up

crazy_functions/CodeInterpreter.py

@@ -0,0 +1,231 @@
+from collections.abc import Callable, Iterable, Mapping
+from typing import Any
+from toolbox import CatchException, update_ui, gen_time_str, trimmed_format_exc, promote_file_to_downloadzone, clear_file_downloadzone
+from .crazy_utils import request_gpt_model_in_new_thread_with_ui_alive
+from .crazy_utils import input_clipping, try_install_deps
+from multiprocessing import Process, Pipe
+import os
+import time
+
+templete = """
+```python
+import ...  # Put dependencies here, e.g. import numpy as np
+
+class TerminalFunction(object): # Do not change the name of the class, The name of the class must be `TerminalFunction`
+
+    def run(self, path):    # The name of the function must be `run`, it takes only a positional argument.
+        # rewrite the function you have just written here 
+        ...
+        return generated_file_path
+```
+"""
+
+def inspect_dependency(chatbot, history):
+    yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+    return True
+
+def get_code_block(reply):
+    import re
+    pattern = r"```([\s\S]*?)```" # regex pattern to match code blocks
+    matches = re.findall(pattern, reply) # find all code blocks in text
+    if len(matches) == 1: 
+        return matches[0].strip('python') #  code block
+    for match in matches:
+        if 'class TerminalFunction' in match:
+            return match.strip('python') #  code block
+    raise RuntimeError("GPT is not generating proper code.")
+
+def gpt_interact_multi_step(txt, file_type, llm_kwargs, chatbot, history):
+    # 输入
+    prompt_compose = [
+        f'Your job:\n'
+        f'1. write a single Python function, which takes a path of a `{file_type}` file as the only argument and returns a `string` containing the result of analysis or the path of generated files. \n',
+        f"2. You should write this function to perform following task: " + txt + "\n",
+        f"3. Wrap the output python function with markdown codeblock."
+    ]
+    i_say = "".join(prompt_compose)
+    demo = []
+
+    # 第一步
+    gpt_say = yield from request_gpt_model_in_new_thread_with_ui_alive(
+        inputs=i_say, inputs_show_user=i_say, 
+        llm_kwargs=llm_kwargs, chatbot=chatbot, history=demo, 
+        sys_prompt= r"You are a programmer."
+    )
+    history.extend([i_say, gpt_say])
+    yield from update_ui(chatbot=chatbot, history=history) # 刷新界面 # 界面更新
+
+    # 第二步
+    prompt_compose = [
+        "If previous stage is successful, rewrite the function you have just written to satisfy following templete: \n",
+        templete
+    ]
+    i_say = "".join(prompt_compose); inputs_show_user = "If previous stage is successful, rewrite the function you have just written to satisfy executable templete. "
+    gpt_say = yield from request_gpt_model_in_new_thread_with_ui_alive(
+        inputs=i_say, inputs_show_user=inputs_show_user, 
+        llm_kwargs=llm_kwargs, chatbot=chatbot, history=history, 
+        sys_prompt= r"You are a programmer."
+    )
+    code_to_return = gpt_say
+    history.extend([i_say, gpt_say])
+    yield from update_ui(chatbot=chatbot, history=history) # 刷新界面 # 界面更新
+    
+    # # 第三步
+    # i_say = "Please list to packages to install to run the code above. Then show me how to use `try_install_deps` function to install them."
+    # i_say += 'For instance. `try_install_deps(["opencv-python", "scipy", "numpy"])`'
+    # installation_advance = yield from request_gpt_model_in_new_thread_with_ui_alive(
+    #     inputs=i_say, inputs_show_user=inputs_show_user, 
+    #     llm_kwargs=llm_kwargs, chatbot=chatbot, history=history, 
+    #     sys_prompt= r"You are a programmer."
+    # )
+    # # # 第三步  
+    # i_say = "Show me how to use `pip` to install packages to run the code above. "
+    # i_say += 'For instance. `pip install -r opencv-python scipy numpy`'
+    # installation_advance = yield from request_gpt_model_in_new_thread_with_ui_alive(
+    #     inputs=i_say, inputs_show_user=i_say, 
+    #     llm_kwargs=llm_kwargs, chatbot=chatbot, history=history, 
+    #     sys_prompt= r"You are a programmer."
+    # )
+    installation_advance = ""
+    
+    return code_to_return, installation_advance, txt, file_type, llm_kwargs, chatbot, history
+
+def make_module(code):
+    module_file = 'gpt_fn_' + gen_time_str().replace('-','_')
+    with open(f'gpt_log/{module_file}.py', 'w', encoding='utf8') as f:
+        f.write(code)
+
+    def get_class_name(class_string):
+        import re
+        # Use regex to extract the class name
+        class_name = re.search(r'class (\w+)\(', class_string).group(1)
+        return class_name
+
+    class_name = get_class_name(code)
+    return f"gpt_log.{module_file}->{class_name}"
+
+def init_module_instance(module):
+    import importlib
+    module_, class_ = module.split('->')
+    init_f = getattr(importlib.import_module(module_), class_)
+    return init_f()
+
+def for_immediate_show_off_when_possible(file_type, fp, chatbot):
+    if file_type in ['png', 'jpg']:
+        image_path = os.path.abspath(fp)
+        chatbot.append(['这是一张图片, 展示如下:',  
+            f'本地文件地址: <br/>`{image_path}`<br/>'+
+            f'本地文件预览: <br/><div align="center"><img src="file={image_path}"></div>'
+        ])
+    return chatbot
+
+def subprocess_worker(instance, file_path, return_dict):
+    return_dict['result'] = instance.run(file_path)
+
+def have_any_recent_upload_files(chatbot):
+    _5min = 5 * 60
+    if not chatbot: return False    # chatbot is None
+    most_recent_uploaded = chatbot._cookies.get("most_recent_uploaded", None)
+    if not most_recent_uploaded: return False   # most_recent_uploaded is None
+    if time.time() - most_recent_uploaded["time"] < _5min: return True # most_recent_uploaded is new
+    else: return False  # most_recent_uploaded is too old
+
+def get_recent_file_prompt_support(chatbot):
+    most_recent_uploaded = chatbot._cookies.get("most_recent_uploaded", None)
+    path = most_recent_uploaded['path']
+    return path
+
+@CatchException
+def 虚空终端CodeInterpreter(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, web_port):
+    """
+    txt             输入栏用户输入的文本，例如需要翻译的一段话，再例如一个包含了待处理文件的路径
+    llm_kwargs      gpt模型参数，如温度和top_p等，一般原样传递下去就行
+    plugin_kwargs   插件模型的参数，暂时没有用武之地
+    chatbot         聊天显示框的句柄，用于显示给用户
+    history         聊天历史，前情提要
+    system_prompt   给gpt的静默提醒
+    web_port        当前软件运行的端口号
+    """
+    raise NotImplementedError
+
+    # 清空历史，以免输入溢出
+    history = []; clear_file_downloadzone(chatbot)
+
+    # 基本信息：功能、贡献者
+    chatbot.append([
+        "函数插件功能？",
+        "CodeInterpreter开源版, 此插件处于开发阶段, 建议暂时不要使用, 插件初始化中 ..."
+    ])
+    yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+
+    if have_any_recent_upload_files(chatbot):
+        file_path = get_recent_file_prompt_support(chatbot)
+    else:
+        chatbot.append(["文件检索", "没有发现任何近期上传的文件。"])
+        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+
+    # 读取文件
+    if ("recently_uploaded_files" in plugin_kwargs) and (plugin_kwargs["recently_uploaded_files"] == ""): plugin_kwargs.pop("recently_uploaded_files")
+    recently_uploaded_files = plugin_kwargs.get("recently_uploaded_files", None)
+    file_path = recently_uploaded_files[-1]
+    file_type = file_path.split('.')[-1]
+
+    # 粗心检查
+    if 'private_upload' in txt:
+        chatbot.append([
+            "...",
+            f"请在输入框内填写需求，然后再次点击该插件（文件路径 {file_path} 已经被记忆）"
+        ])
+        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+        return
+    
+    # 开始干正事
+    for j in range(5):  # 最多重试5次
+        try:
+            code, installation_advance, txt, file_type, llm_kwargs, chatbot, history = \
+                yield from gpt_interact_multi_step(txt, file_type, llm_kwargs, chatbot, history)
+            code = get_code_block(code)
+            res = make_module(code)
+            instance = init_module_instance(res)
+            break
+        except Exception as e:
+            chatbot.append([f"第{j}次代码生成尝试，失败了", f"错误追踪\n```\n{trimmed_format_exc()}\n```\n"])
+            yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+
+    # 代码生成结束, 开始执行
+    try:
+        import multiprocessing
+        manager = multiprocessing.Manager()
+        return_dict = manager.dict()
+
+        p = multiprocessing.Process(target=subprocess_worker, args=(instance, file_path, return_dict))
+        # only has 10 seconds to run
+        p.start(); p.join(timeout=10)
+        if p.is_alive(): p.terminate(); p.join()
+        p.close()
+        res = return_dict['result']
+        # res = instance.run(file_path)
+    except Exception as e:
+        chatbot.append(["执行失败了", f"错误追踪\n```\n{trimmed_format_exc()}\n```\n"])
+        # chatbot.append(["如果是缺乏依赖，请参考以下建议", installation_advance])
+        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+        return
+
+    # 顺利完成，收尾
+    res = str(res)
+    if os.path.exists(res):
+        chatbot.append(["执行成功了，结果是一个有效文件", "结果：" + res])
+        new_file_path = promote_file_to_downloadzone(res, chatbot=chatbot)
+        chatbot = for_immediate_show_off_when_possible(file_type, new_file_path, chatbot)
+        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面 # 界面更新
+    else:
+        chatbot.append(["执行成功了，结果是一个字符串", "结果：" + res])
+        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面 # 界面更新   
+
+"""
+测试：
+    裁剪图像，保留下半部分
+    交换图像的蓝色通道和红色通道
+    将图像转为灰度图像
+    将csv文件转excel表格
+"""

