add the '/Fit' destination

* Add arm64 support

* Bug fix

* Some build bug fix

* Add arm support

* 分离arm和x86构建

* 改进构建文档

* update tags

* Update build-with-latex-arm.yml

* Revert "Update build-with-latex-arm.yml"

This reverts commit 9af92549b5.

* Update

* Add

* httpx

* Addison

* Update GithubAction+NoLocal+Latex

* Update docker-compose.yml and GithubAction+NoLocal+Latex

* Update README.md

* test math anim generation

* solve the pdf concatenate error. (#2006)

* solve the pdf concatenate error.

* add legacy fallback option

---------

Co-authored-by: binary-husky <qingxu.fu@outlook.com>

---------

Co-authored-by: binary-husky <96192199+binary-husky@users.noreply.github.com>
Co-authored-by: binary-husky <qingxu.fu@outlook.com>
Co-authored-by: wsg1873 <wsg0326@163.com>

.github/workflows/build-with-jittorllms.yml

@@ -1,44 +0,0 @@
-# https://docs.github.com/en/actions/publishing-packages/publishing-docker-images#publishing-images-to-github-packages
-name: build-with-jittorllms
-
-on:
-  push:
-    branches:
-      - 'master'
-
-env:
-  REGISTRY: ghcr.io
-  IMAGE_NAME: ${{ github.repository }}_jittorllms
-
-jobs:
-  build-and-push-image:
-    runs-on: ubuntu-latest
-    permissions:
-      contents: read
-      packages: write
-
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v3
-
-      - name: Log in to the Container registry
-        uses: docker/login-action@v2
-        with:
-          registry: ${{ env.REGISTRY }}
-          username: ${{ github.actor }}
-          password: ${{ secrets.GITHUB_TOKEN }}
-
-      - name: Extract metadata (tags, labels) for Docker
-        id: meta
-        uses: docker/metadata-action@v4
-        with:
-          images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
-
-      - name: Build and push Docker image
-        uses: docker/build-push-action@v4
-        with:
-          context: .
-          push: true
-          file: docs/GithubAction+JittorLLMs
-          tags: ${{ steps.meta.outputs.tags }}
-          labels: ${{ steps.meta.outputs.labels }}

.github/workflows/build-with-latex-arm.yml

@@ -1,14 +1,14 @@
 # https://docs.github.com/en/actions/publishing-packages/publishing-docker-images#publishing-images-to-github-packages
-name: build-with-all-capacity-beta
+name: build-with-latex-arm
 
 on:
   push:
     branches:
-      - 'master'
+      - "master"
 
 env:
   REGISTRY: ghcr.io
-  IMAGE_NAME: ${{ github.repository }}_with_all_capacity_beta
+  IMAGE_NAME: ${{ github.repository }}_with_latex_arm
 
 jobs:
   build-and-push-image:
@@ -18,11 +18,17 @@ jobs:
       packages: write
 
     steps:
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v3
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
       - name: Checkout repository
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Log in to the Container registry
-        uses: docker/login-action@v2
+        uses: docker/login-action@v3
         with:
           registry: ${{ env.REGISTRY }}
           username: ${{ github.actor }}
@@ -35,10 +41,11 @@ jobs:
           images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
 
       - name: Build and push Docker image
-        uses: docker/build-push-action@v4
+        uses: docker/build-push-action@v6
         with:
           context: .
           push: true
-          file: docs/GithubAction+AllCapacityBeta
+          platforms: linux/arm64
+          file: docs/GithubAction+NoLocal+Latex
           tags: ${{ steps.meta.outputs.tags }}
-          labels: ${{ steps.meta.outputs.labels }}
+          labels: ${{ steps.meta.outputs.labels }}

README.md

@@ -1,5 +1,6 @@
 > [!IMPORTANT]
-> 2024.6.1: 版本3.80加入插件二级菜单功能（详见wiki）  
+> 2024.10.10: 突发停电，紧急恢复了提供[whl包](https://drive.google.com/file/d/19U_hsLoMrjOlQSzYS3pzWX9fTzyusArP/view?usp=sharing)的文件服务器  
+> 2024.10.8: 版本3.90加入对llama-index的初步支持，版本3.80加入插件二级菜单功能（详见wiki）  
 > 2024.5.1: 加入Doc2x翻译PDF论文的功能，[查看详情](https://github.com/binary-husky/gpt_academic/wiki/Doc2x)  
 > 2024.3.11: 全力支持Qwen、GLM、DeepseekCoder等中文大语言模型！ SoVits语音克隆模块，[查看详情](https://www.bilibili.com/video/BV1Rp421S7tF/) 
 > 2024.1.17: 安装依赖时，请选择`requirements.txt`中**指定的版本**。 安装命令：`pip install -r requirements.txt`。本项目完全开源免费，您可通过订阅[在线服务](https://github.com/binary-husky/gpt_academic/wiki/online)的方式鼓励本项目的发展。

config.py

@@ -57,9 +57,9 @@ EMBEDDING_MODEL = "text-embedding-3-small"
 #   "yi-34b-chat-0205","yi-34b-chat-200k","yi-large","yi-medium","yi-spark","yi-large-turbo","yi-large-preview",
 # ]
 # --- --- --- ---
-# 此外，您还可以在接入one-api/vllm/ollama时，
-# 使用"one-api-*","vllm-*","ollama-*"前缀直接使用非标准方式接入的模型，例如
-# AVAIL_LLM_MODELS = ["one-api-claude-3-sonnet-20240229(max_token=100000)", "ollama-phi3(max_token=4096)"]
+# 此外，您还可以在接入one-api/vllm/ollama/Openroute时，
+# 使用"one-api-*","vllm-*","ollama-*","openrouter-*"前缀直接使用非标准方式接入的模型，例如
+# AVAIL_LLM_MODELS = ["one-api-claude-3-sonnet-20240229(max_token=100000)", "ollama-phi3(max_token=4096)","openrouter-openai/gpt-4o-mini","openrouter-openai/chatgpt-4o-latest"]
 # --- --- --- ---
 
 

core_functional.py

@@ -17,7 +17,7 @@ def get_core_functions():
                             text_show_english=
                                 r"Below is a paragraph from an academic paper. Polish the writing to meet the academic style, "
                                 r"improve the spelling, grammar, clarity, concision and overall readability. When necessary, rewrite the whole sentence. "
-                                r"Firstly, you should provide the polished paragraph. "
+                                r"Firstly, you should provide the polished paragraph (in English). "
                                 r"Secondly, you should list all your modification and explain the reasons to do so in markdown table.",
                             text_show_chinese=
                                 r"作为一名中文学术论文写作改进助理，你的任务是改进所提供文本的拼写、语法、清晰、简洁和整体可读性，"

crazy_functional.py

@@ -6,7 +6,6 @@ from loguru import logger
 def get_crazy_functions():
     from crazy_functions.读文章写摘要 import 读文章写摘要
     from crazy_functions.生成函数注释 import 批量生成函数注释
-    from crazy_functions.Rag_Interface import Rag问答
     from crazy_functions.SourceCode_Analyse import 解析项目本身
     from crazy_functions.SourceCode_Analyse import 解析一个Python项目
     from crazy_functions.SourceCode_Analyse import 解析一个Matlab项目
@@ -52,13 +51,6 @@ def get_crazy_functions():
     from crazy_functions.SourceCode_Comment import 注释Python项目
 
     function_plugins = {
-        "Rag智能召回": {
-            "Group": "对话",
-            "Color": "stop",
-            "AsButton": False,
-            "Info": "将问答数据记录到向量库中，作为长期参考。",
-            "Function": HotReload(Rag问答),
-        },
         "虚空终端": {
             "Group": "对话|编程|学术|智能体",
             "Color": "stop",
@@ -707,6 +699,31 @@ def get_crazy_functions():
         logger.error(trimmed_format_exc())
         logger.error("Load function plugin failed")
 
+    try:
+        from crazy_functions.Rag_Interface import Rag问答
+
+        function_plugins.update(
+            {
+                "Rag智能召回": {
+                    "Group": "对话",
+                    "Color": "stop",
+                    "AsButton": False,
+                    "Info": "将问答数据记录到向量库中，作为长期参考。",
+                    "Function": HotReload(Rag问答),
+                },
+            }
+        )
+    except:
+        logger.error(trimmed_format_exc())
+        logger.error("Load function plugin failed")
+
+
+    
+
+
+
+
+
     # try:
     #     from crazy_functions.高级功能函数模板 import 测试图表渲染
     #     function_plugins.update({

crazy_functions/Rag_Interface.py

@@ -2,20 +2,7 @@ from toolbox import CatchException, update_ui, get_conf, get_log_folder, update_
 from crazy_functions.crazy_utils import input_clipping
 from crazy_functions.crazy_utils import request_gpt_model_in_new_thread_with_ui_alive
 
-VECTOR_STORE_TYPE = "Milvus"
-
-if VECTOR_STORE_TYPE == "Milvus":
-    try:
-        from crazy_functions.rag_fns.milvus_worker import MilvusRagWorker as LlamaIndexRagWorker
-    except:
-        VECTOR_STORE_TYPE = "Simple"
-
-if VECTOR_STORE_TYPE == "Simple":
-    from crazy_functions.rag_fns.llama_index_worker import LlamaIndexRagWorker
-
-
 RAG_WORKER_REGISTER = {}
-
 MAX_HISTORY_ROUND = 5
 MAX_CONTEXT_TOKEN_LIMIT = 4096
 REMEMBER_PREVIEW = 1000
@@ -23,6 +10,16 @@ REMEMBER_PREVIEW = 1000
 @CatchException
 def Rag问答(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_request):
 
