0.15.1

Update ImageNode.py
LoadImage 的mask保存到appinfo
2024-02-06 23:06:33 +08:00 · 2024-02-06 22:22:56 +08:00 · 2024-02-06 19:01:41 +08:00 · 2024-02-05 22:44:04 +08:00 · 2024-02-05 22:25:25 +08:00 · 2024-02-05 22:24:34 +08:00
23 changed files with 2209 additions and 471 deletions
@@ -1,5 +1,7 @@
 > 适配了最新版comfyui的py3.11 ，torch 2.1.2+cu121

+> [discord](https://discord.gg/cXs9vZSqeK)
+
 ## 🚀🚗🚚🏃 Workflow-to-APP 
 - 新增AppInfo节点，可以通过简单的配置，把workflow转变为一个Web APP。
 - 支持多个web app 切换
@@ -115,6 +117,11 @@ https://github.com/shadowcz007/comfyui-mixlab-nodes/assets/12645064/e7e77f90-e43

 - [Added DynamicDelayByText, enabling delayed execution based on input text length.](./workflow/audio-chatgpt-workflow.json)

+- [使用CkptNames 对比不同的模型效果](./workflow/ckpts-image-workflow.json)
+
+- [CkptNames compare the effects of different models.](./workflow/ckpts-image-workflow.json)
+
+

 ## Other Nodes

@@ -130,15 +137,6 @@ https://github.com/shadowcz007/comfyui-mixlab-nodes/assets/12645064/e7e77f90-e43
 ![TransparentImage](./assets/TransparentImage.png)


-> Consistency Decoder
-
-[openai Consistency Decoder]( https://github.com/openai/consistencydecoder)
-
-![Consistency](./assets/consistency.png)
-After downloading the OpenAI VAE model, place it in the "model/vae" directory for use.
-https://openaipublic.azureedge.net/diff-vae/c9cebd3132dd9c42936d803e33424145a748843c8f716c0814838bdc8a2fe7cb/decoder.pt
-
-
 > FeatheredMask、SmoothMask

 Add edges to an image.
@@ -173,7 +171,7 @@ An improvement has been made to directly redirect to GitHub to search for missin

 [Download Salesforce/blip-image-captioning-base](https://huggingface.co/Salesforce/blip-image-captioning-base), move to : models/clip_interrogator/Salesforce/blip-image-captioning-base

-[Download succinctly/text2image-prompt-generator](https://huggingface.co/succinctly/text2image-prompt-generator/tree/main),move to:text_generator/text2image-prompt-generator
+[Download succinctly/text2image-prompt-generator](https://huggingface.co/succinctly/text2image-prompt-generator/tree/main),move to:prompt_generator/text2image-prompt-generator

 [Download Helsinki-NLP/opus-mt-zh-en](https://huggingface.co/Helsinki-NLP/opus-mt-zh-en/tree/main),move to:prompt_generator/opus-mt-zh-en

@@ -206,6 +204,11 @@ If you are using a venv, make sure you have it activated before installation and
 pip3 install -r requirements.txt
 ```

+#### 
+[comfyui-ultralytics-yolo](https://github.com/shadowcz007/comfyui-ultralytics-yolo)
+[comfyui-moondream](https://github.com/shadowcz007/comfyui-moondream)
+
+
 #### Chinese community
 访问 [www.mixcomfy.com](https://www.mixcomfy.com)，获得更多内测功能，关注微信公众号：Mixlab无界社区

@@ -192,7 +192,7 @@ def read_workflow_json_files(folder_path ):
 def get_workflows():
    # print("#####path::", current_path)
    workflow_path=os.path.join(current_path, "workflow")
-    print('workflow_path: ',workflow_path)
+    # print('workflow_path: ',workflow_path)
    if not os.path.exists(workflow_path):
        # 使用mkdir()方法创建新目录
        os.mkdir(workflow_path)
@@ -222,7 +222,8 @@ def get_my_workflow_for_app(filename="my_workflow_app.json",category="",is_all=F
            # print(item)
            try:
                x=item["data"]
-                if i==0:
+                # 管理员模式，读取全部数据
+                if i==0 or is_all:
                    apps.append({
                        "filename":item["filename"],
                        # "category":item['category'],
@@ -231,8 +232,14 @@ def get_my_workflow_for_app(filename="my_workflow_app.json",category="",is_all=F
                    })
                else:
                    category=''
+                    input=None
+                    output=None
                    if 'category' in x['app']:
                        category=x['app']['category']
+                    if 'input' in x['app']:
+                        input=x['app']['input']
+                    if 'output' in x['app']:
+                        output=x['app']['output']
                    apps.append({
                        "filename":item["filename"],
                        "category":category,
@@ -244,6 +251,8 @@ def get_my_workflow_for_app(filename="my_workflow_app.json",category="",is_all=F
                                "icon":(x['app']['icon'] if 'icon' in x['app'] else None),
                                "name":x['app']['name'],
                                "version":x['app']['version'],
+                                "input":input,
+                                "output":output
                            }
                        },
                        "date":item["date"]
@@ -271,8 +280,14 @@ def get_my_workflow_for_app(filename="my_workflow_app.json",category="",is_all=F
                # print(apps[0]['filename'] ,item["filename"])
                if apps[0]['filename']!=item["filename"]:
                    category=''
+                    input=None
+                    output=None
                    if 'category' in x['app']:
                        category=x['app']['category']
+                    if 'input' in x['app']:
+                        input=x['app']['input']
+                    if 'output' in x['app']:
+                        output=x['app']['output']
                    apps.append({
                        "filename":item["filename"],
                        # "category":category,
@@ -284,6 +299,8 @@ def get_my_workflow_for_app(filename="my_workflow_app.json",category="",is_all=F
                                "icon":(x['app']['icon'] if 'icon' in x['app'] else None),
                                "name":x['app']['name'],
                                "version":x['app']['version'],
+                                "input":input,
+                                "output":output
                            }
                        },
                        "date":item["date"]
@@ -414,7 +431,7 @@ async def new_start(self, address, port, verbose=True, call_on_start=None):
 PromptServer.start=new_start

 # 创建路由表
-routes = web.RouteTableDef()
+routes = PromptServer.instance.routes

@routes.post('/mixlab')
 async def mixlab_hander(request):
@@ -500,27 +517,6 @@ async def nodes_map_hander(request):

    return web.json_response(result)

-# 把插件自定义的路由添加到comfyui server里
-def new_add_routes(self):
-        import nodes
-        try:
-            self.user_manager.add_routes(self.routes)
-        except:
-            print('pls update')
-        
-        self.app.add_routes(routes)
-        self.app.add_routes(self.routes)
-        for name, dir in nodes.EXTENSION_WEB_DIRS.items():
-            self.app.add_routes([
-                web.static('/extensions/' + urllib.parse.quote(name), dir, follow_symlinks=True),
-            ])
-        self.app.add_routes([
-            web.static('/', self.web_root, follow_symlinks=True),
-        ])
-
-PromptServer.add_routes=new_add_routes
-
-

 # 扩展api接口
 # from server import PromptServer
@@ -534,15 +530,15 @@ PromptServer.add_routes=new_add_routes


 # 导入节点
-from .nodes.PromptNode import EmbeddingPrompt,RandomPrompt,PromptSlide,PromptSimplification,PromptImage
-from .nodes.ImageNode import GetImageSize_,MirroredImage,ImageColorTransfer,NoiseImage,TransparentImage,GradientImage,LoadImagesFromPath,LoadImagesFromURL,ResizeImage,TextImage,SvgImage,Image3D,ShowLayer,NewLayer,MergeLayers,AreaToMask,SmoothMask,FeatheredMask,SplitLongMask,ImageCropByAlpha,EnhanceImage,FaceToMask
-from .nodes.Vae import VAELoader,VAEDecode
+from .nodes.PromptNode import EmbeddingPrompt,RandomPrompt,PromptSlide,PromptSimplification,PromptImage,JoinWithDelimiter
+from .nodes.ImageNode import SaveImageToLocal,SplitImage,GridOutput,GetImageSize_,MirroredImage,ImageColorTransfer,NoiseImage,TransparentImage,GradientImage,LoadImagesFromPath,LoadImagesFromURL,ResizeImage,TextImage,SvgImage,Image3D,ShowLayer,NewLayer,MergeLayers,CenterImage,AreaToMask,SmoothMask,SplitLongMask,ImageCropByAlpha,EnhanceImage,FaceToMask
+# from .nodes.Vae import VAELoader,VAEDecode
 from .nodes.ScreenShareNode import ScreenShareNode,FloatingVideo
 from .nodes.Clipseg import CLIPSeg,CombineMasks
 from .nodes.ChatGPT import ChatGPTNode,ShowTextForGPT,CharacterInText
 from .nodes.Audio import GamePal,SpeechRecognition,SpeechSynthesis
-from .nodes.Utils import TESTNODE_,AppInfo,IntNumber,FloatSlider,TextInput,ColorInput,FontInput,TextToNumber,DynamicDelayProcessor,LimitNumber,SwitchByIndex,MultiplicationNode
-from .nodes.Mask import OutlineMask
+from .nodes.Utils import CreateLoraNames,CreateSampler_names,CreateCkptNames,CreateSeedNode,TESTNODE_,AppInfo,IntNumber,FloatSlider,TextInput,ColorInput,FontInput,TextToNumber,DynamicDelayProcessor,LimitNumber,SwitchByIndex,MultiplicationNode
+from .nodes.Mask import OutlineMask,FeatheredMask


 # 要导出的所有节点及其名称的字典
@@ -569,6 +565,9 @@ NODE_CLASS_MAPPINGS = {
    "ImageColorTransfer":ImageColorTransfer,
    "ShowLayer":ShowLayer,
    "NewLayer":NewLayer,
+    "SplitImage":SplitImage,
+    "CenterImage":CenterImage,
+    "GridOutput":GridOutput,
    "MergeLayers":MergeLayers,
    "SplitLongMask":SplitLongMask,
    "FeatheredMask":FeatheredMask,
@@ -576,8 +575,9 @@ NODE_CLASS_MAPPINGS = {
    "FaceToMask":FaceToMask,
    "AreaToMask":AreaToMask,
    "ImageCropByAlpha":ImageCropByAlpha,
-    "VAELoaderConsistencyDecoder":VAELoader,
-    "VAEDecodeConsistencyDecoder":VAEDecode,
+    # "VAELoaderConsistencyDecoder":VAELoader,
+    "SaveImageToLocal":SaveImageToLocal,
+    # "VAEDecodeConsistencyDecoder":VAEDecode,
    "ScreenShare":ScreenShareNode,
    "FloatingVideo":FloatingVideo,
    "CLIPSeg_":CLIPSeg,
@@ -598,7 +598,12 @@ NODE_CLASS_MAPPINGS = {
    "GetImageSize_":GetImageSize_,
    "SwitchByIndex":SwitchByIndex,
    "LimitNumber":LimitNumber, 
-    "OutlineMask":OutlineMask
+    "OutlineMask":OutlineMask,
+    "JoinWithDelimiter":JoinWithDelimiter,
+    "Seed_":CreateSeedNode,
+    "CkptNames_":CreateCkptNames,
+    "SamplerNames_":CreateSampler_names,
+    "LoraNames_":CreateLoraNames
    # "LaMaInpainting":LaMaInpainting
    # "GamePal":GamePal
 }
@@ -625,7 +630,8 @@ NODE_DISPLAY_NAME_MAPPINGS = {
    "PromptGenerate_Mix":"PromptGenerate ♾️Mixlab",
    "ChinesePrompt_Mix":"ChinesePrompt ♾️Mixlab",
    "GamePal":"GamePal ♾️Mixlab",
-    "RembgNode_Mix":"Removebg"
+    "RembgNode_Mix":"Removebg",
+    "LoraNames_":"LoraName_TriggerWords.safetensors"
 }

 # web ui的节点功能
@@ -4761,7 +4761,10 @@
    ],
    "https://github.com/shadowcz007/comfyui-mixlab-nodes": [
        [
+            "GridOutput",
+            "SplitImage",
            "PromptGenerate_Mix",
+            "JoinWithDelimiter",
            "ChinesePrompt_Mix",
            "3DImage",
            "AppInfo",
@@ -4770,11 +4773,16 @@
            "ResizeImage",
            "NoiseImage",
            "PromptImage",
+            "SaveImageToLocal",
            "AreaToMask",
            "CLIPSeg_",
            "CharacterInText",
            "ChatGPTOpenAI",
            "Color",
+            "Seed_",
+            "CkptNames_",
+            "SamplerNames_",
+            "LoraNames_",
            "CombineMasks_",
            "EnhanceImage",
            "GradientImage",
@@ -4787,6 +4795,7 @@
            "LoadImagesFromURL",
            "MergeLayers",
            "NewLayer",
+            "CenterImage",
            "RandomPrompt",
            "PromptSlide",
            "PromptSimplification",
@@ -4802,12 +4811,11 @@
            "TextImage",
            "ResizeImageMixlab",
            "TransparentImage",
-            "VAEDecodeConsistencyDecoder",
-            "VAELoaderConsistencyDecoder",
            "TextToNumber",
            "TextInput_",
            "DynamicDelayProcessor",
-            "LaMaInpainting"
+            "LaMaInpainting",
+            "Moondream"
        ],
        {
            "title_aux": "comfyui-mixlab-nodes"
@@ -1,7 +1,20 @@
 import openai 
 import time
 import urllib.error
-import re,json
+import re,json,os,string,random
+import folder_paths
+import hashlib
+
+def get_unique_hash(string):
+    hash_object = hashlib.sha1(string.encode())
+    unique_hash = hash_object.hexdigest()
+    return unique_hash
+
+def generate_random_string(length):
+    letters = string.ascii_letters + string.digits
+    return ''.join(random.choice(letters) for _ in range(length))
+
+

 #  判断是否是azure服务
 def is_azure_url(url):
@@ -73,8 +86,8 @@ class ChatGPTNode:
    def INPUT_TYPES(cls):
        return {
            "required": {
-                "api_key":("KEY", {"default": "", "multiline": True}),
-                "api_url":("URL", {"default": "", "multiline": True}),
+                "api_key":("KEY", {"default": "", "multiline": True,"dynamicPrompts": False}),
+                "api_url":("URL", {"default": "", "multiline": True,"dynamicPrompts": False}),
                "prompt": ("STRING", {"multiline": True,"dynamicPrompts": False}),
                "system_content": ("STRING", 
                                   {
@@ -168,7 +181,10 @@ class ShowTextForGPT:
        return {
            "required": {
                "text": ("STRING", {"forceInput": True,"dynamicPrompts": False}),
-            }
+            },
+            "optional":{ 
+                    "output_dir": ("STRING",{"forceInput": True,"default": "","multiline": True,"dynamicPrompts": False}), 
+                }
        }

    INPUT_IS_LIST = True
@@ -179,7 +195,51 @@ class ShowTextForGPT:

    CATEGORY = "♾️Mixlab/GPT"

-    def run(self, text):
+    def run(self, text,output_dir=[""]):
+
+        if len(output_dir)==1 and (output_dir[0]=='' or os.path.dirname(output_dir[0])==''):
+            t='\n'.join(text)
+            output_dir=[
+                os.path.join(folder_paths.get_temp_directory(),
+                             get_unique_hash(t)+'.txt'
+                             )
+            ]
+        elif len(output_dir)==1:
+            base=os.path.basename(output_dir[0])
+            t='\n'.join(text)
+            if base=='' or os.path.splitext(base)[1]=='':
+                base=get_unique_hash(t)+'.txt'
+            output_dir=[
+                os.path.join(output_dir[0],
+                             base
+                             )
+            ]
+        # elif len(output_dir)>1:
+
+        
+
+        if len(output_dir)==1 and len(text)>1:
+            output_dir=[output_dir[0] for _ in range(len(text))]
+        
+        for i in range(len(text)):
+
+            o_fp=output_dir[i]
+            dirp=os.path.dirname(o_fp)
+            if dirp=='':
+                dirp=folder_paths.get_temp_directory()
+                o_fp=os.path.join(folder_paths.get_temp_directory(),o_fp
+                             )
+
+            if not os.path.exists(dirp):
+                os.mkdir(dirp)
+
+            if not os.path.splitext(o_fp)[1].lower()=='.txt':
+                o_fp=o_fp+'.txt'
+
+            t=text[i]
+            with open(o_fp, 'w') as file:
+                file.write(t)
+
        # print(text)
        return {"ui": {"text": text}, "result": (text,)}
        
@@ -24,6 +24,7 @@ def is_installed(package):
        return False
    return spec is not None

+
 try:
    if is_installed('clip_interrogator')==False:
        import subprocess
@@ -36,20 +37,25 @@ try:
        #检查命令执行结果
        if result.returncode == 0:
            print("#install success")
-            from transformers import AutoProcessor, BlipForConditionalGeneration
            from clip_interrogator import Config, Interrogator
            _available=True
        else:
            print("#install error")
        
    else:
-        from transformers import AutoProcessor, BlipForConditionalGeneration
        from clip_interrogator import Config, Interrogator
        _available=True

 except:
    _available=False

+try:
+    from transformers import AutoProcessor, BlipForConditionalGeneration
+except:
+    _available=False
+    print('pls check  transformers.__version__>=4.36.0:: AutoProcessor, BlipForConditionalGeneration')
+
+

 def load_caption_model(model_path,config,t='blip-base'):
    dtype=torch.float16 if config.device == 'cuda' else torch.float32
@@ -9,11 +9,22 @@ from io import BytesIO
 import folder_paths
 import json,io
 from comfy.cli_args import args
-import cv2 
-import math
+import cv2
+import string
+import math,glob
 from .Watcher import FolderWatcher

+def generate_random_string(length):
+    letters = string.ascii_letters + string.digits
+    return ''.join(random.choice(letters) for _ in range(length))

+class AnyType(str):
+  """A special class that is always equal in not equal comparisons. Credit to pythongosssss"""
+
+  def __ne__(self, __value: object) -> bool:
+    return False
+
+any_type = AnyType("*")


 FONT_PATH= os.path.abspath(os.path.join(os.path.dirname(__file__),'../assets/王汉宗颜楷体繁.ttf'))
@@ -129,10 +140,10 @@ def naive_cutout(img, mask,invert=True):
    image using the mask.