crazy_functions/Latex输出PDF结果.py

@@ -6,7 +6,7 @@ pj = os.path.join
 ARXIV_CACHE_DIR = os.path.expanduser(f"~/arxiv_cache/")
 
 # =================================== 工具函数 ===============================================
-专业词汇声明  = 'If the term "agent" is used in this section, it should be translated to "智能体". '
+# 专业词汇声明  = 'If the term "agent" is used in this section, it should be translated to "智能体". '
 def switch_prompt(pfg, mode, more_requirement):
     """
     Generate prompts and system prompts based on the mode for proofreading or translating.
@@ -109,7 +109,7 @@ def arxiv_download(chatbot, history, txt):
 
     url_ = txt   # https://arxiv.org/abs/1707.06690
     if not txt.startswith('https://arxiv.org/abs/'): 
-        msg = f"解析arxiv网址失败, 期望格式例如: https://arxiv.org/abs/1707.06690。实际得到格式: {url_}"
+        msg = f"解析arxiv网址失败, 期望格式例如: https://arxiv.org/abs/1707.06690。实际得到格式: {url_}。"
         yield from update_ui_lastest_msg(msg, chatbot=chatbot, history=history) # 刷新界面
         return msg, None
     # <-------------- set format ------------->
@@ -255,7 +255,7 @@ def Latex翻译中文并重新编译PDF(txt, llm_kwargs, plugin_kwargs, chatbot,
         project_folder = txt
     else:
         if txt == "": txt = '空空如也的输入栏'
-        report_execption(chatbot, history, a = f"解析项目: {txt}", b = f"找不到本地项目或无权访问: {txt}")
+        report_execption(chatbot, history, a = f"解析项目: {txt}", b = f"找不到本地项目或无法处理: {txt}")
         yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
         return
     
@@ -291,7 +291,7 @@ def Latex翻译中文并重新编译PDF(txt, llm_kwargs, plugin_kwargs, chatbot,
         yield from update_ui(chatbot=chatbot, history=history); time.sleep(1) # 刷新界面
         promote_file_to_downloadzone(file=zip_res, chatbot=chatbot)
     else:
-        chatbot.append((f"失败了", '虽然PDF生成失败了, 但请查收结果（压缩包）, 内含已经翻译的Tex文档, 也是可读的, 您可以到Github Issue区, 用该压缩包+对话历史存档进行反馈 ...'))
+        chatbot.append((f"失败了", '虽然PDF生成失败了, 但请查收结果（压缩包）, 内含已经翻译的Tex文档, 您可以到Github Issue区, 用该压缩包进行反馈。如系统是Linux，请检查系统字体（见Github wiki） ...'))
         yield from update_ui(chatbot=chatbot, history=history); time.sleep(1) # 刷新界面
         promote_file_to_downloadzone(file=zip_res, chatbot=chatbot)
 

crazy_functions/crazy_utils.py

@@ -591,11 +591,16 @@ def get_files_from_everything(txt, type): # type='.md'
         # 网络的远程文件
         import requests
         from toolbox import get_conf
+        from toolbox import get_log_folder, gen_time_str
         proxies, = get_conf('proxies')
-        r = requests.get(txt, proxies=proxies)
-        with open('./gpt_log/temp'+type, 'wb+') as f: f.write(r.content)
-        project_folder = './gpt_log/'
-        file_manifest = ['./gpt_log/temp'+type]
+        try:
+            r = requests.get(txt, proxies=proxies)
+        except:
+            raise ConnectionRefusedError(f"无法下载资源{txt}，请检查。")
+        path = os.path.join(get_log_folder(plugin_name='web_download'), gen_time_str()+type)
+        with open(path, 'wb+') as f: f.write(r.content)
+        project_folder = get_log_folder(plugin_name='web_download')
+        file_manifest = [path]
     elif txt.endswith(type):
         # 直接给定文件
         file_manifest = [txt]

crazy_functions/json_fns/pydantic_io.py

@@ -0,0 +1,111 @@
+"""
+https://github.com/langchain-ai/langchain/blob/master/docs/extras/modules/model_io/output_parsers/pydantic.ipynb
+
+Example 1.
+
+# Define your desired data structure.
+class Joke(BaseModel):
+    setup: str = Field(description="question to set up a joke")
+    punchline: str = Field(description="answer to resolve the joke")
+
+    # You can add custom validation logic easily with Pydantic.
+    @validator("setup")
+    def question_ends_with_question_mark(cls, field):
+        if field[-1] != "?":
+            raise ValueError("Badly formed question!")
+        return field
+
+
+Example 2.
+
+# Here's another example, but with a compound typed field.
+class Actor(BaseModel):
+    name: str = Field(description="name of an actor")
+    film_names: List[str] = Field(description="list of names of films they starred in")
+"""
+
+import json, re, logging
+
+
+PYDANTIC_FORMAT_INSTRUCTIONS = """The output should be formatted as a JSON instance that conforms to the JSON schema below.
+
+As an example, for the schema {{"properties": {{"foo": {{"title": "Foo", "description": "a list of strings", "type": "array", "items": {{"type": "string"}}}}}}, "required": ["foo"]}}
+the object {{"foo": ["bar", "baz"]}} is a well-formatted instance of the schema. The object {{"properties": {{"foo": ["bar", "baz"]}}}} is not well-formatted.
+
+Here is the output schema:
+```
+{schema}
+```"""
+
+
+PYDANTIC_FORMAT_INSTRUCTIONS_SIMPLE = """The output should be formatted as a JSON instance that conforms to the JSON schema below.
+```
+{schema}
+```"""
+
+class JsonStringError(Exception): ...
+
+class GptJsonIO():
+
+    def __init__(self, schema, example_instruction=True):
+        self.pydantic_object = schema
+        self.example_instruction = example_instruction
+        self.format_instructions = self.generate_format_instructions()
+
+    def generate_format_instructions(self):
+        schema = self.pydantic_object.schema()
+
+        # Remove extraneous fields.
+        reduced_schema = schema
+        if "title" in reduced_schema:
+            del reduced_schema["title"]
+        if "type" in reduced_schema:
+            del reduced_schema["type"]
+        # Ensure json in context is well-formed with double quotes.
+        if self.example_instruction:
+            schema_str = json.dumps(reduced_schema)
+            return PYDANTIC_FORMAT_INSTRUCTIONS.format(schema=schema_str)
+        else:
+            return PYDANTIC_FORMAT_INSTRUCTIONS_SIMPLE.format(schema=schema_str)
+
+    def generate_output(self, text):
+        # Greedy search for 1st json candidate.
+        match = re.search(
+            r"\{.*\}", text.strip(), re.MULTILINE | re.IGNORECASE | re.DOTALL
+        )
+        json_str = ""
+        if match: json_str = match.group()
+        json_object = json.loads(json_str, strict=False)
+        final_object = self.pydantic_object.parse_obj(json_object)
+        return final_object
+
+    def generate_repair_prompt(self, broken_json, error):
+        prompt = "Fix a broken json string.\n\n" + \
+                 "(1) The broken json string need to fix is: \n\n" + \
+                 "```" + "\n" + \
+                 broken_json + "\n" + \
+                 "```" + "\n\n" + \
+                 "(2) The error message is: \n\n" + \
+                 error + "\n\n" + \
+                "Now, fix this json string. \n\n"
+        return prompt
+    
+    def generate_output_auto_repair(self, response, gpt_gen_fn):
+        """
+        response: string containing canidate json
+        gpt_gen_fn: gpt_gen_fn(inputs, sys_prompt)
+        """
+        try:
+            result = self.generate_output(response)
+        except Exception as e:
+            try:
+                logging.info(f'Repairing json：{response}')
+                repair_prompt = self.generate_repair_prompt(broken_json = response, error=repr(e))
+                result = self.generate_output(gpt_gen_fn(repair_prompt, self.format_instructions))
+                logging.info('Repaire json success.')
+            except Exception as e:
+                # 没辙了，放弃治疗
+                logging.info('Repaire json fail.')
+                raise JsonStringError('Cannot repair json.', str(e))
+        return result
+

crazy_functions/live_audio/aliyunASR.py

@@ -1,4 +1,4 @@
-import time, threading, json
+import time, logging, json
 
 
 class AliyunASR():
@@ -12,14 +12,14 @@ class AliyunASR():
         message = json.loads(message)
         self.parsed_sentence = message['payload']['result']
         self.event_on_entence_end.set()
-        print(self.parsed_sentence)
+        # print(self.parsed_sentence)
 
     def test_on_start(self, message, *args):
         # print("test_on_start:{}".format(message))
         pass
 
     def test_on_error(self, message, *args):
-        print("on_error args=>{}".format(args))
+        logging.error("on_error args=>{}".format(args))
         pass
 
     def test_on_close(self, *args):
@@ -36,7 +36,6 @@ class AliyunASR():
         # print("on_completed:args=>{} message=>{}".format(args, message))
         pass
 