+    # import vector store lib
+    VECTOR_STORE_TYPE = "Milvus"
+    if VECTOR_STORE_TYPE == "Milvus":
+        try:
+            from crazy_functions.rag_fns.milvus_worker import MilvusRagWorker as LlamaIndexRagWorker
+        except:
+            VECTOR_STORE_TYPE = "Simple"
+    if VECTOR_STORE_TYPE == "Simple":
+        from crazy_functions.rag_fns.llama_index_worker import LlamaIndexRagWorker
+
     # 1. we retrieve rag worker from global context
     user_name = chatbot.get_user()
     checkpoint_dir = get_log_folder(user_name, plugin_name='experimental_rag')

crazy_functions/Social_Helper.py

@@ -1,7 +1,13 @@
+import pickle, os, random
 from toolbox import CatchException, update_ui, get_conf, get_log_folder, update_ui_lastest_msg
 from crazy_functions.crazy_utils import input_clipping
 from crazy_functions.crazy_utils import request_gpt_model_in_new_thread_with_ui_alive
-import pickle, os
+from request_llms.bridge_all import predict_no_ui_long_connection
+from crazy_functions.json_fns.select_tool import structure_output, select_tool
+from pydantic import BaseModel, Field
+from loguru import logger
+from typing import List
+
 
 SOCIAL_NETWOK_WORKER_REGISTER = {}
 
@@ -9,7 +15,7 @@ class SocialNetwork():
     def __init__(self):
         self.people = []
 
-class SocialNetworkWorker():
+class SaveAndLoad():
     def __init__(self, user_name, llm_kwargs, auto_load_checkpoint=True, checkpoint_dir=None) -> None:
         self.user_name = user_name
         self.checkpoint_dir = checkpoint_dir
@@ -41,8 +47,105 @@ class SocialNetworkWorker():
             return SocialNetwork()
 
 
+class Friend(BaseModel):
+    friend_name: str = Field(description="name of a friend")
+    friend_description: str = Field(description="description of a friend (everything about this friend)")
+    friend_relationship: str = Field(description="The relationship with a friend (e.g. friend, family, colleague)")
+
+class FriendList(BaseModel):
+    friends_list: List[Friend] = Field(description="The list of friends")
+
+
+class SocialNetworkWorker(SaveAndLoad):
+    def ai_socail_advice(self, prompt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, run_gpt_fn, intention_type):
+        pass
+
+    def ai_remove_friend(self, prompt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, run_gpt_fn, intention_type):
+        pass
+
+    def ai_list_friends(self, prompt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, run_gpt_fn, intention_type):
+        pass
+
+    def ai_add_multi_friends(self, prompt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, run_gpt_fn, intention_type):
+        friend, err_msg = structure_output(
+            txt=prompt,
+            prompt="根据提示, 解析多个联系人的身份信息\n\n",
+            err_msg=f"不能理解该联系人",
+            run_gpt_fn=run_gpt_fn,
+            pydantic_cls=FriendList
+        )
+        if friend.friends_list:
+            for f in friend.friends_list: 
+                self.add_friend(f)
+            msg = f"成功添加{len(friend.friends_list)}个联系人: {str(friend.friends_list)}"
+            yield from update_ui_lastest_msg(lastmsg=msg, chatbot=chatbot, history=history, delay=0)
+
+
+    def run(self, txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_request):
+        prompt = txt
+        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=[])
+        self.tools_to_select = {
+            "SocialAdvice":{
+                "explain_to_llm": "如果用户希望获取社交指导，调用SocialAdvice生成一些社交建议",
+                "callback": self.ai_socail_advice,
+            },
+            "AddFriends":{
+                "explain_to_llm": "如果用户给出了联系人，调用AddMultiFriends把联系人添加到数据库",
+                "callback": self.ai_add_multi_friends,
+            },
+            "RemoveFriend":{
+                "explain_to_llm": "如果用户希望移除某个联系人，调用RemoveFriend",
+                "callback": self.ai_remove_friend,
+            },
+            "ListFriends":{
+                "explain_to_llm": "如果用户列举联系人，调用ListFriends",
+                "callback": self.ai_list_friends,
+            }
+        }
+
+        try:
+            Explaination = '\n'.join([f'{k}: {v["explain_to_llm"]}' for k, v in self.tools_to_select.items()])
+            class UserSociaIntention(BaseModel):
+                intention_type: str = Field(
+                    description=
+                        f"The type of user intention. You must choose from {self.tools_to_select.keys()}.\n\n" 
+                        f"Explaination:\n{Explaination}", 
+                    default="SocialAdvice"
+                )
+            pydantic_cls_instance, err_msg = select_tool(
+                prompt=txt,
+                run_gpt_fn=run_gpt_fn,
+                pydantic_cls=UserSociaIntention
+            )
+        except Exception as e:
+            yield from update_ui_lastest_msg(
+                lastmsg=f"无法理解用户意图 {err_msg}", 
+                chatbot=chatbot, 
+                history=history, 
+                delay=0
+            )
+            return
+
+        intention_type = pydantic_cls_instance.intention_type
+        intention_callback = self.tools_to_select[pydantic_cls_instance.intention_type]['callback']
+        yield from intention_callback(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, run_gpt_fn, intention_type)
+
+
+    def add_friend(self, friend):
+        # check whether the friend is already in the social network
+        for f in self.social_network.people:
+            if f.friend_name == friend.friend_name:
+                f.friend_description = friend.friend_description
+                f.friend_relationship = friend.friend_relationship
+                logger.info(f"Repeated friend, update info: {friend}")
+                return
+        logger.info(f"Add a new friend: {friend}")
+        self.social_network.people.append(friend)
+        return
+
+
 @CatchException
-def I人助手(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_request, num_day=5):
+def I人助手(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_request):
 
     # 1. we retrieve worker from global context
     user_name = chatbot.get_user()
@@ -58,8 +161,7 @@ def I人助手(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt,
         )
 
     # 2. save
-    social_network_worker.social_network.people.append("张三")
+    yield from social_network_worker.run(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_request)
     social_network_worker.save_to_checkpoint(checkpoint_dir)
-    chatbot.append(["good", "work"])
     yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
 

crazy_functions/json_fns/select_tool.py

@@ -0,0 +1,26 @@
+from crazy_functions.json_fns.pydantic_io import GptJsonIO, JsonStringError
+
+def structure_output(txt, prompt, err_msg, run_gpt_fn, pydantic_cls):
+    gpt_json_io = GptJsonIO(pydantic_cls)
+    analyze_res = run_gpt_fn(
+        txt, 
+        sys_prompt=prompt + gpt_json_io.format_instructions
+    )
+    try:
+        friend = gpt_json_io.generate_output_auto_repair(analyze_res, run_gpt_fn)
+    except JsonStringError as e:
+        return None, err_msg
+
+    err_msg = ""
+    return friend, err_msg
+
+
+def select_tool(prompt, run_gpt_fn, pydantic_cls):
+    pydantic_cls_instance, err_msg = structure_output(
+        txt=prompt,
+        prompt="根据提示, 分析应该调用哪个工具函数\n\n",
+        err_msg=f"不能理解该联系人",
+        run_gpt_fn=run_gpt_fn,
+        pydantic_cls=pydantic_cls
+    )
+    return pydantic_cls_instance, err_msg

crazy_functions/latex_fns/latex_toolbox.py

@@ -644,6 +644,213 @@ def run_in_subprocess(func):
 