    """

-    img=img.convert("RGBA")
+    # img=img.convert("RGBA")
    mask=mask.convert("RGBA")

-    empty = Image.new("RGBA", (img.size), 0)
+    empty = Image.new("RGBA", (mask.size), 0)

    red, green, blue, alpha = mask.split()

@@ -369,6 +380,7 @@ def get_images_filepath(f,white_bg=False):
        for root, dirs, files in os.walk(f):
            for file in files:
                file_path = os.path.join(root, file)
+                file_name=os.path.basename(file_path)
                try:
                    imgs=load_image(file_path,white_bg)
                    for img in imgs:
@@ -376,6 +388,7 @@ def get_images_filepath(f,white_bg=False):
                            "image":img['image'],
                            "mask":img['mask'],
                            "file_path":file_path,
+                            "file_name":file_name,
                            "psd":len(imgs)>1
                        })
                except:
@@ -383,12 +396,15 @@ def get_images_filepath(f,white_bg=False):
 
    elif os.path.isfile(f):
        try:
+            file_path = os.path.join(root, f)
+            file_name=os.path.basename(file_path)
            imgs=load_image(f,white_bg)
            for img in imgs:
                images.append({
                    "image":img['image'],
                    "mask":img['mask'],
                    "file_path":file_path,
+                    "file_name":file_name,
                    "psd":len(imgs)>1
                })
        except:
@@ -847,79 +863,6 @@ class SmoothMask:



-class FeatheredMask:
-    @classmethod
-    def INPUT_TYPES(s):
-        return {
-                "required": {
-                                "mask": ("MASK",),
-                                "start_offset":("INT", {"default": 1, 
-                                                        "min": -150, 
-                                                        "max": 150, 
-                                                        "step": 1,
-                                                        "display": "slider"}),
-                                "feathering_weight":("FLOAT", {"default": 0.1,
-                                                                "min": 0.0,
-                                                                "max": 1,
-                                                                "step": 0.1,
-                                                                "display": "slider"})
-                            }
-            }
-    
-    RETURN_TYPES = ('MASK',)
-
-    FUNCTION = "run"
-
-    CATEGORY = "♾️Mixlab/Mask"
-
-    OUTPUT_IS_LIST = (False,)
-  
-    # 运行的函数
-    def run(self,mask,start_offset, feathering_weight):
-        # print(mask.shape,mask.size())
-        
-        image=tensor2pil(mask)
-
-        # Open the image using PIL
-        image = image.convert("L")
-        if start_offset>0:
-            image=ImageOps.invert(image)
-
-        # Convert the image to a numpy array
-        image_np = np.array(image)
-
-        # Use Canny edge detection to get black contours
-        edges = cv2.Canny(image_np, 30, 150)
-
-        for i in range(0,abs(start_offset)):
-            # int(100*feathering_weight)
-            a=int(abs(start_offset)*0.1*i)
-            # Dilate the black contours to make them wider
-            kernel = np.ones((a, a), np.uint8)
-
-            dilated_edges = cv2.dilate(edges, kernel, iterations=1)
-            # dilated_edges = cv2.erode(edges, kernel, iterations=1)
-            # Smooth the dilated edges using Gaussian blur
-            smoothed_edges = cv2.GaussianBlur(dilated_edges, (5, 5), 0)
-
-            # Adjust the feathering weight
-            feathering_weight = max(0, min(feathering_weight, 1))
-
-            # Blend the smoothed edges with the original image to achieve feathering effect
-            image_np = cv2.addWeighted(image_np, 1, smoothed_edges, feathering_weight, feathering_weight)
-
-        # Convert the result back to PIL image
-        result_image = Image.fromarray(np.uint8(image_np))
-        result_image=result_image.convert("L")
-
-        if start_offset>0:
-            result_image=ImageOps.invert(result_image)
-        
-        mask=pil2tensor(result_image)
-        # print(mask.shape,mask.size())
-        return mask
-
-


 class SplitLongMask:
@@ -1034,6 +977,7 @@ class TransparentImage:
        # ui.images 节点里显示图片，和 传参，image_path自定义的数据，需要写节点的自定义ui
        # result 里输出给下个节点的数据 
        # print('TransparentImage',len(images_rgb))
+        
        return {"ui":{"images": ui_images,"image_paths":image_paths},"result": (image_paths,images_rgb,images_rgba)}


@@ -1120,14 +1064,15 @@ class LoadImagesFromPath:
                }
            }
    
-    RETURN_TYPES = ('IMAGE','MASK','STRING',)
+    RETURN_TYPES = ('IMAGE','MASK','STRING','STRING',)
+    RETURN_NAMES = ("IMAGE","MASK","prompt_for_FloatingVideo","filepaths",)

    FUNCTION = "run"

    CATEGORY = "♾️Mixlab/Image"

    # INPUT_IS_LIST = True
-    OUTPUT_IS_LIST = (True,True,False,)
+    OUTPUT_IS_LIST = (True,True,False,True,)
  
    global watcher_folder
    watcher_folder=None
@@ -1163,10 +1108,12 @@ class LoadImagesFromPath:

        imgs=[]
        masks=[]
+        file_names=[]

        for im in sorted_files:
            imgs.append(im['image'])
            masks.append(im['mask'])
+            file_names.append(im['file_name'])
        
        # print('index_variable',index_variable)
        
@@ -1174,11 +1121,12 @@ class LoadImagesFromPath:
            if index_variable!=-1:
                imgs=[imgs[index_variable]] if index_variable < len(imgs) else None
                masks=[masks[index_variable]] if index_variable < len(masks) else None
+                file_names=[file_names[index_variable]] if index_variable < len(file_names) else None
        except Exception as e:
            print("发生了一个未知的错误：", str(e))

        # print('#prompt::::',prompt)
-        return  {"ui": {"seed": [1]}, "result":(imgs,masks,prompt,)}
+        return  {"ui": {"seed": [1]}, "result":(imgs,masks,prompt,file_names,)}


 # TODO 扩大选区的功能,重新输出mask
@@ -1209,7 +1157,12 @@ class ImageCropByAlpha:

        # print(RGBA)
        im=tensor2pil(RGBA)
-        im=naive_cutout(im,im)
+
+        # 要把im的alpha通道转为mask
+        im=im.convert('RGBA')
+        red, green, blue, alpha = im.split()
+
+        im=naive_cutout(bf_im,alpha)
        x, y, w, h=get_not_transparent_area(im)
        # print('#ForImageCrop:',w, h,x, y,)

@@ -1224,11 +1177,12 @@ class ImageCropByAlpha:
        height_1=h

        img = image[:,y:to_y, x:to_x, :]
-
+        # tensor2pil(img).save('test2.png')

        # 原图的mask
        ori=RGBA[:,y:to_y, x:to_x, :]
        ori=tensor2pil(ori)
+        # ori.save('test.png')

        # 创建一个新的图像对象，大小和模式与原始图像相同
        new_image = Image.new("RGBA", ori.size)
@@ -1249,7 +1203,7 @@ class ImageCropByAlpha:
                if a != 0:
                    new_pixel_data[x, y] = (255, 255, 255, 255)
                else:
-                    new_pixel_data[x, y] = (r, g, b, a)
+                    new_pixel_data[x, y] = (0,0,0,0)

        # 保存修改后的图像
        # new_image.save("output.png")
@@ -1323,6 +1277,9 @@ class LoadImagesFromURL:
        return {"required": { 
                    "url": ("STRING",{"multiline": True,"default": "https://","dynamicPrompts": False}),
                             },
+                "optional":{
+                    "seed": (any_type,  {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
+                }
                }
    
    RETURN_TYPES = ("IMAGE","MASK",)
@@ -1339,7 +1296,7 @@ class LoadImagesFromURL:
    global urls_image
    urls_image={}

-    def run(self,url):
+    def run(self,url,seed=0):
        global urls_image
        print(urls_image)
        def filter_http_urls(urls):
@@ -1671,6 +1628,201 @@ class NewLayer:

        return (layer_n,)

+
+
+def createMask(image,x,y,w,h):
+    mask = Image.new("L", image.size)
+    pixels = mask.load()
+    # 遍历指定区域的像素，将其设置为黑色（0 表示黑色）
+    for i in range(int(x), int(x + w)):
+        for j in range(int(y), int(y + h)):
+            pixels[i, j] = 255
+    # mask.save("mask.png")
+    return mask
+
+
+def splitImage(image, num):
+    width, height = image.size
+
+    num_rows = int(num ** 0.5)
+    num_cols = int(num / num_rows)
+    
+    grid_width = width // num_cols
+    grid_height = height // num_rows
+
+    grid_coordinates = []
+    for i in range(num_rows):
+        for j in range(num_cols):
+            x = j * grid_width
+            y = i * grid_height
+            grid_coordinates.append((x, y, grid_width, grid_height))
+
+    return grid_coordinates
+
+
+def centerImage(margin,canvas):
+    w,h=canvas.size
+
+    l,t,r,b=margin
+
+    x=l
+    y=t
+    width=w-r-l
+    height=h-t-b
+
+    return (x,y,width,height)
+
+# # 读取图片
+# image = Image.open("path_to_your_image.jpg")
+
+# # 定义要切割的区域数量
+# num = 9
+
+# # 切割图片
+# grid_coordinates = splitImage(image, num)
+
+# # 输出切割区域坐标
+# for i, coordinates in enumerate(grid_coordinates):
+#     print(f"Region {i + 1}: x={coordinates[0]}, y={coordinates[1]}, width={coordinates[2]}, height={coordinates[3]}")
+
+
+class SplitImage:
+    @classmethod
+    def INPUT_TYPES(s):
+        return { 
+            "required": { 
+                "image": ("IMAGE",), 
+                "num": ("INT",{
+                    "default": 4, 
+                    "min": 1, #Minimum value
+                    "max": 500, #Maximum value
+                    "step": 1, #Slider's step
+                    "display": "number" # Cosmetic only: display as "number" or "slider"
+                }),
+                "seed": ("INT",{
+                    "default": 4, 
+                    "min": 1, #Minimum value
+                    "max": 500, #Maximum value
+                    "step": 1, #Slider's step
+                    "display": "number" # Cosmetic only: display as "number" or "slider"
+                }),
+            }
+                }
+    
+    RETURN_TYPES = ("_GRID","_GRID","MASK",)
+    RETURN_NAMES = ("grids","grid","mask",)
+
+    FUNCTION = "run"
+
+    CATEGORY = "♾️Mixlab/Layer"
+
+    INPUT_IS_LIST = False
+    # OUTPUT_IS_LIST = (True,)
+
+    def run(self,image,num,seed):
+        image=tensor2pil(image)
+        
+        grids=splitImage(image,num)
+
+        if seed>num:
+            num=int(seed / 500 * num)-1
+        else:
+            num=seed-1
+
+        num=max(0,num) 
+
+        g=grids[num]
+
+        x,y,w,h=g
+        mask=createMask(image, x,y,w,h)
+        mask=pil2tensor(mask)
+
+        return (grids,g,mask,)
+
+
+
+class CenterImage:
+    @classmethod
+    def INPUT_TYPES(s):
+        return { 
+            "required": { 
+                "canvas": ("IMAGE",), 
+                "left": ("INT",{
+                    "default":24, 
+                    "min": 0, #Minimum value
+                    "max": 5000, #Maximum value
+                    "step": 1, #Slider's step
+                    "display": "number" # Cosmetic only: display as "number" or "slider"
+                }),
+               "top": ("INT",{
+                    "default":24, 
+                    "min": 0, #Minimum value
+                    "max": 5000, #Maximum value
+                    "step": 1, #Slider's step
+                    "display": "number" # Cosmetic only: display as "number" or "slider"
+                }),
+                "right": ("INT",{
+                    "default": 24, 
+                    "min": 0, #Minimum value
+                    "max": 5000, #Maximum value
+                    "step": 1, #Slider's step
+                    "display": "number" # Cosmetic only: display as "number" or "slider"
+                }),
+                 "bottom": ("INT",{
+                    "default": 24, 
+                    "min": 0, #Minimum value
+                    "max": 5000, #Maximum value
+                    "step": 1, #Slider's step
+                    "display": "number" # Cosmetic only: display as "number" or "slider"
+                }),
+            }
+                }
+    
+    RETURN_TYPES = ("_GRID","MASK",)
+    RETURN_NAMES = ("grid","mask",)
+
+    FUNCTION = "run"
+
+    CATEGORY = "♾️Mixlab/Layer"
+
+    INPUT_IS_LIST = False
+    # OUTPUT_IS_LIST = (True,)
+
+    def run(self,canvas,left,top,right,bottom):
+        canvas=tensor2pil(canvas)
+
+        grid=centerImage((left,top,right,bottom),canvas)
+
+        mask=createMask(canvas,left,top,canvas.width-left-right,canvas.height-top-bottom)
+
+        return (grid,pil2tensor(mask),)
+
+
+
+
+class GridOutput:
+    @classmethod
+    def INPUT_TYPES(s):
+        return { 
+            "required": {
+                "grid": ("_GRID",)
+            }
+                }
+    
+    RETURN_TYPES = ("INT","INT","INT","INT",)
+    RETURN_NAMES = ("x","y","width","height",)
+
+    FUNCTION = "run"
+
+    CATEGORY = "♾️Mixlab/Layer"
+
+    INPUT_IS_LIST = False
+    # OUTPUT_IS_LIST = (True,)
+
+    def run(self,grid):
+        x,y,w,h=grid
+        return (x,y,w,h,)
+
 class ShowLayer:
    @classmethod
    def INPUT_TYPES(s):
@@ -2007,14 +2159,14 @@ class ResizeImage:
                    "default": 512, 
                    "min": 1, #Minimum value
                    "max": 8192, #Maximum value
