import gradio as gr import hashlib moldP_list = [ "温和内敛", "热情奔放", "率真耿直", "沉着稳重", "古灵精怪" ] moldN_list = [ "暴躁易怒", "情感淡漠", "优柔寡断", "沉默寡言", "胆小怯懦", "敏感脆弱" ] traitsP_list = [ "富有想象力，擅长战略性规划", "富有创造力，且充满好奇心", "富有个人魅力，善于鼓舞、领导他人", "意志力强大，擅长解决问题", "聪明好奇，热衷于发现和挑战", "善于观察总结，能提出有效的建议", "生性浪漫，善良且乐于助人", "风趣幽默，乐于社交且自由自在", "重情重义，喜欢和家人、朋友在一起", "言出必行，十分可靠", "善于倾听，有化解冲突的能力", "直觉敏锐，且善于实践", "有独到的审美造诣，擅于用创作表达" ] traitsN_list = [ "无法独立生活，十分依赖于他人", "缺乏同理心，有较高犯罪天赋", "偏执任性、顽固不化", "意志力薄弱、难以保持原则", "生性多疑、难以信任他人", "不善于表达、时常会无意间伤害他人", "过分追求理念、时常苛责他人", "斤斤计较、患得患失，让人难以接近", "习惯性逃避，会把问题堆积到无可挽回的地步", "有很强的控制欲，时常会伤害到周围的人" ] def get_mold_and_trait(name): # 计算name的sha256哈希值 sha256_hash = hashlib.sha256(name.encode()).hexdigest() # 根据sha256第1个字符ascii的奇偶性选择mold列表 mold_list = moldP_list if int(sha256_hash[0], 16) % 2 == 0 else moldN_list mold_polarity = 'P' if mold_list == moldP_list else 'N' # 根据sha256第2、3字符代表的hex转成十进制选择mold mold_index = int(sha256_hash[1:3], 16) % len(mold_list) mold = mold_list[mold_index] # 根据sha256第4个字符ascii的奇偶性选择traits列表 traits_list = traitsP_list if int(sha256_hash[3], 16) % 2 == 0 else traitsN_list traits_polarity = 'P' if traits_list == traitsP_list else 'N' # 根据sha256第5、6字符代表的hex转成十进制选择traits traits_index = int(sha256_hash[4:6], 16) % len(traits_list) traits = traits_list[traits_index] # 根据mold和traits的极性拼接字符串 if mold_polarity == traits_polarity: connector = "，且" else: connector = "，但" result = f"{mold}{connector}{traits}" return result def select_story_type(name_A, name_B): # 拼接两个字符串 concatenated_name = name_A + name_B # 计算SHA-256哈希值 sha256_hash = hashlib.sha256(concatenated_name.encode()).hexdigest() # 将哈希值转换为十进制 decimal_hash = int(sha256_hash, 16) # 定义故事类型列表 story_types = [ "相爱相杀冤家型", "青梅竹马养成型", "平平淡淡温情型", "一见倾心梦幻型", "轰轰烈烈激情型", "青涩懵懂青春型", "互相讨厌培养型", "互相折磨病娇型", "又甜又真日常型", "催人泪下悲情型" ] # 使用mod操作符选择故事类型 selected_story_type = story_types[decimal_hash % len(story_types)] return selected_story_type def calculate_cp_point(name_A,name_B): B = 1.7 A = 3 * B hash1 = hashlib.sha256(name_A.encode()).hexdigest() hash2 = hashlib.sha256(name_B.encode()).hexdigest() hamming_distance = sum(bit1 != bit2 for bit1, bit2 in zip(hash1, hash2)) hamming_point = 64 - hamming_distance distance_list = [abs(ord(bit1) - ord(bit2)) if bit1.isalpha() and bit2.isalpha() else 0 for bit1, bit2 in zip(hash1, hash2)] distance_sum = sum(distance_list) cp_point = A * hamming_point + B * distance_sum return [hash1,hash2,cp_point] def calculate_story(name_A,name_B): resultA = get_mold_and_trait(name_A) resultB = get_mold_and_trait(name_B) story = select_story_type(name_A, name_B) output_string = f"{name_A}:{resultA}\n{name_B}:{resultB}\n请根据他们的性格特点，写一个关于{story}的故事。" return output_string with gr.Blocks() as demo: with gr.Row(): gr.Markdown("# Cyber CP Demo") with gr.Row(): with gr.Column(scale=1): name_A = gr.Textbox(label="A的名字") hash_A = gr.Textbox(label="Hash_A") with gr.Column(scale=1): name_B = gr.Textbox(label="B的名字") hash_B = gr.Textbox(label="Hash_B") with gr.Row(): with gr.Column(scale=1): cp_button = gr.Button("计算cp值") story_button = gr.Button("生成故事Prompt") with gr.Row(): with gr.Column(scale=1): cp_point = gr.Number(label="cp值") story_output = gr.Textbox(label="故事Prompt",show_label=True,show_copy_button=True) cp_button.click(calculate_cp_point, [name_A,name_B], [hash_A,hash_B,cp_point]) story_button.click(calculate_story, [name_A,name_B], story_output) gr.Markdown("请将上述Prompt复制后交给Kimi生成故事：[Kimi.ai - 帮你看更大的世界](https://kimi.moonshot.cn/chat/)\nhttps://kimi.moonshot.cn/chat/") demo.queue() demo.launch(inbrowser=True)