 
 def _merge_pdfs(pdf1_path, pdf2_path, output_path):
+    try:
+        logger.info("Merging PDFs using _merge_pdfs_ng")
+        _merge_pdfs_ng(pdf1_path, pdf2_path, output_path)
+    except:
+        logger.info("Merging PDFs using _merge_pdfs_legacy")
+        _merge_pdfs_legacy(pdf1_path, pdf2_path, output_path)
+
+
+def _merge_pdfs_ng(pdf1_path, pdf2_path, output_path):
+    import PyPDF2  # PyPDF2这个库有严重的内存泄露问题，把它放到子进程中运行，从而方便内存的释放
+    from PyPDF2.generic import NameObject, TextStringObject, ArrayObject, FloatObject, NumberObject
+
+    Percent = 1
+    # raise RuntimeError('PyPDF2 has a serious memory leak problem, please use other tools to merge PDF files.')
+    # Open the first PDF file
+    with open(pdf1_path, "rb") as pdf1_file:
+        pdf1_reader = PyPDF2.PdfFileReader(pdf1_file)
+        # Open the second PDF file
+        with open(pdf2_path, "rb") as pdf2_file:
+            pdf2_reader = PyPDF2.PdfFileReader(pdf2_file)
+            # Create a new PDF file to store the merged pages
+            output_writer = PyPDF2.PdfFileWriter()
+            # Determine the number of pages in each PDF file
+            num_pages = max(pdf1_reader.numPages, pdf2_reader.numPages)
+            # Merge the pages from the two PDF files
+            for page_num in range(num_pages):
+                # Add the page from the first PDF file
+                if page_num < pdf1_reader.numPages:
+                    page1 = pdf1_reader.getPage(page_num)
+                else:
+                    page1 = PyPDF2.PageObject.createBlankPage(pdf1_reader)
+                # Add the page from the second PDF file
+                if page_num < pdf2_reader.numPages:
+                    page2 = pdf2_reader.getPage(page_num)
+                else:
+                    page2 = PyPDF2.PageObject.createBlankPage(pdf1_reader)
+                # Create a new empty page with double width
+                new_page = PyPDF2.PageObject.createBlankPage(
+                    width=int(
+                        int(page1.mediaBox.getWidth())
+                        + int(page2.mediaBox.getWidth()) * Percent
+                    ),
+                    height=max(page1.mediaBox.getHeight(), page2.mediaBox.getHeight()),
+                )
+                new_page.mergeTranslatedPage(page1, 0, 0)
+                new_page.mergeTranslatedPage(
+                    page2,
+                    int(
+                        int(page1.mediaBox.getWidth())
+                        - int(page2.mediaBox.getWidth()) * (1 - Percent)
+                    ),
+                    0,
+                )
+                if "/Annots" in new_page:
+                    annotations = new_page["/Annots"]
+                    for i, annot in enumerate(annotations):
+                        annot_obj = annot.get_object()
+
+                        # 检查注释类型是否是链接（/Link）
+                        if annot_obj.get("/Subtype") == "/Link":
+                            # 检查是否为内部链接跳转（/GoTo）或外部URI链接（/URI）
+                            action = annot_obj.get("/A")
+                            if action:
+
+                                if "/S" in action and action["/S"] == "/GoTo":
+                                    # 内部链接：跳转到文档中的某个页面
+                                    dest = action.get("/D")  # 目标页或目标位置
+                                    # if dest and annot.idnum in page2_annot_id:
+                                    if dest in pdf2_reader.named_destinations:
+                                        # 获取原始文件中跳转信息，包括跳转页面
+                                        destination = pdf2_reader.named_destinations[
+                                            dest
+                                        ]
+                                        page_number = (
+                                            pdf2_reader.get_destination_page_number(
+                                                destination
+                                            )
+                                        )
+                                        # 更新跳转信息，跳转到对应的页面和，指定坐标 (100, 150)，缩放比例为 100%
+                                        # “/D”:[10,'/XYZ',100,100,0]
+                                        if destination.dest_array[1] == "/XYZ":
+                                            annot_obj["/A"].update(
+                                                {
+                                                    NameObject("/D"): ArrayObject(
+                                                        [
+                                                            NumberObject(page_number),
+                                                            destination.dest_array[1],
+                                                            FloatObject(
+                                                                destination.dest_array[
+                                                                    2
+                                                                ]
+                                                                + int(
+                                                                    page1.mediaBox.getWidth()
+                                                                )
+                                                            ),
+                                                            destination.dest_array[3],
+                                                            destination.dest_array[4],
+                                                        ]
+                                                    )  # 确保键和值是 PdfObject
+                                                }
+                                            )
+                                        else:
+                                            annot_obj["/A"].update(
+                                                {
+                                                    NameObject("/D"): ArrayObject(
+                                                        [
+                                                            NumberObject(page_number),
+                                                            destination.dest_array[1],
+                                                        ]
+                                                    )  # 确保键和值是 PdfObject
+                                                }
+                                            )
+
+                                        rect = annot_obj.get("/Rect")
+                                        # 更新点击坐标
+                                        rect = ArrayObject(
+                                            [
+                                                FloatObject(
+                                                    rect[0]
+                                                    + int(page1.mediaBox.getWidth())
+                                                ),
+                                                rect[1],
+                                                FloatObject(
+                                                    rect[2]
+                                                    + int(page1.mediaBox.getWidth())
+                                                ),
+                                                rect[3],
+                                            ]
+                                        )
+                                        annot_obj.update(
+                                            {
+                                                NameObject(
+                                                    "/Rect"
+                                                ): rect  # 确保键和值是 PdfObject
+                                            }
+                                        )
+                                    # if dest and annot.idnum in page1_annot_id:
+                                    if dest in pdf1_reader.named_destinations:
+
+                                        # 获取原始文件中跳转信息，包括跳转页面
+                                        destination = pdf1_reader.named_destinations[
+                                            dest
+                                        ]
+                                        page_number = (
+                                            pdf1_reader.get_destination_page_number(
+                                                destination
+                                            )
+                                        )
+                                        # 更新跳转信息，跳转到对应的页面和，指定坐标 (100, 150)，缩放比例为 100%
+                                        # “/D”:[10,'/XYZ',100,100,0]
+                                        if destination.dest_array[1] == "/XYZ":
+                                            annot_obj["/A"].update(
+                                                {
+                                                    NameObject("/D"): ArrayObject(
+                                                        [
+                                                            NumberObject(page_number),
+                                                            destination.dest_array[1],
+                                                            FloatObject(
+                                                                destination.dest_array[
+                                                                    2
+                                                                ]
+                                                            ),
+                                                            destination.dest_array[3],
+                                                            destination.dest_array[4],
+                                                        ]
+                                                    )  # 确保键和值是 PdfObject
+                                                }
+                                            )
+                                        else:
+                                            annot_obj["/A"].update(
+                                                {
+                                                    NameObject("/D"): ArrayObject(
+                                                        [
+                                                            NumberObject(page_number),
+                                                            destination.dest_array[1],
+                                                        ]
+                                                    )  # 确保键和值是 PdfObject
+                                                }
+                                            )
+
+                                        rect = annot_obj.get("/Rect")
+                                        rect = ArrayObject(
+                                            [
+                                                FloatObject(rect[0]),
+                                                rect[1],
+                                                FloatObject(rect[2]),
+                                                rect[3],
+                                            ]
+                                        )
+                                        annot_obj.update(
+                                            {
+                                                NameObject(
+                                                    "/Rect"
+                                                ): rect  # 确保键和值是 PdfObject
+                                            }
+                                        )
+
+                                elif "/S" in action and action["/S"] == "/URI":
+                                    # 外部链接：跳转到某个URI
+                                    uri = action.get("/URI")
+                output_writer.addPage(new_page)
+            # Save the merged PDF file
+            with open(output_path, "wb") as output_file:
+                output_writer.write(output_file)
+
+
+def _merge_pdfs_legacy(pdf1_path, pdf2_path, output_path):
     import PyPDF2  # PyPDF2这个库有严重的内存泄露问题，把它放到子进程中运行，从而方便内存的释放
 
     Percent = 0.95

docs/Dockerfile+JittorLLM

@@ -1 +0,0 @@
-# 此Dockerfile不再维护，请前往docs/GithubAction+JittorLLMs

docs/GithubAction+AllCapacityBeta

@@ -1,57 +0,0 @@
-# docker build -t gpt-academic-all-capacity -f docs/GithubAction+AllCapacity  --network=host --build-arg http_proxy=http://localhost:10881 --build-arg https_proxy=http://localhost:10881 .
-# docker build -t gpt-academic-all-capacity -f docs/GithubAction+AllCapacityBeta  --network=host .
-# docker run -it --net=host gpt-academic-all-capacity  bash
-
-# 从NVIDIA源，从而支持显卡（检查宿主的nvidia-smi中的cuda版本必须>=11.3）
-FROM fuqingxu/11.3.1-runtime-ubuntu20.04-with-texlive:latest
-
-# edge-tts需要的依赖，某些pip包所需的依赖
-RUN apt update && apt install ffmpeg build-essential -y
-
-# use python3 as the system default python
-WORKDIR /gpt
-RUN curl -sS https://bootstrap.pypa.io/get-pip.py | python3.8
-
-# # 非必要步骤，更换pip源 （以下三行，可以删除）
-# RUN echo '[global]' > /etc/pip.conf && \
-#     echo 'index-url = https://mirrors.aliyun.com/pypi/simple/' >> /etc/pip.conf && \
-#     echo 'trusted-host = mirrors.aliyun.com' >> /etc/pip.conf
-
-# 下载pytorch
-RUN python3 -m pip install torch torchvision --extra-index-url https://download.pytorch.org/whl/cu113
-# 准备pip依赖
-RUN python3 -m pip install openai numpy arxiv rich
-RUN python3 -m pip install colorama Markdown pygments pymupdf
-RUN python3 -m pip install python-docx moviepy pdfminer
-RUN python3 -m pip install zh_langchain==0.2.1 pypinyin
-RUN python3 -m pip install rarfile py7zr
-RUN python3 -m pip install aliyun-python-sdk-core==2.13.3 pyOpenSSL webrtcvad scipy git+https://github.com/aliyun/alibabacloud-nls-python-sdk.git
-# 下载分支
-WORKDIR /gpt
-RUN git clone --depth=1 https://github.com/binary-husky/gpt_academic.git
-WORKDIR /gpt/gpt_academic
-RUN git clone --depth=1 https://github.com/OpenLMLab/MOSS.git request_llms/moss
-
-RUN python3 -m pip install -r requirements.txt
-RUN python3 -m pip install -r request_llms/requirements_moss.txt
-RUN python3 -m pip install -r request_llms/requirements_qwen.txt
-RUN python3 -m pip install -r request_llms/requirements_chatglm.txt
-RUN python3 -m pip install -r request_llms/requirements_newbing.txt
-RUN python3 -m pip install nougat-ocr
-
-
-# 预热Tiktoken模块
-RUN python3  -c 'from check_proxy import warm_up_modules; warm_up_modules()'
-
-# 安装知识库插件的额外依赖
-RUN apt-get update && apt-get install libgl1 -y
-RUN pip3 install transformers protobuf langchain sentence-transformers  faiss-cpu nltk beautifulsoup4 bitsandbytes tabulate icetk --upgrade
-RUN pip3 install unstructured[all-docs] --upgrade
-RUN python3  -c 'from check_proxy import warm_up_vectordb; warm_up_vectordb()'
-RUN rm -rf /usr/local/lib/python3.8/dist-packages/tests
-
-
-# COPY .cache /root/.cache
-# COPY config_private.py config_private.py
-# 启动
-CMD ["python3", "-u", "main.py"]

docs/GithubAction+NoLocal+Latex

@@ -3,33 +3,19 @@
 # - 2 构建 docker build -t gpt-academic-nolocal-latex -f docs/GithubAction+NoLocal+Latex .
 # - 3 运行 docker run -v /home/fuqingxu/arxiv_cache:/root/arxiv_cache --rm -it --net=host gpt-academic-nolocal-latex
 
-FROM fuqingxu/python311_texlive_ctex:latest
-ENV PATH "$PATH:/usr/local/texlive/2022/bin/x86_64-linux"
-ENV PATH "$PATH:/usr/local/texlive/2023/bin/x86_64-linux"
-ENV PATH "$PATH:/usr/local/texlive/2024/bin/x86_64-linux"
-ENV PATH "$PATH:/usr/local/texlive/2025/bin/x86_64-linux"
-ENV PATH "$PATH:/usr/local/texlive/2026/bin/x86_64-linux"
-
-# 指定路径
+FROM menghuan1918/ubuntu_uv_ctex:latest
+ENV DEBIAN_FRONTEND=noninteractive
+SHELL ["/bin/bash", "-c"]
 WORKDIR /gpt
-
-RUN pip3 install openai numpy arxiv rich
-RUN pip3 install colorama Markdown pygments pymupdf
-RUN pip3 install python-docx pdfminer
-RUN pip3 install nougat-ocr
-
-# 装载项目文件
 COPY . .
-
-
-# 安装依赖
-RUN pip3 install -r requirements.txt
-
-# edge-tts需要的依赖
-RUN apt update && apt install ffmpeg -y
+RUN /root/.cargo/bin/uv venv --seed \
+    && source .venv/bin/activate \
+    && /root/.cargo/bin/uv pip install openai numpy arxiv rich colorama Markdown pygments pymupdf python-docx pdfminer \
+    && /root/.cargo/bin/uv pip install -r requirements.txt \
+    && /root/.cargo/bin/uv clean
 
 # 可选步骤，用于预热模块
-RUN python3  -c 'from check_proxy import warm_up_modules; warm_up_modules()'
+RUN .venv/bin/python3 -c 'from check_proxy import warm_up_modules; warm_up_modules()'
 
 # 启动
-CMD ["python3", "-u", "main.py"]
+CMD [".venv/bin/python3", "-u", "main.py"]

docs/WithFastapi.md

@@ -4,7 +4,7 @@ We currently support fastapi in order to solve sub-path deploy issue.
 