-                    "step": 1, #Slider's step
+                    "step": 8, #Slider's step
                    "display": "number" # Cosmetic only: display as "number" or "slider"
                }),
                "height": ("INT",{
                    "default": 512, 
                    "min": 1, #Minimum value
                    "max": 8192, #Maximum value
-                    "step": 1, #Slider's step
+                    "step": 8, #Slider's step
                    "display": "number" # Cosmetic only: display as "number" or "slider"
                }),
                "scale_option": (["width","height",'overall','center'],),
@@ -2025,20 +2177,21 @@ class ResizeImage:
                    "image": ("IMAGE",),
                    "average_color": (["on",'off'],),
                    "fill_color":("STRING",{"multiline": False,"default": "#FFFFFF","dynamicPrompts": False}),
+                    "mask": ("MASK",),
            }
                }
    
-    RETURN_TYPES = ("IMAGE","IMAGE","STRING",)
-    RETURN_NAMES = ("image","average_image","average_hex",)
+    RETURN_TYPES = ("IMAGE","IMAGE","STRING","MASK",)
+    RETURN_NAMES = ("image","average_image","average_hex","mask",)

    FUNCTION = "run"

    CATEGORY = "♾️Mixlab/Image"

    INPUT_IS_LIST = True
-    OUTPUT_IS_LIST = (True,True,True,)
+    OUTPUT_IS_LIST = (True,True,True,True,)

-    def run(self,width,height,scale_option,image=None,average_color=['on'],fill_color=["#FFFFFF"]):
+    def run(self,width,height,scale_option,image=None,average_color=['on'],fill_color=["#FFFFFF"],mask=None):
        
        w=width[0]
        h=height[0]
@@ -2047,6 +2200,7 @@ class ResizeImage:
        fill_color=fill_color[0]

        imgs=[]
+        masks=[]
        average_images=[]
        hexs=[]

@@ -2075,8 +2229,20 @@ class ResizeImage:
                    a_im=pil2tensor(a_im)
                    average_images.append(a_im)
                    hexs.append(hex)
+
+            try:
+                for mas in mask:
+                    for ma in mas:
+                        ma=tensor2pil(ma)
+                        ma=ma.convert('RGB')
+                        ma=resize_image(ma,scale_option,w,h,fill_color)
+                        ma=ma.convert('L')
+                        ma=pil2tensor(ma)
+                        masks.append(ma)
+            except:
+                print('')
        
-        return (imgs,average_images,hexs,)
+        return (imgs,average_images,hexs,masks,)


 class MirroredImage:
@@ -2120,19 +2286,41 @@ class GetImageSize_:
        return {
            "required": {
                "image": ("IMAGE",),
-            }
+            },
+             "optional":{
+                    "min_width":("INT", {
+                        "default": 512, 
+                        "min":1, #Minimum value
+                        "max": 2048, #Maximum value
+                        "step": 8, #Slider's step
+                        "display": "number" # Cosmetic only: display as "number" or "slider"
+                    })
+                },
        }

-    RETURN_TYPES = ("INT", "INT")
-    RETURN_NAMES = ("width", "height")
+    RETURN_TYPES = ("INT", "INT","INT", "INT",)
+    RETURN_NAMES = ("width", "height","min_width", "min_height",)

    FUNCTION = "get_size"

    CATEGORY = "♾️Mixlab/Image"

-    def get_size(self, image):
+    def get_size(self, image,min_width):
        _, height, width, _ = image.shape
-        return (width, height)
+        
+        # 如果比min_widht,还小，则输出 min width
+        if min_width>width:
+            im=tensor2pil(image)
+            im=resize_image(im,'width',min_width,min_width,"white")
+            im=im.convert('RGB')
+
+            min_width,min_height=im.size
+
+        else:
+            min_width=width
+            min_height=height
+
+        return (width, height,min_width,min_height,)



@@ -2179,3 +2367,104 @@ class ImageColorTransfer:



+class SaveImageToLocal:
+    def __init__(self):
+        self.output_dir = folder_paths.get_output_directory()
+        self.type = "output"
+        self.compress_level = 4
+
+    @classmethod
+    def INPUT_TYPES(s):
+        return {"required": 
+                    {"images": ("IMAGE", ),
+                     "file_path": ("STRING",{"multiline": True,"default": "","dynamicPrompts": False}),
+                     },
+                "hidden": {"prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO"},
+                
+                }
+
+    RETURN_TYPES = ()
+    FUNCTION = "save_images"
+
+    OUTPUT_NODE = True
+
+    CATEGORY = "♾️Mixlab/Image"
+
+    def save_images(self, images,file_path , prompt=None, extra_pnginfo=None):
+        filename_prefix = os.path.basename(file_path)
+        if file_path=='':
+            filename_prefix="ComfyUI"
+        
+        filename_prefix, _ = os.path.splitext(filename_prefix)
+
+        _, extension = os.path.splitext(file_path)
+
+        if extension:
+            # 是文件名，需要处理
+            file_path=os.path.dirname(file_path)
+            # filename_prefix=
+
+            
+        full_output_folder, filename, counter, subfolder, filename_prefix = folder_paths.get_save_image_path(filename_prefix, self.output_dir, images[0].shape[1], images[0].shape[0])
+        
+    
+        if not os.path.exists(file_path) and not extension:
+            # 使用os.makedirs函数创建新目录
+            os.makedirs(file_path)
+            print("目录已创建")
+        else:
+            print("目录已存在")
+
+        # 使用glob模块获取当前目录下的所有文件
+        if file_path=="":
+            files = glob.glob(full_output_folder + '/*')
+        else:
+            files = glob.glob(file_path + '/*')
+        # 统计文件数量
+        file_count = len(files)
+        counter+=file_count
+        print('统计文件数量',file_count,counter)
+
+        results = list()
+        for image in images:
+            i = 255. * image.cpu().numpy()
+            img = Image.fromarray(np.clip(i, 0, 255).astype(np.uint8))
+            metadata = None
+            if not args.disable_metadata:
+                metadata = PngInfo()
+                if prompt is not None:
+                    metadata.add_text("prompt", json.dumps(prompt))
+                if extra_pnginfo is not None:
+                    for x in extra_pnginfo:
+                        metadata.add_text(x, json.dumps(extra_pnginfo[x]))
+
+            file = f"{filename}_{counter:05}_.png"
+            
+            if file_path=="":
+                fp=os.path.join(full_output_folder, file)
+                if os.path.exists(fp):
+                    file = f"{filename}_{counter:05}_{generate_random_string(8)}.png"
+                    fp=os.path.join(full_output_folder, file)
+                img.save(fp, pnginfo=metadata, compress_level=self.compress_level)
+                results.append({
+                    "filename": file,
+                    "subfolder": subfolder,
+                    "type": self.type
+                })
+            
+            else:
+
+                fp=os.path.join(file_path, file)
+                if os.path.exists(fp):
+                    file = f"{filename}_{counter:05}_{generate_random_string(8)}.png"
+                    fp=os.path.join(file_path, file)
+
+                img.save(os.path.join(file_path, file), pnginfo=metadata, compress_level=self.compress_level)
+                results.append({
+                    "filename": file,
+                    "subfolder": file_path,
+                    "type": self.type
+                })
+            counter += 1
+
+        return ()
@@ -1,10 +1,26 @@
-import numpy as np
+ 
 import scipy.ndimage
 import torch
-import comfy.utils

 from nodes import MAX_RESOLUTION

+import numpy as np 
+# from PIL import Image, ImageDraw
+from PIL import Image, ImageOps 
+
+from comfy.cli_args import args
+import cv2
+
+
+
+# Tensor to PIL
+def tensor2pil(image):
+    return Image.fromarray(np.clip(255. * image.cpu().numpy().squeeze(), 0, 255).astype(np.uint8))
+
+# Convert PIL to Tensor
+def pil2tensor(image):
+    return torch.from_numpy(np.array(image).astype(np.float32) / 255.0).unsqueeze(0)
+

 def grow(mask, expand, tapered_corners):
    c = 0 if tapered_corners else 1
@@ -68,4 +84,88 @@ class OutlineMask:

        m3=combine(m1,m2,0,0)

-        return (m3,)
+        return (m3,)
+    
+
+
+
+class FeatheredMask:
+    @classmethod
+    def INPUT_TYPES(s):
+        return {
+                "required": {
+                                "mask": ("MASK",),
+                                "start_offset":("INT", {"default": 1, 
+                                                        "min": -150, 
+                                                        "max": 150, 
+                                                        "step": 1,
+                                                        "display": "slider"}),
+                                "feathering_weight":("FLOAT", {"default": 0.1,
+                                                                "min": 0.0,
+                                                                "max": 1,
+                                                                "step": 0.1,
+                                                                "display": "slider"})
+                            }
+            }
+    
+    RETURN_TYPES = ('MASK',)
+
+    FUNCTION = "run"
+
+    CATEGORY = "♾️Mixlab/Mask"
+
+    OUTPUT_IS_LIST = (True,)
+  
+    # 运行的函数
+    def run(self,mask,start_offset, feathering_weight):
+        # print(mask.shape,mask.size())
+        
+        num,_,_=mask.size()
+
+        masks=[]
+
+        for i in range(num):
+            mm=mask[i]
+            image=tensor2pil(mm)
+
+            # Open the image using PIL
+            image = image.convert("L")
+            if start_offset>0:
+                image=ImageOps.invert(image)
+
+            # Convert the image to a numpy array
+            image_np = np.array(image)
+
+            # Use Canny edge detection to get black contours
+            edges = cv2.Canny(image_np, 30, 150)
+
+            for i in range(0,abs(start_offset)):
+                # int(100*feathering_weight)
+                a=int(abs(start_offset)*0.1*i)
+                # Dilate the black contours to make them wider
+                kernel = np.ones((a, a), np.uint8)
+
+                dilated_edges = cv2.dilate(edges, kernel, iterations=1)
+                # dilated_edges = cv2.erode(edges, kernel, iterations=1)
+                # Smooth the dilated edges using Gaussian blur
+                smoothed_edges = cv2.GaussianBlur(dilated_edges, (5, 5), 0)
+
+                # Adjust the feathering weight
+                feathering_weight = max(0, min(feathering_weight, 1))
+
+                # Blend the smoothed edges with the original image to achieve feathering effect
+                image_np = cv2.addWeighted(image_np, 1, smoothed_edges, feathering_weight, feathering_weight)
+
+            # Convert the result back to PIL image
+            result_image = Image.fromarray(np.uint8(image_np))
+            result_image=result_image.convert("L")
+
+            if start_offset>0:
+                result_image=ImageOps.invert(result_image)
+            
+            result_image=result_image.convert("L")
+            mt=pil2tensor(result_image)
+            masks.append(mt)
+         
+            # print( mt.size())
+        return (masks,)
@@ -23,6 +23,19 @@ def get_files_with_extension(directory, extension):
                file_list.append(file_name)
    return file_list

+def join_with_(text_list,delimiter):
+    joined_text = delimiter.join(text_list)
+    return joined_text
+
+
+class AnyType(str):
+  """A special class that is always equal in not equal comparisons. Credit to pythongosssss"""
+
+  def __ne__(self, __value: object) -> bool:
+    return False
+
+any_type = AnyType("*")
+

 default_prompt1='''Swing
                                                Slide
@@ -65,7 +78,7 @@ def addWeight(text, weight=1):
    if weight == 1:
        return text
    else:
-        return f"({text}:{round(weight,2)})"
+        return f"({text}:{round(weight,3)})"

 def prompt_delete_words(sentence, new_words_length):
    # 使用逗号分割句子，并去除空格
@@ -308,6 +321,10 @@ class RandomPrompt:
                            "default": 'sticker, Cartoon, ``'
                          }),
                "random_sample": (["enable", "disable"],),
+                # "seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff, "step": 1}),
+                },
+            "optional":{
+                    "seed": (any_type,  {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
                }
            }
    
@@ -324,7 +341,7 @@ class RandomPrompt:


    # 运行的函数
-    def run(self,max_count,mutable_prompt,immutable_prompt,random_sample):
+    def run(self,max_count,mutable_prompt,immutable_prompt,random_sample,seed=0):
        # print('#运行的函数',mutable_prompt,immutable_prompt,max_count,random_sample)
        
        # Split the text into an array of words
@@ -394,6 +411,7 @@ class EmbeddingPrompt:

    # 运行的函数
    def run(self,embedding,weight):
+        weight = round(weight, 3)
        prompt='embedding:'+embedding
        if weight!=1:
            prompt='('+prompt+':'+str(weight)+')'
@@ -401,3 +419,33 @@ class EmbeddingPrompt:
