will be used.
     * <p>
     * If the temporal object does not contain an instant, but does contain
     * an offset then an additional check is made. If the normalized override
     * zone is an offset that differs from the offset of the temporal, then
     * a {@code DateTimeException} is thrown. In all other cases, the override
     * zone is added to the temporal, replacing any previous zone, but without
     * changing the date/time.
     * <p>
     * When parsing, there are two distinct cases to consider.
     * If a zone has been parsed directly from the text, perhaps because
     * {@link DateTimeFormatterBuilder#appendZoneId()} was used, then
     * this override zone has no effect.
     * If no zone has been parsed, then this override zone will be included in
     * the result of the parse where it can be used to build instants and date-times.
     * <p>
     * This instance is immutable and unaffected by this method call.
     *
     * @param zone  the new override zone, null if no override
     * @return a formatter based on this formatter with the requested override zone, not null
     */
    public DateTimeFormatter withZone(ZoneId zone) {
        if (Objects.equals(this.zone, zone)) {
            return this;
        }
        return new DateTimeFormatter(printerParser, locale, decimalStyle, resolverStyle, resolverFields, chrono, zone);
    }

    //-----------------------------------------------------------------------
    /**
     * Gets the resolver style to use during parsing.
     * <p>
     * This returns the resolver style, used during the second phase of parsing
     * when fields are resolved into dates and times.
     * By default, a formatter has the {@link ResolverStyle#SMART SMART} resolver style.
     * See {@link #withResolverStyle(ResolverStyle)} for more details.
     *
     * @return the resolver style of this formatter, not null
     */
    public ResolverStyle getResolverStyle() {
        return resolverStyle;
    }

    /**
     * Returns a copy of this formatter with a new resolver style.
     * <p>
     * This returns a formatter with similar state to this formatter but
     * with the resolver style set. By default, a formatter has the
     * {@link ResolverStyle#SMART SMART} resolver style.
     * <p>
     * Changing the resolver style only has an effect during parsing.
     * Parsing a text string occurs in two phases.
     * Phase 1 is a basic text parse according to the fields added to the builder.
     * Phase 2 resolves the parsed field-value pairs into date and/or time objects.
     * The resolver style is used to control how phase 2, resolving, happens.
     * See {@code ResolverStyle} for more information on the options available.
     * <p>
     * This instance is immutable and unaffected by this method call.
     *
     * @param resolverStyle  the new resolver style, not null
     * @return a formatter based on this formatter with the requested resolver style, not null
     */
    public DateTimeFormatter withResolverStyle(ResolverStyle resolverStyle) {
        Objects.requireNonNull(resolverStyle, "resolverStyle");
        if (Objects.equals(this.resolverStyle, resolverStyle)) {
            return this;
        }
        return new DateTimeFormatter(printerParser, locale, decimalStyle, resolverStyle, resolverFields, chrono, zone);
    }

    //-----------------------------------------------------------------------
    /**
     * Gets the resolver fields to use during parsing.
     * <p>
     * This returns the resolver fields, used during the second phase of parsing
     * when fields are resolved into dates and times.
     * By default, a formatter has no resolver fields, and thus returns null.
     * See {@link #withResolverFields(Set)} for more details.
     *
     * @return the immutable set of resolver fields of this formatter, null if no fields
     */
    public Set<TemporalField> getResolverFields() {
        return resolverFields;
    }