 1. change CUSTOM_PATH setting in `config.py`
 
-``` sh
+```sh
 nano config.py
 ```
 
@@ -35,9 +35,8 @@ if __name__ == "__main__":
     main()
 ```
 
-
 3. Go!
 
-``` sh
+```sh
 python main.py
 ```

docs/translate_std.json

@@ -108,5 +108,22 @@
     "解析PDF_简单拆解": "ParsePDF_simpleDecomposition",
     "解析PDF_DOC2X_单文件": "ParsePDF_DOC2X_singleFile",
     "注释Python项目": "CommentPythonProject",
-    "注释源代码": "CommentSourceCode"
+    "注释源代码": "CommentSourceCode",
+    "log亮黄": "log_yellow",
+    "log亮绿": "log_green",
+    "log亮红": "log_red",
+    "log亮紫": "log_purple",
+    "log亮蓝": "log_blue",
+    "Rag问答": "RagQA",
+    "sprint红": "sprint_red",
+    "sprint绿": "sprint_green",
+    "sprint黄": "sprint_yellow",
+    "sprint蓝": "sprint_blue",
+    "sprint紫": "sprint_purple",
+    "sprint靛": "sprint_indigo",
+    "sprint亮红": "sprint_bright_red",
+    "sprint亮绿": "sprint_bright_green",
+    "sprint亮黄": "sprint_bright_yellow",
+    "sprint亮蓝": "sprint_bright_blue",
+    "sprint亮紫": "sprint_bright_purple"
 }

request_llms/bridge_all.py

@@ -256,6 +256,8 @@ model_info = {
         "max_token": 128000,
         "tokenizer": tokenizer_gpt4,
         "token_cnt": get_token_num_gpt4,
+        "openai_disable_system_prompt": True,
+        "openai_disable_stream": True,
     },
     "o1-mini": {
         "fn_with_ui": chatgpt_ui,
@@ -264,6 +266,8 @@ model_info = {
         "max_token": 128000,
         "tokenizer": tokenizer_gpt4,
         "token_cnt": get_token_num_gpt4,
+        "openai_disable_system_prompt": True,
+        "openai_disable_stream": True,
     },
 
     "gpt-4-turbo": {
@@ -1116,6 +1120,24 @@ if len(AZURE_CFG_ARRAY) > 0:
         if azure_model_name not in AVAIL_LLM_MODELS:
             AVAIL_LLM_MODELS += [azure_model_name]
 
+# -=-=-=-=-=-=- Openrouter模型对齐支持 -=-=-=-=-=-=-
+# 为了更灵活地接入Openrouter路由，设计了此接口
+for model in [m for m in AVAIL_LLM_MODELS if m.startswith("openrouter-")]:
+    from request_llms.bridge_openrouter import predict_no_ui_long_connection as openrouter_noui
+    from request_llms.bridge_openrouter import predict as openrouter_ui
+    model_info.update({
+        model: {
+            "fn_with_ui": openrouter_ui,
+            "fn_without_ui": openrouter_noui,
+            # 以下参数参考gpt-4o-mini的配置, 请根据实际情况修改
+            "endpoint": openai_endpoint,
+            "has_multimodal_capacity": True,
+            "max_token": 128000,
+            "tokenizer": tokenizer_gpt4,
+            "token_cnt": get_token_num_gpt4,
+        },
+    })
+
 
 # -=-=-=-=-=-=--=-=-=-=-=-=--=-=-=-=-=-=--=-=-=-=-=-=-=-=
 # -=-=-=-=-=-=-=-=-=- ☝️ 以上是模型路由 -=-=-=-=-=-=-=-=-=
@@ -1261,5 +1283,6 @@ def predict(inputs:str, llm_kwargs:dict, plugin_kwargs:dict, chatbot,
     if additional_fn: # 根据基础功能区 ModelOverride 参数调整模型类型
         llm_kwargs, additional_fn, method = execute_model_override(llm_kwargs, additional_fn, method)
 
+    # 更新一下llm_kwargs的参数，否则会出现参数不匹配的问题
     yield from method(inputs, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, stream, additional_fn)
 

request_llms/bridge_chatgpt.py

@@ -134,22 +134,33 @@ def predict_no_ui_long_connection(inputs:str, llm_kwargs:dict, history:list=[],
     observe_window = None：
         用于负责跨越线程传递已经输出的部分，大部分时候仅仅为了fancy的视觉效果，留空即可。observe_window[0]：观测窗。observe_window[1]：看门狗
     """
+    from request_llms.bridge_all import model_info
+
     watch_dog_patience = 5 # 看门狗的耐心, 设置5秒即可
-    headers, payload = generate_payload(inputs, llm_kwargs, history, system_prompt=sys_prompt, stream=True)
+
+    if model_info[llm_kwargs['llm_model']].get('openai_disable_stream', False): stream = False
+    else: stream = True
+
+    headers, payload = generate_payload(inputs, llm_kwargs, history, system_prompt=sys_prompt, stream=stream)
     retry = 0
     while True:
         try:
             # make a POST request to the API endpoint, stream=False
-            from .bridge_all import model_info
             endpoint = verify_endpoint(model_info[llm_kwargs['llm_model']]['endpoint'])
             response = requests.post(endpoint, headers=headers, proxies=proxies,
-                                    json=payload, stream=True, timeout=TIMEOUT_SECONDS); break
+                                    json=payload, stream=stream, timeout=TIMEOUT_SECONDS); break
         except requests.exceptions.ReadTimeout as e:
             retry += 1
             traceback.print_exc()
             if retry > MAX_RETRY: raise TimeoutError
             if MAX_RETRY!=0: logger.error(f'请求超时，正在重试 ({retry}/{MAX_RETRY}) ……')
 
+    if not stream:
+        # 该分支仅适用于不支持stream的o1模型，其他情形一律不适用
+        chunkjson = json.loads(response.content.decode())
+        gpt_replying_buffer = chunkjson['choices'][0]["message"]["content"]
+        return gpt_replying_buffer
+
     stream_response = response.iter_lines()
     result = ''
     json_data = None
@@ -181,7 +192,7 @@ def predict_no_ui_long_connection(inputs:str, llm_kwargs:dict, history:list=[],
         if (not has_content) and (not has_role): continue # raise RuntimeError("发现不标准的第三方接口："+delta)
         if has_content: # has_role = True/False
             result += delta["content"]
-            if not console_slience: logger.info(delta["content"], end='')
+            if not console_slience: print(delta["content"], end='')
             if observe_window is not None:
                 # 观测窗，把已经获取的数据显示出去
                 if len(observe_window) >= 1:
@@ -209,7 +220,7 @@ def predict(inputs:str, llm_kwargs:dict, plugin_kwargs:dict, chatbot:ChatBotWith
     chatbot 为WebUI中显示的对话列表，修改它，然后yeild出去，可以直接修改对话界面内容
     additional_fn代表点击的哪个按钮，按钮见functional.py
     """
-    from .bridge_all import model_info
+    from request_llms.bridge_all import model_info
     if is_any_api_key(inputs):
         chatbot._cookies['api_key'] = inputs
         chatbot.append(("输入已识别为openai的api_key", what_keys(inputs)))
@@ -238,6 +249,10 @@ def predict(inputs:str, llm_kwargs:dict, plugin_kwargs:dict, chatbot:ChatBotWith
     chatbot.append((_inputs, ""))
     yield from update_ui(chatbot=chatbot, history=history, msg="等待响应") # 刷新界面
 
+    # 禁用stream的特殊模型处理
+    if model_info[llm_kwargs['llm_model']].get('openai_disable_stream', False): stream = False
+    else: stream = True
+
     # check mis-behavior
     if is_the_upload_folder(user_input):
         chatbot[-1] = (inputs, f"[Local Message] 检测到操作错误！当您上传文档之后，需点击“**函数插件区**”按钮进行处理，请勿点击“提交”按钮或者“基础功能区”按钮。")
@@ -271,7 +286,7 @@ def predict(inputs:str, llm_kwargs:dict, plugin_kwargs:dict, chatbot:ChatBotWith
         try:
             # make a POST request to the API endpoint, stream=True
             response = requests.post(endpoint, headers=headers, proxies=proxies,
-                                    json=payload, stream=True, timeout=TIMEOUT_SECONDS);break
+                                    json=payload, stream=stream, timeout=TIMEOUT_SECONDS);break
         except:
             retry += 1
             chatbot[-1] = ((chatbot[-1][0], timeout_bot_msg))
@@ -279,10 +294,15 @@ def predict(inputs:str, llm_kwargs:dict, plugin_kwargs:dict, chatbot:ChatBotWith
             yield from update_ui(chatbot=chatbot, history=history, msg="请求超时"+retry_msg) # 刷新界面
             if retry > MAX_RETRY: raise TimeoutError
 
-    gpt_replying_buffer = ""
 
-    is_head_of_the_stream = True
+    if not stream:
+        # 该分支仅适用于不支持stream的o1模型，其他情形一律不适用
+        yield from handle_o1_model_special(response, inputs, llm_kwargs, chatbot, history)
+        return
+
     if stream:
+        gpt_replying_buffer = ""
+        is_head_of_the_stream = True
         stream_response =  response.iter_lines()
         while True:
             try:
@@ -343,12 +363,24 @@ def predict(inputs:str, llm_kwargs:dict, plugin_kwargs:dict, chatbot:ChatBotWith
                     chunk_decoded = chunk.decode()
                     error_msg = chunk_decoded
                     chatbot, history = handle_error(inputs, llm_kwargs, chatbot, history, chunk_decoded, error_msg)
-                    yield from update_ui(chatbot=chatbot, history=history, msg="Json异常" + error_msg) # 刷新界面
+                    yield from update_ui(chatbot=chatbot, history=history, msg="Json解析异常" + error_msg) # 刷新界面
                     logger.error(error_msg)
                     return
+        return  # return from stream-branch
+
+def handle_o1_model_special(response, inputs, llm_kwargs, chatbot, history):
+    try:
+        chunkjson = json.loads(response.content.decode())
+        gpt_replying_buffer = chunkjson['choices'][0]["message"]["content"]