-
     def audio_convertion_thread(self, uuid):
         # 在一个异步线程中采集音频
         import nls  # pip install git+https://github.com/aliyun/alibabacloud-nls-python-sdk.git

crazy_functions/pdf_fns/parse_pdf.py

@@ -20,6 +20,11 @@ def get_avail_grobid_url():
 def parse_pdf(pdf_path, grobid_url):
     import scipdf   # pip install scipdf_parser
     if grobid_url.endswith('/'): grobid_url = grobid_url.rstrip('/')
-    article_dict = scipdf.parse_pdf_to_dict(pdf_path, grobid_url=grobid_url)
+    try:
+        article_dict = scipdf.parse_pdf_to_dict(pdf_path, grobid_url=grobid_url)
+    except GROBID_OFFLINE_EXCEPTION:
+        raise GROBID_OFFLINE_EXCEPTION("GROBID服务不可用，请修改config中的GROBID_URL，可修改成本地GROBID服务。")
+    except:
+        raise RuntimeError("解析PDF失败，请检查PDF是否损坏。")
     return article_dict
 

crazy_functions/vt_fns/vt_call_plugin.py

@@ -0,0 +1,114 @@
+from pydantic import BaseModel, Field
+from typing import List
+from toolbox import update_ui_lastest_msg, disable_auto_promotion
+from request_llm.bridge_all import predict_no_ui_long_connection
+from crazy_functions.json_fns.pydantic_io import GptJsonIO, JsonStringError
+import copy, json, pickle, os, sys, time
+
+
+def read_avail_plugin_enum():
+    from crazy_functional import get_crazy_functions
+    plugin_arr = get_crazy_functions()
+    # remove plugins with out explaination
+    plugin_arr = {k:v for k, v in plugin_arr.items() if 'Info' in v}
+    plugin_arr_info = {"F_{:04d}".format(i):v["Info"] for i, v in enumerate(plugin_arr.values(), start=1)}
+    plugin_arr_dict = {"F_{:04d}".format(i):v for i, v in enumerate(plugin_arr.values(), start=1)}
+    plugin_arr_dict_parse = {"F_{:04d}".format(i):v for i, v in enumerate(plugin_arr.values(), start=1)}
+    plugin_arr_dict_parse.update({f"F_{i}":v for i, v in enumerate(plugin_arr.values(), start=1)})
+    prompt = json.dumps(plugin_arr_info, ensure_ascii=False, indent=2)
+    prompt = "\n\nThe defination of PluginEnum:\nPluginEnum=" + prompt
+    return prompt, plugin_arr_dict, plugin_arr_dict_parse
+
+def wrap_code(txt):
+    txt = txt.replace('```','')
+    return f"\n```\n{txt}\n```\n"
+
+def have_any_recent_upload_files(chatbot):
+    _5min = 5 * 60
+    if not chatbot: return False    # chatbot is None
+    most_recent_uploaded = chatbot._cookies.get("most_recent_uploaded", None)
+    if not most_recent_uploaded: return False   # most_recent_uploaded is None
+    if time.time() - most_recent_uploaded["time"] < _5min: return True # most_recent_uploaded is new
+    else: return False  # most_recent_uploaded is too old
+
+def get_recent_file_prompt_support(chatbot):
+    most_recent_uploaded = chatbot._cookies.get("most_recent_uploaded", None)
+    path = most_recent_uploaded['path']
+    prompt =   "\nAdditional Information:\n"
+    prompt =   "In case that this plugin requires a path or a file as argument," 
+    prompt += f"it is important for you to know that the user has recently uploaded a file, located at: `{path}`" 
+    prompt += f"Only use it when necessary, otherwise, you can ignore this file." 
+    return prompt
+
+def get_inputs_show_user(inputs, plugin_arr_enum_prompt):
+    # remove plugin_arr_enum_prompt from inputs string
+    inputs_show_user = inputs.replace(plugin_arr_enum_prompt, "")
+    inputs_show_user += plugin_arr_enum_prompt[:200] + '...'
+    inputs_show_user += '\n...\n'
+    inputs_show_user += '...\n'
+    inputs_show_user += '...}'
+    return inputs_show_user
+
+def execute_plugin(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_intention):
+    plugin_arr_enum_prompt, plugin_arr_dict, plugin_arr_dict_parse = read_avail_plugin_enum()
+    class Plugin(BaseModel):
+        plugin_selection: str = Field(description="The most related plugin from one of the PluginEnum.", default="F_0000")
+        reason_of_selection: str = Field(description="The reason why you should select this plugin.", default="This plugin satisfy user requirement most")
+    # ⭐ ⭐ ⭐ 选择插件
+    yield from update_ui_lastest_msg(lastmsg=f"正在执行任务: {txt}\n\n查找可用插件中...", chatbot=chatbot, history=history, delay=0)
+    gpt_json_io = GptJsonIO(Plugin)
+    gpt_json_io.format_instructions = "The format of your output should be a json that can be parsed by json.loads.\n"
+    gpt_json_io.format_instructions += """Output example: {"plugin_selection":"F_1234", "reason_of_selection":"F_1234 plugin satisfy user requirement most"}\n"""
+    gpt_json_io.format_instructions += "The plugins you are authorized to use are listed below:\n"
+    gpt_json_io.format_instructions += plugin_arr_enum_prompt
+    inputs = "Choose the correct plugin according to user requirements, the user requirement is: \n\n" + \
+             ">> " + txt.rstrip('\n').replace('\n','\n>> ') + '\n\n' + gpt_json_io.format_instructions
+
+    run_gpt_fn = lambda inputs, sys_prompt: predict_no_ui_long_connection(
+        inputs=inputs, llm_kwargs=llm_kwargs, history=[], sys_prompt=sys_prompt, observe_window=[])
+    try:
+        gpt_reply = run_gpt_fn(inputs, "")
+        plugin_sel = gpt_json_io.generate_output_auto_repair(gpt_reply, run_gpt_fn)
+    except JsonStringError:
+        msg = f"抱歉, {llm_kwargs['llm_model']}无法理解您的需求。"
+        msg += "请求的Prompt为：\n" + wrap_code(get_inputs_show_user(inputs, plugin_arr_enum_prompt))
+        msg += "语言模型回复为：\n" + wrap_code(gpt_reply)
+        msg += "\n但您可以尝试再试一次\n"
+        yield from update_ui_lastest_msg(lastmsg=msg, chatbot=chatbot, history=history, delay=2)
+        return
+    if plugin_sel.plugin_selection not in plugin_arr_dict_parse:
+        msg = f"抱歉, 找不到合适插件执行该任务, 或者{llm_kwargs['llm_model']}无法理解您的需求。"
+        msg += f"语言模型{llm_kwargs['llm_model']}选择了不存在的插件：\n" + wrap_code(gpt_reply)
+        msg += "\n但您可以尝试再试一次\n"
+        yield from update_ui_lastest_msg(lastmsg=msg, chatbot=chatbot, history=history, delay=2)
+        return
+    
+    # ⭐ ⭐ ⭐ 确认插件参数
+    if not have_any_recent_upload_files(chatbot):
+        appendix_info = ""
+    else:
+        appendix_info = get_recent_file_prompt_support(chatbot)
+
+    plugin = plugin_arr_dict_parse[plugin_sel.plugin_selection]
+    yield from update_ui_lastest_msg(lastmsg=f"正在执行任务: {txt}\n\n提取插件参数...", chatbot=chatbot, history=history, delay=0)
+    class PluginExplicit(BaseModel):
+        plugin_selection: str = plugin_sel.plugin_selection
+        plugin_arg: str = Field(description="The argument of the plugin.", default="")
+    gpt_json_io = GptJsonIO(PluginExplicit)
+    gpt_json_io.format_instructions += "The information about this plugin is:" + plugin["Info"]
+    inputs = f"A plugin named {plugin_sel.plugin_selection} is selected, " + \
+             "you should extract plugin_arg from the user requirement, the user requirement is: \n\n" + \
+             ">> " + (txt + appendix_info).rstrip('\n').replace('\n','\n>> ') + '\n\n' + \
+             gpt_json_io.format_instructions 
+    run_gpt_fn = lambda inputs, sys_prompt: predict_no_ui_long_connection(
+        inputs=inputs, llm_kwargs=llm_kwargs, history=[], sys_prompt=sys_prompt, observe_window=[])
+    plugin_sel = gpt_json_io.generate_output_auto_repair(run_gpt_fn(inputs, ""), run_gpt_fn)
+
+
+    # ⭐ ⭐ ⭐ 执行插件
+    fn = plugin['Function']
+    fn_name = fn.__name__
+    msg = f'{llm_kwargs["llm_model"]}为您选择了插件: `{fn_name}`\n\n插件说明：{plugin["Info"]}\n\n插件参数：{plugin_sel.plugin_arg}\n\n假如偏离了您的要求，按停止键终止。'
+    yield from update_ui_lastest_msg(lastmsg=msg, chatbot=chatbot, history=history, delay=2)
+    yield from fn(plugin_sel.plugin_arg, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, -1)
+    return

