镜像自地址
https://github.com/binary-husky/gpt_academic.git
已同步 2025-12-06 06:26:47 +00:00
243 行
9.7 KiB
Python
243 行
9.7 KiB
Python
from toolbox import CatchException, update_ui
|
||
from .crazy_utils import request_gpt_model_in_new_thread_with_ui_alive
|
||
from .crazy_utils import read_and_clean_pdf_text
|
||
import datetime
|
||
|
||
#暂时只写了这几种的PROMPT
|
||
SELECT_PROMPT = """
|
||
“{subject}”
|
||
=============
|
||
以上是从文章中提取的摘要,将会使用这些摘要绘制图表。请你选择一个合适的图表类型:
|
||
1 流程图
|
||
2 序列图
|
||
3 类图
|
||
4 饼图
|
||
5 甘特图
|
||
6 状态图
|
||
7 实体关系图
|
||
8 象限提示图
|
||
不需要解释原因，仅需要输出单个不带任何标点符号的数字。
|
||
"""
|
||
#流程图
|
||
PROMPT_1 = """
|
||
请你给出围绕“{subject}”的逻辑关系图，使用mermaid语法，mermaid语法举例：
|
||
```mermaid
|
||
graph TD
|
||
P(编程) --> L1(Python)
|
||
P(编程) --> L2(C)
|
||
P(编程) --> L3(C++)
|
||
P(编程) --> L4(Javascipt)
|
||
P(编程) --> L5(PHP)
|
||
```
|
||
"""
|
||
#序列图
|
||
PROMPT_2 = """
|
||
请你给出围绕“{subject}”的序列图，使用mermaid语法，mermaid语法举例：
|
||
```mermaid
|
||
sequenceDiagram
|
||
participant A as 用户
|
||
participant B as 系统
|
||
A->>B: 登录请求
|
||
B->>A: 登录成功
|
||
A->>B: 获取数据
|
||
B->>A: 返回数据
|
||
```
|
||
"""
|
||
#类图
|
||
PROMPT_3 = """
|
||
请你给出围绕“{subject}”的类图，使用mermaid语法，mermaid语法举例：
|
||
```mermaid
|
||
classDiagram
|
||
Class01 <|-- AveryLongClass : Cool
|
||
Class03 *-- Class04
|
||
Class05 o-- Class06
|
||
Class07 .. Class08
|
||
Class09 --> C2 : Where am i?
|
||
Class09 --* C3
|
||
Class09 --|> Class07
|
||
Class07 : equals()
|
||
Class07 : Object[] elementData
|
||
Class01 : size()
|
||
Class01 : int chimp
|
||
Class01 : int gorilla
|
||
Class08 <--> C2: Cool label
|
||
```
|
||
"""
|
||
#饼图
|
||
PROMPT_4 = """
|
||
请你给出围绕“{subject}”的饼图，使用mermaid语法，mermaid语法举例：
|
||
```mermaid
|
||
pie title Pets adopted by volunteers
|
||
"狗" : 386
|
||
"猫" : 85
|
||
"兔子" : 15
|
||
```
|
||
"""
|
||
#甘特图
|
||
PROMPT_5 = """
|
||
请你给出围绕“{subject}”的甘特图，使用mermaid语法，mermaid语法举例：
|
||
```mermaid
|
||
gantt
|
||
title 项目开发流程
|
||
dateFormat YYYY-MM-DD
|
||
section 设计
|
||
需求分析 :done, des1, 2024-01-06,2024-01-08
|
||
原型设计 :active, des2, 2024-01-09, 3d
|
||
UI设计 : des3, after des2, 5d
|
||
section 开发
|
||
前端开发 :2024-01-20, 10d
|
||
后端开发 :2024-01-20, 10d
|
||
```
|
||
"""
|
||
#状态图
|
||
PROMPT_6 = """
|
||