+        log_chat(llm_model=llm_kwargs["llm_model"], input_str=inputs, output_str=gpt_replying_buffer)
+        history[-1] = gpt_replying_buffer
+        chatbot[-1] = (history[-2], history[-1])
+        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+    except Exception as e:
+        yield from update_ui(chatbot=chatbot, history=history, msg="Json解析异常" + response.text) # 刷新界面
 
 def handle_error(inputs, llm_kwargs, chatbot, history, chunk_decoded, error_msg):
-    from .bridge_all import model_info
+    from request_llms.bridge_all import model_info
     openai_website = ' 请登录OpenAI查看详情 https://platform.openai.com/signup'
     if "reduce the length" in error_msg:
         if len(history) >= 2: history[-1] = ""; history[-2] = "" # 清除当前溢出的输入：history[-2] 是本次输入, history[-1] 是本次输出
@@ -381,6 +413,8 @@ def generate_payload(inputs:str, llm_kwargs:dict, history:list, system_prompt:st
     """
     整合所有信息，选择LLM模型，生成http请求，为发送请求做准备
     """
+    from request_llms.bridge_all import model_info
+
     if not is_any_api_key(llm_kwargs['api_key']):
         raise AssertionError("你提供了错误的API_KEY。\n\n1. 临时解决方案：直接在输入区键入api_key，然后回车提交。\n\n2. 长效解决方案：在config.py中配置。")
 
@@ -409,10 +443,16 @@ def generate_payload(inputs:str, llm_kwargs:dict, history:list, system_prompt:st
     else:
         enable_multimodal_capacity = False
 
+    conversation_cnt = len(history) // 2
+    openai_disable_system_prompt = model_info[llm_kwargs['llm_model']].get('openai_disable_system_prompt', False)
+
+    if openai_disable_system_prompt:
+        messages = [{"role": "user", "content": system_prompt}]
+    else:
+        messages = [{"role": "system", "content": system_prompt}]
+
     if not enable_multimodal_capacity:
         # 不使用多模态能力
-        conversation_cnt = len(history) // 2
-        messages = [{"role": "system", "content": system_prompt}]
         if conversation_cnt:
             for index in range(0, 2*conversation_cnt, 2):
                 what_i_have_asked = {}
@@ -434,8 +474,6 @@ def generate_payload(inputs:str, llm_kwargs:dict, history:list, system_prompt:st
         messages.append(what_i_ask_now)
     else:
         # 多模态能力
-        conversation_cnt = len(history) // 2
-        messages = [{"role": "system", "content": system_prompt}]
         if conversation_cnt:
             for index in range(0, 2*conversation_cnt, 2):
                 what_i_have_asked = {}

request_llms/bridge_cohere.py

@@ -111,7 +111,7 @@ def predict_no_ui_long_connection(inputs:str, llm_kwargs:dict, history:list=[],
         if chunkjson['event_type'] == 'stream-start': continue
         if chunkjson['event_type'] == 'text-generation':
             result += chunkjson["text"]
-            if not console_slience: logger.info(chunkjson["text"], end='')
+            if not console_slience: print(chunkjson["text"], end='')
             if observe_window is not None:
                 # 观测窗，把已经获取的数据显示出去
                 if len(observe_window) >= 1:

request_llms/bridge_ollama.py

@@ -99,7 +99,7 @@ def predict_no_ui_long_connection(inputs, llm_kwargs, history=[], sys_prompt="",
                     logger.info(f'[response] {result}')
                     break
                 result += chunkjson['message']["content"]
-                if not console_slience: logger.info(chunkjson['message']["content"], end='')
+                if not console_slience: print(chunkjson['message']["content"], end='')
                 if observe_window is not None:
                     # 观测窗，把已经获取的数据显示出去
                     if len(observe_window) >= 1:

request_llms/bridge_openrouter.py

@@ -0,0 +1,541 @@
+"""
+    该文件中主要包含三个函数
+
+    不具备多线程能力的函数：
+    1. predict: 正常对话时使用，具备完备的交互功能，不可多线程
+
+    具备多线程调用能力的函数
+    2. predict_no_ui_long_connection：支持多线程
+"""
+
+import json
+import os
+import re
+import time
+import traceback
+import requests
+import random
+from loguru import logger
+
+# config_private.py放自己的秘密如API和代理网址
+# 读取时首先看是否存在私密的config_private配置文件（不受git管控），如果有，则覆盖原config文件
+from toolbox import get_conf, update_ui, is_any_api_key, select_api_key, what_keys, clip_history
+from toolbox import trimmed_format_exc, is_the_upload_folder, read_one_api_model_name, log_chat
+from toolbox import ChatBotWithCookies, have_any_recent_upload_image_files, encode_image
+proxies, TIMEOUT_SECONDS, MAX_RETRY, API_ORG, AZURE_CFG_ARRAY = \
+    get_conf('proxies', 'TIMEOUT_SECONDS', 'MAX_RETRY', 'API_ORG', 'AZURE_CFG_ARRAY')
+
+timeout_bot_msg = '[Local Message] Request timeout. Network error. Please check proxy settings in config.py.' + \
+                  '网络错误，检查代理服务器是否可用，以及代理设置的格式是否正确，格式须是[协议]://[地址]:[端口]，缺一不可。'
+
+def get_full_error(chunk, stream_response):
+    """
+        获取完整的从Openai返回的报错
+    """
+    while True:
+        try:
+            chunk += next(stream_response)
+        except:
+            break
+    return chunk
+
+def make_multimodal_input(inputs, image_paths):
+    image_base64_array = []
+    for image_path in image_paths:
+        path = os.path.abspath(image_path)
+        base64 = encode_image(path)
+        inputs = inputs + f'<br/><br/><div align="center"><img src="file={path}" base64="{base64}"></div>'
+        image_base64_array.append(base64)
+    return inputs, image_base64_array
+
+def reverse_base64_from_input(inputs):
+    # 定义一个正则表达式来匹配 Base64 字符串（假设格式为 base64="<Base64编码>"）
+    # pattern = re.compile(r'base64="([^"]+)"></div>')
+    pattern = re.compile(r'<br/><br/><div align="center"><img[^<>]+base64="([^"]+)"></div>')
+    # 使用 findall 方法查找所有匹配的 Base64 字符串
+    base64_strings = pattern.findall(inputs)
+    # 返回反转后的 Base64 字符串列表
+    return base64_strings
+
+def contain_base64(inputs):
+    base64_strings = reverse_base64_from_input(inputs)
+    return len(base64_strings) > 0
+
+def append_image_if_contain_base64(inputs):
+    if not contain_base64(inputs):
+        return inputs
+    else:
+        image_base64_array = reverse_base64_from_input(inputs)
+        pattern = re.compile(r'<br/><br/><div align="center"><img[^><]+></div>')
+        inputs = re.sub(pattern, '', inputs)
+        res = []
+        res.append({
+            "type": "text",
+            "text": inputs
+        })
+        for image_base64 in image_base64_array:
+            res.append({
+                "type": "image_url",
+                "image_url": {
+                    "url": f"data:image/jpeg;base64,{image_base64}"
+                }
+            })
+        return res
+
+def remove_image_if_contain_base64(inputs):
+    if not contain_base64(inputs):
+        return inputs
+    else:
+        pattern = re.compile(r'<br/><br/><div align="center"><img[^><]+></div>')
+        inputs = re.sub(pattern, '', inputs)
+        return inputs
+
+def decode_chunk(chunk):
+    # 提前读取一些信息 （用于判断异常）
+    chunk_decoded = chunk.decode()
+    chunkjson = None
+    has_choices = False
+    choice_valid = False
+    has_content = False
+    has_role = False
+    try:
+        chunkjson = json.loads(chunk_decoded[6:])
+        has_choices = 'choices' in chunkjson
+        if has_choices: choice_valid = (len(chunkjson['choices']) > 0)
+        if has_choices and choice_valid: has_content = ("content" in chunkjson['choices'][0]["delta"])
+        if has_content: has_content = (chunkjson['choices'][0]["delta"]["content"] is not None)
+        if has_choices and choice_valid: has_role = "role" in chunkjson['choices'][0]["delta"]
+    except:
+        pass
+    return chunk_decoded, chunkjson, has_choices, choice_valid, has_content, has_role
+
+from functools import lru_cache
+@lru_cache(maxsize=32)
+def verify_endpoint(endpoint):
+    """
+        检查endpoint是否可用
+    """
+    if "你亲手写的api名称" in endpoint:
+        raise ValueError("Endpoint不正确, 请检查AZURE_ENDPOINT的配置! 当前的Endpoint为:" + endpoint)
+    return endpoint
+
+def predict_no_ui_long_connection(inputs:str, llm_kwargs:dict, history:list=[], sys_prompt:str="", observe_window:list=None, console_slience:bool=False):
+    """
+    发送至chatGPT，等待回复，一次性完成，不显示中间过程。但内部用stream的方法避免中途网线被掐。
+    inputs：
+        是本次问询的输入
+    sys_prompt:
+        系统静默prompt
+    llm_kwargs：
+        chatGPT的内部调优参数
+    history：
+        是之前的对话列表
+    observe_window = None：
+        用于负责跨越线程传递已经输出的部分，大部分时候仅仅为了fancy的视觉效果，留空即可。observe_window[0]：观测窗。observe_window[1]：看门狗
+    """
+    from request_llms.bridge_all import model_info
+
+    watch_dog_patience = 5 # 看门狗的耐心, 设置5秒即可
+
+    if model_info[llm_kwargs['llm_model']].get('openai_disable_stream', False): stream = False
