nodes.py

import torch

import os
import sys
import json
import hashlib
import traceback
import math
import time
import struct
from io import BytesIO

from PIL import Image, ImageOps
from PIL.PngImagePlugin import PngInfo
import numpy as np
import safetensors.torch


sys.path.insert(0, os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy"))


import comfy.diffusers_load
import comfy.samplers
import comfy.sample
import comfy.sd
import comfy.utils
from comfy.cli_args import args, LatentPreviewMethod
from comfy.taesd.taesd import TAESD

import comfy.clip_vision

import comfy.model_management
import importlib

import folder_paths


class LatentPreviewer:
    def decode_latent_to_preview(self, device, x0):
        pass


class Latent2RGBPreviewer(LatentPreviewer):
    def __init__(self):
        self.latent_rgb_factors = torch.tensor([
                    #   R        G        B
                    [0.298, 0.207, 0.208],  # L1
                    [0.187, 0.286, 0.173],  # L2
                    [-0.158, 0.189, 0.264],  # L3
                    [-0.184, -0.271, -0.473],  # L4
                ], device="cpu")

    def decode_latent_to_preview(self, device, x0):
        latent_image = x0[0].permute(1, 2, 0).cpu() @ self.latent_rgb_factors

        latents_ubyte = (((latent_image + 1) / 2)
                            .clamp(0, 1)  # change scale from -1..1 to 0..1
                            .mul(0xFF)  # to 0..255
                            .byte()).cpu()

        return Image.fromarray(latents_ubyte.numpy())


def before_node_execution():
    comfy.model_management.throw_exception_if_processing_interrupted()

def interrupt_processing(value=True):
    comfy.model_management.interrupt_current_processing(value)

MAX_RESOLUTION=8192
MAX_PREVIEW_RESOLUTION = 512

class CLIPTextEncode:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"text": ("STRING", {"multiline": True}), "clip": ("CLIP", )}}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "encode"

    CATEGORY = "conditioning"

    def encode(self, clip, text):
        return ([[clip.encode(text), {}]], )

class ConditioningCombine:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning_1": ("CONDITIONING", ), "conditioning_2": ("CONDITIONING", )}}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "combine"

    CATEGORY = "conditioning"

    def combine(self, conditioning_1, conditioning_2):
        return (conditioning_1 + conditioning_2, )

class ConditioningAverage :
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning_to": ("CONDITIONING", ), "conditioning_from": ("CONDITIONING", ),
                              "conditioning_to_strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.01})
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "addWeighted"

    CATEGORY = "conditioning"

    def addWeighted(self, conditioning_to, conditioning_from, conditioning_to_strength):
        out = []

        if len(conditioning_from) > 1:
            print("Warning: ConditioningAverage conditioning_from contains more than 1 cond, only the first one will actually be applied to conditioning_to.")

        cond_from = conditioning_from[0][0]

        for i in range(len(conditioning_to)):
            t1 = conditioning_to[i][0]
            t0 = cond_from[:,:t1.shape[1]]
            if t0.shape[1] < t1.shape[1]:
                t0 = torch.cat([t0] + [torch.zeros((1, (t1.shape[1] - t0.shape[1]), t1.shape[2]))], dim=1)

            tw = torch.mul(t1, conditioning_to_strength) + torch.mul(t0, (1.0 - conditioning_to_strength))
            n = [tw, conditioning_to[i][1].copy()]
            out.append(n)
        return (out, )

class ConditioningSetArea:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", ),
                              "width": ("INT", {"default": 64, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                              "height": ("INT", {"default": 64, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                              "x": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              "y": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "append"

    CATEGORY = "conditioning"

    def append(self, conditioning, width, height, x, y, strength):
        c = []
        for t in conditioning:
            n = [t[0], t[1].copy()]
            n[1]['area'] = (height // 8, width // 8, y // 8, x // 8)
            n[1]['strength'] = strength
            n[1]['set_area_to_bounds'] = False
            c.append(n)
        return (c, )

class ConditioningSetMask:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", ),
                              "mask": ("MASK", ),
                              "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01}),
                              "set_cond_area": (["default", "mask bounds"],),
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "append"

    CATEGORY = "conditioning"

    def append(self, conditioning, mask, set_cond_area, strength):
        c = []
        set_area_to_bounds = False
        if set_cond_area != "default":
            set_area_to_bounds = True
        if len(mask.shape) < 3:
            mask = mask.unsqueeze(0)
        for t in conditioning:
            n = [t[0], t[1].copy()]
            _, h, w = mask.shape
            n[1]['mask'] = mask
            n[1]['set_area_to_bounds'] = set_area_to_bounds
            n[1]['mask_strength'] = strength
            c.append(n)
        return (c, )

class VAEDecode:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT", ), "vae": ("VAE", )}}
    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "decode"

    CATEGORY = "latent"

    def decode(self, vae, samples):
        return (vae.decode(samples["samples"]), )

class VAEDecodeTiled:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT", ), "vae": ("VAE", )}}
    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "decode"

    CATEGORY = "_for_testing"

    def decode(self, vae, samples):
        return (vae.decode_tiled(samples["samples"]), )

class VAEEncode:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "pixels": ("IMAGE", ), "vae": ("VAE", )}}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "encode"

    CATEGORY = "latent"

    @staticmethod
    def vae_encode_crop_pixels(pixels):
        x = (pixels.shape[1] // 8) * 8
        y = (pixels.shape[2] // 8) * 8
        if pixels.shape[1] != x or pixels.shape[2] != y:
            x_offset = (pixels.shape[1] % 8) // 2
            y_offset = (pixels.shape[2] % 8) // 2
            pixels = pixels[:, x_offset:x + x_offset, y_offset:y + y_offset, :]
        return pixels

    def encode(self, vae, pixels):
        pixels = self.vae_encode_crop_pixels(pixels)
        t = vae.encode(pixels[:,:,:,:3])
        return ({"samples":t}, )

class VAEEncodeTiled:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "pixels": ("IMAGE", ), "vae": ("VAE", )}}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "encode"

