import logging
import os
import shutil
from optparse import Values
from typing import List

from pip._internal.cache import WheelCache
from pip._internal.cli import cmdoptions
from pip._internal.cli.req_command import RequirementCommand, with_cleanup
from pip._internal.cli.status_codes import SUCCESS
from pip._internal.exceptions import CommandError
from pip._internal.operations.build.build_tracker import get_build_tracker
from pip._internal.req.req_install import (
    InstallRequirement,
    check_legacy_setup_py_options,
)
from pip._internal.utils.misc import ensure_dir, normalize_path
from pip._internal.utils.temp_dir import TempDirectory
from pip._internal.wheel_builder import build, should_build_for_wheel_command

logger = logging.getLogger(__name__)


class WheelCommand(RequirementCommand):
    """
    Build Wheel archives for your requirements and dependencies.

    Wheel is a built-package format, and offers the advantage of not
    recompiling your software during every install. For more details, see the
    wheel docs: https://wheel.readthedocs.io/en/latest/

    'pip wheel' uses the build system interface as described here:
    https://pip.pypa.io/en/stable/reference/build-system/

    """

    usage = """
      %prog [options] <requirement specifier> ...
      %prog [options] -r <requirements file> ...
      %prog [options] [-e] <vcs project url> ...
      %prog [options] [-e] <local project path> ...
      %prog [options] <archive url/path> ..."""

    def add_options(self) -> None:
        self.cmd_opts.add_option(
            "-w",
            "--wheel-dir",
            dest="wheel_dir",
            metavar="dir",
            default=os.curdir,
            help=(
                "Build wheels into <dir>, where the default is the "
                "current working directory."
            ),
        )
        self.cmd_opts.add_option(cmdoptions.no_binary())
        self.cmd_opts.add_option(cmdoptions.only_binary())
        self.cmd_opts.add_option(cmdoptions.prefer_binary())
        self.cmd_opts.add_option(cmdoptions.no_build_isolation())
        self.cmd_opts.add_option(cmdoptions.use_pep517())
        self.cmd_opts.add_option(cmdoptions.no_use_pep517())
        self.cmd_opts.add_option(cmdoptions.check_build_deps())
        self.cmd_opts.add_option(cmdoptions.constraints())
        self.cmd_opts.add_option(cmdoptions.editable())
        self.cmd_opts.add_option(cmdoptions.requirements())
        self.cmd_opts.add_option(cmdoptions.src())
        self.cmd_opts.add_option(cmdoptions.ignore_requires_python())
        self.cmd_opts.add_option(cmdoptions.no_deps())
        self.cmd_opts.add_option(cmdoptions.progress_bar())

        self.cmd_opts.add_option(
            "--no-verify",
            dest="no_verify",
            action="store_true",
            default=False,
            help="Don't verify if built wheel is valid.",
        )

        self.cmd_opts.add_option(cmdoptions.config_settings())
        self.cmd_opts.add_option(cmdoptions.build_options())
        self.cmd_opts.add_option(cmdoptions.global_options())

        self.cmd_opts.add_option(
            "--pre",
            action="store_true",
            default=False,
            help=(
                "Include pre-release and development versions. By default, "
                "pip only finds stable versions."
            ),
        )

        self.cmd_opts.add_option(cmdoptions.require_hashes())

        index_opts = cmdoptions.make_option_group(
            cmdoptions.index_group,
            self.parser,
        )

        self.parser.insert_option_group(0, index_opts)
        self.parser.insert_option_group(0, self.cmd_opts)

    @with_cleanup
    def run(self, options: Values, args: List[str]) -> int:
        session = self.get_default_session(options)

        finder = self._build_package_finder(options, session)

        options.wheel_dir = normalize_path(options.wheel_dir)
        ensure_dir(options.wheel_dir)

        build_tracker = self.enter_context(get_build_tracker())

        directory = TempDirectory(
            delete=not options.no_clean,
            kind="wheel",
            globally_managed=True,
        )

        reqs = self.get_requirements(args, options, finder, session)
        check_legacy_setup_py_options(options, reqs)

        wheel_cache = WheelCache(options.cache_dir)

        preparer = self.make_requirement_preparer(
            temp_build_dir=directory,
            options=options,
            build_tracker=build_tracker,
            session=session,
            finder=finder,
            download_dir=options.wheel_dir,
            use_user_site=False,
            verbosity=self.verbosity,
        )

        resolver = self.make_resolver(
            preparer=preparer,
            finder=finder,
            options=options,
            wheel_cache=wheel_cache,
            ignore_requires_python=options.ignore_requires_python,
            use_pep517=options.use_pep517,
        )

        self.trace_basic_info(finder)

        requirement_set = resolver.resolve(reqs, check_supported_wheels=True)

        reqs_to_build: List[InstallRequirement] = []
        for req in requirement_set.requirements.values():
            if req.is_wheel:
                preparer.save_linked_requirement(req)
            elif should_build_for_wheel_command(req):
                reqs_to_build.append(req)

        preparer.prepare_linked_requirements_more(requirement_set.requirements.values())

        # build wheels
        build_successes, build_failures = build(
            reqs_to_build,
            wheel_cache=wheel_cache,
            verify=(not options.no_verify),
            build_options=options.build_options or [],
            global_options=options.global_options or [],
        )
        for req in build_successes:
            assert req.link and req.link.is_wheel
            assert req.local_file_path
            # copy from cache to target directory
            try:
                shutil.copy(req.local_file_path, options.wheel_dir)
            except OSError as e:
                logger.warning(
                    "Building wheel for %s failed: %s",
                    req.name,
                    e,
                )
                build_failures.append(req)
        if len(build_failures) != 0:
            raise CommandError("Failed to build one or more wheels")

        return SUCCESS

"""
    pygments.formatters.pangomarkup
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

    Formatter for Pango markup output.

    :copyright: Copyright 2006-2024 by the Pygments team, see AUTHORS.
    :license: BSD, see LICENSE for details.
"""

from pip._vendor.pygments.formatter import Formatter


__all__ = ['PangoMarkupFormatter']


_escape_table = {
    ord('&'): '&',
    ord('<'): '&lt;',
}


def escape_special_chars(text, table=_escape_table):
    """Escape & and < for Pango Markup."""
    return text.translate(table)


class PangoMarkupFormatter(Formatter):
    """
    Format tokens as Pango Markup code. It can then be rendered to an SVG.

    .. versionadded:: 2.9
    """

    name = 'Pango Markup'
    aliases = ['pango', 'pangomarkup']
    filenames = []

    def __init__(self, **options):
        Formatter.__init__(self, **options)

        self.styles = {}

        for token, style in self.style:
            start = ''
            end = ''
            if style['color']:
                start += '<span fgcolor="#{}">'.format(style['color'])
                end = '</span>' + end
            if style['bold']:
                start += '<b>'
                end = '</b>' + end
            if style['italic']:
                start += '<i>'
                end = '</i>' + end
            if style['underline']:
                start += '<u>'
                end = '</u>' + end
            self.styles[token] = (start, end)

    def format_unencoded(self, tokensource, outfile):
        lastval = ''
        lasttype = None

        outfile.write('<tt>')

        for ttype, value in tokensource:
            while ttype not in self.styles:
                ttype = ttype.parent
            if ttype == lasttype:
                lastval += escape_special_chars(value)
            else:
                if lastval:
                    stylebegin, styleend = self.styles[lasttype]
                    outfile.write(stylebegin + lastval + styleend)
                lastval = escape_special_chars(value)
                lasttype = ttype

        if lastval:
            stylebegin, styleend = self.styles[lasttype]
            outfile.write(stylebegin + lastval + styleend)

        outfile.write('</tt>')

import sys
from optparse import Values
from typing import AbstractSet, List

from pip._internal.cli import cmdoptions
from pip._internal.cli.base_command import Command
from pip._internal.cli.status_codes import SUCCESS
from pip._internal.operations.freeze import freeze
from pip._internal.utils.compat import stdlib_pkgs


def _should_suppress_build_backends() -> bool:
    return sys.version_info < (3, 12)


def _dev_pkgs() -> AbstractSet[str]:
    pkgs = {"pip"}

    if _should_suppress_build_backends():
        pkgs |= {"setuptools", "distribute", "wheel"}

    return pkgs


class FreezeCommand(Command):
    """
    Output installed packages in requirements format.

    packages are listed in a case-insensitive sorted order.
    """

    ignore_require_venv = True
    usage = """
      %prog [options]"""
    log_streams = ("ext://sys.stderr", "ext://sys.stderr")

    def add_options(self) -> None:
        self.cmd_opts.add_option(
            "-r",
            "--requirement",
            dest="requirements",
            action="append",
            default=[],
            metavar="file",
            help=(
                "Use the order in the given requirements file and its "
                "comments when generating output. This option can be "
                "used multiple times."
            ),
        )
        self.cmd_opts.add_option(
            "-l",
            "--local",
            dest="local",
            action="store_true",
            default=False,
            help=(
                "If in a virtualenv that has global access, do not output "
                "globally-installed packages."
            ),
        )
        self.cmd_opts.add_option(
            "--user",
            dest="user",
            action="store_true",
            default=False,
            help="Only output packages installed in user-site.",
        )
        self.cmd_opts.add_option(cmdoptions.list_path())
        self.cmd_opts.add_option(
            "--all",
            dest="freeze_all",
            action="store_true",
            help=(
                "Do not skip these packages in the output:"
                " {}".format(", ".join(_dev_pkgs()))
            ),
        )
        self.cmd_opts.add_option(
            "--exclude-editable",
            dest="exclude_editable",
            action="store_true",
            help="Exclude editable package from output.",
        )
        self.cmd_opts.add_option(cmdoptions.list_exclude())

        self.parser.insert_option_group(0, self.cmd_opts)

    def run(self, options: Values, args: List[str]) -> int:
        skip = set(stdlib_pkgs)
        if not options.freeze_all:
            skip.update(_dev_pkgs())

        if options.excludes:
            skip.update(options.excludes)

        cmdoptions.check_list_path_option(options)

        for line in freeze(
            requirement=options.requirements,
            local_only=options.local,
            user_only=options.user,
            paths=options.path,
            isolated=options.isolated_mode,
            skip=skip,
            exclude_editable=options.exclude_editable,
        ):
            sys.stdout.write(line + "\n")
        return SUCCESS

"""
    pygments.lexer
    ~~~~~~~~~~~~~~

    Base lexer classes.