    /**
     * Returns a copy of this formatter with a new set of resolver fields.
     * <p>
     * This returns a formatter with similar state to this formatter but with
     * the resolver fields set. By default, a formatter has no resolver fields.
     * <p>
     * Changing the resolver fields only has an effect during parsing.
     * Parsing a text string occurs in two phases.
     * Phase 1 is a basic text parse according to the fields added to the builder.
     * Phase 2 resolves the parsed field-value pairs into date and/or time objects.
     * The resolver fields are used to filter the field-value pairs between phase 1 and 2.
     * <p>
     * This can be used to select between two or more ways that a date or time might
     * be resolved. For example, if the formatter consists of year, month, day-of-month
     * and day-of-year, then there are two ways to resolve a date.
     * Calling this method with the arguments {@link ChronoField#YEAR YEAR} and
     * {@link ChronoField#DAY_OF_YEAR DAY_OF_YEAR} will ensure that the date is
     * resolved using the year and day-of-year, effectively meaning that the month
     * and day-of-month are ignored during the resolving phase.
     * <p>
     * In a similar manner, this method can be used to ignore secondary fields that
     * would otherwise be cross-checked. For example, if the formatter consists of year,
     * month, day-of-month and day-of-week, then there is only one way to resolve a
     * date, but the parsed value for day-of-week will be cross-checked against the
     * resolved date. Calling this method with the arguments {@link ChronoField#YEAR YEAR},
     * {@link ChronoField#MONTH_OF_YEAR MONTH_OF_YEAR} and
     * {@link ChronoField#DAY_OF_MONTH DAY_OF_MONTH} will ensure that the date is
     * resolved correctly, but without any cross-check for the day-of-week.
     * <p>
     * In implementation terms, this method behaves as follows. The result of the
     * parsing phase can be considered to be a map of field to value. The behavior
     * of this method is to cause that map to be filtered between phase 1 and 2,
     * removing all fields other than those specified as arguments to this method.
     * <p>
     * This instance is immutable and unaffected by this method call.
     *
     * @param resolverFields  the new set of resolver fields, null if no fields
     * @return a formatter based on this formatter with the requested resolver style, not null
     */
    public DateTimeFormatter withResolverFields(TemporalField... resolverFields) {
        Set<TemporalField> fields = null;
        if (resolverFields != null) {
            // Set.of cannot be used because it is hostile to nulls and duplicate elements
            fields = Collections.unmodifiableSet(new HashSet<>(Arrays.asList(resolverFields)));
        }
        if (Objects.equals(this.resolverFields, fields)) {
            return this;
        }
        return new DateTimeFormatter(printerParser, locale, decimalStyle, resolverStyle, fields, chrono, zone);
    }

    /**
     * Returns a copy of this formatter with a new set of resolver fields.
     * <p>
     * This returns a formatter with similar state to this formatter but with
     * the resolver fields set. By default, a formatter has no resolver fields.
     * <p>
     * Changing the resolver fields only has an effect during parsing.
     * Parsing a text string occurs in two phases.
     * Phase 1 is a basic text parse according to the fields added to the builder.
     * Phase 2 resolves the parsed field-value pairs into date and/or time objects.
     * The resolver fields are used to filter the field-value pairs between phase 1 and 2.
     * <p>
     * This can be used to select between two or more ways that a date or time might
     * be resolved. For example, if the formatter consists of year, month, day-of-month
     * and day-of-year, then there are two ways to resolve a date.
     * Calling this method with the arguments {@link ChronoField#YEAR YEAR} and
     * {@link ChronoField#DAY_OF_YEAR DAY_OF_YEAR} will ensure that the date is
     * resolved using the year and day-of-year, effectively meaning that the month
     * and day-of-month are ignored during the resolving phase.
     * <p>
     * In a similar manner, this method can be used to ignore secondary fields that
     * would otherwise be cross-checked. For example, if the formatter consists of year,
     * month, day-of-month and day-of-week, then there is only one way to resolve a
     * date, but the parsed value for day-of-week will be cross-checked against the
     * resolved date. Calling this method with the arguments {@link ChronoField#YEAR YEAR},
     * {@link ChronoField#MONTH_OF_YEAR MONTH_OF_YEAR} and
     * {@link ChronoField#DAY_OF_MONTH DAY_OF_MONTH} will ensure that the date is
     * resolved correctly, but without any cross-check for the day-of-week.
     * <p>
     * In implementation terms, this method behaves as follows. The result of the
     * parsing phase can be considered to be a map of field to value. The behavior
     * of this method is to cause that map to be filtered between phase 1 and 2,
     * removing all fields other than those specified as arguments to this method.
     * <p>
     * This instance is immutable and unaffected by this method call.
     *
     * @param resolverFields  the new set of resolver fields, null if no fields
     * @return a formatter based on this formatter with the requested resolver style, not null
     */
    public DateTimeFormatter withResolverFields(Set<TemporalField> resolverFields) {
        if (Objects.equals(this.resolverFields, resolverFields)) {
            return this;
        }
        if (resolverFields != null) {
            resolverFields = Collections.unmodifiableSet(new HashSet<>(resolverFields));
        }
        return new DateTimeFormatter(printerParser, locale, decimalStyle, resolverStyle, resolverFields, chrono, zone);
    }

