llm-dialog-manager 0.4.3__tar.gz → 0.4.4__tar.gz

{llm_dialog_manager-0.4.3 → llm_dialog_manager-0.4.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: llm_dialog_manager
-Version: 0.4.3
+Version: 0.4.4
 Summary: A Python package for managing LLM chat conversation history
 Author-email: xihajun <work@2333.fun>
 License: MIT

{llm_dialog_manager-0.4.3 → llm_dialog_manager-0.4.4}/llm_dialog_manager/__init__.py

@@ -1,4 +1,4 @@
 from .chat_history import ChatHistory
 from .agent import Agent
 
-__version__ = "0.4.3" 
+__version__ = "0.4.4" 

{llm_dialog_manager-0.4.3 → llm_dialog_manager-0.4.4}/llm_dialog_manager/agent.py

@@ -38,6 +38,10 @@ def load_env_vars():
 
 load_env_vars()
 
+def encode_image(image_path):
+    with open(image_path, "rb") as image_file:
+        return base64.b64encode(image_file.read()).decode("utf-8")
+
 def format_messages_for_gemini(messages):
     """
     将标准化的消息格式转化为 Gemini 格式。
@@ -393,22 +397,42 @@ class Agent:
                 # For Gemini, load as PIL.Image
                 image_pil = Image.open(image_path)
                 image_block = image_pil
-            else:
+            elif "claude" in self.model_name and "openai" not in self.model_name:
                 # For Claude and others, use base64 encoding
                 with open(image_path, "rb") as img_file:
                     image_data = base64.standard_b64encode(img_file.read()).decode("utf-8")
                 image_block = {
-                    "type": "image_base64",
-                    "image_base64": {
+                    "type": "image",
+                    "source": {
+                        "type": "base64",
                         "media_type": media_type,
-                        "data": image_data
-                    }
+                        "data": image_data,
+                    },
+                }
+            else:
+                # openai format
+                base64_image = encode_image(image_path)
+                image_block = {
+                    "type": "image_url",
+                    "image_url": {"url": f"data:image/jpeg;base64,{base64_image}"},
                 }
         else:
             # If image_url is provided
             if "gemini" in self.model_name and "openai" not in self.model_name:
                 # For Gemini, you can pass image URLs directly
                 image_block = {"type": "image_url", "image_url": {"url": image_url}}
+            elif "claude" in self.model_name and "openai" not in self.model_name:
+                import httpx
+                media_type = "image/jpeg"
+                image_data = base64.standard_b64encode(httpx.get(image_url).content).decode("utf-8")
+                image_block = {
+                    "type": "image",
+                    "source": {
+                        "type": "base64",
+                        "media_type": media_type,
+                        "data": image_data,
+                    },
+                }
             else:
                 # For Claude and others, use image URLs
                 image_block = {
@@ -525,7 +549,7 @@ if __name__ == "__main__":
     agent = Agent("gemini-1.5-flash", "you are Jack101", memory_enabled=True)
     
     # Add an image
-    agent.add_image(image_path="/Users/junfan/Projects/Personal/oneapi/dialog_manager/example.png")
+    agent.add_image(image_path="example.png")
     
     # Add a user message
     agent.add_message("user", "Who are you? What's in this image?")

{llm_dialog_manager-0.4.3 → llm_dialog_manager-0.4.4}/llm_dialog_manager.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: llm_dialog_manager
-Version: 0.4.3
+Version: 0.4.4
 Summary: A Python package for managing LLM chat conversation history
 Author-email: xihajun <work@2333.fun>
 License: MIT

{llm_dialog_manager-0.4.3 → llm_dialog_manager-0.4.4}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "llm_dialog_manager"
-version = "0.4.3"
+version = "0.4.4"
 description = "A Python package for managing LLM chat conversation history"
 readme = "README.md"
 classifiers = [ "Development Status :: 3 - Alpha", "Intended Audience :: Developers", "License :: OSI Approved :: MIT License", "Operating System :: OS Independent", "Programming Language :: Python :: 3.8", "Programming Language :: Python :: 3.9", "Programming Language :: Python :: 3.10", "Topic :: Software Development :: Libraries :: Python Modules", "Topic :: Scientific/Engineering :: Artificial Intelligence",]