import itertools
import logging
import os
import posixpath
import urllib.parse
from dataclasses import dataclass
from typing import List

from pip._vendor.packaging.utils import canonicalize_name

from pip._internal.models.index import PyPI
from pip._internal.utils.compat import has_tls
from pip._internal.utils.misc import normalize_path, redact_auth_from_url

logger = logging.getLogger(__name__)


@dataclass(frozen=True)
class SearchScope:
    """
    Encapsulates the locations that pip is configured to search.
    """

    __slots__ = ["find_links", "index_urls", "no_index"]

    find_links: List[str]
    index_urls: List[str]
    no_index: bool

    @classmethod
    def create(
        cls,
        find_links: List[str],
        index_urls: List[str],
        no_index: bool,
    ) -> "SearchScope":
        """
        Create a SearchScope object after normalizing the `find_links`.
        """
        # Build find_links. If an argument starts with ~, it may be
        # a local file relative to a home directory. So try normalizing
        # it and if it exists, use the normalized version.
        # This is deliberately conservative - it might be fine just to
        # blindly normalize anything starting with a ~...
        built_find_links: List[str] = []
        for link in find_links:
            if link.startswith("~"):
                new_link = normalize_path(link)
                if os.path.exists(new_link):
                    link = new_link
            built_find_links.append(link)

        # If we don't have TLS enabled, then WARN if anyplace we're looking
        # relies on TLS.
        if not has_tls():
            for link in itertools.chain(index_urls, built_find_links):
                parsed = urllib.parse.urlparse(link)
                if parsed.scheme == "https":
                    logger.warning(
                        "pip is configured with locations that require "
                        "TLS/SSL, however the ssl module in Python is not "
                        "available."
                    )
                    break

        return cls(
            find_links=built_find_links,
            index_urls=index_urls,
            no_index=no_index,
        )

    def get_formatted_locations(self) -> str:
        lines = []
        redacted_index_urls = []
        if self.index_urls and self.index_urls != [PyPI.simple_url]:
            for url in self.index_urls:
                redacted_index_url = redact_auth_from_url(url)

                # Parse the URL
                purl = urllib.parse.urlsplit(redacted_index_url)

                # URL is generally invalid if scheme and netloc is missing
                # there are issues with Python and URL parsing, so this test
                # is a bit crude. See bpo-20271, bpo-23505. Python doesn't
                # always parse invalid URLs correctly - it should raise
                # exceptions for malformed URLs
                if not purl.scheme and not purl.netloc:
                    logger.warning(
                        'The index url "%s" seems invalid, please provide a scheme.',
                        redacted_index_url,
                    )

                redacted_index_urls.append(redacted_index_url)

            lines.append(
                "Looking in indexes: {}".format(", ".join(redacted_index_urls))
            )

        if self.find_links:
            lines.append(
                "Looking in links: {}".format(
                    ", ".join(redact_auth_from_url(url) for url in self.find_links)
                )
            )
        return "\n".join(lines)

    def get_index_urls_locations(self, project_name: str) -> List[str]:
        """Returns the locations found via self.index_urls

        Checks the url_name on the main (first in the list) index and
        use this url_name to produce all locations
        """

        def mkurl_pypi_url(url: str) -> str:
            loc = posixpath.join(
                url, urllib.parse.quote(canonicalize_name(project_name))
            )
            # For maximum compatibility with easy_install, ensure the path
            # ends in a trailing slash.  Although this isn't in the spec
            # (and PyPI can handle it without the slash) some other index
            # implementations might break if they relied on easy_install's
            # behavior.
            if not loc.endswith("/"):
                loc = loc + "/"
            return loc

        return [mkurl_pypi_url(url) for url in self.index_urls]

import logging
import mimetypes
import os
from collections import defaultdict
from typing import Callable, Dict, Iterable, List, Optional, Tuple

from pip._vendor.packaging.utils import (
    InvalidSdistFilename,
    InvalidWheelFilename,
    canonicalize_name,
    parse_sdist_filename,
    parse_wheel_filename,
)

from pip._internal.models.candidate import InstallationCandidate
from pip._internal.models.link import Link
from pip._internal.utils.urls import path_to_url, url_to_path
from pip._internal.vcs import is_url

logger = logging.getLogger(__name__)