crazy_functions/vt_fns/vt_modify_config.py

@@ -0,0 +1,81 @@
+from pydantic import BaseModel, Field
+from typing import List
+from toolbox import update_ui_lastest_msg, get_conf
+from request_llm.bridge_all import predict_no_ui_long_connection
+from crazy_functions.json_fns.pydantic_io import GptJsonIO
+import copy, json, pickle, os, sys
+
+
+def modify_configuration_hot(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_intention):
+    ALLOW_RESET_CONFIG, = get_conf('ALLOW_RESET_CONFIG')
+    if not ALLOW_RESET_CONFIG:
+        yield from update_ui_lastest_msg(
+            lastmsg=f"当前配置不允许被修改！如需激活本功能，请在config.py中设置ALLOW_RESET_CONFIG=True后重启软件。", 
+            chatbot=chatbot, history=history, delay=2
+        )
+        return
+
+    # ⭐ ⭐ ⭐ 读取可配置项目条目
+    names = {}
+    from enum import Enum
+    import config
+    for k, v in config.__dict__.items():
+        if k.startswith('__'): continue
+        names.update({k:k})
+        # if len(names) > 20: break   # 限制最多前10个配置项，如果太多了会导致gpt无法理解
+
+    ConfigOptions = Enum('ConfigOptions', names)
+    class ModifyConfigurationIntention(BaseModel):
+        which_config_to_modify: ConfigOptions = Field(description="the name of the configuration to modify, you must choose from one of the ConfigOptions enum.", default=None)
+        new_option_value: str = Field(description="the new value of the option", default=None)
+
+    # ⭐ ⭐ ⭐ 分析用户意图
+    yield from update_ui_lastest_msg(lastmsg=f"正在执行任务: {txt}\n\n读取新配置中", chatbot=chatbot, history=history, delay=0)
+    gpt_json_io = GptJsonIO(ModifyConfigurationIntention)
+    inputs = "Analyze how to change configuration according to following user input, answer me with json: \n\n" + \
+             ">> " + txt.rstrip('\n').replace('\n','\n>> ') + '\n\n' + \
+             gpt_json_io.format_instructions
+    
+    run_gpt_fn = lambda inputs, sys_prompt: predict_no_ui_long_connection(
+        inputs=inputs, llm_kwargs=llm_kwargs, history=[], sys_prompt=sys_prompt, observe_window=[])
+    user_intention = gpt_json_io.generate_output_auto_repair(run_gpt_fn(inputs, ""), run_gpt_fn)
+
+    explicit_conf = user_intention.which_config_to_modify.value
+
+    ok = (explicit_conf in txt)
+    if ok:
+        yield from update_ui_lastest_msg(
+            lastmsg=f"正在执行任务: {txt}\n\n新配置{explicit_conf}={user_intention.new_option_value}", 
+            chatbot=chatbot, history=history, delay=1
+        )
+        yield from update_ui_lastest_msg(
+            lastmsg=f"正在执行任务: {txt}\n\n新配置{explicit_conf}={user_intention.new_option_value}\n\n正在修改配置中", 
+            chatbot=chatbot, history=history, delay=2
+        )
+
+        # ⭐ ⭐ ⭐ 立即应用配置
+        from toolbox import set_conf
+        set_conf(explicit_conf, user_intention.new_option_value)
+
+        yield from update_ui_lastest_msg(
+            lastmsg=f"正在执行任务: {txt}\n\n配置修改完成，重新页面即可生效。", chatbot=chatbot, history=history, delay=1
+        )
+    else:
+        yield from update_ui_lastest_msg(
+            lastmsg=f"失败，如果需要配置{explicit_conf}，您需要明确说明并在指令中提到它。", chatbot=chatbot, history=history, delay=5
+        )
+
+def modify_configuration_reboot(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_intention):
+    ALLOW_RESET_CONFIG, = get_conf('ALLOW_RESET_CONFIG')
+    if not ALLOW_RESET_CONFIG:
+        yield from update_ui_lastest_msg(
+            lastmsg=f"当前配置不允许被修改！如需激活本功能，请在config.py中设置ALLOW_RESET_CONFIG=True后重启软件。", 
+            chatbot=chatbot, history=history, delay=2
+        )
+        return
+
+    yield from modify_configuration_hot(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_intention)
+    yield from update_ui_lastest_msg(
+        lastmsg=f"正在执行任务: {txt}\n\n配置修改完成，五秒后即将重启！若出现报错请无视即可。", chatbot=chatbot, history=history, delay=5
+    )
+    os.execl(sys.executable, sys.executable, *sys.argv)

crazy_functions/vt_fns/vt_state.py

@@ -0,0 +1,28 @@
+import pickle
+
+class VoidTerminalState():
+    def __init__(self):
+        self.reset_state()
+
+    def reset_state(self):
+        self.has_provided_explaination = False
+ 
+    def lock_plugin(self, chatbot):
+        chatbot._cookies['lock_plugin'] = 'crazy_functions.虚空终端->虚空终端'
+        chatbot._cookies['plugin_state'] = pickle.dumps(self)
+
+    def unlock_plugin(self, chatbot):
+        self.reset_state()
+        chatbot._cookies['lock_plugin'] = None
+        chatbot._cookies['plugin_state'] = pickle.dumps(self)
+
+    def set_state(self, chatbot, key, value):
+        setattr(self, key, value)
+        chatbot._cookies['plugin_state'] = pickle.dumps(self)
+
+    def get_state(chatbot):
+        state = chatbot._cookies.get('plugin_state', None)
+        if state is not None:   state = pickle.loads(state)
+        else:                   state = VoidTerminalState()
+        state.chatbot = chatbot
+        return state

crazy_functions/批量Markdown翻译.py

@@ -145,6 +145,8 @@ def get_files_from_everything(txt, preference=''):
         project_folder = txt
         file_manifest = [f for f in glob.glob(f'{project_folder}/**/*.md', recursive=True)]
     else:
+        project_folder = None
+        file_manifest = []
         success = False
 