    :copyright: Copyright 2006-2024 by the Pygments team, see AUTHORS.
    :license: BSD, see LICENSE for details.
"""

import re
import sys
import time

from pip._vendor.pygments.filter import apply_filters, Filter
from pip._vendor.pygments.filters import get_filter_by_name
from pip._vendor.pygments.token import Error, Text, Other, Whitespace, _TokenType
from pip._vendor.pygments.util import get_bool_opt, get_int_opt, get_list_opt, \
    make_analysator, Future, guess_decode
from pip._vendor.pygments.regexopt import regex_opt

__all__ = ['Lexer', 'RegexLexer', 'ExtendedRegexLexer', 'DelegatingLexer',
           'LexerContext', 'include', 'inherit', 'bygroups', 'using', 'this',
           'default', 'words', 'line_re']

line_re = re.compile('.*?\n')

_encoding_map = [(b'\xef\xbb\xbf', 'utf-8'),
                 (b'\xff\xfe\0\0', 'utf-32'),
                 (b'\0\0\xfe\xff', 'utf-32be'),
                 (b'\xff\xfe', 'utf-16'),
                 (b'\xfe\xff', 'utf-16be')]

_default_analyse = staticmethod(lambda x: 0.0)


class LexerMeta(type):
    """
    This metaclass automagically converts ``analyse_text`` methods into
    static methods which always return float values.
    """

    def __new__(mcs, name, bases, d):
        if 'analyse_text' in d:
            d['analyse_text'] = make_analysator(d['analyse_text'])
        return type.__new__(mcs, name, bases, d)


class Lexer(metaclass=LexerMeta):
    """
    Lexer for a specific language.

    See also :doc:`lexerdevelopment`, a high-level guide to writing
    lexers.

    Lexer classes have attributes used for choosing the most appropriate
    lexer based on various criteria.

    .. autoattribute:: name
       :no-value:
    .. autoattribute:: aliases
       :no-value:
    .. autoattribute:: filenames
       :no-value:
    .. autoattribute:: alias_filenames
    .. autoattribute:: mimetypes
       :no-value:
    .. autoattribute:: priority

    Lexers included in Pygments should have two additional attributes:

    .. autoattribute:: url
       :no-value:
    .. autoattribute:: version_added
       :no-value:

    Lexers included in Pygments may have additional attributes:

    .. autoattribute:: _example
       :no-value:

    You can pass options to the constructor. The basic options recognized
    by all lexers and processed by the base `Lexer` class are:

    ``stripnl``
        Strip leading and trailing newlines from the input (default: True).
    ``stripall``
        Strip all leading and trailing whitespace from the input
        (default: False).
    ``ensurenl``
        Make sure that the input ends with a newline (default: True).  This
        is required for some lexers that consume input linewise.

        .. versionadded:: 1.3

    ``tabsize``
        If given and greater than 0, expand tabs in the input (default: 0).
    ``encoding``
        If given, must be an encoding name. This encoding will be used to
        convert the input string to Unicode, if it is not already a Unicode
        string (default: ``'guess'``, which uses a simple UTF-8 / Locale /
        Latin1 detection.  Can also be ``'chardet'`` to use the chardet
        library, if it is installed.
    ``inencoding``
        Overrides the ``encoding`` if given.
    """

    #: Full name of the lexer, in human-readable form
    name = None

    #: A list of short, unique identifiers that can be used to look
    #: up the lexer from a list, e.g., using `get_lexer_by_name()`.
    aliases = []

    #: A list of `fnmatch` patterns that match filenames which contain
    #: content for this lexer. The patterns in this list should be unique among
    #: all lexers.
    filenames = []

    #: A list of `fnmatch` patterns that match filenames which may or may not
    #: contain content for this lexer. This list is used by the
    #: :func:`.guess_lexer_for_filename()` function, to determine which lexers
    #: are then included in guessing the correct one. That means that
    #: e.g. every lexer for HTML and a template language should include
    #: ``\*.html`` in this list.
    alias_filenames = []

    #: A list of MIME types for content that can be lexed with this lexer.
    mimetypes = []

    #: Priority, should multiple lexers match and no content is provided
    priority = 0

    #: URL of the language specification/definition. Used in the Pygments
    #: documentation. Set to an empty string to disable.
    url = None

    #: Version of Pygments in which the lexer was added.
    version_added = None

    #: Example file name. Relative to the ``tests/examplefiles`` directory.
    #: This is used by the documentation generator to show an example.
    _example = None

    def __init__(self, **options):
        """
        This constructor takes arbitrary options as keyword arguments.
        Every subclass must first process its own options and then call
        the `Lexer` constructor, since it processes the basic
        options like `stripnl`.

        An example looks like this:

        .. sourcecode:: python

           def __init__(self, **options):
               self.compress = options.get('compress', '')
               Lexer.__init__(self, **options)

        As these options must all be specifiable as strings (due to the
        command line usage), there are various utility functions
        available to help with that, see `Utilities`_.
        """
        self.options = options
        self.stripnl = get_bool_opt(options, 'stripnl', True)
        self.stripall = get_bool_opt(options, 'stripall', False)
        self.ensurenl = get_bool_opt(options, 'ensurenl', True)
        self.tabsize = get_int_opt(options, 'tabsize', 0)
        self.encoding = options.get('encoding', 'guess')
        self.encoding = options.get('inencoding') or self.encoding
        self.filters = []
        for filter_ in get_list_opt(options, 'filters', ()):
            self.add_filter(filter_)

    def __repr__(self):
        if self.options:
            return f'<pygments.lexers.{self.__class__.__name__} with {self.options!r}>'
        else:
            return f'<pygments.lexers.{self.__class__.__name__}>'

    def add_filter(self, filter_, **options):
        """
        Add a new stream filter to this lexer.
        """
        if not isinstance(filter_, Filter):
            filter_ = get_filter_by_name(filter_, **options)
        self.filters.append(filter_)

    def analyse_text(text):
        """
        A static method which is called for lexer guessing.

        It should analyse the text and return a float in the range
        from ``0.0`` to ``1.0``.  If it returns ``0.0``, the lexer
        will not be selected as the most probable one, if it returns
        ``1.0``, it will be selected immediately.  This is used by
        `guess_lexer`.

        The `LexerMeta` metaclass automatically wraps this function so
        that it works like a static method (no ``self`` or ``cls``
        parameter) and the return value is automatically converted to
        `float`. If the return value is an object that is boolean `False`
        it's the same as if the return values was ``0.0``.
        """

    def _preprocess_lexer_input(self, text):
        """Apply preprocessing such as decoding the input, removing BOM and normalizing newlines."""

        if not isinstance(text, str):
            if self.encoding == 'guess':
                text, _ = guess_decode(text)
            elif self.encoding == 'chardet':
                try:
                    # pip vendoring note: this code is not reachable by pip,
                    # removed import of chardet to make it clear.
                    raise ImportError('chardet is not vendored by pip')
                except ImportError as e:
                    raise ImportError('To enable chardet encoding guessing, '
                                      'please install the chardet library '
                                      'from http://chardet.feedparser.org/') from e
                # check for BOM first
                decoded = None
                for bom, encoding in _encoding_map:
                    if text.startswith(bom):
                        decoded = text[len(bom):].decode(encoding, 'replace')
                        break
                # no BOM found, so use chardet
                if decoded is None:
                    enc = chardet.detect(text[:1024])  # Guess using first 1KB
                    decoded = text.decode(enc.get('encoding') or 'utf-8',
                                          'replace')
                text = decoded
            else:
                text = text.decode(self.encoding)
                if text.startswith('\ufeff'):
                    text = text[len('\ufeff'):]
        else:
            if text.startswith('\ufeff'):
                text = text[len('\ufeff'):]

        # text now *is* a unicode string
        text = text.replace('\r\n', '\n')
        text = text.replace('\r', '\n')
        if self.stripall:
            text = text.strip()
        elif self.stripnl:
            text = text.strip('\n')
        if self.tabsize > 0:
            text = text.expandtabs(self.tabsize)
        if self.ensurenl and not text.endswith('\n'):
            text += '\n'

        return text

    def get_tokens(self, text, unfiltered=False):
        """
        This method is the basic interface of a lexer. It is called by
        the `highlight()` function. It must process the text and return an
        iterable of ``(tokentype, value)`` pairs from `text`.