    //-----------------------------------------------------------------------
    /**
     * Formats a date-time object using this formatter.
     * <p>
     * This formats the date-time to a String using the rules of the formatter.
     *
     * @param temporal  the temporal object to format, not null
     * @return the formatted string, not null
     * @throws DateTimeException if an error occurs during formatting
     */
    public String format(TemporalAccessor temporal) {
        StringBuilder buf = new StringBuilder(32);
        formatTo(temporal, buf);
        return buf.toString();
    }

    //-----------------------------------------------------------------------
    /**
     * Formats a date-time object to an {@code Appendable} using this formatter.
     * <p>
     * This outputs the formatted date-time to the specified destination.
     * {@link Appendable} is a general purpose interface that is implemented by all
     * key character output classes including {@code StringBuffer}, {@code StringBuilder},
     * {@code PrintStream} and {@code Writer}.
     * <p>
     * Although {@code Appendable} methods throw an {@code IOException}, this method does not.
     * Instead, any {@code IOException} is wrapped in a runtime exception.
     *
     * @param temporal  the temporal object to format, not null
     * @param appendable  the appendable to format to, not null
     * @throws DateTimeException if an error occurs during formatting
     */
    public void formatTo(TemporalAccessor temporal, Appendable appendable) {
        Objects.requireNonNull(temporal, "temporal");
        Objects.requireNonNull(appendable, "appendable");
        try {
            DateTimePrintContext context = new DateTimePrintContext(temporal, this);
            if (appendable instanceof StringBuilder) {
                printerParser.format(context, (StringBuilder) appendable);
            } else {
                // buffer output to avoid writing to appendable in case of error
                StringBuilder buf = new StringBuilder(32);
                printerParser.format(context, buf);
                appendable.append(buf);
            }
        } catch (IOException ex) {
            throw new DateTimeException(ex.getMessage(), ex);
        }
    }

    //-----------------------------------------------------------------------
    /**
     * Fully parses the text producing a temporal object.
     * <p>
     * This parses the entire text producing a temporal object.
     * It is typically more useful to use {@link #parse(CharSequence, TemporalQuery)}.
     * The result of this method is {@code TemporalAccessor} which has been resolved,
     * applying basic validation checks to help ensure a valid date-time.
     * <p>
     * If the parse completes without reading the entire length of the text,
     * or a problem occurs during parsing or merging, then an exception is thrown.
     *
     * @param text  the text to parse, not null
     * @return the parsed temporal object, not null
     * @throws DateTimeParseException if unable to parse the requested result
     */
    public TemporalAccessor parse(CharSequence text) {
        Objects.requireNonNull(text, "text");
        try {
            return parseResolved0(text, null);
        } catch (DateTimeParseException ex) {
            throw ex;
        } catch (RuntimeException ex) {
            throw createError(text, ex);
        }
    }

    /**
     * Parses the text using this formatter, providing control over the text position.
     * <p>
     * This parses the text without requiring the parse to start from the beginning
     * of the string or finish at the end.
     * The result of this method is {@code TemporalAccessor} which has been resolved,
     * applying basic validation checks to help ensure a valid date-time.
     * <p>
     * The text will be parsed from the specified start {@code ParsePosition}.
     * The entire length of the text does not have to be parsed, the {@code ParsePosition}
     * will be updated with the index at the end of parsing.
     * <p>
     * The operation of this method is slightly different to similar methods using
     * {@code ParsePosition} on {@code java.text.Format}. That class will return
     * errors using the error index on the {@code ParsePosition}. By contrast, this
     * method will throw a {@link DateTimeParseException} if an error occurs, with
     * the exception containing the error index.
     * This change in behavior is necessary due to the increased complexity of
     * parsing and resolving dates/times in this API.
     * <p>
     * If the formatter parses the same field more than once with different values,
     * the result will be an error.
     *
     * @param text  the text to parse, not null
     * @param position  the position to parse from, updated with length parsed
     *  and the index of any error, not null
     * @return the parsed temporal object, not null
     * @throws DateTimeParseException if unable to parse the requested result
     * @throws IndexOutOfBoundsException if the position is invalid
     */
    public TemporalAccessor parse(CharSequence text, ParsePosition position) {
        Objects.requireNonNull(text, "text");
        Objects.requireNonNull(position, "position");
        try {
            return parseResolved0(text, position);
        } catch (DateTimeParseException | IndexOutOfBoundsException ex) {
            throw ex;
        } catch (RuntimeException ex) {
            throw createError(text, ex);
        }
    }

