add files upload in webapp and human control example

examples/run.py

@@ -145,4 +145,4 @@ def main():
 
 
 if __name__ == "__main__":
-    main()
+    main()

examples/run_human_control.py

@@ -0,0 +1,152 @@
+# ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
+import sys
+import pathlib
+from dotenv import load_dotenv
+from camel.models import ModelFactory
+from camel.toolkits import (
+    AudioAnalysisToolkit,
+    CodeExecutionToolkit,
+    ExcelToolkit,
+    ImageAnalysisToolkit,
+    SearchToolkit,
+    VideoAnalysisToolkit,
+    BrowserToolkit,
+    FileWriteToolkit,
+    HumanToolkit,
+    TerminalToolkit,
+)
+from camel.types import ModelPlatformType, ModelType
+from camel.logger import set_log_level
+from camel.societies import RolePlaying
+
+from owl.utils import run_society, DocumentProcessingToolkit
+
+base_dir = pathlib.Path(__file__).parent.parent
+env_path = base_dir / "owl" / ".env"
+load_dotenv(dotenv_path=str(env_path))
+
+set_log_level(level="DEBUG")
+
+
+def construct_society(question: str) -> RolePlaying:
+    r"""Construct a society of agents based on the given question.
+
+    Args:
+        question (str): The task or question to be addressed by the society.
+
+    Returns:
+        RolePlaying: A configured society of agents ready to address the question.
+    """
+
+    # Create models for different components
+    models = {
+        "user": ModelFactory.create(
+            model_platform=ModelPlatformType.OPENAI,
+            model_type=ModelType.GPT_4O,
+            model_config_dict={"temperature": 0},
+        ),
+        "assistant": ModelFactory.create(
+            model_platform=ModelPlatformType.OPENAI,
+            model_type=ModelType.GPT_4O,
+            model_config_dict={"temperature": 0},
+        ),
+        "browsing": ModelFactory.create(
+            model_platform=ModelPlatformType.OPENAI,
+            model_type=ModelType.GPT_4O,
+            model_config_dict={"temperature": 0},
+        ),
+        "planning": ModelFactory.create(
+            model_platform=ModelPlatformType.OPENAI,
+            model_type=ModelType.GPT_4O,
+            model_config_dict={"temperature": 0},
+        ),
+        "video": ModelFactory.create(
+            model_platform=ModelPlatformType.OPENAI,
+            model_type=ModelType.GPT_4O,
+            model_config_dict={"temperature": 0},
+        ),
+        "image": ModelFactory.create(
+            model_platform=ModelPlatformType.OPENAI,
+            model_type=ModelType.GPT_4O,
+            model_config_dict={"temperature": 0},
+        ),
+        "document": ModelFactory.create(
+            model_platform=ModelPlatformType.OPENAI,
+            model_type=ModelType.GPT_4O,
+            model_config_dict={"temperature": 0},
+        ),
+    }
+
+    # Configure toolkits
+    tools = [
+        *HumanToolkit().get_tools(),
+        *TerminalToolkit().get_tools(),
+        *BrowserToolkit(
+            headless=False,  # Set to True for headless mode (e.g., on remote servers)
+            web_agent_model=models["browsing"],
+            planning_agent_model=models["planning"],
+        ).get_tools(),
+        *VideoAnalysisToolkit(model=models["video"]).get_tools(),
+        *AudioAnalysisToolkit().get_tools(),  # This requires OpenAI Key
+        *CodeExecutionToolkit(sandbox="subprocess", verbose=True).get_tools(),
+        *ImageAnalysisToolkit(model=models["image"]).get_tools(),
+        SearchToolkit().search_duckduckgo,
+        SearchToolkit().search_google,  # Comment this out if you don't have google search
+        SearchToolkit().search_wiki,
+        *ExcelToolkit().get_tools(),
+        *DocumentProcessingToolkit(model=models["document"]).get_tools(),
+        *FileWriteToolkit(output_dir="./").get_tools(),
+    ]
+
+    # Configure agent roles and parameters
+    user_agent_kwargs = {"model": models["user"]}
+    assistant_agent_kwargs = {"model": models["assistant"], "tools": tools}
+
+    # Configure task parameters
+    task_kwargs = {
+        "task_prompt": question,
+        "with_task_specify": False,
+    }
+
+    # Create and return the society
+    society = RolePlaying(
+        **task_kwargs,
+        user_role_name="user",
+        user_agent_kwargs=user_agent_kwargs,
+        assistant_role_name="assistant",
+        assistant_agent_kwargs=assistant_agent_kwargs,
+    )
+
+    return society
+
+
+def main():
+    r"""Main function to run the OWL system with an example question."""
+    # Default research question
+    default_task = "Use plot to draw a heart-shaped graph, save it locally, and ask for my opinion using HumanToolkit before taking each step."
+
+    # Override default task if command line argument is provided
+    task = sys.argv[1] if len(sys.argv) > 1 else default_task
+
+    # Construct and run the society
+    society = construct_society(task)
+    answer, chat_history, token_count = run_society(society)
+
+    # Output the result
+    print(f"\033[94mAnswer: {answer}\033[0m")
+
+
+if __name__ == "__main__":
+    main()