        Normally, you don't need to override this method. The default
        implementation processes the options recognized by all lexers
        (`stripnl`, `stripall` and so on), and then yields all tokens
        from `get_tokens_unprocessed()`, with the ``index`` dropped.

        If `unfiltered` is set to `True`, the filtering mechanism is
        bypassed even if filters are defined.
        """
        text = self._preprocess_lexer_input(text)

        def streamer():
            for _, t, v in self.get_tokens_unprocessed(text):
                yield t, v
        stream = streamer()
        if not unfiltered:
            stream = apply_filters(stream, self.filters, self)
        return stream

    def get_tokens_unprocessed(self, text):
        """
        This method should process the text and return an iterable of
        ``(index, tokentype, value)`` tuples where ``index`` is the starting
        position of the token within the input text.

        It must be overridden by subclasses. It is recommended to
        implement it as a generator to maximize effectiveness.
        """
        raise NotImplementedError


class DelegatingLexer(Lexer):
    """
    This lexer takes two lexer as arguments. A root lexer and
    a language lexer. First everything is scanned using the language
    lexer, afterwards all ``Other`` tokens are lexed using the root
    lexer.

    The lexers from the ``template`` lexer package use this base lexer.
    """

    def __init__(self, _root_lexer, _language_lexer, _needle=Other, **options):
        self.root_lexer = _root_lexer(**options)
        self.language_lexer = _language_lexer(**options)
        self.needle = _needle
        Lexer.__init__(self, **options)

    def get_tokens_unprocessed(self, text):
        buffered = ''
        insertions = []
        lng_buffer = []
        for i, t, v in self.language_lexer.get_tokens_unprocessed(text):
            if t is self.needle:
                if lng_buffer:
                    insertions.append((len(buffered), lng_buffer))
                    lng_buffer = []
                buffered += v
            else:
                lng_buffer.append((i, t, v))
        if lng_buffer:
            insertions.append((len(buffered), lng_buffer))
        return do_insertions(insertions,
                             self.root_lexer.get_tokens_unprocessed(buffered))


# ------------------------------------------------------------------------------
# RegexLexer and ExtendedRegexLexer
#


class include(str):  # pylint: disable=invalid-name
    """
    Indicates that a state should include rules from another state.
    """
    pass


class _inherit:
    """
    Indicates the a state should inherit from its superclass.
    """
    def __repr__(self):
        return 'inherit'

inherit = _inherit()  # pylint: disable=invalid-name


class combined(tuple):  # pylint: disable=invalid-name
    """
    Indicates a state combined from multiple states.
    """

    def __new__(cls, *args):
        return tuple.__new__(cls, args)

    def __init__(self, *args):
        # tuple.__init__ doesn't do anything
        pass


class _PseudoMatch:
    """
    A pseudo match object constructed from a string.
    """

    def __init__(self, start, text):
        self._text = text
        self._start = start

    def start(self, arg=None):
        return self._start

    def end(self, arg=None):
        return self._start + len(self._text)

    def group(self, arg=None):
        if arg:
            raise IndexError('No such group')
        return self._text

    def groups(self):
        return (self._text,)

    def groupdict(self):
        return {}


def bygroups(*args):
    """
    Callback that yields multiple actions for each group in the match.
    """
    def callback(lexer, match, ctx=None):
        for i, action in enumerate(args):
            if action is None:
                continue
            elif type(action) is _TokenType:
                data = match.group(i + 1)
                if data:
                    yield match.start(i + 1), action, data
            else:
                data = match.group(i + 1)
                if data is not None:
                    if ctx:
                        ctx.pos = match.start(i + 1)
                    for item in action(lexer,
                                       _PseudoMatch(match.start(i + 1), data), ctx):
                        if item:
                            yield item
        if ctx:
            ctx.pos = match.end()
    return callback


class _This:
    """
    Special singleton used for indicating the caller class.
    Used by ``using``.
    """

this = _This()


def using(_other, **kwargs):
    """
    Callback that processes the match with a different lexer.

    The keyword arguments are forwarded to the lexer, except `state` which
    is handled separately.

    `state` specifies the state that the new lexer will start in, and can
    be an enumerable such as ('root', 'inline', 'string') or a simple
    string which is assumed to be on top of the root state.

    Note: For that to work, `_other` must not be an `ExtendedRegexLexer`.
    """
    gt_kwargs = {}
    if 'state' in kwargs:
        s = kwargs.pop('state')
        if isinstance(s, (list, tuple)):
            gt_kwargs['stack'] = s
        else:
            gt_kwargs['stack'] = ('root', s)

    if _other is this:
        def callback(lexer, match, ctx=None):
            # if keyword arguments are given the callback
            # function has to create a new lexer instance
            if kwargs:
                # XXX: cache that somehow
                kwargs.update(lexer.options)
                lx = lexer.__class__(**kwargs)
            else:
                lx = lexer
            s = match.start()
            for i, t, v in lx.get_tokens_unprocessed(match.group(), **gt_kwargs):
                yield i + s, t, v
            if ctx:
                ctx.pos = match.end()
    else:
        def callback(lexer, match, ctx=None):
            # XXX: cache that somehow
            kwargs.update(lexer.options)
            lx = _other(**kwargs)

            s = match.start()
            for i, t, v in lx.get_tokens_unprocessed(match.group(), **gt_kwargs):
                yield i + s, t, v
            if ctx:
                ctx.pos = match.end()
    return callback


class default:
    """
    Indicates a state or state action (e.g. #pop) to apply.
    For example default('#pop') is equivalent to ('', Token, '#pop')
    Note that state tuples may be used as well.

    .. versionadded:: 2.0
    """
    def __init__(self, state):
        self.state = state


class words(Future):
    """
    Indicates a list of literal words that is transformed into an optimized
    regex that matches any of the words.

    .. versionadded:: 2.0
    """
    def __init__(self, words, prefix='', suffix=''):
        self.words = words
        self.prefix = prefix
        self.suffix = suffix

    def get(self):
        return regex_opt(self.words, prefix=self.prefix, suffix=self.suffix)


class RegexLexerMeta(LexerMeta):
    """
    Metaclass for RegexLexer, creates the self._tokens attribute from
    self.tokens on the first instantiation.
    """

    def _process_regex(cls, regex, rflags, state):
        """Preprocess the regular expression component of a token definition."""
        if isinstance(regex, Future):
            regex = regex.get()
        return re.compile(regex, rflags).match

    def _process_token(cls, token):
        """Preprocess the token component of a token definition."""
        assert type(token) is _TokenType or callable(token), \
            f'token type must be simple type or callable, not {token!r}'
        return token

    def _process_new_state(cls, new_state, unprocessed, processed):
        """Preprocess the state transition action of a token definition."""
        if isinstance(new_state, str):
            # an existing state
            if new_state == '#pop':
                return -1
            elif new_state in unprocessed:
                return (new_state,)
            elif new_state == '#push':
                return new_state
            elif new_state[:5] == '#pop:':
                return -int(new_state[5:])
            else:
                assert False, f'unknown new state {new_state!r}'
        elif isinstance(new_state, combined):
            # combine a new state from existing ones
            tmp_state = '_tmp_%d' % cls._tmpname
            cls._tmpname += 1
            itokens = []
            for istate in new_state:
                assert istate != new_state, f'circular state ref {istate!r}'
                itokens.extend(cls._process_state(unprocessed,
                                                  processed, istate))
            processed[tmp_state] = itokens
            return (tmp_state,)
        elif isinstance(new_state, tuple):
            # push more than one state
            for istate in new_state:
                assert (istate in unprocessed or
                        istate in ('#pop', '#push')), \
                    'unknown new state ' + istate
            return new_state
        else:
            assert False, f'unknown new state def {new_state!r}'

    def _process_state(cls, unprocessed, processed, state):
        """Preprocess a single state definition."""
        assert isinstance(state, str), f"wrong state name {state!r}"
        assert state[0] != '#', f"invalid state name {state!r}"
        if state in processed:
            return processed[state]
        tokens = processed[state] = []
        rflags = cls.flags
        for tdef in unprocessed[state]:
            if isinstance(tdef, include):
                # it's a state reference
                assert tdef != state, f"circular state reference {state!r}"
                tokens.extend(cls._process_state(unprocessed, processed,
                                                 str(tdef)))
                continue
            if isinstance(tdef, _inherit):
                # should be processed already, but may not in the case of:
                # 1. the state has no counterpart in any parent
                # 2. the state includes more than one 'inherit'
                continue
            if isinstance(tdef, default):
                new_state = cls._process_new_state(tdef.state, unprocessed, processed)
                tokens.append((re.compile('').match, None, new_state))
                continue

            assert type(tdef) is tuple, f"wrong rule def {tdef!r}"

            try:
                rex = cls._process_regex(tdef[0], rflags, state)
            except Exception as err:
                raise ValueError(f"uncompilable regex {tdef[0]!r} in state {state!r} of {cls!r}: {err}") from err

            token = cls._process_token(tdef[1])

            if len(tdef) == 2:
                new_state = None
            else:
                new_state = cls._process_new_state(tdef[2],
                                                   unprocessed, processed)

            tokens.append((rex, token, new_state))
        return tokens

    def process_tokendef(cls, name, tokendefs=None):
        """Preprocess a dictionary of token definitions."""
        processed = cls._all_tokens[name] = {}
        tokendefs = tokendefs or cls.tokens[name]
        for state in list(tokendefs):
            cls._process_state(tokendefs, processed, state)
        return processed

    def get_tokendefs(cls):
        """
        Merge tokens from superclasses in MRO order, returning a single tokendef
        dictionary.