     return success, file_manifest, project_folder

crazy_functions/批量翻译PDF文档_NOUGAT.py

@@ -0,0 +1,271 @@
+from toolbox import CatchException, report_execption, gen_time_str
+from toolbox import update_ui, promote_file_to_downloadzone, update_ui_lastest_msg, disable_auto_promotion
+from toolbox import write_history_to_file, get_log_folder
+from .crazy_utils import request_gpt_model_in_new_thread_with_ui_alive
+from .crazy_utils import request_gpt_model_multi_threads_with_very_awesome_ui_and_high_efficiency
+from .crazy_utils import read_and_clean_pdf_text
+from .pdf_fns.parse_pdf import parse_pdf, get_avail_grobid_url
+from colorful import *
+import os
+import math
+import logging
+
+def markdown_to_dict(article_content):
+    import markdown
+    from bs4 import BeautifulSoup
+    cur_t = ""
+    cur_c = ""
+    results = {}
+    for line in article_content:
+        if line.startswith('#'):
+            if cur_t!="":
+                if cur_t not in results:
+                    results.update({cur_t:cur_c.lstrip('\n')})
+                else:
+                    # 处理重名的章节
+                    results.update({cur_t + " " + gen_time_str():cur_c.lstrip('\n')})
+            cur_t = line.rstrip('\n')
+            cur_c = ""
+        else:
+            cur_c += line
+    results_final = {}
+    for k in list(results.keys()):
+        if k.startswith('# '):
+            results_final['title'] = k.split('# ')[-1]
+            results_final['authors'] = results.pop(k).lstrip('\n')
+        if k.startswith('###### Abstract'):
+            results_final['abstract'] = results.pop(k).lstrip('\n')
+
+    results_final_sections = []
+    for k,v in results.items():
+        results_final_sections.append({
+            'heading':k.lstrip("# "),
+            'text':v if len(v) > 0 else f"The beginning of {k.lstrip('# ')} section."
+        })
+    results_final['sections'] = results_final_sections
+    return results_final
+
+
+@CatchException
+def 批量翻译PDF文档(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, web_port):
+
+    disable_auto_promotion(chatbot)
+    # 基本信息：功能、贡献者
+    chatbot.append([
+        "函数插件功能？",
+        "批量翻译PDF文档。函数插件贡献者: Binary-Husky"])
+    yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+
+    # 尝试导入依赖，如果缺少依赖，则给出安装建议
+    try:
+        import nougat
+        import tiktoken
+    except:
+        report_execption(chatbot, history,
+                         a=f"解析项目: {txt}",
+                         b=f"导入软件依赖失败。使用该模块需要额外依赖，安装方法```pip install --upgrade nougat-ocr tiktoken```。")
+        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+        return
+
+    # 清空历史，以免输入溢出
+    history = []
+
+    from .crazy_utils import get_files_from_everything
+    success, file_manifest, project_folder = get_files_from_everything(txt, type='.pdf')
+    # 检测输入参数，如没有给定输入参数，直接退出
+    if not success:
+        if txt == "": txt = '空空如也的输入栏'
+
+    # 如果没找到任何文件
+    if len(file_manifest) == 0:
+        report_execption(chatbot, history,
+                         a=f"解析项目: {txt}", b=f"找不到任何.tex或.pdf文件: {txt}")
+        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+        return
+
+    # 开始正式执行任务
+    yield from 解析PDF_基于NOUGAT(file_manifest, project_folder, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt)
+
+    
+def nougat_with_timeout(command, cwd, timeout=3600):
+    import subprocess
+    process = subprocess.Popen(command, shell=True, cwd=cwd)
+    try:
+        stdout, stderr = process.communicate(timeout=timeout)
+    except subprocess.TimeoutExpired:
+        process.kill()
+        stdout, stderr = process.communicate()
+        print("Process timed out!")
+        return False
+    return True
+
+
+def NOUGAT_parse_pdf(fp):
+    import glob
+    from toolbox import get_log_folder, gen_time_str
+    dst = os.path.join(get_log_folder(plugin_name='nougat'), gen_time_str())
+    os.makedirs(dst)
+    nougat_with_timeout(f'nougat --out "{os.path.abspath(dst)}" "{os.path.abspath(fp)}"', os.getcwd())
+    res = glob.glob(os.path.join(dst,'*.mmd'))
+    if len(res) == 0:
+        raise RuntimeError("Nougat解析论文失败。")
+    return res[0]
+
+
+def 解析PDF_基于NOUGAT(file_manifest, project_folder, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt):
+    import copy
+    import tiktoken
+    TOKEN_LIMIT_PER_FRAGMENT = 1280
+    generated_conclusion_files = []
+    generated_html_files = []
+    DST_LANG = "中文"
+    for index, fp in enumerate(file_manifest):
+        chatbot.append(["当前进度：", f"正在解析论文，请稍候。（第一次运行时，需要花费较长时间下载NOUGAT参数）"]); yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+        fpp = NOUGAT_parse_pdf(fp)
+
+        with open(fpp, 'r', encoding='utf8') as f:
+            article_content = f.readlines()
+        article_dict = markdown_to_dict(article_content)
+        logging.info(article_dict)
+
+        prompt = "以下是一篇学术论文的基本信息:\n"
+        # title
+        title = article_dict.get('title', '无法获取 title'); prompt += f'title:{title}\n\n'
+        # authors
+        authors = article_dict.get('authors', '无法获取 authors'); prompt += f'authors:{authors}\n\n'
+        # abstract
+        abstract = article_dict.get('abstract', '无法获取 abstract'); prompt += f'abstract:{abstract}\n\n'
+        # command
+        prompt += f"请将题目和摘要翻译为{DST_LANG}。"
+        meta = [f'# Title:\n\n', title, f'# Abstract:\n\n', abstract ]
+
+        # 单线，获取文章meta信息
+        paper_meta_info = yield from request_gpt_model_in_new_thread_with_ui_alive(
+            inputs=prompt,
+            inputs_show_user=prompt,
+            llm_kwargs=llm_kwargs,
+            chatbot=chatbot, history=[],
+            sys_prompt="You are an academic paper reader。",
+        )
+
+        # 多线，翻译
+        inputs_array = []
+        inputs_show_user_array = []
+
+        # get_token_num
+        from request_llm.bridge_all import model_info
+        enc = model_info[llm_kwargs['llm_model']]['tokenizer']
+        def get_token_num(txt): return len(enc.encode(txt, disallowed_special=()))
+        from .crazy_utils import breakdown_txt_to_satisfy_token_limit_for_pdf
+
+        def break_down(txt):
+            raw_token_num = get_token_num(txt)
+            if raw_token_num <= TOKEN_LIMIT_PER_FRAGMENT:
+                return [txt]
+            else:
+                # raw_token_num > TOKEN_LIMIT_PER_FRAGMENT
+                # find a smooth token limit to achieve even seperation
+                count = int(math.ceil(raw_token_num / TOKEN_LIMIT_PER_FRAGMENT))
+                token_limit_smooth = raw_token_num // count + count
+                return breakdown_txt_to_satisfy_token_limit_for_pdf(txt, get_token_fn=get_token_num, limit=token_limit_smooth)
+
+        for section in article_dict.get('sections'):
+            if len(section['text']) == 0: continue
+            section_frags = break_down(section['text'])
+            for i, fragment in enumerate(section_frags):
+                heading = section['heading']
+                if len(section_frags) > 1: heading += f' Part-{i+1}'
+                inputs_array.append(
+                    f"你需要翻译{heading}章节，内容如下: \n\n{fragment}"
+                )
+                inputs_show_user_array.append(
+                    f"# {heading}\n\n{fragment}"
+                )
+
+        gpt_response_collection = yield from request_gpt_model_multi_threads_with_very_awesome_ui_and_high_efficiency(
+            inputs_array=inputs_array,
+            inputs_show_user_array=inputs_show_user_array,
+            llm_kwargs=llm_kwargs,
+            chatbot=chatbot,
+            history_array=[meta for _ in inputs_array],
+            sys_prompt_array=[
+                "请你作为一个学术翻译，负责把学术论文准确翻译成中文。注意文章中的每一句话都要翻译。" for _ in inputs_array],
+        )
+        res_path = write_history_to_file(meta +  ["# Meta Translation" , paper_meta_info] + gpt_response_collection, file_basename=None, file_fullname=None)
+        promote_file_to_downloadzone(res_path, rename_file=os.path.basename(fp)+'.md', chatbot=chatbot)
+        generated_conclusion_files.append(res_path)
+
+        ch = construct_html() 
+        orig = ""
+        trans = ""
+        gpt_response_collection_html = copy.deepcopy(gpt_response_collection)
+        for i,k in enumerate(gpt_response_collection_html): 
+            if i%2==0:
+                gpt_response_collection_html[i] = inputs_show_user_array[i//2]
+            else:
+                gpt_response_collection_html[i] = gpt_response_collection_html[i]
+
+        final = ["", "", "一、论文概况",  "", "Abstract", paper_meta_info,  "二、论文翻译",  ""]
+        final.extend(gpt_response_collection_html)
+        for i, k in enumerate(final): 
+            if i%2==0:
+                orig = k
+            if i%2==1:
+                trans = k
+                ch.add_row(a=orig, b=trans)
+        create_report_file_name = f"{os.path.basename(fp)}.trans.html"
+        html_file = ch.save_file(create_report_file_name)
+        generated_html_files.append(html_file)
+        promote_file_to_downloadzone(html_file, rename_file=os.path.basename(html_file), chatbot=chatbot)
+
+    chatbot.append(("给出输出文件清单", str(generated_conclusion_files + generated_html_files)))
+    yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+
+
+
+class construct_html():
+    def __init__(self) -> None:
+        self.css = """
+.row {
+  display: flex;
+  flex-wrap: wrap;
+}
+
+.column {
+  flex: 1;
+  padding: 10px;
+}
+
+.table-header {
+  font-weight: bold;
+  border-bottom: 1px solid black;
+}
+
+.table-row {
+  border-bottom: 1px solid lightgray;
+}
+
+.table-cell {
+  padding: 5px;
+}
+        """
+        self.html_string = f'<!DOCTYPE html><head><meta charset="utf-8"><title>翻译结果</title><style>{self.css}</style></head>'
+
+
+    def add_row(self, a, b):
+        tmp = """
+<div class="row table-row">
+    <div class="column table-cell">REPLACE_A</div>
+    <div class="column table-cell">REPLACE_B</div>
+</div>
+        """
+        from toolbox import markdown_convertion
+        tmp = tmp.replace('REPLACE_A', markdown_convertion(a))
+        tmp = tmp.replace('REPLACE_B', markdown_convertion(b))
+        self.html_string += tmp
+
+
+    def save_file(self, file_name):
+        with open(os.path.join(get_log_folder(), file_name), 'w', encoding='utf8') as f:
+            f.write(self.html_string.encode('utf-8', 'ignore').decode())
+        return os.path.join(get_log_folder(), file_name)

crazy_functions/批量翻译PDF文档_多线程.py

@@ -24,10 +24,11 @@ def 批量翻译PDF文档(txt, llm_kwargs, plugin_kwargs, chatbot, history, syst
     try:
         import fitz
         import tiktoken
+        import scipdf
     except:
         report_execption(chatbot, history,
                          a=f"解析项目: {txt}",
-                         b=f"导入软件依赖失败。使用该模块需要额外依赖，安装方法```pip install --upgrade pymupdf tiktoken```。")
+                         b=f"导入软件依赖失败。使用该模块需要额外依赖，安装方法```pip install --upgrade pymupdf tiktoken scipdf_parser```。")
         yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
         return
 