    CATEGORY = "_for_testing"

    def encode(self, vae, pixels):
        pixels = VAEEncode.vae_encode_crop_pixels(pixels)
        t = vae.encode_tiled(pixels[:,:,:,:3])
        return ({"samples":t}, )

class VAEEncodeForInpaint:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "pixels": ("IMAGE", ), "vae": ("VAE", ), "mask": ("MASK", ), "grow_mask_by": ("INT", {"default": 6, "min": 0, "max": 64, "step": 1}),}}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "encode"

    CATEGORY = "latent/inpaint"

    def encode(self, vae, pixels, mask, grow_mask_by=6):
        x = (pixels.shape[1] // 8) * 8
        y = (pixels.shape[2] // 8) * 8
        mask = torch.nn.functional.interpolate(mask.reshape((-1, 1, mask.shape[-2], mask.shape[-1])), size=(pixels.shape[1], pixels.shape[2]), mode="bilinear")

        pixels = pixels.clone()
        if pixels.shape[1] != x or pixels.shape[2] != y:
            x_offset = (pixels.shape[1] % 8) // 2
            y_offset = (pixels.shape[2] % 8) // 2
            pixels = pixels[:,x_offset:x + x_offset, y_offset:y + y_offset,:]
            mask = mask[:,:,x_offset:x + x_offset, y_offset:y + y_offset]

        #grow mask by a few pixels to keep things seamless in latent space
        if grow_mask_by == 0:
            mask_erosion = mask
        else:
            kernel_tensor = torch.ones((1, 1, grow_mask_by, grow_mask_by))
            padding = math.ceil((grow_mask_by - 1) / 2)

            mask_erosion = torch.clamp(torch.nn.functional.conv2d(mask.round(), kernel_tensor, padding=padding), 0, 1)

        m = (1.0 - mask.round()).squeeze(1)
        for i in range(3):
            pixels[:,:,:,i] -= 0.5
            pixels[:,:,:,i] *= m
            pixels[:,:,:,i] += 0.5
        t = vae.encode(pixels)

        return ({"samples":t, "noise_mask": (mask_erosion[:,:,:x,:y].round())}, )

class TAESDPreviewerImpl(LatentPreviewer):
    def __init__(self, taesd):
        self.taesd = taesd

    def decode_latent_to_preview(self, device, x0):
        x_sample = self.taesd.decoder(x0.to(device))[0].detach()
        # x_sample = self.taesd.unscale_latents(x_sample).div(4).add(0.5)  # returns value in [-2, 2]
        x_sample = x_sample.sub(0.5).mul(2)

        x_sample = torch.clamp((x_sample + 1.0) / 2.0, min=0.0, max=1.0)
        x_sample = 255. * np.moveaxis(x_sample.cpu().numpy(), 0, 2)
        x_sample = x_sample.astype(np.uint8)

        preview_image = Image.fromarray(x_sample)
        return preview_image

class SaveLatent:
    def __init__(self):
        self.output_dir = folder_paths.get_output_directory()

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT", ),
                              "filename_prefix": ("STRING", {"default": "latents/ComfyUI"})},
                "hidden": {"prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO"},
                }
    RETURN_TYPES = ()
    FUNCTION = "save"

    OUTPUT_NODE = True

    CATEGORY = "_for_testing"

    def save(self, samples, filename_prefix="ComfyUI", prompt=None, extra_pnginfo=None):
        full_output_folder, filename, counter, subfolder, filename_prefix = folder_paths.get_save_image_path(filename_prefix, self.output_dir)

        # support save metadata for latent sharing
        prompt_info = ""
        if prompt is not None:
            prompt_info = json.dumps(prompt)

        metadata = {"prompt": prompt_info}
        if extra_pnginfo is not None:
            for x in extra_pnginfo:
                metadata[x] = json.dumps(extra_pnginfo[x])

        file = f"{filename}_{counter:05}_.latent"
        file = os.path.join(full_output_folder, file)

        output = {}
        output["latent_tensor"] = samples["samples"]

        safetensors.torch.save_file(output, file, metadata=metadata)

        return {}


class LoadLatent:
    @classmethod
    def INPUT_TYPES(s):
        input_dir = folder_paths.get_input_directory()
        files = [f for f in os.listdir(input_dir) if os.path.isfile(os.path.join(input_dir, f)) and f.endswith(".latent")]
        return {"required": {"latent": [sorted(files), ]}, }

    CATEGORY = "_for_testing"

    RETURN_TYPES = ("LATENT", )
    FUNCTION = "load"

    def load(self, latent):
        latent_path = folder_paths.get_annotated_filepath(latent)
        latent = safetensors.torch.load_file(latent_path, device="cpu")
        samples = {"samples": latent["latent_tensor"].float()}
        return (samples, )

    @classmethod
    def IS_CHANGED(s, latent):
        image_path = folder_paths.get_annotated_filepath(latent)
        m = hashlib.sha256()
        with open(image_path, 'rb') as f:
            m.update(f.read())
        return m.digest().hex()

    @classmethod
    def VALIDATE_INPUTS(s, latent):
        if not folder_paths.exists_annotated_filepath(latent):
            return "Invalid latent file: {}".format(latent)
        return True


class CheckpointLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "config_name": (folder_paths.get_filename_list("configs"), ),
                              "ckpt_name": (folder_paths.get_filename_list("checkpoints"), )}}
    RETURN_TYPES = ("MODEL", "CLIP", "VAE")
    FUNCTION = "load_checkpoint"