FoundCandidates = Iterable[InstallationCandidate]
FoundLinks = Iterable[Link]
CandidatesFromPage = Callable[[Link], Iterable[InstallationCandidate]]
PageValidator = Callable[[Link], bool]


class LinkSource:
    @property
    def link(self) -> Optional[Link]:
        """Returns the underlying link, if there's one."""
        raise NotImplementedError()

    def page_candidates(self) -> FoundCandidates:
        """Candidates found by parsing an archive listing HTML file."""
        raise NotImplementedError()

    def file_links(self) -> FoundLinks:
        """Links found by specifying archives directly."""
        raise NotImplementedError()


def _is_html_file(file_url: str) -> bool:
    return mimetypes.guess_type(file_url, strict=False)[0] == "text/html"


class _FlatDirectoryToUrls:
    """Scans directory and caches results"""

    def __init__(self, path: str) -> None:
        self._path = path
        self._page_candidates: List[str] = []
        self._project_name_to_urls: Dict[str, List[str]] = defaultdict(list)
        self._scanned_directory = False

    def _scan_directory(self) -> None:
        """Scans directory once and populates both page_candidates
        and project_name_to_urls at the same time
        """
        for entry in os.scandir(self._path):
            url = path_to_url(entry.path)
            if _is_html_file(url):
                self._page_candidates.append(url)
                continue

            # File must have a valid wheel or sdist name,
            # otherwise not worth considering as a package
            try:
                project_filename = parse_wheel_filename(entry.name)[0]
            except InvalidWheelFilename:
                try:
                    project_filename = parse_sdist_filename(entry.name)[0]
                except InvalidSdistFilename:
                    continue

            self._project_name_to_urls[project_filename].append(url)
        self._scanned_directory = True

    @property
    def page_candidates(self) -> List[str]:
        if not self._scanned_directory:
            self._scan_directory()

        return self._page_candidates

    @property
    def project_name_to_urls(self) -> Dict[str, List[str]]:
        if not self._scanned_directory:
            self._scan_directory()

        return self._project_name_to_urls


class _FlatDirectorySource(LinkSource):
    """Link source specified by ``--find-links=<path-to-dir>``.

    This looks the content of the directory, and returns:

    * ``page_candidates``: Links listed on each HTML file in the directory.
    * ``file_candidates``: Archives in the directory.
    """

    _paths_to_urls: Dict[str, _FlatDirectoryToUrls] = {}

    def __init__(
        self,
        candidates_from_page: CandidatesFromPage,
        path: str,
        project_name: str,
    ) -> None:
        self._candidates_from_page = candidates_from_page
        self._project_name = canonicalize_name(project_name)

        # Get existing instance of _FlatDirectoryToUrls if it exists
        if path in self._paths_to_urls:
            self._path_to_urls = self._paths_to_urls[path]
        else:
            self._path_to_urls = _FlatDirectoryToUrls(path=path)
            self._paths_to_urls[path] = self._path_to_urls

    @property
    def link(self) -> Optional[Link]:
        return None

    def page_candidates(self) -> FoundCandidates:
        for url in self._path_to_urls.page_candidates:
            yield from self._candidates_from_page(Link(url))

    def file_links(self) -> FoundLinks:
        for url in self._path_to_urls.project_name_to_urls[self._project_name]:
            yield Link(url)


class _LocalFileSource(LinkSource):
    """``--find-links=<path-or-url>`` or ``--[extra-]index-url=<path-or-url>``.

    If a URL is supplied, it must be a ``file:`` URL. If a path is supplied to
    the option, it is converted to a URL first. This returns:

    * ``page_candidates``: Links listed on an HTML file.
    * ``file_candidates``: The non-HTML file.
    """

    def __init__(
        self,
        candidates_from_page: CandidatesFromPage,
        link: Link,
    ) -> None:
        self._candidates_from_page = candidates_from_page
        self._link = link

    @property
    def link(self) -> Optional[Link]:
        return self._link

    def page_candidates(self) -> FoundCandidates:
        if not _is_html_file(self._link.url):
            return
        yield from self._candidates_from_page(self._link)

    def file_links(self) -> FoundLinks:
        if _is_html_file(self._link.url):
            return
        yield self._link


class _RemoteFileSource(LinkSource):
    """``--find-links=<url>`` or ``--[extra-]index-url=<url>``.

