from pypinyin import pinyin, Style
import argparse
import re

def chinese_to_pinyin(text):
    # Use regular expression to extract Chinese characters
    chinese_text = ''.join(re.findall(r'[\u4e00-\u9fa5]', text))
    
    out = "static const sr_cmd_t sr_commands[] = {\n"
    enum = "enum {\n"
    
    cmd_id = 0
    text_list = chinese_text.split(";")
    
    for item in text_list:
        item = item.strip()  # Remove leading and trailing spaces
        if not item:
            continue
        
        item = item.split(",")
        phrase_id = 0
        pinyin_list_all = []
        for phrase in item:
            phrase = phrase.strip()  # Remove leading and trailing spaces
            if not phrase:
                continue
            
            # Convert Chinese phrase to pinyin with spaces between each syllable
            pinyin_list = pinyin(phrase, style=Style.NORMAL, heteronym=False)
            # Join pinyin list with spaces
            phoneme = ' '.join([py[0] for py in pinyin_list])
            
            out += f'  {{ {cmd_id}, "{phoneme}", "{phoneme}" }},\n'
            
            if phrase_id == 0:
                enum += f'  SR_CMD_{phoneme.upper().replace(" ", "_")},\n'
            
            phrase_id += 1
        
        cmd_id += 1
    
    out += "};\n"
    enum += "};\n"
    
    print(enum)
    print(out)

    return out

if __name__ == "__main__":
    parser = argparse.ArgumentParser(prog="Chinese Speech Commands to Pinyin")
    parser.add_argument("text", type=str, default=None, help="input text")
    args = parser.parse_args()

    if args.text is not None:
        chinese_to_pinyin(args.text)