    CATEGORY = "advanced/loaders"

    def load_checkpoint(self, config_name, ckpt_name, output_vae=True, output_clip=True):
        config_path = folder_paths.get_full_path("configs", config_name)
        ckpt_path = folder_paths.get_full_path("checkpoints", ckpt_name)
        return comfy.sd.load_checkpoint(config_path, ckpt_path, output_vae=True, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings"))

class CheckpointLoaderSimple:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "ckpt_name": (folder_paths.get_filename_list("checkpoints"), ),
                             }}
    RETURN_TYPES = ("MODEL", "CLIP", "VAE")
    FUNCTION = "load_checkpoint"

    CATEGORY = "loaders"

    def load_checkpoint(self, ckpt_name, output_vae=True, output_clip=True):
        ckpt_path = folder_paths.get_full_path("checkpoints", ckpt_name)
        out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae=True, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings"))
        return out

class DiffusersLoader:
    @classmethod
    def INPUT_TYPES(cls):
        paths = []
        for search_path in folder_paths.get_folder_paths("diffusers"):
            if os.path.exists(search_path):
                for root, subdir, files in os.walk(search_path, followlinks=True):
                    if "model_index.json" in files:
                        paths.append(os.path.relpath(root, start=search_path))

        return {"required": {"model_path": (paths,), }}
    RETURN_TYPES = ("MODEL", "CLIP", "VAE")
    FUNCTION = "load_checkpoint"

    CATEGORY = "advanced/loaders"

    def load_checkpoint(self, model_path, output_vae=True, output_clip=True):
        for search_path in folder_paths.get_folder_paths("diffusers"):
            if os.path.exists(search_path):
                path = os.path.join(search_path, model_path)
                if os.path.exists(path):
                    model_path = path
                    break

        return comfy.diffusers_load.load_diffusers(model_path, fp16=comfy.model_management.should_use_fp16(), output_vae=output_vae, output_clip=output_clip, embedding_directory=folder_paths.get_folder_paths("embeddings"))


class unCLIPCheckpointLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "ckpt_name": (folder_paths.get_filename_list("checkpoints"), ),
                             }}
    RETURN_TYPES = ("MODEL", "CLIP", "VAE", "CLIP_VISION")
    FUNCTION = "load_checkpoint"

    CATEGORY = "loaders"

    def load_checkpoint(self, ckpt_name, output_vae=True, output_clip=True):
        ckpt_path = folder_paths.get_full_path("checkpoints", ckpt_name)
        out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae=True, output_clip=True, output_clipvision=True, embedding_directory=folder_paths.get_folder_paths("embeddings"))
        return out

class CLIPSetLastLayer:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "clip": ("CLIP", ),
                              "stop_at_clip_layer": ("INT", {"default": -1, "min": -24, "max": -1, "step": 1}),
                              }}
    RETURN_TYPES = ("CLIP",)
    FUNCTION = "set_last_layer"

    CATEGORY = "conditioning"

    def set_last_layer(self, clip, stop_at_clip_layer):
        clip = clip.clone()
        clip.clip_layer(stop_at_clip_layer)
        return (clip,)

class LoraLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "model": ("MODEL",),
                              "clip": ("CLIP", ),
                              "lora_name": (folder_paths.get_filename_list("loras"), ),
                              "strength_model": ("FLOAT", {"default": 1.0, "min": -10.0, "max": 10.0, "step": 0.01}),
                              "strength_clip": ("FLOAT", {"default": 1.0, "min": -10.0, "max": 10.0, "step": 0.01}),
                              }}
    RETURN_TYPES = ("MODEL", "CLIP")
    FUNCTION = "load_lora"

    CATEGORY = "loaders"

    def load_lora(self, model, clip, lora_name, strength_model, strength_clip):
        if strength_model == 0 and strength_clip == 0:
            return (model, clip)

        lora_path = folder_paths.get_full_path("loras", lora_name)
        model_lora, clip_lora = comfy.sd.load_lora_for_models(model, clip, lora_path, strength_model, strength_clip)
        return (model_lora, clip_lora)

class TomePatchModel:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "model": ("MODEL",),
                              "ratio": ("FLOAT", {"default": 0.3, "min": 0.0, "max": 1.0, "step": 0.01}),
                              }}
    RETURN_TYPES = ("MODEL",)
    FUNCTION = "patch"

    CATEGORY = "_for_testing"

    def patch(self, model, ratio):
        m = model.clone()
        m.set_model_tomesd(ratio)
        return (m, )

class VAELoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "vae_name": (folder_paths.get_filename_list("vae"), )}}
    RETURN_TYPES = ("VAE",)
    FUNCTION = "load_vae"

    CATEGORY = "loaders"

    #TODO: scale factor?
    def load_vae(self, vae_name):
        vae_path = folder_paths.get_full_path("vae", vae_name)
        vae = comfy.sd.VAE(ckpt_path=vae_path)
        return (vae,)

class ControlNetLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "control_net_name": (folder_paths.get_filename_list("controlnet"), )}}

    RETURN_TYPES = ("CONTROL_NET",)
    FUNCTION = "load_controlnet"

    CATEGORY = "loaders"

    def load_controlnet(self, control_net_name):
        controlnet_path = folder_paths.get_full_path("controlnet", control_net_name)
        controlnet = comfy.sd.load_controlnet(controlnet_path)
        return (controlnet,)

class DiffControlNetLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "model": ("MODEL",),
                              "control_net_name": (folder_paths.get_filename_list("controlnet"), )}}

    RETURN_TYPES = ("CONTROL_NET",)
    FUNCTION = "load_controlnet"

    CATEGORY = "loaders"

    def load_controlnet(self, model, control_net_name):
        controlnet_path = folder_paths.get_full_path("controlnet", control_net_name)
        controlnet = comfy.sd.load_controlnet(controlnet_path, model)
        return (controlnet,)


class ControlNetApply:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", ),
                             "control_net": ("CONTROL_NET", ),
                             "image": ("IMAGE", ),
                             "strength": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.01})
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "apply_controlnet"