    This returns:

    * ``page_candidates``: Links listed on an HTML file.
    * ``file_candidates``: The non-HTML file.
    """

    def __init__(
        self,
        candidates_from_page: CandidatesFromPage,
        page_validator: PageValidator,
        link: Link,
    ) -> None:
        self._candidates_from_page = candidates_from_page
        self._page_validator = page_validator
        self._link = link

    @property
    def link(self) -> Optional[Link]:
        return self._link

    def page_candidates(self) -> FoundCandidates:
        if not self._page_validator(self._link):
            return
        yield from self._candidates_from_page(self._link)

    def file_links(self) -> FoundLinks:
        yield self._link


class _IndexDirectorySource(LinkSource):
    """``--[extra-]index-url=<path-to-directory>``.

    This is treated like a remote URL; ``candidates_from_page`` contains logic
    for this by appending ``index.html`` to the link.
    """

    def __init__(
        self,
        candidates_from_page: CandidatesFromPage,
        link: Link,
    ) -> None:
        self._candidates_from_page = candidates_from_page
        self._link = link

    @property
    def link(self) -> Optional[Link]:
        return self._link

    def page_candidates(self) -> FoundCandidates:
        yield from self._candidates_from_page(self._link)

    def file_links(self) -> FoundLinks:
        return ()


def build_source(
    location: str,
    *,
    candidates_from_page: CandidatesFromPage,
    page_validator: PageValidator,
    expand_dir: bool,
    cache_link_parsing: bool,
    project_name: str,
) -> Tuple[Optional[str], Optional[LinkSource]]:
    path: Optional[str] = None
    url: Optional[str] = None
    if os.path.exists(location):  # Is a local path.
        url = path_to_url(location)
        path = location
    elif location.startswith("file:"):  # A file: URL.
        url = location
        path = url_to_path(location)
    elif is_url(location):
        url = location

    if url is None:
        msg = (
            "Location '%s' is ignored: "
            "it is either a non-existing path or lacks a specific scheme."
        )
        logger.warning(msg, location)
        return (None, None)

    if path is None:
        source: LinkSource = _RemoteFileSource(
            candidates_from_page=candidates_from_page,
            page_validator=page_validator,
            link=Link(url, cache_link_parsing=cache_link_parsing),
        )
        return (url, source)

    if os.path.isdir(path):
        if expand_dir:
            source = _FlatDirectorySource(
                candidates_from_page=candidates_from_page,
                path=path,
                project_name=project_name,
            )
        else:
            source = _IndexDirectorySource(
                candidates_from_page=candidates_from_page,
                link=Link(url, cache_link_parsing=cache_link_parsing),
            )
        return (url, source)
    elif os.path.isfile(path):
        source = _LocalFileSource(
            candidates_from_page=candidates_from_page,
            link=Link(url, cache_link_parsing=cache_link_parsing),
        )
        return (url, source)
    logger.warning(
        "Location '%s' is ignored: it is neither a file nor a directory.",
        location,
    )
    return (url, None)

"""
    pygments.console
    ~~~~~~~~~~~~~~~~

    Format colored console output.

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

esc = "\x1b["

codes = {}
codes[""] = ""
codes["reset"] = esc + "39;49;00m"

codes["bold"] = esc + "01m"
codes["faint"] = esc + "02m"
codes["standout"] = esc + "03m"
codes["underline"] = esc + "04m"
codes["blink"] = esc + "05m"
codes["overline"] = esc + "06m"

dark_colors = ["black", "red", "green", "yellow", "blue",
               "magenta", "cyan", "gray"]
light_colors = ["brightblack", "brightred", "brightgreen", "brightyellow", "brightblue",
                "brightmagenta", "brightcyan", "white"]

x = 30
for dark, light in zip(dark_colors, light_colors):
    codes[dark] = esc + "%im" % x
    codes[light] = esc + "%im" % (60 + x)
    x += 1

del dark, light, x

codes["white"] = codes["bold"]


def reset_color():
    return codes["reset"]


def colorize(color_key, text):
    return codes[color_key] + text + codes["reset"]


def ansiformat(attr, text):
    """
    Format ``text`` with a color and/or some attributes::

        color       normal color
        *color*     bold color
        _color_     underlined color
        +color+     blinking color
    """
    result = []
    if attr[:1] == attr[-1:] == '+':
        result.append(codes['blink'])
        attr = attr[1:-1]
    if attr[:1] == attr[-1:] == '*':
        result.append(codes['bold'])
        attr = attr[1:-1]
    if attr[:1] == attr[-1:] == '_':
        result.append(codes['underline'])
        attr = attr[1:-1]
    result.append(codes[attr])
    result.append(text)
    result.append(codes['reset'])
    return ''.join(result)