        Any state that is not defined by a subclass will be inherited
        automatically.  States that *are* defined by subclasses will, by
        default, override that state in the superclass.  If a subclass wishes to
        inherit definitions from a superclass, it can use the special value
        "inherit", which will cause the superclass' state definition to be
        included at that point in the state.
        """
        tokens = {}
        inheritable = {}
        for c in cls.__mro__:
            toks = c.__dict__.get('tokens', {})

            for state, items in toks.items():
                curitems = tokens.get(state)
                if curitems is None:
                    # N.b. because this is assigned by reference, sufficiently
                    # deep hierarchies are processed incrementally (e.g. for
                    # A(B), B(C), C(RegexLexer), B will be premodified so X(B)
                    # will not see any inherits in B).
                    tokens[state] = items
                    try:
                        inherit_ndx = items.index(inherit)
                    except ValueError:
                        continue
                    inheritable[state] = inherit_ndx
                    continue

                inherit_ndx = inheritable.pop(state, None)
                if inherit_ndx is None:
                    continue

                # Replace the "inherit" value with the items
                curitems[inherit_ndx:inherit_ndx+1] = items
                try:
                    # N.b. this is the index in items (that is, the superclass
                    # copy), so offset required when storing below.
                    new_inh_ndx = items.index(inherit)
                except ValueError:
                    pass
                else:
                    inheritable[state] = inherit_ndx + new_inh_ndx

        return tokens

    def __call__(cls, *args, **kwds):
        """Instantiate cls after preprocessing its token definitions."""
        if '_tokens' not in cls.__dict__:
            cls._all_tokens = {}
            cls._tmpname = 0
            if hasattr(cls, 'token_variants') and cls.token_variants:
                # don't process yet
                pass
            else:
                cls._tokens = cls.process_tokendef('', cls.get_tokendefs())

        return type.__call__(cls, *args, **kwds)


class RegexLexer(Lexer, metaclass=RegexLexerMeta):
    """
    Base for simple stateful regular expression-based lexers.
    Simplifies the lexing process so that you need only
    provide a list of states and regular expressions.
    """

    #: Flags for compiling the regular expressions.
    #: Defaults to MULTILINE.
    flags = re.MULTILINE

    #: At all time there is a stack of states. Initially, the stack contains
    #: a single state 'root'. The top of the stack is called "the current state".
    #:
    #: Dict of ``{'state': [(regex, tokentype, new_state), ...], ...}``
    #:
    #: ``new_state`` can be omitted to signify no state transition.
    #: If ``new_state`` is a string, it is pushed on the stack. This ensure
    #: the new current state is ``new_state``.
    #: If ``new_state`` is a tuple of strings, all of those strings are pushed
    #: on the stack and the current state will be the last element of the list.
    #: ``new_state`` can also be ``combined('state1', 'state2', ...)``
    #: to signify a new, anonymous state combined from the rules of two
    #: or more existing ones.
    #: Furthermore, it can be '#pop' to signify going back one step in
    #: the state stack, or '#push' to push the current state on the stack
    #: again. Note that if you push while in a combined state, the combined
    #: state itself is pushed, and not only the state in which the rule is
    #: defined.
    #:
    #: The tuple can also be replaced with ``include('state')``, in which
    #: case the rules from the state named by the string are included in the
    #: current one.
    tokens = {}

    def get_tokens_unprocessed(self, text, stack=('root',)):
        """
        Split ``text`` into (tokentype, text) pairs.

        ``stack`` is the initial stack (default: ``['root']``)
        """
        pos = 0
        tokendefs = self._tokens
        statestack = list(stack)
        statetokens = tokendefs[statestack[-1]]
        while 1:
            for rexmatch, action, new_state in statetokens:
                m = rexmatch(text, pos)
                if m:
                    if action is not None:
                        if type(action) is _TokenType:
                            yield pos, action, m.group()
                        else:
                            yield from action(self, m)
                    pos = m.end()
                    if new_state is not None:
                        # state transition
                        if isinstance(new_state, tuple):
                            for state in new_state:
                                if state == '#pop':
                                    if len(statestack) > 1:
                                        statestack.pop()
                                elif state == '#push':
                                    statestack.append(statestack[-1])
                                else:
                                    statestack.append(state)
                        elif isinstance(new_state, int):
                            # pop, but keep at least one state on the stack
                            # (random code leading to unexpected pops should
                            # not allow exceptions)
                            if abs(new_state) >= len(statestack):
                                del statestack[1:]
                            else:
                                del statestack[new_state:]
                        elif new_state == '#push':
                            statestack.append(statestack[-1])
                        else:
                            assert False, f"wrong state def: {new_state!r}"
                        statetokens = tokendefs[statestack[-1]]
                    break
            else:
                # We are here only if all state tokens have been considered
                # and there was not a match on any of them.
                try:
                    if text[pos] == '\n':
                        # at EOL, reset state to "root"
                        statestack = ['root']
                        statetokens = tokendefs['root']
                        yield pos, Whitespace, '\n'
                        pos += 1
                        continue
                    yield pos, Error, text[pos]
                    pos += 1
                except IndexError:
                    break


class LexerContext:
    """
    A helper object that holds lexer position data.
    """

    def __init__(self, text, pos, stack=None, end=None):
        self.text = text
        self.pos = pos
        self.end = end or len(text)  # end=0 not supported ;-)
        self.stack = stack or ['root']

    def __repr__(self):
        return f'LexerContext({self.text!r}, {self.pos!r}, {self.stack!r})'


class ExtendedRegexLexer(RegexLexer):
    """
    A RegexLexer that uses a context object to store its state.
    """

    def get_tokens_unprocessed(self, text=None, context=None):
        """
        Split ``text`` into (tokentype, text) pairs.
        If ``context`` is given, use this lexer context instead.
        """
        tokendefs = self._tokens
        if not context:
            ctx = LexerContext(text, 0)
            statetokens = tokendefs['root']
        else:
            ctx = context
            statetokens = tokendefs[ctx.stack[-1]]
            text = ctx.text
        while 1:
            for rexmatch, action, new_state in statetokens:
                m = rexmatch(text, ctx.pos, ctx.end)
                if m:
                    if action is not None:
                        if type(action) is _TokenType:
                            yield ctx.pos, action, m.group()
                            ctx.pos = m.end()
                        else:
                            yield from action(self, m, ctx)
                            if not new_state:
                                # altered the state stack?
                                statetokens = tokendefs[ctx.stack[-1]]
                    # CAUTION: callback must set ctx.pos!
                    if new_state is not None:
                        # state transition
                        if isinstance(new_state, tuple):
                            for state in new_state:
                                if state == '#pop':
                                    if len(ctx.stack) > 1:
                                        ctx.stack.pop()
                                elif state == '#push':
                                    ctx.stack.append(ctx.stack[-1])
                                else:
                                    ctx.stack.append(state)
                        elif isinstance(new_state, int):
                            # see RegexLexer for why this check is made
                            if abs(new_state) >= len(ctx.stack):
                                del ctx.stack[1:]
                            else:
                                del ctx.stack[new_state:]
                        elif new_state == '#push':
                            ctx.stack.append(ctx.stack[-1])
                        else:
                            assert False, f"wrong state def: {new_state!r}"
                        statetokens = tokendefs[ctx.stack[-1]]
                    break
            else:
                try:
                    if ctx.pos >= ctx.end:
                        break
                    if text[ctx.pos] == '\n':
                        # at EOL, reset state to "root"
                        ctx.stack = ['root']
                        statetokens = tokendefs['root']
                        yield ctx.pos, Text, '\n'
                        ctx.pos += 1
                        continue
                    yield ctx.pos, Error, text[ctx.pos]
                    ctx.pos += 1
                except IndexError:
                    break


def do_insertions(insertions, tokens):
    """
    Helper for lexers which must combine the results of several
    sublexers.

    ``insertions`` is a list of ``(index, itokens)`` pairs.
    Each ``itokens`` iterable should be inserted at position
    ``index`` into the token stream given by the ``tokens``
    argument.

    The result is a combined token stream.