    CATEGORY = "conditioning"

    def apply_controlnet(self, conditioning, control_net, image, strength):
        if strength == 0:
            return (conditioning, )

        c = []
        control_hint = image.movedim(-1,1)
        for t in conditioning:
            n = [t[0], t[1].copy()]
            c_net = control_net.copy().set_cond_hint(control_hint, strength)
            if 'control' in t[1]:
                c_net.set_previous_controlnet(t[1]['control'])
            n[1]['control'] = c_net
            c.append(n)
        return (c, )

class CLIPLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "clip_name": (folder_paths.get_filename_list("clip"), ),
                             }}
    RETURN_TYPES = ("CLIP",)
    FUNCTION = "load_clip"

    CATEGORY = "loaders"

    def load_clip(self, clip_name):
        clip_path = folder_paths.get_full_path("clip", clip_name)
        clip = comfy.sd.load_clip(ckpt_path=clip_path, embedding_directory=folder_paths.get_folder_paths("embeddings"))
        return (clip,)

class CLIPVisionLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "clip_name": (folder_paths.get_filename_list("clip_vision"), ),
                             }}
    RETURN_TYPES = ("CLIP_VISION",)
    FUNCTION = "load_clip"

    CATEGORY = "loaders"

    def load_clip(self, clip_name):
        clip_path = folder_paths.get_full_path("clip_vision", clip_name)
        clip_vision = comfy.clip_vision.load(clip_path)
        return (clip_vision,)

class CLIPVisionEncode:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "clip_vision": ("CLIP_VISION",),
                              "image": ("IMAGE",)
                             }}
    RETURN_TYPES = ("CLIP_VISION_OUTPUT",)
    FUNCTION = "encode"

    CATEGORY = "conditioning"

    def encode(self, clip_vision, image):
        output = clip_vision.encode_image(image)
        return (output,)

class StyleModelLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "style_model_name": (folder_paths.get_filename_list("style_models"), )}}

    RETURN_TYPES = ("STYLE_MODEL",)
    FUNCTION = "load_style_model"

    CATEGORY = "loaders"

    def load_style_model(self, style_model_name):
        style_model_path = folder_paths.get_full_path("style_models", style_model_name)
        style_model = comfy.sd.load_style_model(style_model_path)
        return (style_model,)


class StyleModelApply:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", ),
                             "style_model": ("STYLE_MODEL", ),
                             "clip_vision_output": ("CLIP_VISION_OUTPUT", ),
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "apply_stylemodel"

    CATEGORY = "conditioning/style_model"

    def apply_stylemodel(self, clip_vision_output, style_model, conditioning):
        cond = style_model.get_cond(clip_vision_output)
        c = []
        for t in conditioning:
            n = [torch.cat((t[0], cond), dim=1), t[1].copy()]
            c.append(n)
        return (c, )

class unCLIPConditioning:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning": ("CONDITIONING", ),
                             "clip_vision_output": ("CLIP_VISION_OUTPUT", ),
                             "strength": ("FLOAT", {"default": 1.0, "min": -10.0, "max": 10.0, "step": 0.01}),
                             "noise_augmentation": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.01}),
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "apply_adm"

    CATEGORY = "conditioning"

    def apply_adm(self, conditioning, clip_vision_output, strength, noise_augmentation):
        if strength == 0:
            return (conditioning, )

        c = []
        for t in conditioning:
            o = t[1].copy()
            x = (clip_vision_output, strength, noise_augmentation)
            if "adm" in o:
                o["adm"] = o["adm"][:] + [x]
            else:
                o["adm"] = [x]
            n = [t[0], o]
            c.append(n)
        return (c, )

class GLIGENLoader:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "gligen_name": (folder_paths.get_filename_list("gligen"), )}}

    RETURN_TYPES = ("GLIGEN",)
    FUNCTION = "load_gligen"

    CATEGORY = "loaders"

    def load_gligen(self, gligen_name):
        gligen_path = folder_paths.get_full_path("gligen", gligen_name)
        gligen = comfy.sd.load_gligen(gligen_path)
        return (gligen,)

class GLIGENTextBoxApply:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": {"conditioning_to": ("CONDITIONING", ),
                              "clip": ("CLIP", ),
                              "gligen_textbox_model": ("GLIGEN", ),
                              "text": ("STRING", {"multiline": True}),
                              "width": ("INT", {"default": 64, "min": 8, "max": MAX_RESOLUTION, "step": 8}),
                              "height": ("INT", {"default": 64, "min": 8, "max": MAX_RESOLUTION, "step": 8}),
                              "x": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              "y": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                             }}
    RETURN_TYPES = ("CONDITIONING",)
    FUNCTION = "append"

    CATEGORY = "conditioning/gligen"

    def append(self, conditioning_to, clip, gligen_textbox_model, text, width, height, x, y):
        c = []
        cond, cond_pooled = clip.encode_from_tokens(clip.tokenize(text), return_pooled=True)
        for t in conditioning_to:
            n = [t[0], t[1].copy()]
            position_params = [(cond_pooled, height // 8, width // 8, y // 8, x // 8)]
            prev = []
            if "gligen" in n[1]:
                prev = n[1]['gligen'][2]

            n[1]['gligen'] = ("position", gligen_textbox_model, prev + position_params)
            c.append(n)
        return (c, )

class EmptyLatentImage:
    def __init__(self, device="cpu"):
        self.device = device

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "width": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                              "height": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                              "batch_size": ("INT", {"default": 1, "min": 1, "max": 64})}}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "generate"

    CATEGORY = "latent"

    def generate(self, width, height, batch_size=1):
        latent = torch.zeros([batch_size, 4, height // 8, width // 8])
        return ({"samples":latent}, )


class LatentFromBatch:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "batch_index": ("INT", {"default": 0, "min": 0, "max": 63}),
                              "length": ("INT", {"default": 1, "min": 1, "max": 64}),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "frombatch"