owl/webapp_zh.py

@@ -25,6 +25,7 @@
 import threading
 import queue
 import re  # For regular expression operations
+import tempfile  # 添加此行，用于处理临时文件
 
 os.environ["PYTHONIOENCODING"] = "utf-8"
 
@@ -309,12 +310,13 @@ def validate_input(question: str) -> bool:
     return True
 
 
-def run_owl(question: str, example_module: str) -> Tuple[str, str, str]:
+def run_owl(question: str, example_module: str, uploaded_file=None) -> Tuple[str, str, str]:
     """运行OWL系统并返回结果
 
     Args:
         question: 用户问题
         example_module: 要导入的示例模块名（如 "run_terminal_zh" 或 "run_deep"）
+        uploaded_file: 上传的文件（可选）- 当使用type="binary"时，这是字节数据
 
     Returns:
         Tuple[...]: 回答、令牌计数、状态
@@ -330,6 +332,73 @@ def run_owl(question: str, example_module: str) -> Tuple[str, str, str]:
         # 确保环境变量已加载
         load_dotenv(find_dotenv(), override=True)
         logging.info(f"处理问题: '{question}', 使用模块: {example_module}")
+
+        # 处理上传的文件
+        if uploaded_file is not None:
+            try:
+                # 创建临时目录保存上传的文件
+                tmp_dir = tempfile.mkdtemp(prefix="owl_upload_")
+                logging.info(f"创建临时目录: {tmp_dir}")
+
+                # 生成一个唯一的文件名
+                timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
+
+                # 简单地根据文件内容的前几个字节来猜测文件类型
+                file_ext = ".bin"  # 默认扩展名
+
+                # 简单的文件类型检测，不依赖magic库
+                if len(uploaded_file) > 4:
+                    # PDF 文件通常以 %PDF 开头
+                    if uploaded_file[:4] == b'%PDF':
+                        file_ext = ".pdf"
+                    # ZIP 文件 (可能是DOCX, XLSX等)
+                    elif uploaded_file[:4] == b'PK\x03\x04':
+                        # 尝试根据文件名确定具体类型
+                        file_ext = ".zip"  # 默认为zip
+                    # 纯文本文件检测 (如果前100个字节都是可打印ASCII或常见Unicode)
+                    elif all(c < 128 and c >= 32 or c in (9, 10, 13) for c in uploaded_file[:100]):
+                        file_ext = ".txt"
+
+                file_name = f"uploaded_file_{timestamp}{file_ext}"
+
+                # 生成安全的文件路径
+                file_path = os.path.normpath(os.path.join(tmp_dir, file_name))
+
+                logging.info(f"准备保存文件: {file_path}")
+
+                # 保存文件内容
+                try:
+                    with open(file_path, "wb") as f:
+                        f.write(uploaded_file)
+
+                    # 检查文件是否成功写入
+                    if not os.path.exists(file_path) or os.path.getsize(file_path) == 0:
+                        raise ValueError("文件保存失败或文件为空")
+
+                    logging.info(f"文件已成功保存: {file_path}, 大小: {os.path.getsize(file_path)} 字节")
+
+                    # 将文件路径添加到问题中（使用规范化的路径，避免系统差异）
+                    normalized_path = file_path.replace('\\', '/')
+                    if "文件路径:" not in question and "file path:" not in question.lower():
+                        question = f"{question}\n文件路径: {normalized_path}"
+
+                    logging.info(f"更新后的问题: '{question}'")
+
+                except Exception as e:
+                    logging.error(f"保存文件时出错: {str(e)}")
+                    return (
+                        f"保存文件时出错: {str(e)}",
+                        "0",
+                        f"❌ 错误: 文件保存失败 - {str(e)}",
+                    )
+
+            except Exception as e:
+                logging.error(f"处理上传文件时出错: {str(e)}")
+                return (
+                    f"处理上传文件时出错: {str(e)}",
+                    "0",
+                    f"❌ 错误: 文件处理失败 - {str(e)}",
+                )
 