    //-----------------------------------------------------------------------
    /**
     * Fully parses the text producing an object of the specified type.
     * <p>
     * Most applications should use this method for parsing.
     * It parses the entire text to produce the required date-time.
     * The query is typically a method reference to a {@code from(TemporalAccessor)} method.
     * For example:
     * <pre>
     *  LocalDateTime dt = parser.parse(str, LocalDateTime::from);
     * </pre>
     * If the parse completes without reading the entire length of the text,
     * or a problem occurs during parsing or merging, then an exception is thrown.
     *
     * @param <T> the type of the parsed date-time
     * @param text  the text to parse, not null
     * @param query  the query defining the type to parse to, not null
     * @return the parsed date-time, not null
     * @throws DateTimeParseException if unable to parse the requested result
     */
    public <T> T parse(CharSequence text, TemporalQuery<T> query) {
        Objects.requireNonNull(text, "text");
        Objects.requireNonNull(query, "query");
        try {
            return parseResolved0(text, null).query(query);
        } catch (DateTimeParseException ex) {
            throw ex;
        } catch (RuntimeException ex) {
            throw createError(text, ex);
        }
    }

    /**
     * Fully parses the text producing an object of one of the specified types.
     * <p>
     * This parse method is convenient for use when the parser can handle optional elements.
     * For example, a pattern of 'uuuu-MM-dd HH.mm[ VV]' can be fully parsed to a {@code ZonedDateTime},
     * or partially parsed to a {@code LocalDateTime}.
     * The queries must be specified in order, starting from the best matching full-parse option
     * and ending with the worst matching minimal parse option.
     * The query is typically a method reference to a {@code from(TemporalAccessor)} method.
     * <p>
     * The result is associated with the first type that successfully parses.
     * Normally, applications will use {@code instanceof} to check the result.
     * For example:
     * <pre>
     *  TemporalAccessor dt = parser.parseBest(str, ZonedDateTime::from, LocalDateTime::from);
     *  if (dt instanceof ZonedDateTime) {
     *   ...
     *  } else {
     *   ...
     *  }
     * </pre>
     * If the parse completes without reading the entire length of the text,
     * or a problem occurs during parsing or merging, then an exception is thrown.
     *
     * @param text  the text to parse, not null
     * @param queries  the queries defining the types to attempt to parse to,
     *  must implement {@code TemporalAccessor}, not null
     * @return the parsed date-time, not null
     * @throws IllegalArgumentException if less than 2 types are specified
     * @throws DateTimeParseException if unable to parse the requested result
     */
    public TemporalAccessor parseBest(CharSequence text, TemporalQuery<?>... queries) {
        Objects.requireNonNull(text, "text");
        Objects.requireNonNull(queries, "queries");
        if (queries.length < 2) {
            throw new IllegalArgumentException("At least two queries must be specified");
        }
        try {
            TemporalAccessor resolved = parseResolved0(text, null);
            for (TemporalQuery<?> query : queries) {
                try {
                    return (TemporalAccessor) resolved.query(query);
                } catch (RuntimeException ex) {
                    // continue
                }
            }
            throw new DateTimeException("Unable to convert parsed text using any of the specified queries");
        } catch (DateTimeParseException ex) {
            throw ex;
        } catch (RuntimeException ex) {
            throw createError(text, ex);
        }
    }

    private DateTimeParseException createError(CharSequence text, RuntimeException ex) {
        String abbr;
        if (text.length() > 64) {
            abbr = text.subSequence(0, 64).toString() + "...";
        } else {
            abbr = text.toString();
        }
        return new DateTimeParseException("Text '" + abbr + "' could not be parsed: " + ex.getMessage(), text, 0, ex);
    }