请你给出围绕“{subject}”的状态图，使用mermaid语法，mermaid语法举例：
|
||
```mermaid
|
||
stateDiagram-v2
|
||
[*] --> Still
|
||
Still --> [*]
|
||
Still --> Moving
|
||
Moving --> Still
|
||
Moving --> Crash
|
||
Crash --> [*]
|
||
```
|
||
"""
|
||
#实体关系图
|
||
PROMPT_7 = """
|
||
请你给出围绕“{subject}”的实体关系图，使用mermaid语法，mermaid语法举例：
|
||
```mermaid
|
||
erDiagram
|
||
CUSTOMER ||--o{ ORDER : places
|
||
ORDER ||--|{ LINE-ITEM : contains
|
||
CUSTOMER {
|
||
string name
|
||
string id
|
||
}
|
||
ORDER {
|
||
string orderNumber
|
||
date orderDate
|
||
string customerID
|
||
}
|
||
LINE-ITEM {
|
||
number quantity
|
||
string productID
|
||
}
|
||
```
|
||
"""
|
||
#象限提示图
|
||
PROMPT_8 = """
|
||
请你给出围绕“{subject}”的象限图，使用mermaid语法，mermaid语法举例：
|
||
```mermaid
|
||
graph LR
|
||
A[Hard skill] --> B(Programming)
|
||
A[Hard skill] --> C(Design)
|
||
D[Soft skill] --> E(Coordination)
|
||
D[Soft skill] --> F(Communication)
|
||
```
|
||
"""
|
||
|
||
def 解析历史输入(history,llm_kwargs,chatbot):
|
||
############################## <第 0 步，切割输入> ##################################
|
||
# 借用PDF切割中的函数对文本进行切割
|
||
TOKEN_LIMIT_PER_FRAGMENT = 2500
|
||
txt = str(history).encode('utf-8', 'ignore').decode() # avoid reading non-utf8 chars
|
||
from crazy_functions.pdf_fns.breakdown_txt import breakdown_text_to_satisfy_token_limit
|
||
txt = breakdown_text_to_satisfy_token_limit(txt=txt, limit=TOKEN_LIMIT_PER_FRAGMENT, llm_model=llm_kwargs['llm_model'])
|
||
############################## <第 1 步，迭代地历遍整个文章，提取精炼信息> ##################################
|
||
i_say_show_user = f'首先你从历史记录或文件中提取摘要。'; gpt_say = "[Local Message] 收到。" # 用户提示
|
||
chatbot.append([i_say_show_user, gpt_say]); yield from update_ui(chatbot=chatbot, history=history) # 更新UI
|
||
results = []
|
||
MAX_WORD_TOTAL = 4096
|
||
n_txt = len(txt)
|
||
last_iteration_result = "从以下文本中提取摘要。"
|
||
if n_txt >= 20: print('文章极长，不能达到预期效果')
|
||
for i in range(n_txt):
|
||
NUM_OF_WORD = MAX_WORD_TOTAL // n_txt
|
||
i_say = f"Read this section, recapitulate the content of this section with less than {NUM_OF_WORD} words: {txt[i]}"
|
||
i_say_show_user = f"[{i+1}/{n_txt}] Read this section, recapitulate the content of this section with less than {NUM_OF_WORD} words: {txt[i][:200]} ...."