    CATEGORY = "latent/batch"

    def frombatch(self, samples, batch_index, length):
        s = samples.copy()
        s_in = samples["samples"]
        batch_index = min(s_in.shape[0] - 1, batch_index)
        length = min(s_in.shape[0] - batch_index, length)
        s["samples"] = s_in[batch_index:batch_index + length].clone()
        if "noise_mask" in samples:
            masks = samples["noise_mask"]
            if masks.shape[0] == 1:
                s["noise_mask"] = masks.clone()
            else:
                if masks.shape[0] < s_in.shape[0]:
                    masks = masks.repeat(math.ceil(s_in.shape[0] / masks.shape[0]), 1, 1, 1)[:s_in.shape[0]]
                s["noise_mask"] = masks[batch_index:batch_index + length].clone()
        if "batch_index" not in s:
            s["batch_index"] = [x for x in range(batch_index, batch_index+length)]
        else:
            s["batch_index"] = samples["batch_index"][batch_index:batch_index + length]
        return (s,)
    
class RepeatLatentBatch:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "amount": ("INT", {"default": 1, "min": 1, "max": 64}),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "repeat"

    CATEGORY = "latent/batch"

    def repeat(self, samples, amount):
        s = samples.copy()
        s_in = samples["samples"]
        
        s["samples"] = s_in.repeat((amount, 1,1,1))
        if "noise_mask" in samples and samples["noise_mask"].shape[0] > 1:
            masks = samples["noise_mask"]
            if masks.shape[0] < s_in.shape[0]:
                masks = masks.repeat(math.ceil(s_in.shape[0] / masks.shape[0]), 1, 1, 1)[:s_in.shape[0]]
            s["noise_mask"] = samples["noise_mask"].repeat((amount, 1,1,1))
        if "batch_index" in s:
            offset = max(s["batch_index"]) - min(s["batch_index"]) + 1
            s["batch_index"] = s["batch_index"] + [x + (i * offset) for i in range(1, amount) for x in s["batch_index"]]
        return (s,)

class LatentUpscale:
    upscale_methods = ["nearest-exact", "bilinear", "area", "bislerp"]
    crop_methods = ["disabled", "center"]

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",), "upscale_method": (s.upscale_methods,),
                              "width": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                              "height": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                              "crop": (s.crop_methods,)}}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "upscale"

    CATEGORY = "latent"

    def upscale(self, samples, upscale_method, width, height, crop):
        s = samples.copy()
        s["samples"] = comfy.utils.common_upscale(samples["samples"], width // 8, height // 8, upscale_method, crop)
        return (s,)

class LatentUpscaleBy:
    upscale_methods = ["nearest-exact", "bilinear", "area", "bislerp"]

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",), "upscale_method": (s.upscale_methods,),
                              "scale_by": ("FLOAT", {"default": 1.5, "min": 0.01, "max": 8.0, "step": 0.01}),}}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "upscale"

    CATEGORY = "latent"

    def upscale(self, samples, upscale_method, scale_by):
        s = samples.copy()
        width = round(samples["samples"].shape[3] * scale_by)
        height = round(samples["samples"].shape[2] * scale_by)
        s["samples"] = comfy.utils.common_upscale(samples["samples"], width, height, upscale_method, "disabled")
        return (s,)

class LatentRotate:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "rotation": (["none", "90 degrees", "180 degrees", "270 degrees"],),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "rotate"

    CATEGORY = "latent/transform"

    def rotate(self, samples, rotation):
        s = samples.copy()
        rotate_by = 0
        if rotation.startswith("90"):
            rotate_by = 1
        elif rotation.startswith("180"):
            rotate_by = 2
        elif rotation.startswith("270"):
            rotate_by = 3

        s["samples"] = torch.rot90(samples["samples"], k=rotate_by, dims=[3, 2])
        return (s,)

class LatentFlip:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "flip_method": (["x-axis: vertically", "y-axis: horizontally"],),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "flip"

    CATEGORY = "latent/transform"

    def flip(self, samples, flip_method):
        s = samples.copy()
        if flip_method.startswith("x"):
            s["samples"] = torch.flip(samples["samples"], dims=[2])
        elif flip_method.startswith("y"):
            s["samples"] = torch.flip(samples["samples"], dims=[3])

        return (s,)

class LatentComposite:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples_to": ("LATENT",),
                              "samples_from": ("LATENT",),
                              "x": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              "y": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              "feather": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "composite"

    CATEGORY = "latent"

    def composite(self, samples_to, samples_from, x, y, composite_method="normal", feather=0):
        x =  x // 8
        y = y // 8
        feather = feather // 8
        samples_out = samples_to.copy()
        s = samples_to["samples"].clone()
        samples_to = samples_to["samples"]
        samples_from = samples_from["samples"]
        if feather == 0:
            s[:,:,y:y+samples_from.shape[2],x:x+samples_from.shape[3]] = samples_from[:,:,:samples_to.shape[2] - y, :samples_to.shape[3] - x]
        else:
            samples_from = samples_from[:,:,:samples_to.shape[2] - y, :samples_to.shape[3] - x]
            mask = torch.ones_like(samples_from)
            for t in range(feather):
                if y != 0:
                    mask[:,:,t:1+t,:] *= ((1.0/feather) * (t + 1))

                if y + samples_from.shape[2] < samples_to.shape[2]:
                    mask[:,:,mask.shape[2] -1 -t: mask.shape[2]-t,:] *= ((1.0/feather) * (t + 1))
                if x != 0:
                    mask[:,:,:,t:1+t] *= ((1.0/feather) * (t + 1))
                if x + samples_from.shape[3] < samples_to.shape[3]:
                    mask[:,:,:,mask.shape[3]- 1 - t: mask.shape[3]- t] *= ((1.0/feather) * (t + 1))
            rev_mask = torch.ones_like(mask) - mask
            s[:,:,y:y+samples_from.shape[2],x:x+samples_from.shape[3]] = samples_from[:,:,:samples_to.shape[2] - y, :samples_to.shape[3] - x] * mask + s[:,:,y:y+samples_from.shape[2],x:x+samples_from.shape[3]] * rev_mask
        samples_out["samples"] = s
        return (samples_out,)

class LatentCrop:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "width": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                              "height": ("INT", {"default": 512, "min": 64, "max": MAX_RESOLUTION, "step": 8}),
                              "x": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              "y": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "crop"