    //-----------------------------------------------------------------------
    /**
     * Parses and resolves the specified text.
     * <p>
     * This parses to a {@code TemporalAccessor} ensuring that the text is fully parsed.
     *
     * @param text  the text to parse, not null
     * @param position  the position to parse from, updated with length parsed
     *  and the index of any error, null if parsing whole string
     * @return the resolved result of the parse, not null
     * @throws DateTimeParseException if the parse fails
     * @throws DateTimeException if an error occurs while resolving the date or time
     * @throws IndexOutOfBoundsException if the position is invalid
     */
    private TemporalAccessor parseResolved0(final CharSequence text, final ParsePosition position) {
        ParsePosition pos = (position != null ? position : new ParsePosition(0));
        DateTimeParseContext context = parseUnresolved0(text, pos);
        if (context == null || pos.getErrorIndex() >= 0 || (position == null && pos.getIndex() < text.length())) {
            String abbr;
            if (text.length() > 64) {
                abbr = text.subSequence(0, 64).toString() + "...";
            } else {
                abbr = text.toString();
            }
            if (pos.getErrorIndex() >= 0) {
                throw new DateTimeParseException("Text '" + abbr + "' could not be parsed at index " +
                        pos.getErrorIndex(), text, pos.getErrorIndex());
            } else {
                throw new DateTimeParseException("Text '" + abbr + "' could not be parsed, unparsed text found at index " +
                        pos.getIndex(), text, pos.getIndex());
            }
        }
        return context.toResolved(resolverStyle, resolverFields);
    }

    /**
     * Parses the text using this formatter, without resolving the result, intended
     * for advanced use cases.
     * <p>
     * Parsing is implemented as a two-phase operation.
     * First, the text is parsed using the layout defined by the formatter, producing
     * a {@code Map} of field to value, a {@code ZoneId} and a {@code Chronology}.
     * Second, the parsed data is <em>resolved</em>, by validating, combining and
     * simplifying the various fields into more useful ones.
     * This method performs the parsing stage but not the resolving stage.
     * <p>
     * The result of this method is {@code TemporalAccessor} which represents the
     * data as seen in the input. Values are not validated, thus parsing a date string
     * of '2012-00-65' would result in a temporal with three fields - year of '2012',
     * month of '0' and day-of-month of '65'.
     * <p>
     * The text will be parsed from the specified start {@code ParsePosition}.
     * The entire length of the text does not have to be parsed, the {@code ParsePosition}
     * will be updated with the index at the end of parsing.
     * <p>
     * Errors are returned using the error index field of the {@code ParsePosition}
     * instead of {@code DateTimeParseException}.
     * The returned error index will be set to an index indicative of the error.
     * Callers must check for errors before using the result.
     * <p>
     * If the formatter parses the same field more than once with different values,
     * the result will be an error.
     * <p>
     * This method is intended for advanced use cases that need access to the
     * internal state during parsing. Typical application code should use
     * {@link #parse(CharSequence, TemporalQuery)} or the parse method on the target type.
     *
     * @param text  the text to parse, not null
     * @param position  the position to parse from, updated with length parsed
     *  and the index of any error, not null
     * @return the parsed text, null if the parse results in an error
     * @throws DateTimeException if some problem occurs during parsing
     * @throws IndexOutOfBoundsException if the position is invalid
     */
    public TemporalAccessor parseUnresolved(CharSequence text, ParsePosition position) {
        DateTimeParseContext context = parseUnresolved0(text, position);
        if (context == null) {
            return null;
        }
        return context.toUnresolved();
    }

    private DateTimeParseContext parseUnresolved0(CharSequence text, ParsePosition position) {
        Objects.requireNonNull(text, "text");
        Objects.requireNonNull(position, "position");
        DateTimeParseContext context = new DateTimeParseContext(this);
        int pos = position.getIndex();
        pos = printerParser.parse(context, text, pos);
        if (pos < 0) {
            position.setErrorIndex(~pos);  // index not updated from input
            return null;
        }
        position.setIndex(pos);  // errorIndex not updated from input
        return context;
    }

    //-----------------------------------------------------------------------
    /**
     * Returns the formatter as a composite printer parser.
     *
     * @param optional  whether the printer/parser should be optional
     * @return the printer/parser, not null
     */
    CompositePrinterParser toPrinterParser(boolean optional) {
        return printerParser.withOptional(optional);
    }

    /**
     * Returns this formatter as a {@code java.text.Format} instance.
     * <p>
     * The returned {@link Format} instance will format any {@link TemporalAccessor}
     * and parses to a resolved {@link TemporalAccessor}.
     * <p>
     * Exceptions will follow the definitions of {@code Format}, see those methods
     * for details about {@code IllegalArgumentException} during formatting and
     * {@code ParseException} or null during parsing.
     * The format does not support attributing of the returned format string.
     *
     * @return this formatter as a classic format instance, not null
     */
    public Format toFormat() {
        return new ClassicFormat(this, null);
    }