        # return (new_prompt)
        return (prompt,)

+RETURN_TYPES = (any_type,)
+
+class JoinWithDelimiter:
+    @classmethod
+    def INPUT_TYPES(s):
+        return {"required": {
+                    "text_list": (any_type,),
+                    "delimiter":(["newline","comma"],),
+                             },
+                }
+    
+    RETURN_TYPES = ("STRING",) 
+
+    FUNCTION = "run"
+
+    CATEGORY = "♾️Mixlab/Prompt"
+
+    INPUT_IS_LIST = True # 当true的时候，输入时list，当false的时候，如果输入是list，则会自动包一层for循环调用
+    OUTPUT_IS_LIST = (False,)
+
+    def run(self,text_list,delimiter):
+        delimiter=delimiter[0]
+        if delimiter =='newline':
+            delimiter='\n'
+        elif delimiter=='comma':
+            delimiter=','
+        t=''
+        if isinstance(text_list, list):
+            t=join_with_(text_list,delimiter)
+        return (t,)
@@ -7,6 +7,8 @@ import folder_paths
 import matplotlib.font_manager as fm
 import torch

+
+
 def recursive_search(directory, excluded_dir_names=None):
    if not os.path.isdir(directory):
        return [], {}
@@ -198,14 +200,17 @@ class TextToNumber:
        return {"required": { 
                    "text": ("STRING",{"multiline": False,"default": "1"}),
                    "random_number": (["enable", "disable"],),
-                    "number":("INT", {
-                        "default": 0, 
-                        "min": 0, #Minimum value
+                    "max_num":("INT", {
+                        "default": 10, 
+                        "min":2, #Minimum value
                        "max": 10000000000, #Maximum value
                        "step": 1, #Slider's step
                        "display": "number" # Cosmetic only: display as "number" or "slider"
                    }),
                             },
+                 "optional":{
+                    "seed": (any_type,  {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
+                }
                }
    
    RETURN_TYPES = ("INT",)
@@ -218,7 +223,7 @@ class TextToNumber:
    INPUT_IS_LIST = False
    OUTPUT_IS_LIST = (False,)

-    def run(self,text,random_number,number):
+    def run(self,text,random_number,max_num,seed=0):
        
        numbers = re.findall(r'\d+', text)
        result=0
@@ -227,7 +232,7 @@ class TextToNumber:
            # print(result)
        
        if random_number=='enable' and result>0:
-            result= random.randint(1, 10000000000)
+            result= random.randint(1, max_num)
        return {"ui": {"text": [text],"num":[result]}, "result": (result,)}
    

@@ -341,11 +346,18 @@ class MultiplicationNode:
    def INPUT_TYPES(s):
        return {"required": {
                    "numberA":(any_type,),
-                    "numberB":("FLOAT", {
+                    "multiply_by":("FLOAT", {
                        "default": 0, 
-                        "min": -1, #Minimum value
+                        "min": -2, #Minimum value
                        "max": 0xffffffffffffffff,
-                        "step": 0.1, #Slider's step
+                        "step": 0.01, #Slider's step
+                        "display": "number" # Cosmetic only: display as "number" or "slider"
+                    }),
+                     "add_by":("FLOAT", {
+                        "default": 0, 
+                        "min": -2000, #Minimum value
+                        "max": 0xffffffffffffffff,
+                        "step": 0.01, #Slider's step
                        "display": "number" # Cosmetic only: display as "number" or "slider"
                    })
                             },
@@ -360,9 +372,9 @@ class MultiplicationNode:
    INPUT_IS_LIST = False
    OUTPUT_IS_LIST = (False,False,)

-    def run(self,numberA,numberB):
-        b=int(numberA*numberB)
-        a=float(numberA*numberB)
+    def run(self,numberA,multiply_by,add_by):
+        b=int(numberA*multiply_by+add_by)
+        a=float(numberA*multiply_by+add_by)
        return (a,b,) 

 class TextInput:
@@ -418,7 +430,7 @@ class DynamicDelayProcessor:
            },
            "optional":{
                "any_input":(any_type,),
-                "delay_by_text":("STRING",{"multiline":True,}),
+                "delay_by_text":("STRING",{"multiline":True,"dynamicPrompts": False,}),
                "words_per_seconds":("FLOAT",{ "default":1.50,"min": 0.0,"max": 1000.00,"display":"Chars per second?"}),
                "replace_output": (["disable","enable"],),
                "replace_value":("INT",{ "default":-1,"min": 0,"max": 1000000,"display":"Replacement value"})
@@ -705,3 +717,117 @@ class TESTNODE_:
        result = list_stats.count_types(ANY)
            
        return {"ui": {"data": result,"type":[str(type(ANY[0]))]}, "result": (ANY,)}
+
+
+
+class CreateSeedNode:
+    def __init__(self):
+        pass
+
+    @classmethod
+    def INPUT_TYPES(cls):
+        return {
+            "required": {
+                "seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
+            }
+        }
+
+    RETURN_TYPES = ("INT",)
+    RETURN_NAMES = ("seed",)
+  
+    OUTPUT_NODE = True
+    FUNCTION = "run"
+
+    CATEGORY = "♾️Mixlab/Utils"
+
+    def run(self, seed):
+        return (seed,)
+    
+
+class CreateCkptNames:
+    def __init__(self):
+        pass
+
+    @classmethod
+    def INPUT_TYPES(cls):
+        return {
+            "required": {
+                "ckpt_names": ("STRING",{"multiline": True,"default": "\n".join(folder_paths.get_filename_list("checkpoints")),"dynamicPrompts": False}),
+            }
+        }
+
+    RETURN_TYPES = (any_type,)
+    RETURN_NAMES = ("ckpt_names",)
+    
+    INPUT_IS_LIST = False
+    OUTPUT_IS_LIST = (True,)
+
+    # OUTPUT_NODE = True
+    FUNCTION = "run"
+
+    CATEGORY = "♾️Mixlab/Utils"
+
+    def run(self, ckpt_names):
+        ckpt_names=ckpt_names.split('\n')
+        ckpt_names = [name for name in ckpt_names if name.strip()]
+        return (ckpt_names,)
+
+
+class CreateLoraNames:
+    def __init__(self):
+        pass
+
+    @classmethod
+    def INPUT_TYPES(cls):
+        return {
+            "required": {
+                "lora_names": ("STRING",{"multiline": True,"default": "\n".join(folder_paths.get_filename_list("loras")),"dynamicPrompts": False}),
+            }
+        }
+
+    RETURN_TYPES = (any_type,"STRING",)
+    RETURN_NAMES = ("lora_names","prompt",)
+    
+    INPUT_IS_LIST = False
+    OUTPUT_IS_LIST = (True,True,)
+
+    # OUTPUT_NODE = True
+    FUNCTION = "run"
+
+    CATEGORY = "♾️Mixlab/Utils"
+
+    def run(self, lora_names):
+        lora_names=lora_names.split('\n')
+        lora_names = [name for name in lora_names if name.strip()]
+        prompts=[os.path.splitext(n)[0] for n in lora_names]
+        return (lora_names,prompts,)
+   
+
+
+class CreateSampler_names:
+    def __init__(self):
+        pass
+
+    @classmethod
+    def INPUT_TYPES(cls):
+        return {
+            "required": {
+                "sampler_names": ("STRING",{"multiline": True,"default": "\n".join(comfy.samplers.KSampler.SAMPLERS),"dynamicPrompts": False}),
+            }
+        }
+
+    RETURN_TYPES = (any_type,)
+    RETURN_NAMES = ("sampler_names",)
+    
+    INPUT_IS_LIST = False
+    OUTPUT_IS_LIST = (True,)
+
+    # OUTPUT_NODE = True
+    FUNCTION = "run"
+
+    CATEGORY = "♾️Mixlab/Utils"
+
+    def run(self, sampler_names):
+        sampler_names=sampler_names.split('\n')
+        sampler_names = [name for name in sampler_names if name.strip()]
+        return (sampler_names,)
@@ -1,179 +0,0 @@
-# https://github.com/openai/consistencydecoder/blob/main/consistencydecoder/__init__.py
-
-import folder_paths
-from comfy import model_management
-
-import math
-import torch
-import numpy as np
-from PIL import Image 
-
-
-class ConsistencyDecoderWrapper:
-    def __init__(self, decoder):
-        self.decoder = decoder
-    def decode(self, x):
-        return self.decoder(x)
-
-def _extract_into_tensor(arr, timesteps, broadcast_shape):
-    
-    # from: https://github.com/openai/guided-diffusion/blob/22e0df8183507e13a7813f8d38d51b072ca1e67c/guided_diffusion/gaussian_diffusion.py#L895    """
-    res = arr[timesteps].float()
-    dims_to_append = len(broadcast_shape) - len(res.shape)
-    return res[(...,) + (None,) * dims_to_append]
-
-
-def betas_for_alpha_bar(num_diffusion_timesteps, alpha_bar, max_beta=0.999):
-    # from: https://github.com/openai/guided-diffusion/blob/22e0df8183507e13a7813f8d38d51b072ca1e67c/guided_diffusion/gaussian_diffusion.py#L45
-    betas = []
-    for i in range(num_diffusion_timesteps):
-        t1 = i / num_diffusion_timesteps
-        t2 = (i + 1) / num_diffusion_timesteps
-        betas.append(min(1 - alpha_bar(t2) / alpha_bar(t1), max_beta))
-    return torch.tensor(betas)
-
-class ConsistencyDecoder:
-    def __init__(self, device="cuda:0", download_target=""):
-        self.n_distilled_steps = 64
-        # download_target = _download("https://openaipublic.azureedge.net/diff-vae/c9cebd3132dd9c42936d803e33424145a748843c8f716c0814838bdc8a2fe7cb/decoder.pt", download_root)
-        self.ckpt = torch.jit.load(download_target).to(device)
-        self.device = device 
-        sigma_data = 0.5
-        betas = betas_for_alpha_bar(
-            1024, lambda t: math.cos((t + 0.008) / 1.008 * math.pi / 2) ** 2
-        ).to(device)
-        alphas = 1.0 - betas
-        alphas_cumprod = torch.cumprod(alphas, dim=0)
-        self.sqrt_alphas_cumprod = torch.sqrt(alphas_cumprod)
-        self.sqrt_one_minus_alphas_cumprod = torch.sqrt(1.0 - alphas_cumprod)
-        sqrt_recip_alphas_cumprod = torch.sqrt(1.0 / alphas_cumprod)
-        sigmas = torch.sqrt(1.0 / alphas_cumprod - 1)
-        self.c_skip = (
-            sqrt_recip_alphas_cumprod
-            * sigma_data**2
-            / (sigmas**2 + sigma_data**2)
-        )
-        self.c_out = sigmas * sigma_data / (sigmas**2 + sigma_data**2) ** 0.5
-        self.c_in = sqrt_recip_alphas_cumprod / (sigmas**2 + sigma_data**2) ** 0.5
-
-    @staticmethod
-    def round_timesteps(
-        timesteps, total_timesteps, n_distilled_steps, truncate_start=True
-    ):
-        with torch.no_grad():
-            space = torch.div(total_timesteps, n_distilled_steps, rounding_mode="floor")
-            rounded_timesteps = (
-                torch.div(timesteps, space, rounding_mode="floor") + 1
-            ) * space
-            if truncate_start:
-                rounded_timesteps[rounded_timesteps == total_timesteps] -= space
-            else:
-                rounded_timesteps[rounded_timesteps == total_timesteps] -= space
-                rounded_timesteps[rounded_timesteps == 0] += space
-            return rounded_timesteps
-
-    @staticmethod
-    def ldm_transform_latent(z, extra_scale_factor=1):
-        channel_means = [0.38862467, 0.02253063, 0.07381133, -0.0171294]
-        channel_stds = [0.9654121, 1.0440036, 0.76147926, 0.77022034]
-
-        if len(z.shape) != 4:
-            raise ValueError()
-
-        z = z * 0.18215
-        channels = [z[:, i] for i in range(z.shape[1])]
-
-        channels = [
-            extra_scale_factor * (c - channel_means[i]) / channel_stds[i]
-            for i, c in enumerate(channels)
-        ]
-        return torch.stack(channels, dim=1)
-
-    @torch.no_grad()
-    def __call__(
-        self,
-        features: torch.Tensor,
-        schedule=[1.0, 0.5],
-    ):
-        features = self.ldm_transform_latent(features)
-  
-        ts = self.round_timesteps(
-            torch.arange(0, 1024),
-            1024,
-            self.n_distilled_steps,
-            truncate_start=False,
-        )
-        shape = (
-            features.size(0),
-            3,
-            8 * features.size(2),
-            8 * features.size(3),
-        )
-        
-        x_start = torch.zeros(shape, device=features.device, dtype=features.dtype)
-        schedule_timesteps = [int((1024 - 1) * s) for s in schedule]
-        for i in schedule_timesteps:
-            t = ts[i].item()
-            t_ = torch.tensor([t] * features.shape[0]).to(self.device)
-            noise = torch.randn_like(x_start)
-            
-            x_start = (
-                _extract_into_tensor(self.sqrt_alphas_cumprod, t_, x_start.shape)
-                * x_start
-                + _extract_into_tensor(
-                    self.sqrt_one_minus_alphas_cumprod, t_, x_start.shape
-                )
-                * noise
-            )
-            c_in = _extract_into_tensor(self.c_in, t_, x_start.shape)
-            model_output = self.ckpt(c_in * x_start, t_, features=features)
-            B, C = x_start.shape[:2]
-            model_output, _ = torch.split(model_output, C, dim=1)
-            pred_xstart = (
-                _extract_into_tensor(self.c_out, t_, x_start.shape) * model_output
-                + _extract_into_tensor(self.c_skip, t_, x_start.shape) * x_start
-            ).clamp(-1, 1)
-            x_start = pred_xstart
-        return x_start
-
-
-
-class VAELoader:
-    @classmethod
-    def INPUT_TYPES(s):
-        return {"required": { "vae_name": (folder_paths.get_filename_list("vae"), )}}
-    RETURN_TYPES = ("VAE",)
-    FUNCTION = "load_vae"
-
-    CATEGORY = "♾️Mixlab/__TEST"
-
-    #TODO: scale factor?