+    else: stream = True
+
+    headers, payload = generate_payload(inputs, llm_kwargs, history, system_prompt=sys_prompt, stream=stream)
+    retry = 0
+    while True:
+        try:
+            # make a POST request to the API endpoint, stream=False
+            endpoint = verify_endpoint(model_info[llm_kwargs['llm_model']]['endpoint'])
+            response = requests.post(endpoint, headers=headers, proxies=proxies,
+                                    json=payload, stream=stream, timeout=TIMEOUT_SECONDS); break
+        except requests.exceptions.ReadTimeout as e:
+            retry += 1
+            traceback.print_exc()
+            if retry > MAX_RETRY: raise TimeoutError
+            if MAX_RETRY!=0: logger.error(f'请求超时，正在重试 ({retry}/{MAX_RETRY}) ……')
+
+    if not stream:
+        # 该分支仅适用于不支持stream的o1模型，其他情形一律不适用
+        chunkjson = json.loads(response.content.decode())
+        gpt_replying_buffer = chunkjson['choices'][0]["message"]["content"]
+        return gpt_replying_buffer
+
+    stream_response = response.iter_lines()
+    result = ''
+    json_data = None
+    while True:
+        try: chunk = next(stream_response)
+        except StopIteration:
+            break
+        except requests.exceptions.ConnectionError:
+            chunk = next(stream_response) # 失败了，重试一次？再失败就没办法了。
+        chunk_decoded, chunkjson, has_choices, choice_valid, has_content, has_role = decode_chunk(chunk)
+        if len(chunk_decoded)==0: continue
+        if not chunk_decoded.startswith('data:'):
+            error_msg = get_full_error(chunk, stream_response).decode()
+            if "reduce the length" in error_msg:
+                raise ConnectionAbortedError("OpenAI拒绝了请求:" + error_msg)
+            elif """type":"upstream_error","param":"307""" in error_msg:
+                raise ConnectionAbortedError("正常结束，但显示Token不足，导致输出不完整，请削减单次输入的文本量。")
+            else:
+                raise RuntimeError("OpenAI拒绝了请求：" + error_msg)
+        if ('data: [DONE]' in chunk_decoded): break # api2d 正常完成
+        # 提前读取一些信息 （用于判断异常）
+        if (has_choices and not choice_valid) or ('OPENROUTER PROCESSING' in chunk_decoded):
+            # 一些垃圾第三方接口的出现这样的错误，openrouter的特殊处理
+            continue
+        json_data = chunkjson['choices'][0]
+        delta = json_data["delta"]
+        if len(delta) == 0: break
+        if (not has_content) and has_role: continue
+        if (not has_content) and (not has_role): continue # raise RuntimeError("发现不标准的第三方接口："+delta)
+        if has_content: # has_role = True/False
+            result += delta["content"]
+            if not console_slience: print(delta["content"], end='')
+            if observe_window is not None:
+                # 观测窗，把已经获取的数据显示出去
+                if len(observe_window) >= 1:
+                    observe_window[0] += delta["content"]
+                # 看门狗，如果超过期限没有喂狗，则终止
+                if len(observe_window) >= 2:
+                    if (time.time()-observe_window[1]) > watch_dog_patience:
+                        raise RuntimeError("用户取消了程序。")
+        else: raise RuntimeError("意外Json结构："+delta)
+    if json_data and json_data['finish_reason'] == 'content_filter':
+        raise RuntimeError("由于提问含不合规内容被Azure过滤。")
+    if json_data and json_data['finish_reason'] == 'length':
+        raise ConnectionAbortedError("正常结束，但显示Token不足，导致输出不完整，请削减单次输入的文本量。")
+    return result
+
+
+def predict(inputs:str, llm_kwargs:dict, plugin_kwargs:dict, chatbot:ChatBotWithCookies,
+            history:list=[], system_prompt:str='', stream:bool=True, additional_fn:str=None):
+    """
+    发送至chatGPT，流式获取输出。
+    用于基础的对话功能。
+    inputs 是本次问询的输入
+    top_p, temperature是chatGPT的内部调优参数
+    history 是之前的对话列表（注意无论是inputs还是history，内容太长了都会触发token数量溢出的错误）
+    chatbot 为WebUI中显示的对话列表，修改它，然后yeild出去，可以直接修改对话界面内容
+    additional_fn代表点击的哪个按钮，按钮见functional.py
+    """
+    from request_llms.bridge_all import model_info
+    if is_any_api_key(inputs):
+        chatbot._cookies['api_key'] = inputs
+        chatbot.append(("输入已识别为openai的api_key", what_keys(inputs)))
+        yield from update_ui(chatbot=chatbot, history=history, msg="api_key已导入") # 刷新界面
+        return
+    elif not is_any_api_key(chatbot._cookies['api_key']):
+        chatbot.append((inputs, "缺少api_key。\n\n1. 临时解决方案：直接在输入区键入api_key，然后回车提交。\n\n2. 长效解决方案：在config.py中配置。"))
+        yield from update_ui(chatbot=chatbot, history=history, msg="缺少api_key") # 刷新界面
+        return
+
+    user_input = inputs
+    if additional_fn is not None:
+        from core_functional import handle_core_functionality
+        inputs, history = handle_core_functionality(additional_fn, inputs, history, chatbot)
+
+    # 多模态模型
+    has_multimodal_capacity = model_info[llm_kwargs['llm_model']].get('has_multimodal_capacity', False)
+    if has_multimodal_capacity:
+        has_recent_image_upload, image_paths = have_any_recent_upload_image_files(chatbot, pop=True)
+    else:
+        has_recent_image_upload, image_paths = False, []
+    if has_recent_image_upload:
+        _inputs, image_base64_array = make_multimodal_input(inputs, image_paths)
+    else:
+        _inputs, image_base64_array = inputs, []
+    chatbot.append((_inputs, ""))
+    yield from update_ui(chatbot=chatbot, history=history, msg="等待响应") # 刷新界面
+
+    # 禁用stream的特殊模型处理
+    if model_info[llm_kwargs['llm_model']].get('openai_disable_stream', False): stream = False
+    else: stream = True
+
+    # check mis-behavior
+    if is_the_upload_folder(user_input):
+        chatbot[-1] = (inputs, f"[Local Message] 检测到操作错误！当您上传文档之后，需点击“**函数插件区**”按钮进行处理，请勿点击“提交”按钮或者“基础功能区”按钮。")
+        yield from update_ui(chatbot=chatbot, history=history, msg="正常") # 刷新界面
+        time.sleep(2)
+
+    try:
+        headers, payload = generate_payload(inputs, llm_kwargs, history, system_prompt, image_base64_array, has_multimodal_capacity, stream)
+    except RuntimeError as e:
+        chatbot[-1] = (inputs, f"您提供的api-key不满足要求，不包含任何可用于{llm_kwargs['llm_model']}的api-key。您可能选择了错误的模型或请求源。")
+        yield from update_ui(chatbot=chatbot, history=history, msg="api-key不满足要求") # 刷新界面
+        return
+
+    # 检查endpoint是否合法
+    try:
+        endpoint = verify_endpoint(model_info[llm_kwargs['llm_model']]['endpoint'])
+    except:
+        tb_str = '```\n' + trimmed_format_exc() + '```'
+        chatbot[-1] = (inputs, tb_str)
+        yield from update_ui(chatbot=chatbot, history=history, msg="Endpoint不满足要求") # 刷新界面
+        return
+
+    # 加入历史
+    if has_recent_image_upload:
+        history.extend([_inputs, ""])
+    else:
+        history.extend([inputs, ""])
+
+    retry = 0
+    while True:
+        try:
+            # make a POST request to the API endpoint, stream=True
+            response = requests.post(endpoint, headers=headers, proxies=proxies,
+                                    json=payload, stream=stream, timeout=TIMEOUT_SECONDS);break
+        except:
+            retry += 1
+            chatbot[-1] = ((chatbot[-1][0], timeout_bot_msg))
+            retry_msg = f"，正在重试 ({retry}/{MAX_RETRY}) ……" if MAX_RETRY > 0 else ""
+            yield from update_ui(chatbot=chatbot, history=history, msg="请求超时"+retry_msg) # 刷新界面
+            if retry > MAX_RETRY: raise TimeoutError
+
+
+    if not stream:
+        # 该分支仅适用于不支持stream的o1模型，其他情形一律不适用
+        yield from handle_o1_model_special(response, inputs, llm_kwargs, chatbot, history)
+        return
+
+    if stream:
+        gpt_replying_buffer = ""
+        is_head_of_the_stream = True
+        stream_response =  response.iter_lines()
+        while True:
+            try:
+                chunk = next(stream_response)
+            except StopIteration:
+                # 非OpenAI官方接口的出现这样的报错，OpenAI和API2D不会走这里
+                chunk_decoded = chunk.decode()
+                error_msg = chunk_decoded
+                # 首先排除一个one-api没有done数据包的第三方Bug情形
+                if len(gpt_replying_buffer.strip()) > 0 and len(error_msg) == 0:
+                    yield from update_ui(chatbot=chatbot, history=history, msg="检测到有缺陷的非OpenAI官方接口，建议选择更稳定的接口。")
+                    break
+                # 其他情况，直接返回报错
+                chatbot, history = handle_error(inputs, llm_kwargs, chatbot, history, chunk_decoded, error_msg)
+                yield from update_ui(chatbot=chatbot, history=history, msg="非OpenAI官方接口返回了错误:" + chunk.decode()) # 刷新界面
+                return
+
+            # 提前读取一些信息 （用于判断异常）