    CATEGORY = "latent/transform"

    def crop(self, samples, width, height, x, y):
        s = samples.copy()
        samples = samples['samples']
        x =  x // 8
        y = y // 8

        #enfonce minimum size of 64
        if x > (samples.shape[3] - 8):
            x = samples.shape[3] - 8
        if y > (samples.shape[2] - 8):
            y = samples.shape[2] - 8

        new_height = height // 8
        new_width = width // 8
        to_x = new_width + x
        to_y = new_height + y
        s['samples'] = samples[:,:,y:to_y, x:to_x]
        return (s,)

class SetLatentNoiseMask:
    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "samples": ("LATENT",),
                              "mask": ("MASK",),
                              }}
    RETURN_TYPES = ("LATENT",)
    FUNCTION = "set_mask"

    CATEGORY = "latent/inpaint"

    def set_mask(self, samples, mask):
        s = samples.copy()
        s["noise_mask"] = mask.reshape((-1, 1, mask.shape[-2], mask.shape[-1]))
        return (s,)


def decode_latent_to_preview_image(previewer, device, preview_format, x0):
    preview_image = previewer.decode_latent_to_preview(device, x0)
    preview_image = ImageOps.contain(preview_image, (MAX_PREVIEW_RESOLUTION, MAX_PREVIEW_RESOLUTION), Image.ANTIALIAS)

    preview_type = 1
    if preview_format == "JPEG":
        preview_type = 1
    elif preview_format == "PNG":
        preview_type = 2

    bytesIO = BytesIO()
    header = struct.pack(">I", preview_type)
    bytesIO.write(header)
    preview_image.save(bytesIO, format=preview_format)
    preview_bytes = bytesIO.getvalue()

    return preview_bytes


def common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent, denoise=1.0, disable_noise=False, start_step=None, last_step=None, force_full_denoise=False):
    device = comfy.model_management.get_torch_device()
    latent_image = latent["samples"]

    if disable_noise:
        noise = torch.zeros(latent_image.size(), dtype=latent_image.dtype, layout=latent_image.layout, device="cpu")
    else:
        batch_inds = latent["batch_index"] if "batch_index" in latent else None
        noise = comfy.sample.prepare_noise(latent_image, seed, batch_inds)

    noise_mask = None
    if "noise_mask" in latent:
        noise_mask = latent["noise_mask"]

    preview_format = "JPEG"
    if preview_format not in ["JPEG", "PNG"]:
        preview_format = "JPEG"

    previewer = None
    if not args.disable_previews:
        # TODO previewer methods
        taesd_encoder_path = folder_paths.get_full_path("taesd", "taesd_encoder.pth")
        taesd_decoder_path = folder_paths.get_full_path("taesd", "taesd_decoder.pth")

        method = args.default_preview_method

        if method == LatentPreviewMethod.Auto:
            method = LatentPreviewMethod.Latent2RGB
            if taesd_encoder_path and taesd_encoder_path:
                method = LatentPreviewMethod.TAESD

        if method == LatentPreviewMethod.TAESD:
            if taesd_encoder_path and taesd_encoder_path:
                taesd = TAESD(taesd_encoder_path, taesd_decoder_path).to(device)
                previewer = TAESDPreviewerImpl(taesd)
            else:
                print("Warning: TAESD previews enabled, but could not find models/taesd/taesd_encoder.pth and models/taesd/taesd_decoder.pth")

        if previewer is None:
            previewer = Latent2RGBPreviewer()

    pbar = comfy.utils.ProgressBar(steps)
    def callback(step, x0, x, total_steps):
        preview_bytes = None
        if previewer:
            preview_bytes = decode_latent_to_preview_image(previewer, device, preview_format, x0)
        pbar.update_absolute(step + 1, total_steps, preview_bytes)

    samples = comfy.sample.sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image,
                                  denoise=denoise, disable_noise=disable_noise, start_step=start_step, last_step=last_step,
                                  force_full_denoise=force_full_denoise, noise_mask=noise_mask, callback=callback)
    out = latent.copy()
    out["samples"] = samples
    return (out, )

class KSampler:
    @classmethod
    def INPUT_TYPES(s):
        return {"required":
                    {"model": ("MODEL",),
                    "seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
                    "steps": ("INT", {"default": 20, "min": 1, "max": 10000}),
                    "cfg": ("FLOAT", {"default": 8.0, "min": 0.0, "max": 100.0}),
                    "sampler_name": (comfy.samplers.KSampler.SAMPLERS, ),
                    "scheduler": (comfy.samplers.KSampler.SCHEDULERS, ),
                    "positive": ("CONDITIONING", ),
                    "negative": ("CONDITIONING", ),
                    "latent_image": ("LATENT", ),
                    "denoise": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.01}),
                     }
                }

    RETURN_TYPES = ("LATENT",)
    FUNCTION = "sample"

    CATEGORY = "sampling"

    def sample(self, model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=1.0):
        return common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise)

class KSamplerAdvanced:
    @classmethod
    def INPUT_TYPES(s):
        return {"required":
                    {"model": ("MODEL",),
                    "add_noise": (["enable", "disable"], ),
                    "noise_seed": ("INT", {"default": 0, "min": 0, "max": 0xffffffffffffffff}),
                    "steps": ("INT", {"default": 20, "min": 1, "max": 10000}),
                    "cfg": ("FLOAT", {"default": 8.0, "min": 0.0, "max": 100.0}),
                    "sampler_name": (comfy.samplers.KSampler.SAMPLERS, ),
                    "scheduler": (comfy.samplers.KSampler.SCHEDULERS, ),
                    "positive": ("CONDITIONING", ),
                    "negative": ("CONDITIONING", ),
                    "latent_image": ("LATENT", ),
                    "start_at_step": ("INT", {"default": 0, "min": 0, "max": 10000}),
                    "end_at_step": ("INT", {"default": 10000, "min": 0, "max": 10000}),
                    "return_with_leftover_noise": (["disable", "enable"], ),
                     }
                }

