"""Index entries adapters for sphinx.environment."""

from __future__ import annotations

import re
import unicodedata
from itertools import groupby
from typing import TYPE_CHECKING, Any, Literal

from sphinx.errors import NoUri
from sphinx.locale import _, __
from sphinx.util import logging
from sphinx.util.index_entries import _split_into

if TYPE_CHECKING:
    from sphinx.builders import Builder
    from sphinx.environment import BuildEnvironment

logger = logging.getLogger(__name__)


class IndexEntries:
    def __init__(self, env: BuildEnvironment) -> None:
        self.env = env
        self.builder: Builder

    def create_index(self, builder: Builder, group_entries: bool = True,
                     _fixre: re.Pattern = re.compile(r'(.*) ([(][^()]*[)])'),
                     ) -> list[tuple[str, list[tuple[str, Any]]]]:
        """Create the real index from the collected index entries."""
        new: dict[str, list] = {}

        rel_uri: str | Literal[False]
        index_domain = self.env.domains['index']
        for docname, entries in index_domain.entries.items():
            try:
                rel_uri = builder.get_relative_uri('genindex', docname)
            except NoUri:
                rel_uri = False

            # new entry types must be listed in directives/other.py!
            for entry_type, value, target_id, main, category_key in entries:
                uri = rel_uri is not False and f'{rel_uri}#{target_id}'
                try:
                    if entry_type == 'single':
                        try:
                            entry, sub_entry = _split_into(2, 'single', value)
                        except ValueError:
                            entry, = _split_into(1, 'single', value)
                            sub_entry = ''
                        _add_entry(entry, sub_entry, main,
                                   dic=new, link=uri, key=category_key)
                    elif entry_type == 'pair':
                        first, second = _split_into(2, 'pair', value)
                        _add_entry(first, second, main,
                                   dic=new, link=uri, key=category_key)
                        _add_entry(second, first, main,
                                   dic=new, link=uri, key=category_key)
                    elif entry_type == 'triple':
                        first, second, third = _split_into(3, 'triple', value)
                        _add_entry(first, second + ' ' + third, main,
                                   dic=new, link=uri, key=category_key)
                        _add_entry(second, third + ', ' + first, main,
                                   dic=new, link=uri, key=category_key)
                        _add_entry(third, first + ' ' + second, main,
                                   dic=new, link=uri, key=category_key)
                    elif entry_type == 'see':
                        first, second = _split_into(2, 'see', value)
                        _add_entry(first, _('see %s') % second, None,
                                   dic=new, link=False, key=category_key)
                    elif entry_type == 'seealso':
                        first, second = _split_into(2, 'see', value)
                        _add_entry(first, _('see also %s') % second, None,
                                   dic=new, link=False, key=category_key)
                    else:
                        logger.warning(__('unknown index entry type %r'), entry_type,
                                       location=docname)
                except ValueError as err:
                    logger.warning(str(err), location=docname)

        for (targets, sub_items, _category_key) in new.values():
            targets.sort(key=_key_func_0)
            for (sub_targets, _0, _sub_category_key) in sub_items.values():
                sub_targets.sort(key=_key_func_0)

        new_list = sorted(new.items(), key=_key_func_1)

        if group_entries:
            # fixup entries: transform
            #   func() (in module foo)
            #   func() (in module bar)
            # into
            #   func()
            #     (in module foo)
            #     (in module bar)
            old_key = ''
            old_sub_items: dict[str, list] = {}
            i = 0
            while i < len(new_list):
                key, (targets, sub_items, category_key) = new_list[i]
                # cannot move if it has sub_items; structure gets too complex
                if not sub_items:
                    m = _fixre.match(key)
                    if m:
                        if old_key == m.group(1):
                            # prefixes match: add entry as subitem of the
                            # previous entry
                            old_sub_items.setdefault(
                                m.group(2), [[], {}, category_key])[0].extend(targets)
                            del new_list[i]
                            continue
                        old_key = m.group(1)
                    else:
                        old_key = key
                old_sub_items = sub_items
                i += 1

        return [(key_, list(group))
                for (key_, group) in groupby(new_list, _key_func_3)]


def _add_entry(word: str, subword: str, main: str | None, *,
               dic: dict[str, list], link: str | Literal[False], key: str | None) -> None:
    entry = dic.setdefault(word, [[], {}, key])
    if subword:
        entry = entry[1].setdefault(subword, [[], {}, key])
    if link:
        entry[0].append((main, link))


def _key_func_0(entry: tuple[str, str]) -> tuple[bool, str]:
    """Sort the index entries for same keyword."""
    main, uri = entry
    return not main, uri  # show main entries at first


def _key_func_1(entry: tuple[str, list]) -> tuple[tuple[int, str], str]:
    """Sort the index entries"""
    key, (_targets, _sub_items, category_key) = entry
    if category_key:
        # using the specified category key to sort
        key = category_key
    lc_key = unicodedata.normalize('NFD', key.lower())
    if lc_key.startswith('\N{RIGHT-TO-LEFT MARK}'):
        lc_key = lc_key[1:]

    if not lc_key[0:1].isalpha() and not lc_key.startswith('_'):
        # put symbols at the front of the index (0)
        group = 0
    else:
        # put non-symbol characters at the following group (1)
        group = 1
    # ensure a deterministic order *within* letters by also sorting on
    # the entry itself
    return (group, lc_key), entry[0]


def _key_func_2(entry: tuple[str, list]) -> str:
    """Sort the sub-index entries"""
    key = unicodedata.normalize('NFD', entry[0].lower())
    if key.startswith('\N{RIGHT-TO-LEFT MARK}'):
        key = key[1:]
    if key[0:1].isalpha() or key.startswith('_'):
        key = chr(127) + key
    return key


def _key_func_3(entry: tuple[str, list]) -> str:
    """Group the entries by letter"""
    key, (targets, sub_items, category_key) = entry
    # hack: mutating the sub_items dicts to a list in the key_func
    entry[1][1] = sorted(((sub_key, sub_targets)
                          for (sub_key, (sub_targets, _0, _sub_category_key))
                          in sub_items.items()), key=_key_func_2)

    if category_key is not None:
        return category_key

    # now calculate the key
    if key.startswith('\N{RIGHT-TO-LEFT MARK}'):
        key = key[1:]
    letter = unicodedata.normalize('NFD', key[0])[0].upper()
    if letter.isalpha() or letter == '_':
        return letter

    # get all other symbols under one heading
    return _('Symbols')