|
||
gpt_say = yield from request_gpt_model_in_new_thread_with_ui_alive(i_say, i_say_show_user, # i_say=真正给chatgpt的提问， i_say_show_user=给用户看的提问
|
||
llm_kwargs, chatbot,
|
||
history=["The main content of the previous section is?", last_iteration_result], # 迭代上一次的结果
|
||
sys_prompt="Extracts the main content from the text section where it is located for graphing purposes, answer me with Chinese." # 提示
|
||
)
|
||
results.append(gpt_say)
|
||
last_iteration_result = gpt_say
|
||
############################## <第 2 步，根据整理的摘要选择图表类型> ##################################
|
||
i_say_show_user = f'接下来将判断适合的图表类型,如连续3次判断失败将会使用流程图进行绘制'; gpt_say = "[Local Message] 收到。" # 用户提示
|
||
chatbot.append([i_say_show_user, gpt_say]); yield from update_ui(chatbot=chatbot, history=[]) # 更新UI
|
||
results_txt = '\n'.join(results)
|
||
i_say = SELECT_PROMPT.format(subject=results_txt)
|
||
i_say_show_user = f'请判断适合使用的流程图类型,其中数字对应关系为:1-流程图,2-序列图,3-类图,4-饼图,5-甘特图,6-状态图,7-实体关系图,8-象限提示图'
|
||
for i in range(3):
|
||
gpt_say = yield from request_gpt_model_in_new_thread_with_ui_alive(
|
||
inputs=i_say,
|
||
inputs_show_user=i_say_show_user,
|
||
llm_kwargs=llm_kwargs, chatbot=chatbot, history=[],
|
||
sys_prompt=""
|
||
)
|
||
if gpt_say in ['1','2','3','4','5','6','7','8']: #判断返回是否正确
|
||
break
|
||
if gpt_say not in ['1','2','3','4','5','6','7','8']:
|
||
gpt_say = '1'
|
||
############################## <第 3 步，根据选择的图表类型绘制图表> ##################################
|
||
if gpt_say == '1':
|
||
i_say = PROMPT_1.format(subject=results_txt)
|
||
elif gpt_say == '2':
|
||
i_say = PROMPT_2.format(subject=results_txt)
|
||
elif gpt_say == '3':
|
||
i_say = PROMPT_3.format(subject=results_txt)
|
||
elif gpt_say == '4':
|
||
i_say = PROMPT_4.format(subject=results_txt)
|
||
elif gpt_say == '5':
|
||
i_say = PROMPT_5.format(subject=results_txt)
|
||
elif gpt_say == '6':
|
||
i_say = PROMPT_6.format(subject=results_txt)
|
||
elif gpt_say == '7':
|
||
i_say = PROMPT_7.format(subject=results_txt)
|
||
elif gpt_say == '8':
|
||
i_say = PROMPT_8.format(subject=results_txt)
|
||
i_say_show_user = f'请根据判断结果绘制相应的图表。'
|
||
gpt_say = yield from request_gpt_model_in_new_thread_with_ui_alive(
|
||
inputs=i_say,
|
||
inputs_show_user=i_say_show_user,
|
||
llm_kwargs=llm_kwargs, chatbot=chatbot, history=[],
|
||
sys_prompt=""
|
||
)
|
||
history.append(gpt_say)
|
||
yield from update_ui(chatbot=chatbot, history=history) # 刷新界面 # 界面更新
|
||
|
||
def 输入区文件处理(txt):
|
||
if txt == "": return False, txt
|
||
success = True
|
||
import glob
|
||
from .crazy_utils import get_files_from_everything
|
||
file_pdf,pdf_manifest,folder_pdf = get_files_from_everything(txt, '.pdf')
|
||
if not file_pdf or len(pdf_manifest) == 0:
|
||
return False, txt #如不是pdf文件则返回输入区内容
|
||
final_result = ""
|
||
for index, fp in enumerate(pdf_manifest):
|
||
file_content, page_one = read_and_clean_pdf_text(fp) # （尝试）按照章节切割PDF
|
||
file_content = file_content.encode('utf-8', 'ignore').decode() # avoid reading non-utf8 chars
|
||
final_result += file_content
|
||
return True, final_result
|
||
|
||
@CatchException
|
||
def 生成多种Mermaid图表(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，用于灵活调整复杂功能的各种参数
|
||
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
|
||
"""
|
||
import os
|
||
chatbot.append([
|
||
"函数插件功能？",
|
||
"根据当前聊天历史或PDF中绘制多种mermaid图表的功能，将会首先判断适合的图表类型，随后绘制图表。函数插件贡献者: Menghuan1918"])
|
||
yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
|
||
if os.path.exists(txt): #如输入区无内容则直接解析历史记录
|
||
file_exist, txt = 输入区文件处理(txt)
|
||
history.append(txt) #将解析后的txt传递加入到历史中
|
||
yield from 解析历史输入(history,llm_kwargs,chatbot) |