-    def load_vae(self, vae_name):
-        vae_path = folder_paths.get_full_path("vae", vae_name)
-        device = 'cuda:0'
-        # print('device',device)
-        consistencyDecoder = ConsistencyDecoder(device=device,
-                                 download_target=vae_path) # Model size: 2.49 GB
-        vae = ConsistencyDecoderWrapper(consistencyDecoder)
-        return (vae,)
-    
-
-class VAEDecode:
-    @classmethod
-    def INPUT_TYPES(s):
-        return {"required": { "samples": ("LATENT", ), "vae": ("VAE", )}}
-    RETURN_TYPES = ("IMAGE",)
-    FUNCTION = "decode"
-
-    CATEGORY = "♾️Mixlab/__TEST"
-
-    def decode(self, vae, samples):
-        image = vae.decode(samples["samples"].to("cuda:0"))
-        image = image[0].cpu().numpy()
-        image = (image + 1.0) * 127.5
-        image = image.clip(0, 255).astype(np.uint8)
-        image = Image.fromarray(image.transpose(1, 2, 0))
-        image = image.convert("RGB")
-        image = np.array(image).astype(np.float32) / 255.0
-        image = torch.from_numpy(image)[None,]
-        return (image, )
@@ -4,5 +4,6 @@ watchdog
 opencv-python-headless
 matplotlib
 openai
-# simple-lama-inpainting
-# clip-interrogator==0.6.0
+simple-lama-inpainting
+clip-interrogator==0.6.0
+transformers>=4.36.0
@@ -391,14 +391,14 @@
            <a class="link" href="https://www.mixcomfy.com" target="_blank">ComfyUI中文爱好者社区推荐</a>
        </div>

-        <a target="_blank" href="https://github.com/shadowcz007/comfyui-mixlab-nodes" style="text-decoration: none;
+        <a target="_blank" href="https://www.mixcomfy.com/blog/" style="text-decoration: none;
        color: black;font-size:12px">
            <svg height="32" aria-hidden="true" viewBox="0 0 16 16" version="1.1" width="32" data-view-component="true"
                class="octicon octicon-mark-github v-align-middle color-fg-default">
                <path
                    d="M8 0c4.42 0 8 3.58 8 8a8.013 8.013 0 0 1-5.45 7.59c-.4.08-.55-.17-.55-.38 0-.27.01-1.13.01-2.2 0-.75-.25-1.23-.54-1.48 1.78-.2 3.65-.88 3.65-3.95 0-.88-.31-1.59-.82-2.15.08-.2.36-1.02-.08-2.12 0 0-.67-.22-2.2.82-.64-.18-1.32-.27-2-.27-.68 0-1.36.09-2 .27-1.53-1.03-2.2-.82-2.2-.82-.44 1.1-.16 1.92-.08 2.12-.51.56-.82 1.28-.82 2.15 0 3.06 1.86 3.75 3.64 3.95-.23.2-.44.55-.51 1.07-.46.21-1.61.55-2.33-.66-.15-.24-.6-.83-1.23-.82-.67.01-.27.38.01.53.34.19.73.9.82 1.13.16.45.68 1.31 2.69.94 0 .67.01 1.3.01 1.49 0 .21-.15.45-.55.38A7.995 7.995 0 0 1 0 8c0-4.42 3.58-8 8-8Z">
                </path>
-            </svg> Code
+            </svg> Community

        </a>
    </div>
@@ -643,7 +643,7 @@
            return true
        }

-
+        // 种子的处理
        function randomSeed(seed, data) {
            for (const id in data) {
                if (data[id].inputs.seed != undefined
@@ -1316,6 +1316,7 @@
                        }
                    }), value);

+                    // 选择事件绑定
                    selectDom.addEventListener('change', e => {
                        e.preventDefault();
                        // console.log(selectDom.value)
@@ -1550,6 +1551,7 @@

        }

+        // 创建下拉选择
        function createSelect(options, defaultValue) {
            var selectElement = document.createElement("select");
            selectElement.className = "select"
@@ -1558,7 +1560,7 @@
            for (var i = 0; i < options.length; i++) {
                var option = document.createElement("option");
                option.value = options[i].value;
-                option.text = options[i].text;
+                option.innerText = options[i].text;
                selectElement.appendChild(option);
                // if(options[i].selected) 
            }
@@ -1569,6 +1571,7 @@
            return selectElement
        }

+        // 创建下拉选择 - 带说明
        function createSelectWithOptions(title, options, defaultValue) {

            const div = document.createElement("div");
@@ -67,9 +67,13 @@ async function drawImageToCanvas (imageUrl) {
 }

 function extractInputAndOutputData (jsonData, inputIds = [], outputIds = []) {
-  const data = jsonData
-  const input = []
-  const output = []
+  // workflow
+  // const workflow=jsonData.workflow;
+  // const nodes=workflow.nodes;
+
+  const data = jsonData.output
+  let input = []
+  let output = []
  const seed = {}

  for (const id in data) {
@@ -77,7 +81,7 @@ function extractInputAndOutputData (jsonData, inputIds = [], outputIds = []) {
      let node = app.graph.getNodeById(id)
      if (inputIds.includes(id)) {
        // let node = app.graph.getNodeById(id)
-        let options = []
+        let options = {}
        // 模型
        try {
          if (node.type === 'CheckpointLoaderSimple') {
@@ -113,6 +117,15 @@ function extractInputAndOutputData (jsonData, inputIds = [], outputIds = []) {
        if (node.type == 'Color') {
        }

+        // loadImage的mask支持
+        if (node.type === 'LoadImage') {
+          let output = node.outputs.filter(ot => ot.type == 'MASK')[0]
+          if (output.links) {
+            // 有输出
+            options.hasMask = true
+          }
+        }
+
        input[inputIds.indexOf(id)] = {
          ...data[id],
          title: node.title,
@@ -138,6 +151,10 @@ function extractInputAndOutputData (jsonData, inputIds = [], outputIds = []) {
    }
  }

+  // 修复bug，当节点不存在时
+  input = input.filter(i => i)
+  output = output.filter(i => i)
+
  return { input, output, seed }
 }

@@ -211,8 +228,8 @@ async function save (json, download = false, showInfo = true) {
  try {
    let data = await app.graphToPrompt()

-    const { input, output, seed } = extractInputAndOutputData(
-      data.output,
+    let { input, output, seed } = extractInputAndOutputData(
+      data,
      inputIds,
      outputIds
    )
@@ -399,7 +416,7 @@ app.registerExtension({
    }
  },
  async loadedGraphNode (node, app) {
-    console.log('#loadedGraphNode1111')
+    // console.log('#loadedGraphNode1111')
    window._mixlab_app_json = null //切换workflow需要清空
    if (node.type === 'AppInfo') {
      let auto_save = node.widgets.filter(w => w.name == 'auto_save')[0]
@@ -408,6 +425,9 @@ app.registerExtension({
          auto_save.value = 'enable'
        }
      }
+
+      // app.canvas.centerOnNode(node)
+      // app.canvas.setZoom(0.45)
    }
  }
 })
@@ -3,7 +3,7 @@ import { app } from '../../../scripts/app.js'
 const repoOwner = 'shadowcz007' // 替换为仓库的所有者
 const repoName = 'comfyui-mixlab-nodes' // 替换为仓库的名称

-const version = 'v0.13.0'
+const version = 'v0.15.1'

 fetch(`https://api.github.com/repos/${repoOwner}/${repoName}/releases/latest`)
  .then(response => response.json())
@@ -68,7 +68,7 @@ app.registerExtension({
          size: [128, 32], // a default size
          draw (ctx, node, width, y) {},
          computeSize (...args) {
-            return [128,32] // a method to compute the current size of the widget
+            return [128, 32] // a method to compute the current size of the widget
          },
          async serializeValue (nodeId, widgetIndex) {
            let data = getLocalData('_mixlab_api_key')
@@ -203,76 +203,82 @@ app.registerExtension({

 app.registerExtension({
  name: 'Mixlab.GPT.ShowTextForGPT',
-  async beforeRegisterNodeDef(nodeType, nodeData, app) {
-		if (nodeData.name === "ShowTextForGPT") {
-			function populate(text) {
-				if (this.widgets) {
-          
-					const pos = this.widgets.findIndex((w) => w.name === "text");
-					if (pos !== -1) {
-						for (let i = pos; i < this.widgets.length; i++) {
-							this.widgets[i].onRemove?.();
-						}
-						this.widgets.length = pos;
-					}
-				}
-        // console.log('ShowTextForGPT',text)
-				for (let list of text) {
-					const w = ComfyWidgets["STRING"](this, "text", ["STRING", { multiline: true }], app).widget;
-					w.inputEl.readOnly = true;
-					w.inputEl.style.opacity = 0.6;
+  async beforeRegisterNodeDef (nodeType, nodeData, app) {
+    if (nodeData.name === 'ShowTextForGPT') {
+      function populate (text) {
+        text = text.filter(t => t && t?.trim())

-          try {
-            let data=JSON.parse(list);
-            data=Array.from(data,d=>{
-              return {
-                ...d,
-                content:decodeURIComponent(d.content)
-              }
-            })
-            list=JSON.stringify(data,null,2)
-          } catch (error) {
-            // console.log(error)
+        if (this.widgets) {
+          // const pos = this.widgets.findIndex(w => w.name === 'text')
+          for (let i = 0; i < this.widgets.length; i++) {
+            if (this.widgets[i].name == 'show_text') this.widgets[i].onRemove?.()
          }
+          this.widgets.length = 1
+        }
+        // console.log('ShowTextForGPT',text)
+        for (let list of text) {
+          if (list) {
+            // console.log('#####', list)
+            const w = ComfyWidgets['STRING'](
+              this,
+              'show_text',
+              ['STRING', { multiline: true }],
+              app
+            ).widget
+            w.inputEl.readOnly = true
+            w.inputEl.style.opacity = 0.6

-					w.value =list;
-        
-				}
+            try {
+              if (typeof list != 'string') {
+                let data = JSON.parse(list)
+                data = Array.from(data, d => {
+                  return {
+                    ...d,
+                    content: decodeURIComponent(d.content)
+                  }
+                })
+                list = JSON.stringify(data, null, 2)
+              }
+            } catch (error) {
+              console.log(error)
+            }
+
+            w.value = list
+          }
+        }
        // console.log('ShowTextForGPT',this.widgets.length)
-				requestAnimationFrame(() => {
-					const sz = this.computeSize();
-					if (sz[0] < this.size[0]) {
-						sz[0] = this.size[0];
-					}
-					if (sz[1] < this.size[1]) {
-						sz[1] = this.size[1];
-					}
-					this.onResize?.(sz);
-					app.graph.setDirtyCanvas(true, false);
-				});
-			}
+        requestAnimationFrame(() => {
+          if (this) {
+            const sz = this.computeSize()
+            if (sz[0] < this.size[0]) {
+              sz[0] = this.size[0]
+            }
+            if (sz[1] < this.size[1]) {
+              sz[1] = this.size[1]
+            }
+            this.onResize?.(sz)
+            app.graph.setDirtyCanvas(true, false)
+          }
+        })
+      }

-			// When the node is executed we will be sent the input text, display this in the widget
-			const onExecuted = nodeType.prototype.onExecuted;
-			nodeType.prototype.onExecuted = function (message) {
-				onExecuted?.apply(this, arguments);
-        console.log('##',message.text)
-				populate.call(this, message.text);
-			};
+      // When the node is executed we will be sent the input text, display this in the widget
+      const onExecuted = nodeType.prototype.onExecuted
+      nodeType.prototype.onExecuted = function (message) {
+        onExecuted?.apply(this, arguments)
+        // console.log('##onExecuted', this, message)
+        if (message.text) populate.call(this, message.text)
+      }

-			const onConfigure = nodeType.prototype.onConfigure;
-			nodeType.prototype.onConfigure = function () {
-				onConfigure?.apply(this, arguments);
-				if (this.widgets_values?.length) {
-          
-					populate.call(this, this.widgets_values);
-				}
-			};
+      const onConfigure = nodeType.prototype.onConfigure
+      nodeType.prototype.onConfigure = function () {
+        onConfigure?.apply(this, arguments)
+        if (this.widgets_values?.length) {
+          populate.call(this, this.widgets_values)
+        }
+      }

      this.serialize_widgets = true //需要保存参数
-
-		}
-
-
-	},
+    }
+  }
 })
@@ -156,6 +156,32 @@ const parseSvg = async svgContent => {
  return { data, image: base64, svgElement }
 }

+
+function findImages(nodeId) {
+  // 检查当前节点是否有 imgs 字段
+  const n = app.graph.getNodeById(nodeId)
+  if (n.imgs) {
+    return n.imgs;
+  }
+
+  // 检查当前节点的 inputs 是否有 image 字段
+  if (n.inputs) {
+    for (let i = 0; i < n.inputs.length; i++) {
+      if (n.inputs[i].name==='image'||n.inputs[i].name==='images') {
+        // 获取新的 nodeId，并递归调用 findImages 函数
+        var linkId = n.inputs[i]?.link;
+        var origin_id = app.graph.links[linkId].origin_id
+        return findImages(origin_id);
+      }
+    }
+  }
+
+  // 如果没有找到 imgs 字段或者 image 字段，则返回 null
+  return null;
+}
+
+ 
+
 async function setArea (cw, ch, topBase64, base64, data, fn) {
  let displayHeight = Math.round(window.screen.availHeight * 0.8)
  let div = document.createElement('div')
@@ -571,15 +597,19 @@ app.registerExtension({
              }
            }
            try {
-              console.log('this.inputs', this.inputs)
-              let topLinkId = this.inputs[0].link
-              let topNodeId = app.graph.links[topLinkId].origin_id
-              let topIm = app.graph.getNodeById(topNodeId).imgs[0]
+              console.log('this.inputs', this.id)
+              let imgs=findImages(this.id)
+              
+              // let topLinkId = this.inputs[0].link
+              // let topNodeId = app.graph.links[topLinkId].origin_id
+              let topIm = imgs[0]

              let linkId = this.inputs[3].link
              let nodeId = app.graph.links[linkId].origin_id
              // console.log(linkId,this.inputs)
-              let im = app.graph.getNodeById(nodeId).imgs[0]
+              let imgs2=findImages(nodeId)
+              let im = imgs2[0]
+              console.log(topIm,im)
              // let src = im.src
              setArea(
                im.naturalWidth,
@@ -589,7 +619,9 @@ app.registerExtension({
