#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""提取工程 UI 源码中字符串字面量里的非 ASCII 字符（实际显示的汉字），
输出 <仓库根>/ui_cn_chars.txt，供 make_font.py 生成中文字体子集。

用法（在仓库根目录执行，或任意目录均可，脚本自动定位仓库根）：
    python tools/font_gen/extract_cn.py
"""
import os
import glob

ROOT = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
UI_DIR = os.path.join(ROOT, 'Drivers', 'BSP', 'UI')
TASK_DIR = os.path.join(ROOT, 'App', 'task')
SCAN = (glob.glob(os.path.join(UI_DIR, '*.c')) +
        glob.glob(os.path.join(UI_DIR, '*.h')) +
        [os.path.join(TASK_DIR, 'ui_task.c'), os.path.join(TASK_DIR, 'ui_task.h')])


def collect(text):
    """跳过 /* */ 与 // 注释、字符常量，只收集双引号字符串内的非 ASCII 字符。"""
    out, i, n = set(), 0, len(text)
    blk = line = sstr = cchr = False
    while i < n:
        c = text[i]
        nx = text[i + 1] if i + 1 < n else ''
        if blk:
            if c == '*' and nx == '/':
                blk = False
                i += 2
                continue
            i += 1
            continue
        if line:
            if c == '\n':
                line = False
            i += 1
            continue
        if sstr:
            if c == '\\':
                i += 2
                continue
            if c == '"':
                sstr = False
                i += 1
                continue
            if ord(c) > 0x7F:
                out.add(c)
            i += 1
            continue
        if cchr:
            if c == '\\':
                i += 2
                continue
            if c == "'":
                cchr = False
            i += 1
            continue
        if c == '/' and nx == '*':
            blk = True
            i += 2
            continue
        if c == '/' and nx == '/':
            line = True
            i += 2
            continue
        if c == '"':
            sstr = True
            i += 1
            continue
        if c == "'":
            cchr = True
            i += 1
            continue
        i += 1
    return out


def main():
    chars = set()
    for f in SCAN:
        if not os.path.exists(f):
            print('  skip (not found):', os.path.relpath(f, ROOT))
            continue
        chars |= collect(open(f, encoding='utf-8-sig', errors='ignore').read())
    s = ''.join(sorted(chars))
    out_path = os.path.join(ROOT, 'ui_cn_chars.txt')
    # newline='' 关闭行尾转换：恒输出 LF（与 gen_menu.py / make_font.py 一致），
    # 避免同一脚本在 Windows 出 CRLF、Linux 出 LF 导致提交产物跨平台 diff
    with open(out_path, 'w', encoding='utf-8', newline='') as fp:
        fp.write(s + '\n')
    print('scanned %d files; unique non-ASCII chars: %d' % (len(SCAN), len(s)))
    print('->', out_path)


if __name__ == '__main__':
    main()