功能优化

js/ChildWorkflowNodes.js

@@ -40,12 +40,12 @@ app.registerExtension({
                 this.setProperty("paramList", [])
                 this.setProperty("workflow", {})
                 let thisNode = this
-                const widget = this.widgets.find(w => w.name === "workflowFile");
                 this.addDOMWidget('childJson',"list",$el('span.hidden',{style:{"word-wrap": "break-word"}}),{
                     getValue(){
                         return JSON.stringify(thisNode.properties['workflow'])  
                     }
                 });
+                const widget = this.widgets.find(w => w.name === "workflowFile");
                 widget.callback = async (value) => {
                     if (value) {
                         let props = await getWorkflow(value);

js/ChildWorkflowParameters.js

@@ -79,8 +79,8 @@ function get_position_style(ctx, widget_width, y, node_height) {
     return {  
         transformOrigin: '0 0',  
         transform: transform,  
-        left: `50px`,  
-        top: `35px`,  
+        left: `0px`,  
+        top: `0px`,  
         position: "absolute",  
         maxWidth: `${widget_width - MARGIN*2}px`,  
         maxHeight: `${node_height - MARGIN*2}px`,    // we're assuming we have the whole height of the node  

py/OpenAiPrompt.py

@@ -1,5 +1,15 @@
 import json
 from openai import OpenAI
+import base64
+from PIL import Image
+import io
+import numpy as np
+from lam_tools import tensor2pil,pil2tensor
+
+def encode_image(image_path: str) -> str:
+    """将图像编码为 base64 字符串"""
+    with open(image_path, "rb") as image_file:
+        return base64.b64encode(image_file.read()).decode('utf-8')
 
 class OpenAiPrompt:
     """
@@ -19,12 +29,13 @@ class OpenAiPrompt:
                 "text": ("STRING", {"multiline": True}),
             },
             "optional": {
-                "messages":("LIST",)
+                "messages":("LIST",),
+                "images": ("IMAGE,STRING", )
             }
         }
 
-    RETURN_TYPES = ("STRING",)
-    RETURN_NAMES = ("结果",)
+    RETURN_TYPES = ("STRING","LIST",)
+    RETURN_NAMES = ("结果","messages",)
 
     FUNCTION = "translate"
 
@@ -32,15 +43,30 @@ class OpenAiPrompt:
 
     CATEGORY = "lam"
 
-    def translate(self, server_url,api_key,model_name,system_prompt,text,messages=None):
+    def translate(self, server_url,api_key,model_name,system_prompt,text,messages=None,images=None):
         client = OpenAI(api_key=api_key,base_url=server_url)
         if messages is None:
             messages = []
+        if system_prompt and len(system_prompt.strip())>0:
             messages.append({"role": "system", "content": system_prompt})
-        messages.append({"role": "user", "content": text})
-        completion = client.chat.completions.create(model=model_name,messages=messages,
-                                                    top_p=0.8,
-                                                    temperature=0.7)
+        if images!=None:
+            content=[]
+            content.append({"type": "text", "text": text})
+            if isinstance(images, str):
+                # 图像理解示例
+                image_base64 = encode_image(images)
+                content.append({"type": "image_url", "image_url": {"url":f"data:image/jpeg;base64,{image_base64}"}})
+            else:
+                img = tensor2pil(images)
+                output = io.BytesIO()
+                img.save(output, format="JPEG")
+                image_base64 = base64.b64encode(output.getvalue()).decode('utf-8')
+                content.append({"type": "image_url", "image_url": {"url":f"data:image/jpeg;base64,{image_base64}"}})
+            messages.append({"role": "user", "content": content})
+        else:
+            messages.append({"role": "user", "content": text})
+
+        completion = client.chat.completions.create(model=model_name,messages=messages,top_p=0.8,temperature=0.7)
         messages.append({"role": "assistant", "content": completion.choices[0].message.content})
         return (completion.choices[0].message.content,messages,)