    TODO: clean up the code here.
    """
    insertions = iter(insertions)
    try:
        index, itokens = next(insertions)
    except StopIteration:
        # no insertions
        yield from tokens
        return

    realpos = None
    insleft = True

    # iterate over the token stream where we want to insert
    # the tokens from the insertion list.
    for i, t, v in tokens:
        # first iteration. store the position of first item
        if realpos is None:
            realpos = i
        oldi = 0
        while insleft and i + len(v) >= index:
            tmpval = v[oldi:index - i]
            if tmpval:
                yield realpos, t, tmpval
                realpos += len(tmpval)
            for it_index, it_token, it_value in itokens:
                yield realpos, it_token, it_value
                realpos += len(it_value)
            oldi = index - i
            try:
                index, itokens = next(insertions)
            except StopIteration:
                insleft = False
                break  # not strictly necessary
        if oldi < len(v):
            yield realpos, t, v[oldi:]
            realpos += len(v) - oldi

    # leftover tokens
    while insleft:
        # no normal tokens, set realpos to zero
        realpos = realpos or 0
        for p, t, v in itokens:
            yield realpos, t, v
            realpos += len(v)
        try:
            index, itokens = next(insertions)
        except StopIteration:
            insleft = False
            break  # not strictly necessary


class ProfilingRegexLexerMeta(RegexLexerMeta):
    """Metaclass for ProfilingRegexLexer, collects regex timing info."""

    def _process_regex(cls, regex, rflags, state):
        if isinstance(regex, words):
            rex = regex_opt(regex.words, prefix=regex.prefix,
                            suffix=regex.suffix)
        else:
            rex = regex
        compiled = re.compile(rex, rflags)

        def match_func(text, pos, endpos=sys.maxsize):
            info = cls._prof_data[-1].setdefault((state, rex), [0, 0.0])
            t0 = time.time()
            res = compiled.match(text, pos, endpos)
            t1 = time.time()
            info[0] += 1
            info[1] += t1 - t0
            return res
        return match_func


class ProfilingRegexLexer(RegexLexer, metaclass=ProfilingRegexLexerMeta):
    """Drop-in replacement for RegexLexer that does profiling of its regexes."""

    _prof_data = []
    _prof_sort_index = 4  # defaults to time per call

    def get_tokens_unprocessed(self, text, stack=('root',)):
        # this needs to be a stack, since using(this) will produce nested calls
        self.__class__._prof_data.append({})
        yield from RegexLexer.get_tokens_unprocessed(self, text, stack)
        rawdata = self.__class__._prof_data.pop()
        data = sorted(((s, repr(r).strip('u\'').replace('\\\\', '\\')[:65],
                        n, 1000 * t, 1000 * t / n)
                       for ((s, r), (n, t)) in rawdata.items()),
                      key=lambda x: x[self._prof_sort_index],
                      reverse=True)
        sum_total = sum(x[3] for x in data)

        print()
        print('Profiling result for %s lexing %d chars in %.3f ms' %
              (self.__class__.__name__, len(text), sum_total))
        print('=' * 110)
        print('%-20s %-64s ncalls  tottime  percall' % ('state', 'regex'))
        print('-' * 110)
        for d in data:
            print('%-20s %-65s %5d %8.4f %8.4f' % d)
        print('=' * 110)

from __future__ import annotations

import re
from typing import Iterable

from ._loop import loop_last
from .cells import cell_len, chop_cells

re_word = re.compile(r"\s*\S+\s*")


def words(text: str) -> Iterable[tuple[int, int, str]]:
    """Yields each word from the text as a tuple
    containing (start_index, end_index, word). A "word" in this context may
    include the actual word and any whitespace to the right.
    """
    position = 0
    word_match = re_word.match(text, position)
    while word_match is not None:
        start, end = word_match.span()
        word = word_match.group(0)
        yield start, end, word
        word_match = re_word.match(text, end)


def divide_line(text: str, width: int, fold: bool = True) -> list[int]:
    """Given a string of text, and a width (measured in cells), return a list
    of cell offsets which the string should be split at in order for it to fit
    within the given width.

    Args:
        text: The text to examine.
        width: The available cell width.
        fold: If True, words longer than `width` will be folded onto a new line.

    Returns:
        A list of indices to break the line at.
    """
    break_positions: list[int] = []  # offsets to insert the breaks at
    append = break_positions.append
    cell_offset = 0
    _cell_len = cell_len

    for start, _end, word in words(text):
        word_length = _cell_len(word.rstrip())
        remaining_space = width - cell_offset
        word_fits_remaining_space = remaining_space >= word_length

        if word_fits_remaining_space:
            # Simplest case - the word fits within the remaining width for this line.
            cell_offset += _cell_len(word)
        else:
            # Not enough space remaining for this word on the current line.
            if word_length > width:
                # The word doesn't fit on any line, so we can't simply
                # place it on the next line...
                if fold:
                    # Fold the word across multiple lines.
                    folded_word = chop_cells(word, width=width)
                    for last, line in loop_last(folded_word):
                        if start:
                            append(start)
                        if last:
                            cell_offset = _cell_len(line)
                        else:
                            start += len(line)
                else:
                    # Folding isn't allowed, so crop the word.
                    if start:
                        append(start)
                    cell_offset = _cell_len(word)
            elif cell_offset and start:
                # The word doesn't fit within the remaining space on the current
                # line, but it *can* fit on to the next (empty) line.
                append(start)
                cell_offset = _cell_len(word)

    return break_positions


if __name__ == "__main__":  # pragma: no cover
    from .console import Console

    console = Console(width=10)
    console.print("12345 abcdefghijklmnopqrstuvwyxzABCDEFGHIJKLMNOPQRSTUVWXYZ 12345")
    print(chop_cells("abcdefghijklmnopqrstuvwxyz", 10))

    console = Console(width=20)
    console.rule()
    console.print("TextualはPythonの高速アプリケーション開発フレームワークです")

    console.rule()
    console.print("アプリケーションは1670万色を使用でき")

from __future__ import annotations

import email.feedparser
import email.header
import email.message
import email.parser
import email.policy
import typing
from typing import (
    Any,
    Callable,
    Generic,
    Literal,
    TypedDict,
    cast,
)

from . import requirements, specifiers, utils
from . import version as version_module

T = typing.TypeVar("T")


try:
    ExceptionGroup
except NameError:  # pragma: no cover

    class ExceptionGroup(Exception):
        """A minimal implementation of :external:exc:`ExceptionGroup` from Python 3.11.

        If :external:exc:`ExceptionGroup` is already defined by Python itself,
        that version is used instead.
        """

        message: str
        exceptions: list[Exception]

        def __init__(self, message: str, exceptions: list[Exception]) -> None:
            self.message = message
            self.exceptions = exceptions

        def __repr__(self) -> str:
            return f"{self.__class__.__name__}({self.message!r}, {self.exceptions!r})"

else:  # pragma: no cover
    ExceptionGroup = ExceptionGroup


class InvalidMetadata(ValueError):
    """A metadata field contains invalid data."""

    field: str
    """The name of the field that contains invalid data."""

    def __init__(self, field: str, message: str) -> None:
        self.field = field
        super().__init__(message)


# The RawMetadata class attempts to make as few assumptions about the underlying
# serialization formats as possible. The idea is that as long as a serialization
# formats offer some very basic primitives in *some* way then we can support
# serializing to and from that format.
class RawMetadata(TypedDict, total=False):
    """A dictionary of raw core metadata.

    Each field in core metadata maps to a key of this dictionary (when data is
    provided). The key is lower-case and underscores are used instead of dashes
    compared to the equivalent core metadata field. Any core metadata field that
    can be specified multiple times or can hold multiple values in a single
    field have a key with a plural name. See :class:`Metadata` whose attributes
    match the keys of this dictionary.

    Core metadata fields that can be specified multiple times are stored as a
    list or dict depending on which is appropriate for the field. Any fields
    which hold multiple values in a single field are stored as a list.

    """

    # Metadata 1.0 - PEP 241
    metadata_version: str
    name: str
    version: str
    platforms: list[str]
    summary: str
    description: str
    keywords: list[str]
    home_page: str
    author: str
    author_email: str
    license: str

    # Metadata 1.1 - PEP 314
    supported_platforms: list[str]
    download_url: str
    classifiers: list[str]
    requires: list[str]
    provides: list[str]
    obsoletes: list[str]

    # Metadata 1.2 - PEP 345
    maintainer: str
    maintainer_email: str
    requires_dist: list[str]
    provides_dist: list[str]
    obsoletes_dist: list[str]
    requires_python: str
    requires_external: list[str]
    project_urls: dict[str, str]

    # Metadata 2.0
    # PEP 426 attempted to completely revamp the metadata format
    # but got stuck without ever being able to build consensus on
    # it and ultimately ended up withdrawn.
    #
    # However, a number of tools had started emitting METADATA with
    # `2.0` Metadata-Version, so for historical reasons, this version
    # was skipped.

    # Metadata 2.1 - PEP 566
    description_content_type: str
    provides_extra: list[str]

    # Metadata 2.2 - PEP 643
    dynamic: list[str]

    # Metadata 2.3 - PEP 685
    # No new fields were added in PEP 685, just some edge case were
    # tightened up to provide better interoptability.