                data,
                updateValue
              )
-            } catch (error) {}
+            } catch (error) {
+              console.log(error)
+            }
          })
        }
      }
@@ -1331,7 +1331,13 @@ app.registerExtension({
          let w = 360,
            s = widget.preview.videoWidth / widget.preview.videoHeight,
            h = w / s || w
-          console.log(h)
+          // console.log(h)
+
+          if (!window.documentPictureInPicture) {
+            window.alert(
+              'This feature is available only in secure contexts (HTTPS), in some or all supporting browsers. https://developer.mozilla.org/en-US/docs/Web/API/Document_Picture-in-Picture_API'
+            )
+          }

          const pipWindow = await documentPictureInPicture.requestWindow({
            width: w,
@@ -2137,8 +2143,8 @@ const node = {
  loadedGraphNode (node, app) {
    if (node.type === 'RandomPrompt') {
      try {
-        let max_count = node.widgets.filter(w => w.name === "max_count")[0];
-        max_count.value= node.widgets_values[0]
+        let max_count = node.widgets.filter(w => w.name === 'max_count')[0]
+        max_count.value = node.widgets_values[0]
        // console.log('RandomPrompt',max_count,node.widgets_values[0])
      } catch (error) {
        console.log(error)
@@ -266,16 +266,7 @@ app.registerExtension({
  },
  async loadedGraphNode (node, app) {
    if (node.type === 'RandomPrompt') {
-      // try {
-      //   let mutable_prompt = node.widgets.filter(w => w.name === 'mutable_prompt')[0]
-      //   // let ks = getLocalData(`_mixlab_PromptSlide`)
-      //   let uploadWidget = node.widgets.filter(w => w.name == 'upload')[0]
-      //   // console.log('##widget', uploadWidget.value)
-      //   let keywords = JSON.parse(uploadWidget.value)
-      //   if (keywords && keywords[0]) {
-      //     mutable_prompt.value=keywords.join('\n')
-      //   }
-      // } catch (error) {}
+     
    }
  }
 })
@@ -0,0 +1,302 @@
+const smart_connect_config_input = [
+  {
+    node_type: 'CLIPTextEncode',
+    node_widget_name: 'text',
+    inputNodeName: 'RandomPrompt',
+    inputNode_output_name: 'STRING'
+  },
+  {
+    node_type: 'CLIPTextEncode',
+    node_widget_name: 'text',
+    inputNodeName: 'EmbeddingPrompt',
+    inputNode_output_name: 'STRING'
+  },
+  {
+    node_type: 'CLIPTextEncode',
+    node_widget_name: 'text',
+    inputNodeName: 'ChinesePrompt_Mix',
+    inputNode_output_name: 'prompt'
+  },
+  {
+    node_type: 'CheckpointLoaderSimple',
+    node_widget_name: 'ckpt_name',
+    inputNodeName: 'CkptNames_',
+    inputNode_output_name: 'ckpt_names'
+  },
+  {
+    node_type: 'KSampler',
+    node_widget_name: 'sampler_name',
+    inputNodeName: 'SamplerNames_',
+    inputNode_output_name: 'sampler_names'
+  },
+  {
+    node_type: 'LoraLoaderModelOnly',
+    node_widget_name: 'lora_name',
+    inputNodeName: 'LoraNames_',
+    inputNode_output_name: 'lora_names'
+  },
+  {
+    node_type: 'LoadLoRA',
+    node_widget_name: 'lora_name',
+    inputNodeName: 'LoraNames_',
+    inputNode_output_name: 'lora_names'
+  },
+  {
+    node_type: 'Moondream',
+    node_widget_name: 'image',
+    inputNodeName: 'LoadImage',
+    inputNode_output_name: 'IMAGE'
+  }
+]
+
+const smart_connect_config_output = [
+  {
+    node_type: 'LoadImage',
+    node_output_name: 'IMAGE',
+    outputNodeName: 'ClipInterrogator',
+    outputNode_input_name: 'image'
+  },
+  {
+    node_type: 'VAEDecode',
+    node_output_name: 'IMAGE',
+    outputNodeName: 'PromptImage',
+    outputNode_input_name: 'images'
+  },
+  {
+    node_type: 'VAEDecode',
+    node_output_name: 'IMAGE',
+    outputNodeName: 'PreviewImage',
+    outputNode_input_name: 'images'
+  },
+  {
+    node_type: 'VAEDecode',
+    node_output_name: 'IMAGE',
+    outputNodeName: 'SaveImage',
+    outputNode_input_name: 'images'
+  },
+  {
+    node_type: 'Moondream',
+    node_output_name: 'STRING',
+    outputNodeName: 'ShowTextForGPT',
+    outputNode_input_name: 'text'
+  }
+]
+
+// import {
+//   convertToInput,
+//   getConfig,
+//   isConvertableWidget
+// } from '../../../extensions/core/widgetInputs.js'
+
+const CONVERTED_TYPE = 'converted-widget'
+const GET_CONFIG = Symbol()
+
+function getConfig (widgetName) {
+  const { nodeData } = this.constructor
+  return (
+    nodeData?.input?.required[widgetName] ??
+    nodeData?.input?.optional?.[widgetName]
+  )
+}
+
+function hideWidget (node, widget, suffix = '') {
+  widget.origType = widget.type
+  widget.origComputeSize = widget.computeSize
+  widget.origSerializeValue = widget.serializeValue
+  widget.computeSize = () => [0, -4] // -4 is due to the gap litegraph adds between widgets automatically
+  widget.type = CONVERTED_TYPE + suffix
+  widget.serializeValue = () => {
+    // Prevent serializing the widget if we have no input linked
+    if (!node.inputs) {
+      return undefined
+    }
+    let node_input = node.inputs.find(i => i.widget?.name === widget.name)
+
+    if (!node_input || !node_input.link) {
+      return undefined
+    }
+    return widget.origSerializeValue
+      ? widget.origSerializeValue()
+      : widget.value
+  }
+
+  // Hide any linked widgets, e.g. seed+seedControl
+  if (widget.linkedWidgets) {
+    for (const w of widget.linkedWidgets) {
+      hideWidget(node, w, ':' + widget.name)
+    }
+  }
+}
+
+function convertToInput (node, widget, config) {
+  hideWidget(node, widget)
+
+  const type = config[0]
+
+  // Add input and store widget config for creating on primitive node
+  const sz = node.size
+  node.addInput(widget.name, type, {
+    widget: { name: widget.name, [GET_CONFIG]: () => config }
+  })
+
+  for (const widget of node.widgets) {
+    widget.last_y += LiteGraph.NODE_SLOT_HEIGHT
+  }
+
+  // Restore original size but grow if needed
+  node.setSize([Math.max(sz[0], node.size[0]), Math.max(sz[1], node.size[1])])
+}
+
+export function smart_init () {
+  LGraphCanvas.prototype._createNodeForInput = function (
+    node,
+    widget,
+    inputNodeName,
+    inputNode_slot
+  ) {
+    // console.log(node.pos)
+
+    // var widget = node.widgets.filter(w => w.name === node_widget_name)[0]
+    if (widget) {
+      // 如果有存在的，没有连线输出的，自动连，不新建
+      let input_node = null
+
+      Array.from(app.graph.findNodesByType(inputNodeName), n => {
+        var links = n.outputs.filter(o => o.name === inputNode_slot)[0].links
+        //  console.log(links)
+        if (!links || links?.length === 0) input_node = n
+      })
+      // 新建
+      if (!input_node) {
+        input_node = LiteGraph.createNode(inputNodeName)
+        input_node.pos = [node.pos[0] - node.size[0] - 24, node.pos[1] - 48]
+        app.canvas.graph.add(input_node, false)
+      } else {
+        input_node.pos = [node.pos[0] - node.size[0] - 24, node.pos[1] - 48]
+      }
+
+      const config = getConfig.call(node, widget.name) ?? [
+        widget.type,
+        widget.options || {}
+      ]
+      let node_slotType = config[0]
+      // 如果input没有，则创建
+      if (!node.inputs?.filter(inp => inp.name === widget.name)[0]||!node.inputs)
+        convertToInput(node, widget, config)
+      input_node.connectByType(inputNode_slot, node, node_slotType)
+    }
+  }
+
+  LGraphCanvas.prototype._createNodeForOutput = function (
+    node,
+    widget,
+    outputNodeName,
+    outputNode_slot
+  ) {
+    if (widget) {
+      let output_node
+      Array.from(app.graph.findNodesByType(outputNodeName), n => {
+        var links = n.inputs.filter(o => o.name === outputNode_slot)[0].links
+        //  console.log(links)
+        if (!links || links?.length === 0) output_node = n
+      })
+      console.log('output_node', output_node, widget.name)
+
+      if (!output_node) {
+        // 新建
+        output_node = LiteGraph.createNode(outputNodeName)
+        output_node.pos = [node.pos[0] + node.size[0] + 24, node.pos[1] - 48]
+        app.canvas.graph.add(output_node, false)
+      } else {
+        output_node.pos = [node.pos[0] + node.size[0] + 24, node.pos[1] - 48]
+      }
+
+      const config = getConfig.call(node, widget.name) ?? [
+        widget.type,
+        widget.options || {}
+      ]
+      let node_slotType = config[0]
+      console.log(node_slotType, output_node, outputNode_slot)
+      let type = output_node.inputs.filter(
+        inp => inp.name == outputNode_slot
+      )[0].type
+      node.connectByType(node_slotType, output_node, type)
+    }
+  }
+}
+
+export function addSmartMenu (options, node) {
+  let sopts = []
+
+  for (const sc of smart_connect_config_input) {
+    // 有智能推荐，则出现
+    if (node.type === sc.node_type) {
+      // console.log('smart',node)
+      // 则出现 randomPrompt
+      // CLIPTextEncode 的widget ，name== 'text'
+      let node_widget_name = sc.node_widget_name
+      let widget = node.widgets.filter(w => w.name === node_widget_name)[0]
+      if (!widget) {
+        // 控件没有，则查找inputs
+        widget = node.inputs.filter(w => w.name === node_widget_name)[0]
+      }
+
+      let isLinkNull = true
+      // 如果input里已经有，但是link为空
+      if (node.inputs?.filter(inp => inp.name === node_widget_name)[0]) {
+        isLinkNull =
+          node.inputs.filter(inp => inp.name === node_widget_name)[0].link ===
+          null
+      }
+
+      if (widget && isLinkNull) {
+        sopts.push({
+          content: sc.inputNodeName.split('_')[0] + '➡️',
+          callback: () => {
+            LGraphCanvas.prototype._createNodeForInput(
+              node, //当前node
+              widget, //当前node里需要自动连线的widget
+              sc.inputNodeName, //作为input的node type
+              sc.inputNode_output_name // 作为input的node的outputs的name. the input slot type of the target node
+            )
+          }
+        })
+      }
+    }
+  }
+
+  for (const sc of smart_connect_config_output) {
+    if (node.type === sc.node_type) {
+      let node_output_name = sc.node_output_name
+      const widget = node.outputs.filter(w => w.name === node_output_name)[0]
+
+      let isLinkNull = true
+      // 如果output里 link为空
+      if (node.outputs?.filter(inp => inp.name === node_output_name)[0]) {
+        isLinkNull =
+          node.outputs.filter(inp => inp.name === node_output_name)[0].links
+            ?.length === 0
+        if (!node.outputs.filter(inp => inp.name === node_output_name)[0].links)
+          isLinkNull = true
+      }
+
+      if (widget && isLinkNull) {
+        sopts.push({
+          content: '➡️' + sc.outputNodeName.split('_')[0],
+          callback: () => {
+            LGraphCanvas.prototype._createNodeForOutput(
+              node, //当前node
+              widget, //当前node里需要自动连线的widget
+              sc.outputNodeName, //作为input的node type
+              sc.outputNode_input_name // 作为input的node的outputs的name. the input slot type of the target node
+            )
+          }
+        })
+      }
+    }
+  }
+
+  if (sopts.length > 0) options = [...sopts, null, ...options]
+
+  return options
+}
@@ -1,14 +1,247 @@
 import { app } from '../../../scripts/app.js'
-import { api } from '../../../scripts/api.js'
-import { ComfyWidgets } from '../../../scripts/widgets.js'
-import { $el } from '../../../scripts/ui.js'
 import { closeIcon } from './svg_icons.js'
+import { api } from '../../../scripts/api.js'

 import {
  GroupNodeConfig,
  GroupNodeHandler
 } from '../../../extensions/core/groupNode.js'

+import { smart_init, addSmartMenu } from './smart_connect.js'
+
+let isScriptLoaded = {};
+
+function loadExternalScript(url) {
+  return new Promise((resolve, reject) => {
+    if (isScriptLoaded[url]) {
+      resolve();
+      return;
+    }
+
+    const script = document.createElement('script');
+    script.src = url;
+    script.onload = () => {
+      isScriptLoaded[url]= true;
+      resolve();
+    };
+    script.onerror = reject;
+    document.head.appendChild(script);
+  });
+}
+
+
+//
+
+function createChart(chartDom,nodes){
+
+  var myChart = echarts.init(chartDom);
+  var option;
+
+  console.log(nodes)
+  option = {
+    series: [
+      {
+        type: 'treemap',
+        data: [
+          {
+            name: 'nodeA',
+            value: 10,
+            children: Array.from(nodes,n=>{
+              return {
+                name:n.type,
+                value:n.count
+              }
+            })
+          },
+          
+        ]
+      }
+    ]
+  };
+  
+  option && myChart.setOption(option);
+}
+
+async function createNodesCharts () {
+  await loadExternalScript('/extensions/comfyui-mixlab-nodes/lib/echarts.min.js')
+  const templates = await loadTemplate()
+  var nodes = {}
+  Array.from(templates, t => {
+    let j = JSON.parse(t.data)
+    for (let node of j.nodes) {
+      if (!nodes[node.type]) nodes[node.type] = { type: node.type, count: 0 }
+      nodes[node.type].count++
+    }
+  })
+  nodes = Object.values(nodes).sort((a, b) => b.count - a.count)
+ 
+
+  const menu = document.querySelector('.comfy-menu')
+  const separator = document.createElement('div')
+  separator.style = `margin: 20px 0px;
+  width: 100%;
+  height: 1px;
+  background: var(--border-color);
+  `
+  menu.append(separator)
+
+  const appsButton = document.createElement('button')
+  appsButton.textContent = 'Nodes';
+
+  
+
+  appsButton.onclick = () => { 
+    let div = document.querySelector('#mixlab_apps')
+    if (!div) {
+      div = document.createElement('div')
+      div.id = 'mixlab_apps'
+      document.body.appendChild(div)
+
+      let btn = document.createElement('div')
+      btn.style = `display: flex;
+     width: calc(100% - 24px);
+     justify-content: space-between;
+     align-items: center;