    /**
     * Returns this formatter as a {@code java.text.Format} instance that will
     * parse using the specified query.
     * <p>
     * The returned {@link Format} instance will format any {@link TemporalAccessor}
     * and parses to the type specified.
     * The type must be one that is supported by {@link #parse}.
     * <p>
     * Exceptions will follow the definitions of {@code Format}, see those methods
     * for details about {@code IllegalArgumentException} during formatting and
     * {@code ParseException} or null during parsing.
     * The format does not support attributing of the returned format string.
     *
     * @param parseQuery  the query defining the type to parse to, not null
     * @return this formatter as a classic format instance, not null
     */
    public Format toFormat(TemporalQuery<?> parseQuery) {
        Objects.requireNonNull(parseQuery, "parseQuery");
        return new ClassicFormat(this, parseQuery);
    }

    //-----------------------------------------------------------------------
    /**
     * Returns a description of the underlying formatters.
     *
     * @return a description of this formatter, not null
     */
    @Override
    public String toString() {
        String pattern = printerParser.toString();
        pattern = pattern.startsWith("[") ? pattern : pattern.substring(1, pattern.length() - 1);
        return pattern;
        // TODO: Fix tests to not depend on toString()
//        return "DateTimeFormatter[" + locale +
//                (chrono != null ? "," + chrono : "") +
//                (zone != null ? "," + zone : "") +
//                pattern + "]";
    }

    //-----------------------------------------------------------------------
    /**
     * Implements the classic Java Format API.
     * @serial exclude
     */
    @SuppressWarnings("serial")  // not actually serializable
    static class ClassicFormat extends Format {
        /** The formatter. */
        private final DateTimeFormatter formatter;
        /** The type to be parsed. */
        private final TemporalQuery<?> parseType;
        /** Constructor. */
        public ClassicFormat(DateTimeFormatter formatter, TemporalQuery<?> parseType) {
            this.formatter = formatter;
            this.parseType = parseType;
        }

        @Override
        public StringBuffer format(Object obj, StringBuffer toAppendTo, FieldPosition pos) {
            Objects.requireNonNull(obj, "obj");
            Objects.requireNonNull(toAppendTo, "toAppendTo");
            Objects.requireNonNull(pos, "pos");
            if (obj instanceof TemporalAccessor == false) {
                throw new IllegalArgumentException("Format target must implement TemporalAccessor");
            }
            pos.setBeginIndex(0);
            pos.setEndIndex(0);
            try {
                formatter.formatTo((TemporalAccessor) obj, toAppendTo);
            } catch (RuntimeException ex) {
                throw new IllegalArgumentException(ex.getMessage(), ex);
            }
            return toAppendTo;
        }
        @Override
        public Object parseObject(String text) throws ParseException {
            Objects.requireNonNull(text, "text");
            try {
                if (parseType == null) {
                    return formatter.parseResolved0(text, null);
                }
                return formatter.parse(text, parseType);
            } catch (DateTimeParseException ex) {
                throw new ParseException(ex.getMessage(), ex.getErrorIndex());
            } catch (RuntimeException ex) {
                throw (ParseException) new ParseException(ex.getMessage(), 0).initCause(ex);
            }
        }
        @Override
        public Object parseObject(String text, ParsePosition pos) {
            Objects.requireNonNull(text, "text");
            DateTimeParseContext context;
            try {
                context = formatter.parseUnresolved0(text, pos);
            } catch (IndexOutOfBoundsException ex) {
                if (pos.getErrorIndex() < 0) {
                    pos.setErrorIndex(0);
                }
                return null;
            }
            if (context == null) {
                if (pos.getErrorIndex() < 0) {
                    pos.setErrorIndex(0);
                }
                return null;
            }
            try {
                TemporalAccessor resolved = context.toResolved(formatter.resolverStyle, formatter.resolverFields);
                if (parseType == null) {
                    return resolved;
                }
                return resolved.query(parseType);
            } catch (RuntimeException ex) {
                pos.setErrorIndex(0);
                return null;
            }
        }
    }

}