    RETURN_TYPES = ("LATENT",)
    FUNCTION = "sample"

    CATEGORY = "sampling"

    def sample(self, model, add_noise, noise_seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, start_at_step, end_at_step, return_with_leftover_noise, denoise=1.0):
        force_full_denoise = True
        if return_with_leftover_noise == "enable":
            force_full_denoise = False
        disable_noise = False
        if add_noise == "disable":
            disable_noise = True
        return common_ksampler(model, noise_seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise, disable_noise=disable_noise, start_step=start_at_step, last_step=end_at_step, force_full_denoise=force_full_denoise)

class SaveImage:
    def __init__(self):
        self.output_dir = folder_paths.get_output_directory()
        self.type = "output"

    @classmethod
    def INPUT_TYPES(s):
        return {"required": 
                    {"images": ("IMAGE", ),
                     "filename_prefix": ("STRING", {"default": "ComfyUI"})},
                "hidden": {"prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO"},
                }

    RETURN_TYPES = ()
    FUNCTION = "save_images"

    OUTPUT_NODE = True

    CATEGORY = "image"

    def save_images(self, images, filename_prefix="ComfyUI", prompt=None, extra_pnginfo=None):
        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])
        results = list()
        for image in images:
            i = 255. * image.cpu().numpy()
            img = Image.fromarray(np.clip(i, 0, 255).astype(np.uint8))
            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"
            img.save(os.path.join(full_output_folder, file), pnginfo=metadata, compress_level=4)
            results.append({
                "filename": file,
                "subfolder": subfolder,
                "type": self.type
            })
            counter += 1

        return { "ui": { "images": results } }

class PreviewImage(SaveImage):
    def __init__(self):
        self.output_dir = folder_paths.get_temp_directory()
        self.type = "temp"

    @classmethod
    def INPUT_TYPES(s):
        return {"required":
                    {"images": ("IMAGE", ), },
                "hidden": {"prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO"},
                }

class LoadImage:
    @classmethod
    def INPUT_TYPES(s):
        input_dir = folder_paths.get_input_directory()
        files = [f for f in os.listdir(input_dir) if os.path.isfile(os.path.join(input_dir, f))]
        return {"required":
                    {"image": (sorted(files), )},
                }

    CATEGORY = "image"

    RETURN_TYPES = ("IMAGE", "MASK")
    FUNCTION = "load_image"
    def load_image(self, image):
        image_path = folder_paths.get_annotated_filepath(image)
        i = Image.open(image_path)
        i = ImageOps.exif_transpose(i)
        image = i.convert("RGB")
        image = np.array(image).astype(np.float32) / 255.0
        image = torch.from_numpy(image)[None,]
        if 'A' in i.getbands():
            mask = np.array(i.getchannel('A')).astype(np.float32) / 255.0
            mask = 1. - torch.from_numpy(mask)
        else:
            mask = torch.zeros((64,64), dtype=torch.float32, device="cpu")
        return (image, mask)

    @classmethod
    def IS_CHANGED(s, image):
        image_path = folder_paths.get_annotated_filepath(image)
        m = hashlib.sha256()
        with open(image_path, 'rb') as f:
            m.update(f.read())
        return m.digest().hex()

    @classmethod
    def VALIDATE_INPUTS(s, image):
        if not folder_paths.exists_annotated_filepath(image):
            return "Invalid image file: {}".format(image)

        return True

class LoadImageMask:
    _color_channels = ["alpha", "red", "green", "blue"]
    @classmethod
    def INPUT_TYPES(s):
        input_dir = folder_paths.get_input_directory()
        files = [f for f in os.listdir(input_dir) if os.path.isfile(os.path.join(input_dir, f))]
        return {"required":
                    {"image": (sorted(files), ),
                     "channel": (s._color_channels, ), }
                }

    CATEGORY = "mask"

    RETURN_TYPES = ("MASK",)
    FUNCTION = "load_image"
    def load_image(self, image, channel):
        image_path = folder_paths.get_annotated_filepath(image)
        i = Image.open(image_path)
        i = ImageOps.exif_transpose(i)
        if i.getbands() != ("R", "G", "B", "A"):
            i = i.convert("RGBA")
        mask = None
        c = channel[0].upper()
        if c in i.getbands():
            mask = np.array(i.getchannel(c)).astype(np.float32) / 255.0
            mask = torch.from_numpy(mask)
            if c == 'A':
                mask = 1. - mask
        else:
            mask = torch.zeros((64,64), dtype=torch.float32, device="cpu")
        return (mask,)

    @classmethod
    def IS_CHANGED(s, image, channel):
        image_path = folder_paths.get_annotated_filepath(image)
        m = hashlib.sha256()
        with open(image_path, 'rb') as f:
            m.update(f.read())
        return m.digest().hex()

    @classmethod
    def VALIDATE_INPUTS(s, image, channel):
        if not folder_paths.exists_annotated_filepath(image):
            return "Invalid image file: {}".format(image)

        if channel not in s._color_channels:
            return "Invalid color channel: {}".format(channel)

        return True

class ImageScale:
    upscale_methods = ["nearest-exact", "bilinear", "area"]
    crop_methods = ["disabled", "center"]

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "image": ("IMAGE",), "upscale_method": (s.upscale_methods,),
                              "width": ("INT", {"default": 512, "min": 1, "max": MAX_RESOLUTION, "step": 1}),
                              "height": ("INT", {"default": 512, "min": 1, "max": MAX_RESOLUTION, "step": 1}),
                              "crop": (s.crop_methods,)}}
    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "upscale"