+            chunk_decoded, chunkjson, has_choices, choice_valid, has_content, has_role = decode_chunk(chunk)
+
+            if is_head_of_the_stream and (r'"object":"error"' not in chunk_decoded) and (r"content" not in chunk_decoded):
+                # 数据流的第一帧不携带content
+                is_head_of_the_stream = False; continue
+
+            if chunk:
+                try:
+                    if (has_choices and not choice_valid) or ('OPENROUTER PROCESSING' in chunk_decoded):
+                        # 一些垃圾第三方接口的出现这样的错误, 或者OPENROUTER的特殊处理,因为OPENROUTER的数据流未连接到模型时会出现OPENROUTER PROCESSING
+                        continue
+                    if ('data: [DONE]' not in chunk_decoded) and len(chunk_decoded) > 0 and (chunkjson is None):
+                        # 传递进来一些奇怪的东西
+                        raise ValueError(f'无法读取以下数据，请检查配置。\n\n{chunk_decoded}')
+                    # 前者是API2D的结束条件，后者是OPENAI的结束条件
+                    if ('data: [DONE]' in chunk_decoded) or (len(chunkjson['choices'][0]["delta"]) == 0):
+                        # 判定为数据流的结束，gpt_replying_buffer也写完了
+                        log_chat(llm_model=llm_kwargs["llm_model"], input_str=inputs, output_str=gpt_replying_buffer)
+                        break
+                    # 处理数据流的主体
+                    status_text = f"finish_reason: {chunkjson['choices'][0].get('finish_reason', 'null')}"
+                    # 如果这里抛出异常，一般是文本过长，详情见get_full_error的输出
+                    if has_content:
+                        # 正常情况
+                        gpt_replying_buffer = gpt_replying_buffer + chunkjson['choices'][0]["delta"]["content"]
+                    elif has_role:
+                        # 一些第三方接口的出现这样的错误，兼容一下吧
+                        continue
+                    else:
+                        # 至此已经超出了正常接口应该进入的范围，一些垃圾第三方接口会出现这样的错误
+                        if chunkjson['choices'][0]["delta"]["content"] is None: continue # 一些垃圾第三方接口出现这样的错误，兼容一下吧
+                        gpt_replying_buffer = gpt_replying_buffer + chunkjson['choices'][0]["delta"]["content"]
+
+                    history[-1] = gpt_replying_buffer
+                    chatbot[-1] = (history[-2], history[-1])
+                    yield from update_ui(chatbot=chatbot, history=history, msg=status_text) # 刷新界面
+                except Exception as e:
+                    yield from update_ui(chatbot=chatbot, history=history, msg="Json解析不合常规") # 刷新界面
+                    chunk = get_full_error(chunk, stream_response)
+                    chunk_decoded = chunk.decode()
+                    error_msg = chunk_decoded
+                    chatbot, history = handle_error(inputs, llm_kwargs, chatbot, history, chunk_decoded, error_msg)
+                    yield from update_ui(chatbot=chatbot, history=history, msg="Json解析异常" + error_msg) # 刷新界面
+                    logger.error(error_msg)
+                    return
+        return  # return from stream-branch
+
+def handle_o1_model_special(response, inputs, llm_kwargs, chatbot, history):
+    try:
+        chunkjson = json.loads(response.content.decode())
+        gpt_replying_buffer = chunkjson['choices'][0]["message"]["content"]
+        log_chat(llm_model=llm_kwargs["llm_model"], input_str=inputs, output_str=gpt_replying_buffer)
+        history[-1] = gpt_replying_buffer
+        chatbot[-1] = (history[-2], history[-1])
+        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
+    except Exception as e:
+        yield from update_ui(chatbot=chatbot, history=history, msg="Json解析异常" + response.text) # 刷新界面
+
+def handle_error(inputs, llm_kwargs, chatbot, history, chunk_decoded, error_msg):
+    from request_llms.bridge_all import model_info
+    openai_website = ' 请登录OpenAI查看详情 https://platform.openai.com/signup'
+    if "reduce the length" in error_msg:
+        if len(history) >= 2: history[-1] = ""; history[-2] = "" # 清除当前溢出的输入：history[-2] 是本次输入, history[-1] 是本次输出
+        history = clip_history(inputs=inputs, history=history, tokenizer=model_info[llm_kwargs['llm_model']]['tokenizer'],
+                                               max_token_limit=(model_info[llm_kwargs['llm_model']]['max_token'])) # history至少释放二分之一
+        chatbot[-1] = (chatbot[-1][0], "[Local Message] Reduce the length. 本次输入过长, 或历史数据过长. 历史缓存数据已部分释放, 您可以请再次尝试. (若再次失败则更可能是因为输入过长.)")
+    elif "does not exist" in error_msg:
+        chatbot[-1] = (chatbot[-1][0], f"[Local Message] Model {llm_kwargs['llm_model']} does not exist. 模型不存在, 或者您没有获得体验资格.")
+    elif "Incorrect API key" in error_msg:
+        chatbot[-1] = (chatbot[-1][0], "[Local Message] Incorrect API key. OpenAI以提供了不正确的API_KEY为由, 拒绝服务. " + openai_website)
+    elif "exceeded your current quota" in error_msg:
+        chatbot[-1] = (chatbot[-1][0], "[Local Message] You exceeded your current quota. OpenAI以账户额度不足为由, 拒绝服务." + openai_website)
+    elif "account is not active" in error_msg:
+        chatbot[-1] = (chatbot[-1][0], "[Local Message] Your account is not active. OpenAI以账户失效为由, 拒绝服务." + openai_website)
+    elif "associated with a deactivated account" in error_msg:
+        chatbot[-1] = (chatbot[-1][0], "[Local Message] You are associated with a deactivated account. OpenAI以账户失效为由, 拒绝服务." + openai_website)
+    elif "API key has been deactivated" in error_msg:
+        chatbot[-1] = (chatbot[-1][0], "[Local Message] API key has been deactivated. OpenAI以账户失效为由, 拒绝服务." + openai_website)
+    elif "bad forward key" in error_msg:
+        chatbot[-1] = (chatbot[-1][0], "[Local Message] Bad forward key. API2D账户额度不足.")
+    elif "Not enough point" in error_msg:
+        chatbot[-1] = (chatbot[-1][0], "[Local Message] Not enough point. API2D账户点数不足.")
+    else:
+        from toolbox import regular_txt_to_markdown
+        tb_str = '```\n' + trimmed_format_exc() + '```'
+        chatbot[-1] = (chatbot[-1][0], f"[Local Message] 异常 \n\n{tb_str} \n\n{regular_txt_to_markdown(chunk_decoded)}")
+    return chatbot, history
+
+def generate_payload(inputs:str, llm_kwargs:dict, history:list, system_prompt:str, image_base64_array:list=[], has_multimodal_capacity:bool=False, stream:bool=True):
+    """
+    整合所有信息，选择LLM模型，生成http请求，为发送请求做准备
+    """
+    from request_llms.bridge_all import model_info
+
+    if not is_any_api_key(llm_kwargs['api_key']):
+        raise AssertionError("你提供了错误的API_KEY。\n\n1. 临时解决方案：直接在输入区键入api_key，然后回车提交。\n\n2. 长效解决方案：在config.py中配置。")
+
+    if llm_kwargs['llm_model'].startswith('vllm-'):
+        api_key = 'no-api-key'
+    else:
+        api_key = select_api_key(llm_kwargs['api_key'], llm_kwargs['llm_model'])
+
+    headers = {
+        "Content-Type": "application/json",
+        "Authorization": f"Bearer {api_key}"
+    }
+    if API_ORG.startswith('org-'): headers.update({"OpenAI-Organization": API_ORG})
+    if llm_kwargs['llm_model'].startswith('azure-'):
+        headers.update({"api-key": api_key})
+        if llm_kwargs['llm_model'] in AZURE_CFG_ARRAY.keys():
+            azure_api_key_unshared = AZURE_CFG_ARRAY[llm_kwargs['llm_model']]["AZURE_API_KEY"]
+            headers.update({"api-key": azure_api_key_unshared})
+
+    if has_multimodal_capacity:
+        # 当以下条件满足时，启用多模态能力：
+        # 1. 模型本身是多模态模型（has_multimodal_capacity）
+        # 2. 输入包含图像（len(image_base64_array) > 0）
+        # 3. 历史输入包含图像（ any([contain_base64(h) for h in history]) ）
+        enable_multimodal_capacity = (len(image_base64_array) > 0) or any([contain_base64(h) for h in history])
+    else:
+        enable_multimodal_capacity = False
+
+    conversation_cnt = len(history) // 2
+    openai_disable_system_prompt = model_info[llm_kwargs['llm_model']].get('openai_disable_system_prompt', False)
+
+    if openai_disable_system_prompt:
+        messages = [{"role": "user", "content": system_prompt}]
+    else:
+        messages = [{"role": "system", "content": system_prompt}]
+
+    if not enable_multimodal_capacity:
+        # 不使用多模态能力
+        if conversation_cnt:
+            for index in range(0, 2*conversation_cnt, 2):
+                what_i_have_asked = {}
+                what_i_have_asked["role"] = "user"
+                what_i_have_asked["content"] = remove_image_if_contain_base64(history[index])
+                what_gpt_answer = {}
+                what_gpt_answer["role"] = "assistant"
+                what_gpt_answer["content"] = remove_image_if_contain_base64(history[index+1])
+                if what_i_have_asked["content"] != "":
+                    if what_gpt_answer["content"] == "": continue