_STRING_FIELDS = {
    "author",
    "author_email",
    "description",
    "description_content_type",
    "download_url",
    "home_page",
    "license",
    "maintainer",
    "maintainer_email",
    "metadata_version",
    "name",
    "requires_python",
    "summary",
    "version",
}

_LIST_FIELDS = {
    "classifiers",
    "dynamic",
    "obsoletes",
    "obsoletes_dist",
    "platforms",
    "provides",
    "provides_dist",
    "provides_extra",
    "requires",
    "requires_dist",
    "requires_external",
    "supported_platforms",
}

_DICT_FIELDS = {
    "project_urls",
}


def _parse_keywords(data: str) -> list[str]:
    """Split a string of comma-separate keyboards into a list of keywords."""
    return [k.strip() for k in data.split(",")]


def _parse_project_urls(data: list[str]) -> dict[str, str]:
    """Parse a list of label/URL string pairings separated by a comma."""
    urls = {}
    for pair in data:
        # Our logic is slightly tricky here as we want to try and do
        # *something* reasonable with malformed data.
        #
        # The main thing that we have to worry about, is data that does
        # not have a ',' at all to split the label from the Value. There
        # isn't a singular right answer here, and we will fail validation
        # later on (if the caller is validating) so it doesn't *really*
        # matter, but since the missing value has to be an empty str
        # and our return value is dict[str, str], if we let the key
        # be the missing value, then they'd have multiple '' values that
        # overwrite each other in a accumulating dict.
        #
        # The other potentional issue is that it's possible to have the
        # same label multiple times in the metadata, with no solid "right"
        # answer with what to do in that case. As such, we'll do the only
        # thing we can, which is treat the field as unparseable and add it
        # to our list of unparsed fields.
        parts = [p.strip() for p in pair.split(",", 1)]
        parts.extend([""] * (max(0, 2 - len(parts))))  # Ensure 2 items

        # TODO: The spec doesn't say anything about if the keys should be
        #       considered case sensitive or not... logically they should
        #       be case-preserving and case-insensitive, but doing that
        #       would open up more cases where we might have duplicate
        #       entries.
        label, url = parts
        if label in urls:
            # The label already exists in our set of urls, so this field
            # is unparseable, and we can just add the whole thing to our
            # unparseable data and stop processing it.
            raise KeyError("duplicate labels in project urls")
        urls[label] = url

    return urls


def _get_payload(msg: email.message.Message, source: bytes | str) -> str:
    """Get the body of the message."""
    # If our source is a str, then our caller has managed encodings for us,
    # and we don't need to deal with it.
    if isinstance(source, str):
        payload: str = msg.get_payload()
        return payload
    # If our source is a bytes, then we're managing the encoding and we need
    # to deal with it.
    else:
        bpayload: bytes = msg.get_payload(decode=True)
        try:
            return bpayload.decode("utf8", "strict")
        except UnicodeDecodeError:
            raise ValueError("payload in an invalid encoding")


# The various parse_FORMAT functions here are intended to be as lenient as
# possible in their parsing, while still returning a correctly typed
# RawMetadata.
#
# To aid in this, we also generally want to do as little touching of the
# data as possible, except where there are possibly some historic holdovers
# that make valid data awkward to work with.
#
# While this is a lower level, intermediate format than our ``Metadata``
# class, some light touch ups can make a massive difference in usability.

# Map METADATA fields to RawMetadata.
_EMAIL_TO_RAW_MAPPING = {
    "author": "author",
    "author-email": "author_email",
    "classifier": "classifiers",
    "description": "description",
    "description-content-type": "description_content_type",
    "download-url": "download_url",
    "dynamic": "dynamic",
    "home-page": "home_page",
    "keywords": "keywords",
    "license": "license",
    "maintainer": "maintainer",
    "maintainer-email": "maintainer_email",
    "metadata-version": "metadata_version",
    "name": "name",
    "obsoletes": "obsoletes",
    "obsoletes-dist": "obsoletes_dist",
    "platform": "platforms",
    "project-url": "project_urls",
    "provides": "provides",
    "provides-dist": "provides_dist",
    "provides-extra": "provides_extra",
    "requires": "requires",
    "requires-dist": "requires_dist",
    "requires-external": "requires_external",
    "requires-python": "requires_python",
    "summary": "summary",
    "supported-platform": "supported_platforms",
    "version": "version",
}
_RAW_TO_EMAIL_MAPPING = {raw: email for email, raw in _EMAIL_TO_RAW_MAPPING.items()}


def parse_email(data: bytes | str) -> tuple[RawMetadata, dict[str, list[str]]]:
    """Parse a distribution's metadata stored as email headers (e.g. from ``METADATA``).

    This function returns a two-item tuple of dicts. The first dict is of
    recognized fields from the core metadata specification. Fields that can be
    parsed and translated into Python's built-in types are converted
    appropriately. All other fields are left as-is. Fields that are allowed to
    appear multiple times are stored as lists.

    The second dict contains all other fields from the metadata. This includes
    any unrecognized fields. It also includes any fields which are expected to
    be parsed into a built-in type but were not formatted appropriately. Finally,
    any fields that are expected to appear only once but are repeated are
    included in this dict.

    """
    raw: dict[str, str | list[str] | dict[str, str]] = {}
    unparsed: dict[str, list[str]] = {}

    if isinstance(data, str):
        parsed = email.parser.Parser(policy=email.policy.compat32).parsestr(data)
    else:
        parsed = email.parser.BytesParser(policy=email.policy.compat32).parsebytes(data)

    # We have to wrap parsed.keys() in a set, because in the case of multiple
    # values for a key (a list), the key will appear multiple times in the
    # list of keys, but we're avoiding that by using get_all().
    for name in frozenset(parsed.keys()):
        # Header names in RFC are case insensitive, so we'll normalize to all
        # lower case to make comparisons easier.
        name = name.lower()

        # We use get_all() here, even for fields that aren't multiple use,
        # because otherwise someone could have e.g. two Name fields, and we
        # would just silently ignore it rather than doing something about it.
        headers = parsed.get_all(name) or []

        # The way the email module works when parsing bytes is that it
        # unconditionally decodes the bytes as ascii using the surrogateescape
        # handler. When you pull that data back out (such as with get_all() ),
        # it looks to see if the str has any surrogate escapes, and if it does
        # it wraps it in a Header object instead of returning the string.
        #
        # As such, we'll look for those Header objects, and fix up the encoding.
        value = []
        # Flag if we have run into any issues processing the headers, thus
        # signalling that the data belongs in 'unparsed'.
        valid_encoding = True
        for h in headers:
            # It's unclear if this can return more types than just a Header or
            # a str, so we'll just assert here to make sure.
            assert isinstance(h, (email.header.Header, str))

            # If it's a header object, we need to do our little dance to get
            # the real data out of it. In cases where there is invalid data
            # we're going to end up with mojibake, but there's no obvious, good
            # way around that without reimplementing parts of the Header object
            # ourselves.
            #
            # That should be fine since, if mojibacked happens, this key is
            # going into the unparsed dict anyways.
            if isinstance(h, email.header.Header):
                # The Header object stores it's data as chunks, and each chunk
                # can be independently encoded, so we'll need to check each
                # of them.
                chunks: list[tuple[bytes, str | None]] = []
                for bin, encoding in email.header.decode_header(h):
                    try:
                        bin.decode("utf8", "strict")
                    except UnicodeDecodeError:
                        # Enable mojibake.
                        encoding = "latin1"
                        valid_encoding = False
                    else:
                        encoding = "utf8"
                    chunks.append((bin, encoding))

                # Turn our chunks back into a Header object, then let that
                # Header object do the right thing to turn them into a
                # string for us.
                value.append(str(email.header.make_header(chunks)))
            # This is already a string, so just add it.
            else:
                value.append(h)

        # We've processed all of our values to get them into a list of str,
        # but we may have mojibake data, in which case this is an unparsed
        # field.
        if not valid_encoding:
            unparsed[name] = value
            continue

        raw_name = _EMAIL_TO_RAW_MAPPING.get(name)
        if raw_name is None:
            # This is a bit of a weird situation, we've encountered a key that
            # we don't know what it means, so we don't know whether it's meant
            # to be a list or not.
            #
            # Since we can't really tell one way or another, we'll just leave it
            # as a list, even though it may be a single item list, because that's
            # what makes the most sense for email headers.
            unparsed[name] = value
            continue

        # If this is one of our string fields, then we'll check to see if our
        # value is a list of a single item. If it is then we'll assume that
        # it was emitted as a single string, and unwrap the str from inside
        # the list.
        #
        # If it's any other kind of data, then we haven't the faintest clue
        # what we should parse it as, and we have to just add it to our list
        # of unparsed stuff.
        if raw_name in _STRING_FIELDS and len(value) == 1:
            raw[raw_name] = value[0]
        # If this is one of our list of string fields, then we can just assign
        # the value, since email *only* has strings, and our get_all() call
        # above ensures that this is a list.
        elif raw_name in _LIST_FIELDS:
            raw[raw_name] = value
        # Special Case: Keywords
        # The keywords field is implemented in the metadata spec as a str,
        # but it conceptually is a list of strings, and is serialized using
        # ", ".join(keywords), so we'll do some light data massaging to turn
        # this into what it logically is.
        elif raw_name == "keywords" and len(value) == 1:
            raw[raw_name] = _parse_keywords(value[0])
        # Special Case: Project-URL
        # The project urls is implemented in the metadata spec as a list of
        # specially-formatted strings that represent a key and a value, which
        # is fundamentally a mapping, however the email format doesn't support
        # mappings in a sane way, so it was crammed into a list of strings
        # instead.
        #
        # We will do a little light data massaging to turn this into a map as
        # it logically should be.
        elif raw_name == "project_urls":
            try:
                raw[raw_name] = _parse_project_urls(value)
            except KeyError:
                unparsed[name] = value
        # Nothing that we've done has managed to parse this, so it'll just
        # throw it in our unparseable data and move on.
        else:
            unparsed[name] = value

    # We need to support getting the Description from the message payload in
    # addition to getting it from the the headers. This does mean, though, there
    # is the possibility of it being set both ways, in which case we put both
    # in 'unparsed' since we don't know which is right.
    try:
        payload = _get_payload(parsed, data)
    except ValueError:
        unparsed.setdefault("description", []).append(
            parsed.get_payload(decode=isinstance(data, bytes))
        )
    else:
        if payload:
            # Check to see if we've already got a description, if so then both
            # it, and this body move to unparseable.
            if "description" in raw:
                description_header = cast(str, raw.pop("description"))
                unparsed.setdefault("description", []).extend(
                    [description_header, payload]
                )
            elif "description" in unparsed:
                unparsed["description"].append(payload)
            else:
                raw["description"] = payload

    # We need to cast our `raw` to a metadata, because a TypedDict only support
    # literal key names, but we're computing our key names on purpose, but the
    # way this function is implemented, our `TypedDict` can only have valid key
    # names.
    return cast(RawMetadata, raw), unparsed


_NOT_FOUND = object()


# Keep the two values in sync.
_VALID_METADATA_VERSIONS = ["1.0", "1.1", "1.2", "2.1", "2.2", "2.3"]
_MetadataVersion = Literal["1.0", "1.1", "1.2", "2.1", "2.2", "2.3"]

_REQUIRED_ATTRS = frozenset(["metadata_version", "name", "version"])


class _Validator(Generic[T]):
    """Validate a metadata field.