@@ -58,7 +59,6 @@ def 批量翻译PDF文档(txt, llm_kwargs, plugin_kwargs, chatbot, history, syst
 
 def 解析PDF_基于GROBID(file_manifest, project_folder, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, grobid_url):
     import copy
-    import tiktoken
     TOKEN_LIMIT_PER_FRAGMENT = 1280
     generated_conclusion_files = []
     generated_html_files = []
@@ -66,7 +66,7 @@ def 解析PDF_基于GROBID(file_manifest, project_folder, llm_kwargs, plugin_kwa
     for index, fp in enumerate(file_manifest):
         chatbot.append(["当前进度：", f"正在连接GROBID服务，请稍候: {grobid_url}\n如果等待时间过长，请修改config中的GROBID_URL，可修改成本地GROBID服务。"]); yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
         article_dict = parse_pdf(fp, grobid_url)
-        print(article_dict)
+        if article_dict is None: raise RuntimeError("解析PDF失败，请检查PDF是否损坏。")
         prompt = "以下是一篇学术论文的基本信息:\n"
         # title
         title = article_dict.get('title', '无法获取 title'); prompt += f'title:{title}\n\n'
@@ -113,7 +113,7 @@ def 解析PDF_基于GROBID(file_manifest, project_folder, llm_kwargs, plugin_kwa
             section_frags = break_down(section['text'])
             for i, fragment in enumerate(section_frags):
                 heading = section['heading']
-                if len(section_frags) > 1: heading += f'Part-{i+1}'
+                if len(section_frags) > 1: heading += f' Part-{i+1}'
                 inputs_array.append(
                     f"你需要翻译{heading}章节，内容如下: \n\n{fragment}"
                 )

crazy_functions/联网的ChatGPT.py

@@ -75,7 +75,11 @@ def 连接网络回答问题(txt, llm_kwargs, plugin_kwargs, chatbot, history, s
     proxies, = get_conf('proxies')
     urls = google(txt, proxies)
     history = []
-
+    if len(urls) == 0:
+        chatbot.append((f"结论：{txt}",
+                        "[Local Message] 受到google限制，无法从google获取信息！"))
+        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面 # 由于请求gpt需要一段时间，我们先及时地做一次界面更新
+        return
     # ------------- < 第2步：依次访问网页 > -------------
     max_search_result = 5   # 最多收纳多少个网页的结果
     for index, url in enumerate(urls[:max_search_result]):

crazy_functions/联网的ChatGPT_bing版.py

@@ -75,7 +75,11 @@ def 连接bing搜索回答问题(txt, llm_kwargs, plugin_kwargs, chatbot, histor
     proxies, = get_conf('proxies')
     urls = bing_search(txt, proxies)
     history = []
-
+    if len(urls) == 0:
+        chatbot.append((f"结论：{txt}",
+                        "[Local Message] 受到bing限制，无法从bing获取信息！"))
+        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面 # 由于请求gpt需要一段时间，我们先及时地做一次界面更新
+        return
     # ------------- < 第2步：依次访问网页 > -------------
     max_search_result = 8   # 最多收纳多少个网页的结果
     for index, url in enumerate(urls[:max_search_result]):

crazy_functions/虚空终端.py

@@ -1,119 +1,179 @@
+"""
+Explanation of the Void Terminal Plugin:
+
+Please describe in natural language what you want to do.
+
+1. You can open the plugin's dropdown menu to explore various capabilities of this project, and then describe your needs in natural language, for example:
+- "Please call the plugin to translate a PDF paper for me. I just uploaded the paper to the upload area."
+- "Please use the plugin to translate a PDF paper, with the address being https://www.nature.com/articles/s41586-019-1724-z.pdf."
+- "Generate an image with blooming flowers and lush green grass using the plugin."
+- "Translate the README using the plugin. The GitHub URL is https://github.com/facebookresearch/co-tracker."
+- "Translate an Arxiv paper for me. The Arxiv ID is 1812.10695. Remember to use the plugin and don't do it manually!"
+- "I don't like the current interface color. Modify the configuration and change the theme to THEME="High-Contrast"."
+- "Could you please explain the structure of the Transformer network?"
+
+2. If you use keywords like "call the plugin xxx", "modify the configuration xxx", "please", etc., your intention can be recognized more accurately.
+
+3. Your intention can be recognized more accurately when using powerful models like GPT4. This plugin is relatively new, so please feel free to provide feedback on GitHub.
+
+4. Now, if you need to process a file, please upload the file (drag the file to the file upload area) or describe the path to the file.
+
+5. If you don't need to upload a file, you can simply repeat your command again.
+"""
+explain_msg = """
+## 虚空终端插件说明:
+
+1. 请用**自然语言**描述您需要做什么。例如：
+    - 「请调用插件，为我翻译PDF论文，论文我刚刚放到上传区了」
+    - 「请调用插件翻译PDF论文，地址为https://aaa/bbb/ccc.pdf」
+    - 「把Arxiv论文翻译成中文PDF，arxiv论文的ID是1812.10695，记得用插件！」
+    - 「生成一张图片，图中鲜花怒放，绿草如茵，用插件实现」
+    - 「用插件翻译README，Github网址是https://github.com/facebookresearch/co-tracker」
+    - 「我不喜欢当前的界面颜色，修改配置，把主题THEME更换为THEME="High-Contrast"」
+    - 「请问Transformer网络的结构是怎样的？」
+
+2. 您可以打开插件下拉菜单以了解本项目的各种能力。    
+
+3. 如果您使用「调用插件xxx」、「修改配置xxx」、「请问」等关键词，您的意图可以被识别的更准确。
+
+4. 建议使用 GPT3.5 或更强的模型，弱模型可能无法理解您的想法。该插件诞生时间不长，欢迎您前往Github反馈问题。
+
+5. 现在，如果需要处理文件，请您上传文件（将文件拖动到文件上传区），或者描述文件所在的路径。
+
+6. 如果不需要上传文件，现在您只需要再次重复一次您的指令即可。
+"""
+
+from pydantic import BaseModel, Field
+from typing import List
 from toolbox import CatchException, update_ui, gen_time_str
-from .crazy_utils import request_gpt_model_in_new_thread_with_ui_alive
-from .crazy_utils import input_clipping
-import copy, json
+from toolbox import update_ui_lastest_msg, disable_auto_promotion
+from request_llm.bridge_all import predict_no_ui_long_connection
+from crazy_functions.crazy_utils import request_gpt_model_in_new_thread_with_ui_alive
+from crazy_functions.crazy_utils import input_clipping
+from crazy_functions.json_fns.pydantic_io import GptJsonIO, JsonStringError
+from crazy_functions.vt_fns.vt_state import VoidTerminalState
+from crazy_functions.vt_fns.vt_modify_config import modify_configuration_hot
+from crazy_functions.vt_fns.vt_modify_config import modify_configuration_reboot
+from crazy_functions.vt_fns.vt_call_plugin import execute_plugin
 
