#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""收集剧本字符并生成字体字符集文件。

用法:
    python Tools/collect_font_characters.py
"""

from __future__ import annotations

import os
import sys
import tempfile
from pathlib import Path


PROJECT_ROOT = Path(__file__).resolve().parent.parent
SOURCE_RELATIVE_PATH = Path("Assets/Yarn/FP")
TARGET_RELATIVE_PATH = Path("Assets/Font/字符统计.txt")
LINE_WIDTH = 20
EXCLUDED_CHARACTERS = {"\ufeff", "\r", "\n"}


class CharacterCollectionError(RuntimeError):
    """字符统计失败。"""


def find_source_files(source_root: Path) -> tuple[list[Path], list[Path]]:
    """递归查找 Yarn 和 CSV 文件，并返回稳定排序后的路径。"""
    if not source_root.is_dir():
        raise CharacterCollectionError(f"剧本目录不存在: {source_root}")

    yarn_files: list[Path] = []
    csv_files: list[Path] = []
    for path in source_root.rglob("*"):
        if not path.is_file():
            continue

        suffix = path.suffix.lower()
        if suffix == ".yarn":
            yarn_files.append(path)
        elif suffix == ".csv":
            csv_files.append(path)

    sort_key = lambda path: path.as_posix()
    yarn_files.sort(key=sort_key)
    csv_files.sort(key=sort_key)
    return yarn_files, csv_files


def read_characters(path: Path) -> set[str]:
    """读取 UTF-8 文本并返回其中可写入字符集文件的字符。"""
    try:
        text = path.read_text(encoding="utf-8-sig")
    except (OSError, UnicodeError) as error:
        raise CharacterCollectionError(f"无法读取文件 {path}: {error}") from error

    return {character for character in text if character not in EXCLUDED_CHARACTERS}


def collect_characters(
    source_root: Path, target_path: Path
) -> tuple[set[str], int, int]:
    """合并全部剧本文件与现有字符统计文件中的字符。"""
    yarn_files, csv_files = find_source_files(source_root)
    characters: set[str] = set()

    for path in [*yarn_files, *csv_files]:
        characters.update(read_characters(path))

    if target_path.exists():
        characters.update(read_characters(target_path))

    return characters, len(yarn_files), len(csv_files)


def format_characters(characters: set[str], line_width: int = LINE_WIDTH) -> str:
    """按 Unicode 码点排序，并按固定字符数生成 CRLF 文本。"""
    if line_width <= 0:
        raise ValueError("line_width 必须大于 0")

    ordered = sorted(
        character
        for character in characters
        if character not in EXCLUDED_CHARACTERS
    )
    lines = [
        "".join(ordered[index : index + line_width])
        for index in range(0, len(ordered), line_width)
    ]
    return "\r\n".join(lines) + ("\r\n" if lines else "")


def write_atomic_utf8_bom(target_path: Path, content: str) -> None:
    """以 UTF-8 BOM 原子覆盖目标文件，避免失败时留下半截内容。"""
    if not target_path.parent.is_dir():
        raise CharacterCollectionError(f"输出目录不存在: {target_path.parent}")

    temporary_path: Path | None = None
    try:
        with tempfile.NamedTemporaryFile(
            mode="wb",
            prefix=f".{target_path.name}.",
            suffix=".tmp",
            dir=target_path.parent,
            delete=False,
        ) as temporary_file:
            temporary_path = Path(temporary_file.name)
            temporary_file.write(content.encode("utf-8-sig"))
            temporary_file.flush()
            os.fsync(temporary_file.fileno())

        os.replace(temporary_path, target_path)
        temporary_path = None
    except (OSError, UnicodeError) as error:
        raise CharacterCollectionError(f"无法写入文件 {target_path}: {error}") from error
    finally:
        if temporary_path is not None:
            try:
                temporary_path.unlink(missing_ok=True)
            except OSError:
                pass


def generate_character_file(project_root: Path = PROJECT_ROOT) -> tuple[int, int, int]:
    """收集项目字符、覆盖目标文件，并返回 Yarn/CSV/字符数量。"""
    source_root = project_root / SOURCE_RELATIVE_PATH
    target_path = project_root / TARGET_RELATIVE_PATH
    characters, yarn_count, csv_count = collect_characters(source_root, target_path)
    write_atomic_utf8_bom(target_path, format_characters(characters))
    return yarn_count, csv_count, len(characters)


def main() -> int:
    try:
        yarn_count, csv_count, character_count = generate_character_file()
    except CharacterCollectionError as error:
        print(f"[错误] {error}", file=sys.stderr)
        return 1

    print(f"扫描 Yarn 文件: {yarn_count}")
    print(f"扫描 CSV 文件: {csv_count}")
    print(f"唯一字符数量: {character_count}")
    print(f"已写入: {PROJECT_ROOT / TARGET_RELATIVE_PATH}")
    return 0


if __name__ == "__main__":
    sys.exit(main())