    All _process_*() methods correspond to a core metadata field. The method is
    called with the field's raw value. If the raw value is valid it is returned
    in its "enriched" form (e.g. ``version.Version`` for the ``Version`` field).
    If the raw value is invalid, :exc:`InvalidMetadata` is raised (with a cause
    as appropriate).
    """

    name: str
    raw_name: str
    added: _MetadataVersion

    def __init__(
        self,
        *,
        added: _MetadataVersion = "1.0",
    ) -> None:
        self.added = added

    def __set_name__(self, _owner: Metadata, name: str) -> None:
        self.name = name
        self.raw_name = _RAW_TO_EMAIL_MAPPING[name]

    def __get__(self, instance: Metadata, _owner: type[Metadata]) -> T:
        # With Python 3.8, the caching can be replaced with functools.cached_property().
        # No need to check the cache as attribute lookup will resolve into the
        # instance's __dict__ before __get__ is called.
        cache = instance.__dict__
        value = instance._raw.get(self.name)

        # To make the _process_* methods easier, we'll check if the value is None
        # and if this field is NOT a required attribute, and if both of those
        # things are true, we'll skip the the converter. This will mean that the
        # converters never have to deal with the None union.
        if self.name in _REQUIRED_ATTRS or value is not None:
            try:
                converter: Callable[[Any], T] = getattr(self, f"_process_{self.name}")
            except AttributeError:
                pass
            else:
                value = converter(value)

        cache[self.name] = value
        try:
            del instance._raw[self.name]  # type: ignore[misc]
        except KeyError:
            pass

        return cast(T, value)

    def _invalid_metadata(
        self, msg: str, cause: Exception | None = None
    ) -> InvalidMetadata:
        exc = InvalidMetadata(
            self.raw_name, msg.format_map({"field": repr(self.raw_name)})
        )
        exc.__cause__ = cause
        return exc

    def _process_metadata_version(self, value: str) -> _MetadataVersion:
        # Implicitly makes Metadata-Version required.
        if value not in _VALID_METADATA_VERSIONS:
            raise self._invalid_metadata(f"{value!r} is not a valid metadata version")
        return cast(_MetadataVersion, value)

    def _process_name(self, value: str) -> str:
        if not value:
            raise self._invalid_metadata("{field} is a required field")
        # Validate the name as a side-effect.
        try:
            utils.canonicalize_name(value, validate=True)
        except utils.InvalidName as exc:
            raise self._invalid_metadata(
                f"{value!r} is invalid for {{field}}", cause=exc
            )
        else:
            return value

    def _process_version(self, value: str) -> version_module.Version:
        if not value:
            raise self._invalid_metadata("{field} is a required field")
        try:
            return version_module.parse(value)
        except version_module.InvalidVersion as exc:
            raise self._invalid_metadata(
                f"{value!r} is invalid for {{field}}", cause=exc
            )

    def _process_summary(self, value: str) -> str:
        """Check the field contains no newlines."""
        if "\n" in value:
            raise self._invalid_metadata("{field} must be a single line")
        return value

    def _process_description_content_type(self, value: str) -> str:
        content_types = {"text/plain", "text/x-rst", "text/markdown"}
        message = email.message.EmailMessage()
        message["content-type"] = value

        content_type, parameters = (
            # Defaults to `text/plain` if parsing failed.
            message.get_content_type().lower(),
            message["content-type"].params,
        )
        # Check if content-type is valid or defaulted to `text/plain` and thus was
        # not parseable.
        if content_type not in content_types or content_type not in value.lower():
            raise self._invalid_metadata(
                f"{{field}} must be one of {list(content_types)}, not {value!r}"
            )

        charset = parameters.get("charset", "UTF-8")
        if charset != "UTF-8":
            raise self._invalid_metadata(
                f"{{field}} can only specify the UTF-8 charset, not {list(charset)}"
            )

        markdown_variants = {"GFM", "CommonMark"}
        variant = parameters.get("variant", "GFM")  # Use an acceptable default.
        if content_type == "text/markdown" and variant not in markdown_variants:
            raise self._invalid_metadata(
                f"valid Markdown variants for {{field}} are {list(markdown_variants)}, "
                f"not {variant!r}",
            )
        return value

    def _process_dynamic(self, value: list[str]) -> list[str]:
        for dynamic_field in map(str.lower, value):
            if dynamic_field in {"name", "version", "metadata-version"}:
                raise self._invalid_metadata(
                    f"{value!r} is not allowed as a dynamic field"
                )
            elif dynamic_field not in _EMAIL_TO_RAW_MAPPING:
                raise self._invalid_metadata(f"{value!r} is not a valid dynamic field")
        return list(map(str.lower, value))

    def _process_provides_extra(
        self,
        value: list[str],
    ) -> list[utils.NormalizedName]:
        normalized_names = []
        try:
            for name in value:
                normalized_names.append(utils.canonicalize_name(name, validate=True))
        except utils.InvalidName as exc:
            raise self._invalid_metadata(
                f"{name!r} is invalid for {{field}}", cause=exc
            )
        else:
            return normalized_names

    def _process_requires_python(self, value: str) -> specifiers.SpecifierSet:
        try:
            return specifiers.SpecifierSet(value)
        except specifiers.InvalidSpecifier as exc:
            raise self._invalid_metadata(
                f"{value!r} is invalid for {{field}}", cause=exc
            )

    def _process_requires_dist(
        self,
        value: list[str],
    ) -> list[requirements.Requirement]:
        reqs = []
        try:
            for req in value:
                reqs.append(requirements.Requirement(req))
        except requirements.InvalidRequirement as exc:
            raise self._invalid_metadata(f"{req!r} is invalid for {{field}}", cause=exc)
        else:
            return reqs


class Metadata:
    """Representation of distribution metadata.

    Compared to :class:`RawMetadata`, this class provides objects representing
    metadata fields instead of only using built-in types. Any invalid metadata
    will cause :exc:`InvalidMetadata` to be raised (with a
    :py:attr:`~BaseException.__cause__` attribute as appropriate).
    """

    _raw: RawMetadata

    @classmethod
    def from_raw(cls, data: RawMetadata, *, validate: bool = True) -> Metadata:
        """Create an instance from :class:`RawMetadata`.

        If *validate* is true, all metadata will be validated. All exceptions
        related to validation will be gathered and raised as an :class:`ExceptionGroup`.
        """
        ins = cls()
        ins._raw = data.copy()  # Mutations occur due to caching enriched values.

        if validate:
            exceptions: list[Exception] = []
            try:
                metadata_version = ins.metadata_version
                metadata_age = _VALID_METADATA_VERSIONS.index(metadata_version)
            except InvalidMetadata as metadata_version_exc:
                exceptions.append(metadata_version_exc)
                metadata_version = None

            # Make sure to check for the fields that are present, the required
            # fields (so their absence can be reported).
            fields_to_check = frozenset(ins._raw) | _REQUIRED_ATTRS
            # Remove fields that have already been checked.
            fields_to_check -= {"metadata_version"}

            for key in fields_to_check:
                try:
                    if metadata_version:
                        # Can't use getattr() as that triggers descriptor protocol which
                        # will fail due to no value for the instance argument.
                        try:
                            field_metadata_version = cls.__dict__[key].added
                        except KeyError:
                            exc = InvalidMetadata(key, f"unrecognized field: {key!r}")
                            exceptions.append(exc)
                            continue
                        field_age = _VALID_METADATA_VERSIONS.index(
                            field_metadata_version
                        )
                        if field_age > metadata_age:
                            field = _RAW_TO_EMAIL_MAPPING[key]
                            exc = InvalidMetadata(
                                field,
                                "{field} introduced in metadata version "
                                "{field_metadata_version}, not {metadata_version}",
                            )
                            exceptions.append(exc)
                            continue
                    getattr(ins, key)
                except InvalidMetadata as exc:
                    exceptions.append(exc)

            if exceptions:
                raise ExceptionGroup("invalid metadata", exceptions)

        return ins

    @classmethod
    def from_email(cls, data: bytes | str, *, validate: bool = True) -> Metadata:
        """Parse metadata from email headers.