+     padding: 0 12px;
+     height: 44px;`
+      let btnB = document.createElement('button')
+      let textB = document.createElement('p')
+      btn.appendChild(textB)
+      btn.appendChild(btnB)
+      textB.style.fontSize = '12px'
+      textB.innerText = `Nodes`
+
+      btnB.style = `float: right; border: none; color: var(--input-text);
+     background-color: var(--comfy-input-bg); border-color: var(--border-color);cursor: pointer;`
+      btnB.addEventListener('click', () => {
+        div.style.display = 'none'
+      })
+      btnB.innerText = 'X'
+
+      // 悬浮框拖动事件
+      div.addEventListener('mousedown', function (e) {
+        var startX = e.clientX
+        var startY = e.clientY
+        var offsetX = div.offsetLeft
+        var offsetY = div.offsetTop
+
+        function moveBox (e) {
+          var newX = e.clientX
+          var newY = e.clientY
+          var deltaX = newX - startX
+          var deltaY = newY - startY
+          div.style.left = offsetX + deltaX + 'px'
+          div.style.top = offsetY + deltaY + 'px'
+          localStorage.setItem(
+            'mixlab_app_pannel',
+            JSON.stringify({ x: div.style.left, y: div.style.top })
+          )
+        }
+
+        function stopMoving () {
+          document.removeEventListener('mousemove', moveBox)
+          document.removeEventListener('mouseup', stopMoving)
+        }
+
+        document.addEventListener('mousemove', moveBox)
+        document.addEventListener('mouseup', stopMoving)
+      })
+
+      div.appendChild(btn)
+
+      let chartDom = document.createElement('div')
+      chartDom.style=`height:80vh;width:450px`
+      chartDom.className='chart'
+      div.appendChild(chartDom)
+      
+    }
+    if (div.style.display == 'flex') {
+      div.style.display = 'none'
+    } else {
+      let pos = JSON.parse(
+        localStorage.getItem('mixlab_app_pannel') ||
+          JSON.stringify({ x: 0, y: 0 })
+      )
+
+      div.style = `
+      flex-direction: column;
+      align-items: end;
+      display:flex;
+      position: absolute; 
+      top: ${pos.y}; left: ${pos.x}; width: 450px; 
+      color: var(--descrip-text);
+      background-color: var(--comfy-menu-bg);
+      padding: 10px; 
+      border: 1px solid black;z-index: 999999999;padding-top: 0;`
+    };
+
+   
+    createChart(div.querySelector('.chart'),nodes)
+
+  }
+  menu.append(appsButton)
+
+}
+
+function copyNodeValues (src, dest) {
+  // title
+  dest.title = src.title
+
+  // copy input connections
+
+  for (let i in src.inputs) {
+    let input = src.inputs[i]
+    if (input.link) {
+      let link = app.graph.links[input.link]
+      let src_node = app.graph.getNodeById(link.origin_id)
+      if (dest.inputs.filter(inp => inp.name === input.name).length === 0) {
+        // 没有，name换了
+        let dInp = dest.inputs.filter(inp => inp.type === input.type)
+        if (dInp.length === 1) {
+          src_node.connect(link.origin_slot, dest.id, dInp[0].name)
+        }
+      } else {
+        src_node.connect(link.origin_slot, dest.id, input.name)
+      }
+    }
+  }
+
+  // copy output connections
+  let output_links = {}
+  for (let i in src.outputs) {
+    let output = src.outputs[i]
+    if (output.links) {
+      let links = []
+      for (let j in output.links) {
+        links.push(app.graph.links[output.links[j]])
+      }
+      output_links[output.name] = links
+    }
+  }
+
+  for (let i in dest.outputs) {
+    let links = output_links[dest.outputs[i].name]
+    if (links) {
+      for (let j in links) {
+        let link = links[j]
+        let target_node = app.graph.getNodeById(link.target_id)
+        dest.connect(parseInt(i), target_node, link.target_slot)
+      }
+    }
+  }
+
+  // copy  widgets
+  for (const w of src.widgets) {
+    for (const d of dest.widgets) {
+      if (w.name === d.name) {
+        d.value = w.value
+      }
+    }
+  }
+
+  app.graph.afterChange()
+}
+
 function deepEqual (obj1, obj2) {
  if (typeof obj1 !== typeof obj2) {
    return false
@@ -558,6 +791,39 @@ function createModal (url, markdown, title) {
  div.appendChild(bgElement)
 }

+const loadTemplate = async () => {
+  const id = 'Comfy.NodeTemplates'
+  const file = 'comfy.templates.json'
+
+  let templates = []
+  if (app.storageLocation === 'server') {
+    if (app.isNewUserSession) {
+      // New user so migrate existing templates
+      const json = localStorage.getItem(id)
+      if (json) {
+        templates = JSON.parse(json)
+      }
+      await api.storeUserData(file, json, { stringify: false })
+    } else {
+      const res = await api.getUserData(file)
+      if (res.status === 200) {
+        try {
+          templates = await res.json()
+        } catch (error) {}
+      } else if (res.status !== 404) {
+        console.error(res.status + ' ' + res.statusText)
+      }
+    }
+  } else {
+    const json = localStorage.getItem(id)
+    if (json) {
+      templates = JSON.parse(json)
+    }
+  }
+
+  return templates ?? []
+}
+
 app.registerExtension({
  name: 'Comfy.Mixlab.ui',
  init () {
@@ -575,22 +841,71 @@ app.registerExtension({
      }
    }

+    LGraphCanvas.prototype.fixTheNode = function (node) {
+      let new_node = LiteGraph.createNode(node.comfyClass)
+      new_node.pos = [node.pos[0], node.pos[1]]
+      app.canvas.graph.add(new_node, false)
+      copyNodeValues(node, new_node)
+      app.canvas.graph.remove(node)
+    }
+
+    smart_init()
+
    const getNodeMenuOptions = LGraphCanvas.prototype.getNodeMenuOptions // store the existing method
    LGraphCanvas.prototype.getNodeMenuOptions = function (node) {
      // replace it
      const options = getNodeMenuOptions.apply(this, arguments) // start by calling the stored one
      node.setDirtyCanvas(true, true) // force a redraw of (foreground, background)
+      console.log('getNodeMenuOptions', node.type == 'CLIPTextEncode')

-      return [
+      let opts = [
        {
          content: 'Help ♾️Mixlab', // with a name
          callback: () => {
            LGraphCanvas.prototype.helpAboutNode(node)
          } // and the callback
        },
-        null,
-        ...options
-      ] // and return the options
+        {
+          content: 'Fix node v2', // with a name
+          callback: () => {
+            LGraphCanvas.prototype.fixTheNode(node)
+          }
+        }
+      ]
+
+      opts = addSmartMenu(opts, node)
+
+      // if (node.type == 'CLIPTextEncode') {
+      //   // 则出现 randomPrompt
+      //   // CLIPTextEncode 的widget ，name== 'text'
+      //   let node_widget_name = 'text'
+      //   const widget = node.widgets.filter(w => w.name === node_widget_name)[0]
+
+      //  let mixlab_nodes_smart_connect= [{node_type:'CLIPTextEncode',
+      //   node_widget_name:'text',
+      //   inputNodeName:'RandomPrompt',
+      //   inputNode_output_type:'STRING'}]
+
+      //   if (widget) {
+      //     opts = [
+      //       {
+      //         content: 'RandomPrompt',
+      //         callback: () => {
+      //           LGraphCanvas.prototype._createNodeForInput(
+      //             node, //当前node
+      //             widget,//当前node里需要自动连线的widget
+      //             'RandomPrompt',//作为input的node type
+      //             'STRING'// 作为input的node的outputs的type. the input slot type of the target node
+      //           )
+      //         }
+      //       },
+      //       null,
+      //       ...opts
+      //     ]
+      //   }
+      // }
+
+      return [...opts, null, ...options] // and return the options
    }

    const getGroupMenuOptions = LGraphCanvas.prototype.getGroupMenuOptions // store the existing method
@@ -599,16 +914,6 @@ app.registerExtension({
      const options = getGroupMenuOptions.apply(this, arguments) // start by calling the stored one
      node.setDirtyCanvas(true, true) // force a redraw of (foreground, background)

-      // templete
-      const key = 'Comfy.NodeTemplates'
-      let templates = localStorage.getItem(key)
-      if (templates) {
-        templates = JSON.parse(templates)
-      } else {
-        templates = []
-      }
-      const store = () => localStorage.setItem(key, JSON.stringify(templates))
-
      return [
        {
          content: 'Clone Group ♾️Mixlab', // with a name
@@ -666,7 +971,7 @@ app.registerExtension({
              localStorage.setItem('litegrapheditor_clipboard', old)
            }

-            clipboardAction(() => {
+            clipboardAction(async () => {
              let name = group.title + ' ♾️Mixlab'
              let nodes = group._nodes

@@ -679,6 +984,8 @@ app.registerExtension({
                const nodeData = node.serialize()

                let groupData = GroupNodeHandler.getGroupData(node)
+
+                // console.log('groupData',GroupNodeHandler.isGroupNode(node),groupData)
                if (groupData) {
                  groupData = groupData.nodeData
                  if (!data.groupNodes) {
@@ -689,14 +996,46 @@ app.registerExtension({
                }
              }

-              templates.push({
+              // templete
+              const store = async nt => {
+                const id = 'Comfy.NodeTemplates'
+                const file = 'comfy.templates.json'
+                let templates = await loadTemplate()
+                templates.push(nt)
+                if (app.storageLocation === 'server') {
+                  const ts = JSON.stringify(templates, undefined, 4)
+                  localStorage.setItem(id, ts) // Backwards compatibility
+                  try {
+                    await api.storeUserData(file, ts, {
+                      stringify: false
+                    })
+                  } catch (error) {
+                    console.error(error)
+                    alert(error.message)
+                  }
+                } else {
+                  localStorage.setItem(id, JSON.stringify(templates))
+                }
+              }
+              console.log('data', data)
+              store({
                name,
                data: JSON.stringify(data)
              })
-              store()
            })
          } // and the callback
        },
+        {
+          content: `Remove Group&Nodes ♾️Mixlab`, // with a name
+          callback: async (value, opts, e, menu, group) => {
+            // console.log(group)
+            let nodes = group._nodes
+            for (const node of nodes) {
+              app.graph.remove(node)
+            }
+            app.graph.remove(group)
+          } // and the callback
+        },
        null,
        ...options
      ] // and return the options
@@ -722,7 +1061,7 @@ app.registerExtension({

      const apps = await get_my_app()

-      let apps_map = { '0': [] }
+      let apps_map = { 0: [] }

      for (const app of apps) {
        if (app.category) {
@@ -735,7 +1074,7 @@ app.registerExtension({

      let apps_opts = []
      for (const category in apps_map) {
-        console.log('category',typeof(category))
+        console.log('category', typeof category)
        if (category === '0') {
          apps_opts.push(
            ...Array.from(apps_map[category], a => {
@@ -764,7 +1103,7 @@ app.registerExtension({
        } else {
          // 二级
          apps_opts.push({
-            content:  '🚀 '+category,
+            content: '🚀 ' + category,
            has_submenu: true,
            disabled: false,
            submenu: {
@@ -1047,7 +1386,7 @@ app.registerExtension({
            has_submenu: true,
            disabled: false,
            submenu: {
-              options:apps_opts
+              options: apps_opts
            }
          }
        )
@@ -1055,5 +1394,21 @@ app.registerExtension({
        return options
      }
    }, 1000)
+
+    // createNodesCharts()
+  },
+  async loadedGraphNode (node, app) {
+    // console.log(
+    //   '#ui init',
+    //   app.graph._nodes[app.graph._nodes.length - 1].id,
+    //   node.id
+    // )
+    try {
+      // 用来居中显示节点
+      if ((app.graph._nodes[app.graph._nodes.length - 1].id, node.id)) {
+        app.canvas.centerOnNode(node)
+        app.canvas.setZoom(0.45)
+      }
+    } catch (error) {}
  }
 })
@@ -1,8 +1,5 @@
 import { app } from '../../../scripts/app.js'
-import { api } from '../../../scripts/api.js'
-import { ComfyWidgets } from '../../../scripts/widgets.js'
-import { $el } from '../../../scripts/ui.js'
-import { addValueControlWidget } from '../../../scripts/widgets.js'
+import { $el } from '../../../scripts/ui.js' 

 const getLocalData = key => {
  let data = {}
@@ -0,0 +1,558 @@
+{
+  "last_node_id": 69,
+  "last_link_id": 72,
+  "nodes": [
+    {
+      "id": 37,
+      "type": "CLIPTextEncode",
+      "pos": [
+        6705,
+        -216
+      ],
+      "size": {
+        "0": 400,
+        "1": 200
+      },
+      "flags": {},
+      "order": 5,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "clip",
+          "type": "CLIP",
+          "link": 33
+        }
+      ],
+      "outputs": [
+        {
+          "name": "CONDITIONING",
+          "type": "CONDITIONING",
+          "links": [
+            34
+          ],
+          "shape": 3
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "CLIPTextEncode"
+      },
+      "widgets_values": [
+        "beautiful scenery nature glass bottle landscape, , purple galaxy bottle,"
+      ]
+    },
+    {
+      "id": 5,
+      "type": "CLIPTextEncode",
+      "pos": [
+        6693,
+        61
+      ],
+      "size": {
+        "0": 425.27801513671875,
+        "1": 180.6060791015625
+      },
+      "flags": {},
+      "order": 4,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "clip",
+          "type": "CLIP",