-def get_fn_lib():
-    return {
-        "BatchTranslatePDFDocuments_MultiThreaded": {
-                "module": "crazy_functions.批量翻译PDF文档_多线程",  
-                "function": "批量翻译PDF文档",
-                "description": "Translate PDF Documents",
-                "arg_1_description": "A path containing pdf files.",
-            },
-        "SummarizingWordDocuments": {
-                "module": "crazy_functions.总结word文档",  
-                "function": "总结word文档",
-                "description": "Summarize Word Documents",
-                "arg_1_description": "A path containing Word files.",
-            },
-        "ImageGeneration": {
-                "module": "crazy_functions.图片生成",  
-                "function": "图片生成",
-                "description": "Generate a image that satisfies some description.",
-                "arg_1_description": "Descriptions about the image to be generated.",
-            },
-        "TranslateMarkdownFromEnglishToChinese": {
-                "module": "crazy_functions.批量Markdown翻译",  
-                "function": "Markdown中译英",
-                "description": "Translate Markdown Documents from English to Chinese.",
-                "arg_1_description": "A path containing Markdown files.",
-            },
-        "SummaryAudioVideo": {
-                "module": "crazy_functions.总结音视频",  
-                "function": "总结音视频",
-                "description": "Get text from a piece of audio and summarize this audio.",
-                "arg_1_description": "A path containing audio files.",
-            },
-    }
+class UserIntention(BaseModel):
+    user_prompt: str = Field(description="the content of user input", default="")
+    intention_type: str = Field(description="the type of user intention, choose from ['ModifyConfiguration', 'ExecutePlugin', 'Chat']", default="ExecutePlugin")
+    user_provide_file: bool = Field(description="whether the user provides a path to a file", default=False)
+    user_provide_url: bool = Field(description="whether the user provides a url", default=False)
 
-functions = [
-    {
-        "name": k,
-        "description": v['description'],
-        "parameters": {
-            "type": "object",
-            "properties": {
-                "plugin_arg_1": {
-                    "type": "string",
-                    "description": v['arg_1_description'],
-                },
-            },
-            "required": ["plugin_arg_1"],
-        },
-    } for k, v in get_fn_lib().items()
-]
 
-def inspect_dependency(chatbot, history):
-    return True
+def chat(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_intention):
+    gpt_say = yield from request_gpt_model_in_new_thread_with_ui_alive(
+        inputs=txt, inputs_show_user=txt,
+        llm_kwargs=llm_kwargs, chatbot=chatbot, history=[], 
+        sys_prompt=system_prompt
+    )
+    chatbot[-1] = [txt, gpt_say]
+    history.extend([txt, gpt_say])
+    yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+    pass
 
-def eval_code(code, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, web_port):
-    import importlib
-    try:
-        tmp = get_fn_lib()[code['name']]
-        fp, fn = tmp['module'], tmp['function']
-        fn_plugin = getattr(importlib.import_module(fp, fn), fn)
-        arg = json.loads(code['arguments'])['plugin_arg_1']
-        yield from fn_plugin(arg, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, web_port)
-    except:
-        from toolbox import trimmed_format_exc
-        chatbot.append(["执行错误", f"\n```\n{trimmed_format_exc()}\n```\n"])
-        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
 
-def get_code_block(reply):
-    import re
-    pattern = r"```([\s\S]*?)```" # regex pattern to match code blocks
-    matches = re.findall(pattern, reply) # find all code blocks in text
-    if len(matches) != 1: 
-        raise RuntimeError("GPT is not generating proper code.")
-    return matches[0].strip('python') #  code block
+explain_intention_to_user = {
+    'Chat': "聊天对话",
+    'ExecutePlugin': "调用插件",
+    'ModifyConfiguration': "修改配置",
+}
+
+
+def analyze_intention_with_simple_rules(txt):
+    user_intention = UserIntention()
+    user_intention.user_prompt = txt
+    is_certain = False
+
+    if '请问' in txt:
+        is_certain = True
+        user_intention.intention_type = 'Chat'
+
+    if '用插件' in txt:
+        is_certain = True
+        user_intention.intention_type = 'ExecutePlugin'
+
+    if '修改配置' in txt:
+        is_certain = True
+        user_intention.intention_type = 'ModifyConfiguration'
+
+    return is_certain, user_intention
+
 
 @CatchException
-def 终端(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, web_port):
-    """
-    txt             输入栏用户输入的文本, 例如需要翻译的一段话, 再例如一个包含了待处理文件的路径
-    llm_kwargs      gpt模型参数, 如温度和top_p等, 一般原样传递下去就行
-    plugin_kwargs   插件模型的参数, 暂时没有用武之地
-    chatbot         聊天显示框的句柄, 用于显示给用户
-    history         聊天历史, 前情提要
-    system_prompt   给gpt的静默提醒
-    web_port        当前软件运行的端口号
-    """
-    # 清空历史, 以免输入溢出
-    history = []    
+def 虚空终端(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, web_port):
+    disable_auto_promotion(chatbot=chatbot)
+    # 获取当前虚空终端状态
+    state = VoidTerminalState.get_state(chatbot)
+    appendix_msg = ""
 
-    # 基本信息：功能、贡献者
-    chatbot.append(["虚空终端插件的功能？", "根据自然语言的描述, 执行任意插件的命令."])
-    yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
-    
-    # 输入
-    i_say = txt
-    # 开始
-    llm_kwargs_function_call = copy.deepcopy(llm_kwargs)
-    llm_kwargs_function_call['llm_model'] = 'gpt-call-fn' # 修改调用函数
-    gpt_say = yield from request_gpt_model_in_new_thread_with_ui_alive(
-        inputs=i_say, inputs_show_user=txt, 
-        llm_kwargs=llm_kwargs_function_call, chatbot=chatbot, history=[], 
-        sys_prompt=functions
-    )
-
-    # 将代码转为动画
-    res = json.loads(gpt_say)['choices'][0]
-    if res['finish_reason'] == 'function_call':
-        code = json.loads(gpt_say)['choices'][0]
-        yield from eval_code(code['message']['function_call'], llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, web_port)
+    # 用简单的关键词检测用户意图
+    is_certain, _ = analyze_intention_with_simple_rules(txt)
+    if txt.startswith('private_upload/') and len(txt) == 34:
+        state.set_state(chatbot=chatbot, key='has_provided_explaination', value=False)
+        appendix_msg = "\n\n**很好，您已经上传了文件**，现在请您描述您的需求。"
+        
+    if is_certain or (state.has_provided_explaination):
+        # 如果意图明确，跳过提示环节
+        state.set_state(chatbot=chatbot, key='has_provided_explaination', value=True)
+        state.unlock_plugin(chatbot=chatbot)
+        yield from update_ui(chatbot=chatbot, history=history)
+        yield from 虚空终端主路由(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, web_port)
+        return
     else:
-        chatbot.append(["无法调用相关功能", res])
-        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+        # 如果意图模糊，提示
+        state.set_state(chatbot=chatbot, key='has_provided_explaination', value=True)
+        state.lock_plugin(chatbot=chatbot)
+        chatbot.append(("虚空终端状态:", explain_msg+appendix_msg))
+        yield from update_ui(chatbot=chatbot, history=history)
+        return
 
 
+
+def 虚空终端主路由(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, web_port):
+    history = []
+    chatbot.append(("虚空终端状态: ", f"正在执行任务: {txt}"))
+    yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+
+    # ⭐ ⭐ ⭐ 分析用户意图
+    is_certain, user_intention = analyze_intention_with_simple_rules(txt)
+    if not is_certain:
+        yield from update_ui_lastest_msg(
+            lastmsg=f"正在执行任务: {txt}\n\n分析用户意图中", chatbot=chatbot, history=history, delay=0)
+        gpt_json_io = GptJsonIO(UserIntention)
+        rf_req = "\nchoose from ['ModifyConfiguration', 'ExecutePlugin', 'Chat']"
+        inputs = "Analyze the intention of the user according to following user input: \n\n" + \
+            ">> " + (txt+rf_req).rstrip('\n').replace('\n','\n>> ') + '\n\n' + gpt_json_io.format_instructions
+        run_gpt_fn = lambda inputs, sys_prompt: predict_no_ui_long_connection(
+            inputs=inputs, llm_kwargs=llm_kwargs, history=[], sys_prompt=sys_prompt, observe_window=[])
+        analyze_res = run_gpt_fn(inputs, "")
+        try:
+            user_intention = gpt_json_io.generate_output_auto_repair(analyze_res, run_gpt_fn)
+            lastmsg=f"正在执行任务: {txt}\n\n用户意图理解: 意图={explain_intention_to_user[user_intention.intention_type]}", 
+        except JsonStringError as e:
+            yield from update_ui_lastest_msg(
+                lastmsg=f"正在执行任务: {txt}\n\n用户意图理解: 失败 当前语言模型（{llm_kwargs['llm_model']}）不能理解您的意图", chatbot=chatbot, history=history, delay=0)
+            return
+    else:
+        pass
+
+    yield from update_ui_lastest_msg(
+        lastmsg=f"正在执行任务: {txt}\n\n用户意图理解: 意图={explain_intention_to_user[user_intention.intention_type]}", 
+        chatbot=chatbot, history=history, delay=0)
+
+    # 用户意图: 修改本项目的配置
+    if user_intention.intention_type == 'ModifyConfiguration':
+        yield from modify_configuration_reboot(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_intention)
+
+    # 用户意图: 调度插件
+    if user_intention.intention_type == 'ExecutePlugin':
+        yield from execute_plugin(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_intention)
+
+    # 用户意图: 聊天
+    if user_intention.intention_type == 'Chat':
+        yield from chat(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_intention)
+
+    return
+