        If *validate* is true, the metadata will be validated. All exceptions
        related to validation will be gathered and raised as an :class:`ExceptionGroup`.
        """
        raw, unparsed = parse_email(data)

        if validate:
            exceptions: list[Exception] = []
            for unparsed_key in unparsed:
                if unparsed_key in _EMAIL_TO_RAW_MAPPING:
                    message = f"{unparsed_key!r} has invalid data"
                else:
                    message = f"unrecognized field: {unparsed_key!r}"
                exceptions.append(InvalidMetadata(unparsed_key, message))

            if exceptions:
                raise ExceptionGroup("unparsed", exceptions)

        try:
            return cls.from_raw(raw, validate=validate)
        except ExceptionGroup as exc_group:
            raise ExceptionGroup(
                "invalid or unparsed metadata", exc_group.exceptions
            ) from None

    metadata_version: _Validator[_MetadataVersion] = _Validator()
    """:external:ref:`core-metadata-metadata-version`
    (required; validated to be a valid metadata version)"""
    name: _Validator[str] = _Validator()
    """:external:ref:`core-metadata-name`
    (required; validated using :func:`~packaging.utils.canonicalize_name` and its
    *validate* parameter)"""
    version: _Validator[version_module.Version] = _Validator()
    """:external:ref:`core-metadata-version` (required)"""
    dynamic: _Validator[list[str] | None] = _Validator(
        added="2.2",
    )
    """:external:ref:`core-metadata-dynamic`
    (validated against core metadata field names and lowercased)"""
    platforms: _Validator[list[str] | None] = _Validator()
    """:external:ref:`core-metadata-platform`"""
    supported_platforms: _Validator[list[str] | None] = _Validator(added="1.1")
    """:external:ref:`core-metadata-supported-platform`"""
    summary: _Validator[str | None] = _Validator()
    """:external:ref:`core-metadata-summary` (validated to contain no newlines)"""
    description: _Validator[str | None] = _Validator()  # TODO 2.1: can be in body
    """:external:ref:`core-metadata-description`"""
    description_content_type: _Validator[str | None] = _Validator(added="2.1")
    """:external:ref:`core-metadata-description-content-type` (validated)"""
    keywords: _Validator[list[str] | None] = _Validator()
    """:external:ref:`core-metadata-keywords`"""
    home_page: _Validator[str | None] = _Validator()
    """:external:ref:`core-metadata-home-page`"""
    download_url: _Validator[str | None] = _Validator(added="1.1")
    """:external:ref:`core-metadata-download-url`"""
    author: _Validator[str | None] = _Validator()
    """:external:ref:`core-metadata-author`"""
    author_email: _Validator[str | None] = _Validator()
    """:external:ref:`core-metadata-author-email`"""
    maintainer: _Validator[str | None] = _Validator(added="1.2")
    """:external:ref:`core-metadata-maintainer`"""
    maintainer_email: _Validator[str | None] = _Validator(added="1.2")
    """:external:ref:`core-metadata-maintainer-email`"""
    license: _Validator[str | None] = _Validator()
    """:external:ref:`core-metadata-license`"""
    classifiers: _Validator[list[str] | None] = _Validator(added="1.1")
    """:external:ref:`core-metadata-classifier`"""
    requires_dist: _Validator[list[requirements.Requirement] | None] = _Validator(
        added="1.2"
    )
    """:external:ref:`core-metadata-requires-dist`"""
    requires_python: _Validator[specifiers.SpecifierSet | None] = _Validator(
        added="1.2"
    )
    """:external:ref:`core-metadata-requires-python`"""
    # Because `Requires-External` allows for non-PEP 440 version specifiers, we
    # don't do any processing on the values.
    requires_external: _Validator[list[str] | None] = _Validator(added="1.2")
    """:external:ref:`core-metadata-requires-external`"""
    project_urls: _Validator[dict[str, str] | None] = _Validator(added="1.2")
    """:external:ref:`core-metadata-project-url`"""
    # PEP 685 lets us raise an error if an extra doesn't pass `Name` validation
    # regardless of metadata version.
    provides_extra: _Validator[list[utils.NormalizedName] | None] = _Validator(
        added="2.1",
    )
    """:external:ref:`core-metadata-provides-extra`"""
    provides_dist: _Validator[list[str] | None] = _Validator(added="1.2")
    """:external:ref:`core-metadata-provides-dist`"""
    obsoletes_dist: _Validator[list[str] | None] = _Validator(added="1.2")
    """:external:ref:`core-metadata-obsoletes-dist`"""
    requires: _Validator[list[str] | None] = _Validator(added="1.1")
    """``Requires`` (deprecated)"""
    provides: _Validator[list[str] | None] = _Validator(added="1.1")
    """``Provides`` (deprecated)"""
    obsoletes: _Validator[list[str] | None] = _Validator(added="1.1")
    """``Obsoletes`` (deprecated)"""

from types import TracebackType
from typing import Optional, Type

from .console import Console, RenderableType
from .jupyter import JupyterMixin
from .live import Live
from .spinner import Spinner
from .style import StyleType


class Status(JupyterMixin):
    """Displays a status indicator with a 'spinner' animation.

    Args:
        status (RenderableType): A status renderable (str or Text typically).
        console (Console, optional): Console instance to use, or None for global console. Defaults to None.
        spinner (str, optional): Name of spinner animation (see python -m rich.spinner). Defaults to "dots".
        spinner_style (StyleType, optional): Style of spinner. Defaults to "status.spinner".
        speed (float, optional): Speed factor for spinner animation. Defaults to 1.0.
        refresh_per_second (float, optional): Number of refreshes per second. Defaults to 12.5.
    """

    def __init__(
        self,
        status: RenderableType,
        *,
        console: Optional[Console] = None,
        spinner: str = "dots",
        spinner_style: StyleType = "status.spinner",
        speed: float = 1.0,
        refresh_per_second: float = 12.5,
    ):
        self.status = status
        self.spinner_style = spinner_style
        self.speed = speed
        self._spinner = Spinner(spinner, text=status, style=spinner_style, speed=speed)
        self._live = Live(
            self.renderable,
            console=console,
            refresh_per_second=refresh_per_second,
            transient=True,
        )

    @property
    def renderable(self) -> Spinner:
        return self._spinner

    @property
    def console(self) -> "Console":
        """Get the Console used by the Status objects."""
        return self._live.console

    def update(
        self,
        status: Optional[RenderableType] = None,
        *,
        spinner: Optional[str] = None,
        spinner_style: Optional[StyleType] = None,
        speed: Optional[float] = None,
    ) -> None:
        """Update status.

        Args:
            status (Optional[RenderableType], optional): New status renderable or None for no change. Defaults to None.
            spinner (Optional[str], optional): New spinner or None for no change. Defaults to None.
            spinner_style (Optional[StyleType], optional): New spinner style or None for no change. Defaults to None.
            speed (Optional[float], optional): Speed factor for spinner animation or None for no change. Defaults to None.
        """
        if status is not None:
            self.status = status
        if spinner_style is not None:
            self.spinner_style = spinner_style
        if speed is not None:
            self.speed = speed
        if spinner is not None:
            self._spinner = Spinner(
                spinner, text=self.status, style=self.spinner_style, speed=self.speed
            )
            self._live.update(self.renderable, refresh=True)
        else:
            self._spinner.update(
                text=self.status, style=self.spinner_style, speed=self.speed
            )

    def start(self) -> None:
        """Start the status animation."""
        self._live.start()

    def stop(self) -> None:
        """Stop the spinner animation."""
        self._live.stop()

    def __rich__(self) -> RenderableType:
        return self.renderable

    def __enter__(self) -> "Status":
        self.start()
        return self

    def __exit__(
        self,
        exc_type: Optional[Type[BaseException]],
        exc_val: Optional[BaseException],
        exc_tb: Optional[TracebackType],
    ) -> None:
        self.stop()


if __name__ == "__main__":  # pragma: no cover
    from time import sleep

    from .console import Console

    console = Console()
    with console.status("[magenta]Covid detector booting up") as status:
        sleep(3)
        console.log("Importing advanced AI")
        sleep(3)
        console.log("Advanced Covid AI Ready")
        sleep(3)
        status.update(status="[bold blue] Scanning for Covid", spinner="earth")
        sleep(3)
        console.log("Found 10,000,000,000 copies of Covid32.exe")
        sleep(3)
        status.update(
            status="[bold red]Moving Covid32.exe to Trash",
            spinner="bouncingBall",
            spinner_style="yellow",
        )
        sleep(5)
    console.print("[bold green]Covid deleted successfully")

from typing import Callable, Match, Optional
import re

from ._emoji_codes import EMOJI


_ReStringMatch = Match[str]  # regex match object
_ReSubCallable = Callable[[_ReStringMatch], str]  # Callable invoked by re.sub
_EmojiSubMethod = Callable[[_ReSubCallable, str], str]  # Sub method of a compiled re


def _emoji_replace(
    text: str,
    default_variant: Optional[str] = None,
    _emoji_sub: _EmojiSubMethod = re.compile(r"(:(\S*?)(?:(?:\-)(emoji|text))?:)").sub,
) -> str:
    """Replace emoji code in text."""
    get_emoji = EMOJI.__getitem__
    variants = {"text": "\uFE0E", "emoji": "\uFE0F"}
    get_variant = variants.get
    default_variant_code = variants.get(default_variant, "") if default_variant else ""

    def do_replace(match: Match[str]) -> str:
        emoji_code, emoji_name, variant = match.groups()
        try:
            return get_emoji(emoji_name.lower()) + get_variant(
                variant, default_variant_code
            )
        except KeyError:
            return emoji_code

    return _emoji_sub(do_replace, text)