+                    if what_gpt_answer["content"] == timeout_bot_msg: continue
+                    messages.append(what_i_have_asked)
+                    messages.append(what_gpt_answer)
+                else:
+                    messages[-1]['content'] = what_gpt_answer['content']
+        what_i_ask_now = {}
+        what_i_ask_now["role"] = "user"
+        what_i_ask_now["content"] = inputs
+        messages.append(what_i_ask_now)
+    else:
+        # 多模态能力
+        if conversation_cnt:
+            for index in range(0, 2*conversation_cnt, 2):
+                what_i_have_asked = {}
+                what_i_have_asked["role"] = "user"
+                what_i_have_asked["content"] = append_image_if_contain_base64(history[index])
+                what_gpt_answer = {}
+                what_gpt_answer["role"] = "assistant"
+                what_gpt_answer["content"] = append_image_if_contain_base64(history[index+1])
+                if what_i_have_asked["content"] != "":
+                    if what_gpt_answer["content"] == "": continue
+                    if what_gpt_answer["content"] == timeout_bot_msg: continue
+                    messages.append(what_i_have_asked)
+                    messages.append(what_gpt_answer)
+                else:
+                    messages[-1]['content'] = what_gpt_answer['content']
+        what_i_ask_now = {}
+        what_i_ask_now["role"] = "user"
+        what_i_ask_now["content"] = []
+        what_i_ask_now["content"].append({
+            "type": "text",
+            "text": inputs
+        })
+        for image_base64 in image_base64_array:
+            what_i_ask_now["content"].append({
+                "type": "image_url",
+                "image_url": {
+                    "url": f"data:image/jpeg;base64,{image_base64}"
+                }
+            })
+        messages.append(what_i_ask_now)
+
+
+    model = llm_kwargs['llm_model']
+    if llm_kwargs['llm_model'].startswith('api2d-'):
+        model = llm_kwargs['llm_model'][len('api2d-'):]
+    if llm_kwargs['llm_model'].startswith('one-api-'):
+        model = llm_kwargs['llm_model'][len('one-api-'):]
+        model, _ = read_one_api_model_name(model)
+    if llm_kwargs['llm_model'].startswith('vllm-'):
+        model = llm_kwargs['llm_model'][len('vllm-'):]
+        model, _ = read_one_api_model_name(model)
+    if llm_kwargs['llm_model'].startswith('openrouter-'):
+        model = llm_kwargs['llm_model'][len('openrouter-'):]
+        model= read_one_api_model_name(model)
+    if model == "gpt-3.5-random": # 随机选择, 绕过openai访问频率限制
+        model = random.choice([
+            "gpt-3.5-turbo",
+            "gpt-3.5-turbo-16k",
+            "gpt-3.5-turbo-1106",
+            "gpt-3.5-turbo-0613",
+            "gpt-3.5-turbo-16k-0613",
+            "gpt-3.5-turbo-0301",
+        ])
+
+    payload = {
+        "model": model,
+        "messages": messages,
+        "temperature": llm_kwargs['temperature'],  # 1.0,
+        "top_p": llm_kwargs['top_p'],  # 1.0,
+        "n": 1,
+        "stream": stream,
+    }
+
+    return headers,payload
+
+

request_llms/oai_std_model_template.py

@@ -224,7 +224,7 @@ def get_predict_function(
                 try:
                     if finish_reason == "stop":
                         if not console_slience:
-                            logger.info(f"[response] {result}")
+                            print(f"[response] {result}")
                         break
                     result += response_text
                     if observe_window is not None:

requirements.txt

@@ -2,14 +2,15 @@ https://public.agent-matrix.com/publish/gradio-3.32.10-py3-none-any.whl
 fastapi==0.110
 gradio-client==0.8
 pypdf2==2.12.1
+httpx<=0.25.2
 zhipuai==2.0.1
 tiktoken>=0.3.3
 requests[socks]
-pydantic==2.5.2
-llama-index==0.10
+pydantic==2.9.2
 protobuf==3.20
 transformers>=4.27.1,<4.42
 scipdf_parser>=0.52
+spacy==3.7.4
 anthropic>=0.18.1
 python-markdown-math
 pymdown-extensions
@@ -32,3 +33,14 @@ loguru
 arxiv
 numpy
 rich
+
+
+llama-index-core==0.10.68
+llama-index-legacy==0.9.48
+llama-index-readers-file==0.1.33
+llama-index-readers-llama-parse==0.1.6
+llama-index-embeddings-azure-openai==0.1.10
+llama-index-embeddings-openai==0.1.10
+llama-parse==0.4.9
+mdit-py-plugins>=0.3.3
+linkify-it-py==2.0.3

shared_utils/config_loader.py

@@ -94,7 +94,7 @@ def read_single_conf_with_lru_cache(arg):
         if r is None:
             log亮红('[PROXY] 网络代理状态：未配置。无代理状态下很可能无法访问OpenAI家族的模型。建议：检查USE_PROXY选项是否修改。')
         else:
-            log亮绿('[PROXY] 网络代理状态：已配置。配置信息如下：', r)
+            log亮绿('[PROXY] 网络代理状态：已配置。配置信息如下：', str(r))
             assert isinstance(r, dict), 'proxies格式错误，请注意proxies选项的格式，不要遗漏括号。'
     return r
 

shared_utils/cookie_manager.py

@@ -90,23 +90,6 @@ def make_history_cache():
 
 
 
-# """
-# with gr.Row():
-#     txt = gr.Textbox(show_label=False, placeholder="Input question here.", elem_id='user_input_main').style(container=False)
-#     txtx = gr.Textbox(show_label=False, placeholder="Input question here.", elem_id='user_input_main').style(container=False)
-# with gr.Row():
-#     btn_value = "Test"
-#     elem_id = "TestCase"
-#     variant = "primary"
-#     input_list = [txt, txtx]
-#     output_list = [txt, txtx]
-#     input_name_list = ["txt(input)", "txtx(input)"]
-#     output_name_list = ["txt", "txtx"]
-#     js_callback = """(txt, txtx)=>{console.log(txt); console.log(txtx);}"""
-#     def function(txt, txtx):
-#         return "booo", "goooo"
-#     create_button_with_javascript_callback(btn_value, elem_id, variant, js_callback, input_list, output_list, function, input_name_list, output_name_list)
-# """
 def create_button_with_javascript_callback(btn_value, elem_id, variant, js_callback, input_list, output_list, function, input_name_list, output_name_list):
     import gradio as gr
     middle_ware_component = gr.Textbox(visible=False, elem_id=elem_id+'_buffer')

shared_utils/key_pattern_manager.py

@@ -34,6 +34,9 @@ def is_api2d_key(key):
     API_MATCH_API2D = re.match(r"fk[a-zA-Z0-9]{6}-[a-zA-Z0-9]{32}$", key)
     return bool(API_MATCH_API2D)
 
+def is_openroute_api_key(key):
+    API_MATCH_OPENROUTE = re.match(r"sk-or-v1-[a-zA-Z0-9]{64}$", key)
+    return bool(API_MATCH_OPENROUTE)
 
 def is_cohere_api_key(key):
     API_MATCH_AZURE = re.match(r"[a-zA-Z0-9]{40}$", key)
@@ -89,6 +92,10 @@ def select_api_key(keys, llm_model):
     if llm_model.startswith('cohere-'):
         for k in key_list:
             if is_cohere_api_key(k): avail_key_list.append(k)
+    
+    if llm_model.startswith('openrouter-'):
+        for k in key_list:
+            if is_openroute_api_key(k): avail_key_list.append(k)
 
     if len(avail_key_list) == 0:
         raise RuntimeError(f"您提供的api-key不满足要求，不包含任何可用于{llm_model}的api-key。您可能选择了错误的模型或请求源（左上角更换模型菜单中可切换openai,azure,claude,cohere等请求源）。")

shared_utils/logging.py

@@ -11,7 +11,7 @@ def not_chat_log_filter(record):
 
 def formatter_with_clip(record):
     # Note this function returns the string to be formatted, not the actual message to be logged
-    record["extra"]["serialized"] = "555555"
+    # record["extra"]["serialized"] = "555555"
     max_len = 12
     record['function_x'] = record['function'].center(max_len)
     if len(record['function_x']) > max_len:

tests/test_anim_gen.py

@@ -0,0 +1,12 @@
+"""
+对项目中的各个插件进行测试。运行方法：直接运行 python tests/test_plugins.py
+"""
+
+import init_test
+import os, sys
+
+
+if __name__ == "__main__":
+    from test_utils import plugin_test
+
+    plugin_test(plugin='crazy_functions.数学动画生成manim->动画生成', main_input="A point moving along function culve y=sin(x), starting from x=0 and stop at x=4*\pi.")

tests/test_social_helper.py

@@ -8,4 +8,17 @@ import os, sys
 
 if __name__ == "__main__":
     from test_utils import plugin_test
-    plugin_test(plugin='crazy_functions.Social_Helper->I人助手', main_input="|")
+    plugin_test(
+        plugin='crazy_functions.Social_Helper->I人助手', 
+        main_input="""
+添加联系人：
+艾德·史塔克：我的养父，他是临冬城的公爵。
+凯特琳·史塔克：我的养母，她对我态度冷淡，因为我是私生子。
+罗柏·史塔克：我的哥哥，他是北境的继承人。
+艾莉亚·史塔克：我的妹妹，她和我关系亲密，性格独立坚强。
+珊莎·史塔克：我的妹妹，她梦想成为一位淑女。
+布兰·史塔克：我的弟弟，他有预知未来的能力。
+瑞肯·史塔克：我的弟弟，他是个天真无邪的小孩。
+山姆威尔·塔利：我的朋友，他在守夜人军团中与我并肩作战。
+伊格瑞特：我的恋人，她是野人中的一员。
+        """)

version

@@ -1,5 +1,5 @@
 {
-  "version": 3.83,
+  "version": 3.90,
   "show_feature": true,
-  "new_feature": "增加欢迎页面 <-> 优化图像生成插件 <-> 添加紫东太初大模型支持 <-> 保留主题选择 <-> 支持更复杂的插件框架 <-> 上传文件时显示进度条"
+  "new_feature": "增加RAG组件 <-> 升级多合一主提交键"
 }