crazy_functions/语音助手.py

@@ -80,9 +80,9 @@ class InterviewAssistant(AliyunASR):
     def __init__(self):
         self.capture_interval = 0.5 # second
         self.stop = False
-        self.parsed_text = ""
-        self.parsed_sentence = ""
-        self.buffered_sentence = ""
+        self.parsed_text = ""   # 下个句子中已经说完的部分, 由 test_on_result_chg() 写入
+        self.parsed_sentence = ""   # 某段话的整个句子,由 test_on_sentence_end() 写入
+        self.buffered_sentence = ""    #
         self.event_on_result_chg = threading.Event()
         self.event_on_entence_end = threading.Event()
         self.event_on_commit_question = threading.Event()
@@ -132,7 +132,7 @@ class InterviewAssistant(AliyunASR):
             self.plugin_wd.feed()
 
             if self.event_on_result_chg.is_set(): 
-                # update audio decode result
+                # called when some words have finished
                 self.event_on_result_chg.clear()
                 chatbot[-1] = list(chatbot[-1])
                 chatbot[-1][0] = self.buffered_sentence + self.parsed_text
@@ -144,7 +144,11 @@ class InterviewAssistant(AliyunASR):
                 # called when a sentence has ended
                 self.event_on_entence_end.clear()
                 self.parsed_text = self.parsed_sentence
-                self.buffered_sentence += self.parsed_sentence
+                self.buffered_sentence += self.parsed_text
+                chatbot[-1] = list(chatbot[-1])
+                chatbot[-1][0] = self.buffered_sentence
+                history = chatbot2history(chatbot)
+                yield from update_ui(chatbot=chatbot, history=history)  # 刷新界面
 
             if self.event_on_commit_question.is_set():
                 # called when a question should be commited

crazy_functions/谷歌检索小助手.py

@@ -1,26 +1,75 @@
 from .crazy_utils import request_gpt_model_in_new_thread_with_ui_alive
-from toolbox import CatchException, report_execption, write_results_to_file
-from toolbox import update_ui
+from toolbox import CatchException, report_execption, promote_file_to_downloadzone
+from toolbox import update_ui, update_ui_lastest_msg, disable_auto_promotion, write_history_to_file
+import logging
+import requests
+import time
+import random
+
+ENABLE_ALL_VERSION_SEARCH = True
 
 def get_meta_information(url, chatbot, history):
-    import requests
     import arxiv
     import difflib
+    import re
     from bs4 import BeautifulSoup
     from toolbox import get_conf
+    from urllib.parse import urlparse
+    session = requests.session()
+
     proxies, = get_conf('proxies')
     headers = {
-        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/105.0.0.0 Safari/537.36',
+        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/116.0.0.0 Safari/537.36',
+        'Accept-Encoding': 'gzip, deflate, br', 
+        'Accept-Language': 'en-US,en;q=0.9,zh-CN;q=0.8,zh;q=0.7',
+        'Cache-Control':'max-age=0',
+        'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7', 
+        'Connection': 'keep-alive'
     }
-    # 发送 GET 请求
-    response = requests.get(url, proxies=proxies, headers=headers)
+    session.proxies.update(proxies)
+    session.headers.update(headers)
 
+    response = session.get(url)
     # 解析网页内容
     soup = BeautifulSoup(response.text, "html.parser")
 
     def string_similar(s1, s2):
         return difflib.SequenceMatcher(None, s1, s2).quick_ratio()
 
+    if ENABLE_ALL_VERSION_SEARCH:
+        def search_all_version(url):
+            time.sleep(random.randint(1,5)) # 睡一会防止触发google反爬虫
+            response = session.get(url)
+            soup = BeautifulSoup(response.text, "html.parser")
+
+            for result in soup.select(".gs_ri"):
+                try:
+                    url = result.select_one(".gs_rt").a['href']
+                except:
+                    continue
+                arxiv_id = extract_arxiv_id(url)
+                if not arxiv_id:
+                    continue
+                search = arxiv.Search(
+                    id_list=[arxiv_id],
+                    max_results=1,
+                    sort_by=arxiv.SortCriterion.Relevance,
+                )
+                try: paper = next(search.results())
+                except: paper = None
+                return paper
+
+            return None
+
+        def extract_arxiv_id(url):
+            # 返回给定的url解析出的arxiv_id，如url未成功匹配返回None
+            pattern = r'arxiv.org/abs/([^/]+)'
+            match = re.search(pattern, url)
+            if match:
+                return match.group(1)
+            else:
+                return None
+
     profile = []
     # 获取所有文章的标题和作者
     for result in soup.select(".gs_ri"):
@@ -31,32 +80,45 @@ def get_meta_information(url, chatbot, history):
         except:
             citation = 'cited by 0'
         abstract = result.select_one(".gs_rs").text.strip()  # 摘要在 .gs_rs 中的文本，需要清除首尾空格
+
+        # 首先在arxiv上搜索，获取文章摘要
         search = arxiv.Search(
             query = title,
             max_results = 1,
             sort_by = arxiv.SortCriterion.Relevance,
         )
-        try:
-            paper = next(search.results())
-            if string_similar(title, paper.title) > 0.90: # same paper
-                abstract = paper.summary.replace('\n', ' ')
-                is_paper_in_arxiv = True
-            else:   # different paper
-                abstract = abstract
-                is_paper_in_arxiv = False
-            paper = next(search.results())
-        except:
+        try: paper = next(search.results())
+        except: paper = None
+        
+        is_match = paper is not None and string_similar(title, paper.title) > 0.90
+
+        # 如果在Arxiv上匹配失败，检索文章的历史版本的题目
+        if not is_match and ENABLE_ALL_VERSION_SEARCH:
+            other_versions_page_url = [tag['href'] for tag in result.select_one('.gs_flb').select('.gs_nph') if 'cluster' in tag['href']]
+            if len(other_versions_page_url) > 0:
+                other_versions_page_url = other_versions_page_url[0]
+                paper = search_all_version('http://' + urlparse(url).netloc + other_versions_page_url)
+                is_match = paper is not None and string_similar(title, paper.title) > 0.90
+
+        if is_match:
+            # same paper
+            abstract = paper.summary.replace('\n', ' ')
+            is_paper_in_arxiv = True
+        else:
+            # different paper
             abstract = abstract
             is_paper_in_arxiv = False
-        print(title)
-        print(author)
-        print(citation)
+
+        logging.info('[title]:' + title)
+        logging.info('[author]:' + author)
+        logging.info('[citation]:' + citation)
+
         profile.append({
-            'title':title,
-            'author':author,
-            'citation':citation,
-            'abstract':abstract,
-            'is_paper_in_arxiv':is_paper_in_arxiv,
+            'title': title,
+            'author': author,
+            'citation': citation,
+            'abstract': abstract,
+            'is_paper_in_arxiv': is_paper_in_arxiv,
         })
 
         chatbot[-1] = [chatbot[-1][0], title + f'\n\n是否在arxiv中（不在arxiv中无法获取完整摘要）:{is_paper_in_arxiv}\n\n' + abstract]
@@ -65,6 +127,7 @@ def get_meta_information(url, chatbot, history):
 
 @CatchException
 def 谷歌检索小助手(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, web_port):
+    disable_auto_promotion(chatbot=chatbot)
     # 基本信息：功能、贡献者
     chatbot.append([
         "函数插件功能？",
@@ -86,6 +149,9 @@ def 谷歌检索小助手(txt, llm_kwargs, plugin_kwargs, chatbot, history, syst
     # 清空历史，以免输入溢出
     history = []
     meta_paper_info_list = yield from get_meta_information(txt, chatbot, history)
+    if len(meta_paper_info_list) == 0:
+        yield from update_ui_lastest_msg(lastmsg='获取文献失败，可能触发了google反爬虫机制。',chatbot=chatbot, history=history, delay=0)
+        return
     batchsize = 5
     for batch in range(math.ceil(len(meta_paper_info_list)/batchsize)):
         if len(meta_paper_info_list[:batchsize]) > 0:
@@ -107,6 +173,7 @@ def 谷歌检索小助手(txt, llm_kwargs, plugin_kwargs, chatbot, history, syst
         "已经全部完成，您可以试试让AI写一个Related Works，例如您可以继续输入Write a \"Related Works\" section about \"你搜索的研究领域\" for me."])
     msg = '正常'
     yield from update_ui(chatbot=chatbot, history=history, msg=msg) # 刷新界面
-    res = write_results_to_file(history)
-    chatbot.append(("完成了吗？", res)); 
+    path = write_history_to_file(history)
+    promote_file_to_downloadzone(path, chatbot=chatbot)
+    chatbot.append(("完成了吗？", path)); 
     yield from update_ui(chatbot=chatbot, history=history, msg=msg) # 刷新界面