+          "link": 6
+        },
+        {
+          "name": "text",
+          "type": "STRING",
+          "link": 25,
+          "widget": {
+            "name": "text"
+          }
+        }
+      ],
+      "outputs": [
+        {
+          "name": "CONDITIONING",
+          "type": "CONDITIONING",
+          "links": [
+            3
+          ],
+          "slot_index": 0
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "CLIPTextEncode"
+      },
+      "widgets_values": [
+        "text, watermark"
+      ]
+    },
+    {
+      "id": 3,
+      "type": "EmptyLatentImage",
+      "pos": [
+        6689,
+        306
+      ],
+      "size": {
+        "0": 315,
+        "1": 106
+      },
+      "flags": {},
+      "order": 0,
+      "mode": 0,
+      "outputs": [
+        {
+          "name": "LATENT",
+          "type": "LATENT",
+          "links": [
+            4
+          ],
+          "slot_index": 0
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "EmptyLatentImage"
+      },
+      "widgets_values": [
+        512,
+        512,
+        1
+      ]
+    },
+    {
+      "id": 27,
+      "type": "EmbeddingPrompt",
+      "pos": [
+        6104,
+        23
+      ],
+      "size": {
+        "0": 399.6408996582031,
+        "1": 82
+      },
+      "flags": {},
+      "order": 1,
+      "mode": 0,
+      "outputs": [
+        {
+          "name": "STRING",
+          "type": "STRING",
+          "links": [
+            25
+          ],
+          "shape": 3
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "EmbeddingPrompt"
+      },
+      "widgets_values": [
+        "negative-embed-verybadimagenegative_v1.3",
+        1
+      ]
+    },
+    {
+      "id": 67,
+      "type": "VAEDecode",
+      "pos": [
+        7551,
+        -184
+      ],
+      "size": {
+        "0": 210,
+        "1": 46
+      },
+      "flags": {},
+      "order": 7,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "samples",
+          "type": "LATENT",
+          "link": 70
+        },
+        {
+          "name": "vae",
+          "type": "VAE",
+          "link": 69
+        }
+      ],
+      "outputs": [
+        {
+          "name": "IMAGE",
+          "type": "IMAGE",
+          "links": [
+            71
+          ],
+          "shape": 3
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "VAEDecode"
+      }
+    },
+    {
+      "id": 1,
+      "type": "KSampler",
+      "pos": [
+        7187,
+        -145
+      ],
+      "size": {
+        "0": 315,
+        "1": 262
+      },
+      "flags": {},
+      "order": 6,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "model",
+          "type": "MODEL",
+          "link": 1
+        },
+        {
+          "name": "positive",
+          "type": "CONDITIONING",
+          "link": 34
+        },
+        {
+          "name": "negative",
+          "type": "CONDITIONING",
+          "link": 3
+        },
+        {
+          "name": "latent_image",
+          "type": "LATENT",
+          "link": 4
+        }
+      ],
+      "outputs": [
+        {
+          "name": "LATENT",
+          "type": "LATENT",
+          "links": [
+            70
+          ],
+          "slot_index": 0
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "KSampler"
+      },
+      "widgets_values": [
+        408562451564429,
+        "fixed",
+        20,
+        8,
+        "euler",
+        "normal",
+        1
+      ]
+    },
+    {
+      "id": 61,
+      "type": "PromptImage",
+      "pos": [
+        7853,
+        -238
+      ],
+      "size": [
+        465.6378949342379,
+        760.4568424013569
+      ],
+      "flags": {},
+      "order": 8,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "images",
+          "type": "IMAGE",
+          "link": 71
+        },
+        {
+          "name": "prompts",
+          "type": "STRING",
+          "link": 72,
+          "widget": {
+            "name": "prompts"
+          }
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "PromptImage"
+      },
+      "widgets_values": [
+        "",
+        "disable",
+        {
+          "_images": [
+            [
+              {
+                "filename": "mixlab_PromptImage_0_00027_.png",
+                "subfolder": "",
+                "type": "output"
+              }
+            ],
+            [
+              {
+                "filename": "mixlab_PromptImage_1_00028_.png",
+                "subfolder": "",
+                "type": "output"
+              }
+            ],
+            [
+              {
+                "filename": "mixlab_PromptImage_2_00029_.png",
+                "subfolder": "",
+                "type": "output"
+              }
+            ],
+            [
+              {
+                "filename": "mixlab_PromptImage_3_00030_.png",
+                "subfolder": "",
+                "type": "output"
+              }
+            ],
+            [
+              {
+                "filename": "mixlab_PromptImage_4_00031_.png",
+                "subfolder": "",
+                "type": "output"
+              }
+            ],
+            [
+              {
+                "filename": "mixlab_PromptImage_5_00032_.png",
+                "subfolder": "",
+                "type": "output"
+              }
+            ]
+          ],
+          "prompts": [
+            "512-inpainting-ema.safetensors",
+            "SSD-1B.safetensors",
+            "awportrait_v12.safetensors",
+            "cardosAnime_v20.safetensors",
+            "deliberate_v2.safetensors",
+            "gameIconInstitute_v40.safetensors"
+          ]
+        }
+      ]
+    },
+    {
+      "id": 2,
+      "type": "CheckpointLoaderSimple",
+      "pos": [
+        6105,
+        -139
+      ],
+      "size": {
+        "0": 315,
+        "1": 98
+      },
+      "flags": {},
+      "order": 3,
+      "mode": 0,
+      "inputs": [
+        {
+          "name": "ckpt_name",
+          "type": [
+            "512-inpainting-ema.safetensors",
+            "SSD-1B.safetensors",
+            "awportrait_v12.safetensors",
+            "cardosAnime_v20.safetensors",
+            "deliberate_v2.safetensors",
+            "gameIconInstitute_v40.safetensors",
+            "illuminatiDiffusionV1_v11-unclip-h-fp16.safetensors",
+            "sd_xl_turbo_1.0_fp16.safetensors",
+            "svd.safetensors"
+          ],
+          "link": 64,
+          "widget": {
+            "name": "ckpt_name"
+          }
+        }
+      ],
+      "outputs": [
+        {
+          "name": "MODEL",
+          "type": "MODEL",
+          "links": [
+            1
+          ],
+          "slot_index": 0
+        },
+        {
+          "name": "CLIP",
+          "type": "CLIP",
+          "links": [
+            6,
+            33
+          ],
+          "slot_index": 1
+        },
+        {
+          "name": "VAE",
+          "type": "VAE",
+          "links": [
+            69
+          ],
+          "slot_index": 2
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "CheckpointLoaderSimple"
+      },
+      "widgets_values": [
+        "deliberate_v2.safetensors"
+      ]
+    },
+    {
+      "id": 56,
+      "type": "CkptNames_",
+      "pos": [
+        7860,
+        -494
+      ],
+      "size": {
+        "0": 400,
+        "1": 200
+      },
+      "flags": {},
+      "order": 2,
+      "mode": 0,
+      "outputs": [
+        {
+          "name": "ckpt_names",
+          "type": "*",
+          "links": [
+            64,
+            72
+          ],
+          "shape": 6,
+          "slot_index": 0
+        }
+      ],
+      "properties": {
+        "Node name for S&R": "CkptNames_"
+      },
+      "widgets_values": [
+        "512-inpainting-ema.safetensors\nSSD-1B.safetensors\nawportrait_v12.safetensors\ncardosAnime_v20.safetensors\ndeliberate_v2.safetensors\ngameIconInstitute_v40.safetensors"
+      ]
+    }
+  ],
+  "links": [
+    [
+      1,
+      2,
+      0,
+      1,
+      0,
+      "MODEL"
+    ],
+    [
+      3,
+      5,
+      0,
+      1,
+      2,
+      "CONDITIONING"
+    ],
+    [
+      4,
+      3,
+      0,
+      1,
+      3,
+      "LATENT"
+    ],
+    [
+      6,
+      2,
+      1,
+      5,
+      0,
+      "CLIP"
+    ],
+    [
+      25,
+      27,
+      0,
+      5,
+      1,
+      "STRING"
+    ],
+    [
+      33,
+      2,
+      1,
+      37,
+      0,
+      "CLIP"
+    ],
+    [
+      34,
+      37,
+      0,
+      1,
+      1,
+      "CONDITIONING"
+    ],
+    [
+      64,
+      56,
+      0,
+      2,
+      0,
+      [
+        "512-inpainting-ema.safetensors",
+        "SSD-1B.safetensors",
+        "awportrait_v12.safetensors",
+        "cardosAnime_v20.safetensors",
+        "deliberate_v2.safetensors",
+        "gameIconInstitute_v40.safetensors",
+        "illuminatiDiffusionV1_v11-unclip-h-fp16.safetensors",
+        "sd_xl_turbo_1.0_fp16.safetensors",
+        "svd.safetensors"
+      ]
+    ],
+    [
+      69,
+      2,
+      2,
+      67,
+      1,
+      "VAE"
+    ],
+    [
+      70,
+      1,
+      0,
+      67,
+      0,
+      "LATENT"
+    ],
+    [
+      71,
+      67,
+      0,
+      61,
+      0,
+      "IMAGE"
+    ],
+    [
+      72,
+      56,
+      0,
+      61,
+      1,
+      "STRING"
+    ]
+  ],
+  "groups": [],
+  "config": {},
+  "extra": {},
+  "version": 0.4
+}
Author	SHA1	Message	Date
shadowcz007	e37af93be3	0.15.1	2024-02-06 23:06:33 +08:00
shadowcz007	244c1700e1	Update ImageNode.py	2024-02-06 22:22:56 +08:00
shadowcz007	a3a15473ba	LoadImage 的mask保存到appinfo	2024-02-06 19:01:41 +08:00
shadowcz007	d734b5077c	GetImageSize_ 增加最小尺寸	2024-02-05 22:44:04 +08:00
shadowcz007	5430072b19	Update app_mixlab.js	2024-02-05 22:25:25 +08:00
shadowcz007	465aebaed4	Update app_mixlab.js	2024-02-05 22:24:34 +08:00
shadowcz007	ee0b16c2ea	修复appinfo配置的bug	2024-02-05 17:56:25 +08:00
shadowcz007	21d5eacb41	Update index.html	2024-02-04 23:46:25 +08:00
shadowcz007	c06688eb0b	comfyui-consistency-decoder	2024-02-02 09:47:51 +08:00
shadowcz007	b74bbcd279	fixbug :SaveImageToLocal	2024-02-01 23:52:23 +08:00
shadowcz007	64d8d9b05d	Delete echarts.min.js	2024-02-01 00:48:25 +08:00
shadowcz007	6ab60f281b	Update ImageNode.py	2024-01-31 00:24:23 +08:00
shadowcz007	e35be3b2fa	Update README.md	2024-01-30 22:57:13 +08:00
shadowcz007	0a0c27ac96	fixbug:Save Group as Template	2024-01-30 22:54:39 +08:00
shadowcz007	fb249e84eb	SplitImage增加mask输出	2024-01-30 18:07:21 +08:00
shadowcz007	76ad86fcae	Update __init__.py	2024-01-29 23:18:16 +08:00
shadowcz007	037614d227	showText 可以保存txt到本地目录	2024-01-29 14:37:51 +08:00
shadowcz007	4a50e445fd	从本地读取文件-输出文件名	2024-01-29 13:46:02 +08:00
shadowcz007	6f3c1c4393	update	2024-01-29 11:55:10 +08:00
shadowcz007	c6a9b4b592	Update gpt_mixlab.js	2024-01-29 11:45:53 +08:00
shadowcz007	e915ac4eca	Update ChatGPT.py	2024-01-29 11:37:49 +08:00
shadowcz007	a857793f63	fixbug	2024-01-29 11:34:06 +08:00
shadowcz007	31914f7510	Update ChatGPT.py	2024-01-29 11:12:02 +08:00
shadowcz007	0be859f0ee	fixbug	2024-01-28 21:38:13 +08:00
shadowcz007	14b9c3697b	Update ImageNode.py	2024-01-28 20:38:22 +08:00
shadowcz007	96b66a57bb	showText can save to local	2024-01-28 18:13:50 +08:00
shadowcz007	f13701c489	v0.15.0	2024-01-28 16:40:31 +08:00
shadow	31515b810e	Merge pull request #159 from wfjsw/debloat-init-1 publish routes without having to replicate add_routes	2024-01-28 16:01:17 +08:00
shadowcz007	29e84e08a4	add CenterImage	2024-01-28 15:59:54 +08:00
shadowcz007	0ac9ad9757	修复批量保存本地图片的bug	2024-01-27 22:44:47 +08:00
shadowcz007	9a432e0608	Update Utils.py	2024-01-27 00:58:08 +08:00
shadowcz007	c83ba5fe7f	Update PromptNode.py	2024-01-27 00:57:40 +08:00
shadowcz007	1b55c743ea	增加一些seed来控制节点	2024-01-27 00:33:49 +08:00
shadowcz007	a93579376c	不覆盖文件	2024-01-26 22:47:34 +08:00
shadowcz007	eba49f3c68	add SaveImageToLocal	2024-01-26 12:15:01 +08:00
shadowcz007	3a3da49c69	Update ImageNode.py	2024-01-25 20:10:29 +08:00
shadowcz007	3d68e48219	Update ImageNode.py	2024-01-25 20:07:31 +08:00
shadowcz007	4351fa6a0e	增加mask 的resize	2024-01-25 17:54:51 +08:00
shadowcz007	77222d2808	修复ImageCropByAlpha的bug	2024-01-25 15:40:00 +08:00
Jabasukuriputo Wang	36db7e5a9a	publish routes without having to replicate add_routes	2024-01-25 00:44:51 -06:00
shadowcz007	3572368f16	fixbug	2024-01-25 10:38:23 +08:00
shadowcz007	0755dc1462	CreateLoraNames	2024-01-24 22:59:49 +08:00
shadowcz007	94f81b7102	fixbug	2024-01-24 17:45:15 +08:00
shadowcz007	08f8fe3d7e	Update README.md	2024-01-23 23:44:47 +08:00
shadowcz007	a6cd383d67	add Sampler_names	2024-01-23 14:23:04 +08:00
shadowcz007	10face6ab0	CkptNames	2024-01-23 14:04:26 +08:00
shadowcz007	a6259ff600	add CkptNames	2024-01-23 13:58:59 +08:00
shadowcz007	d7d9e6cbfe	add smart_connect_v1	2024-01-23 12:18:38 +08:00
shadowcz007	8263609470	优化LoadImageURL，增加seed，保证图片加载失败后可以继续	2024-01-23 10:03:55 +08:00
shadowcz007	fc2367de76	centerOnNode & fix node (widgets)	2024-01-21 20:31:57 +08:00
shadowcz007	9a4f2ebc70	Update ui_mixlab.js	2024-01-20 22:39:56 +08:00
shadowcz007	812879610a	v0.14.0 发布新节点splitImage & gridoutput，用于分割图片和随机摆放元素修复若干bug	2024-01-20 21:43:18 +08:00
shadowcz007	c5e521ccc1	add splitImage&gridoutput	2024-01-20 17:39:53 +08:00
shadowcz007	bc1c8fa351	Update index.html	2024-01-19 09:45:03 +08:00
shadowcz007	7271fcf9c1	api	2024-01-18 22:49:13 +08:00
shadowcz007	ace3b7707b	Update README.md	2024-01-18 12:45:29 +08:00
shadowcz007	9eb65cc4ee	Update ClipInterrogator.py	2024-01-18 11:12:03 +08:00
shadowcz007	c5c2bc779c	add JoinWithDelimiter	2024-01-18 11:00:37 +08:00
shadowcz007	ae1751d9c0	Update Utils.py	2024-01-18 00:32:12 +08:00
shadowcz007	d17583ef7d	fixbug	2024-01-17 17:56:23 +08:00