         # 检查模块是否在MODULE_DESCRIPTIONS中
         if example_module not in MODULE_DESCRIPTIONS:
@@ -769,8 +838,7 @@ def clear_log_file():
             logging.error(f"清空日志文件时出错: {str(e)}")
             return ""
 
-    # 创建一个实时日志更新函数
-    def process_with_live_logs(question, module_name):
+    def process_with_live_logs(question, module_name, uploaded_file=None):
         """处理问题并实时更新日志"""
         global CURRENT_PROCESS
 
@@ -782,7 +850,7 @@ def process_with_live_logs(question, module_name):
 
         def process_in_background():
             try:
-                result = run_owl(question, module_name)
+                result = run_owl(question, module_name, uploaded_file)
                 result_queue.put(result)
             except Exception as e:
                 result_queue.put((f"发生错误: {str(e)}", "0", f"❌ 错误: {str(e)}"))
@@ -1062,6 +1130,23 @@ def process_in_background():
                     value="打开百度搜索，总结一下camel-ai的camel框架的github star、fork数目等，并把数字用plot包写成python文件保存到本地，并运行生成的python文件。",
                 )
 
+
+                file_upload = gr.File(
+                    label="上传文件（可选）",
+                    file_types=["pdf", "docx", "txt", "csv", "xlsx", "json", "py", "ipynb"],
+                    file_count="single",
+                    type="binary",  # 使用binary类型
+                )
+
+                # 文件上传说明 - 更新说明以包含Windows兼容性
+                gr.Markdown("""
+                    <div style="background-color: #e7f3fe; border-left: 6px solid #2196F3; padding: 10px; margin-bottom: 15px; border-radius: 4px;">
+                      <strong>文件上传说明：</strong> 上传文件后，系统会自动将文件保存到临时目录并添加文件路径到您的问题中。
+                      支持多种文件格式，包括PDF、Word文档、文本文件、CSV、Excel、Python代码等。
+                      <br><strong>注意：</strong> 文件会被保存在临时目录，系统重启后可能会被清除。
+                    </div>
+                """)
+
                 # 增强版模块选择下拉菜单
                 # 只包含MODULE_DESCRIPTIONS中定义的模块
                 module_dropdown = gr.Dropdown(
@@ -1205,10 +1290,10 @@ def process_in_background():
 
                     refresh_button.click(fn=update_env_table, outputs=[env_table])
 
-        # 设置事件处理
+
         run_button.click(
             fn=process_with_live_logs,
-            inputs=[question_input, module_dropdown],
+            inputs=[question_input, module_dropdown, file_upload],
             outputs=[token_count_output, status_output, log_display2],
         )