    CATEGORY = "image/upscaling"

    def upscale(self, image, upscale_method, width, height, crop):
        samples = image.movedim(-1,1)
        s = comfy.utils.common_upscale(samples, width, height, upscale_method, crop)
        s = s.movedim(1,-1)
        return (s,)

class ImageInvert:

    @classmethod
    def INPUT_TYPES(s):
        return {"required": { "image": ("IMAGE",)}}

    RETURN_TYPES = ("IMAGE",)
    FUNCTION = "invert"

    CATEGORY = "image"

    def invert(self, image):
        s = 1.0 - image
        return (s,)


class ImagePadForOutpaint:

    @classmethod
    def INPUT_TYPES(s):
        return {
            "required": {
                "image": ("IMAGE",),
                "left": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                "top": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                "right": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                "bottom": ("INT", {"default": 0, "min": 0, "max": MAX_RESOLUTION, "step": 8}),
                "feathering": ("INT", {"default": 40, "min": 0, "max": MAX_RESOLUTION, "step": 1}),
            }
        }

    RETURN_TYPES = ("IMAGE", "MASK")
    FUNCTION = "expand_image"

    CATEGORY = "image"

    def expand_image(self, image, left, top, right, bottom, feathering):
        d1, d2, d3, d4 = image.size()

        new_image = torch.zeros(
            (d1, d2 + top + bottom, d3 + left + right, d4),
            dtype=torch.float32,
        )
        new_image[:, top:top + d2, left:left + d3, :] = image

        mask = torch.ones(
            (d2 + top + bottom, d3 + left + right),
            dtype=torch.float32,
        )

        t = torch.zeros(
            (d2, d3),
            dtype=torch.float32
        )

        if feathering > 0 and feathering * 2 < d2 and feathering * 2 < d3:

            for i in range(d2):
                for j in range(d3):
                    dt = i if top != 0 else d2
                    db = d2 - i if bottom != 0 else d2

                    dl = j if left != 0 else d3
                    dr = d3 - j if right != 0 else d3

                    d = min(dt, db, dl, dr)

                    if d >= feathering:
                        continue

                    v = (feathering - d) / feathering

                    t[i, j] = v * v

        mask[top:top + d2, left:left + d3] = t

        return (new_image, mask)


NODE_CLASS_MAPPINGS = {
    "KSampler": KSampler,
    "CheckpointLoaderSimple": CheckpointLoaderSimple,
    "CLIPTextEncode": CLIPTextEncode,
    "CLIPSetLastLayer": CLIPSetLastLayer,
    "VAEDecode": VAEDecode,
    "VAEEncode": VAEEncode,
    "VAEEncodeForInpaint": VAEEncodeForInpaint,
    "VAELoader": VAELoader,
    "EmptyLatentImage": EmptyLatentImage,
    "LatentUpscale": LatentUpscale,
    "LatentUpscaleBy": LatentUpscaleBy,
    "LatentFromBatch": LatentFromBatch,
    "RepeatLatentBatch": RepeatLatentBatch,
    "SaveImage": SaveImage,
    "PreviewImage": PreviewImage,
    "LoadImage": LoadImage,
    "LoadImageMask": LoadImageMask,
    "ImageScale": ImageScale,
    "ImageInvert": ImageInvert,
    "ImagePadForOutpaint": ImagePadForOutpaint,
    "ConditioningAverage ": ConditioningAverage ,
    "ConditioningCombine": ConditioningCombine,
    "ConditioningSetArea": ConditioningSetArea,
    "ConditioningSetMask": ConditioningSetMask,
    "KSamplerAdvanced": KSamplerAdvanced,
    "SetLatentNoiseMask": SetLatentNoiseMask,
    "LatentComposite": LatentComposite,
    "LatentRotate": LatentRotate,
    "LatentFlip": LatentFlip,
    "LatentCrop": LatentCrop,
    "LoraLoader": LoraLoader,
    "CLIPLoader": CLIPLoader,
    "CLIPVisionEncode": CLIPVisionEncode,
    "StyleModelApply": StyleModelApply,
    "unCLIPConditioning": unCLIPConditioning,
    "ControlNetApply": ControlNetApply,
    "ControlNetLoader": ControlNetLoader,
    "DiffControlNetLoader": DiffControlNetLoader,
    "StyleModelLoader": StyleModelLoader,
    "CLIPVisionLoader": CLIPVisionLoader,
    "VAEDecodeTiled": VAEDecodeTiled,
    "VAEEncodeTiled": VAEEncodeTiled,
    "TomePatchModel": TomePatchModel,
    "unCLIPCheckpointLoader": unCLIPCheckpointLoader,
    "GLIGENLoader": GLIGENLoader,
    "GLIGENTextBoxApply": GLIGENTextBoxApply,

    "CheckpointLoader": CheckpointLoader,
    "DiffusersLoader": DiffusersLoader,

    "LoadLatent": LoadLatent,
    "SaveLatent": SaveLatent
}

NODE_DISPLAY_NAME_MAPPINGS = {
    # Sampling
    "KSampler": "KSampler",
    "KSamplerAdvanced": "KSampler (Advanced)",
    # Loaders
    "CheckpointLoader": "Load Checkpoint (With Config)",
    "CheckpointLoaderSimple": "Load Checkpoint",
    "VAELoader": "Load VAE",
    "LoraLoader": "Load LoRA",
    "CLIPLoader": "Load CLIP",
    "ControlNetLoader": "Load ControlNet Model",
    "DiffControlNetLoader": "Load ControlNet Model (diff)",
    "StyleModelLoader": "Load Style Model",
    "CLIPVisionLoader": "Load CLIP Vision",
    "UpscaleModelLoader": "Load Upscale Model",
    # Conditioning
    "CLIPVisionEncode": "CLIP Vision Encode",
    "StyleModelApply": "Apply Style Model",
    "CLIPTextEncode": "CLIP Text Encode (Prompt)",
    "CLIPSetLastLayer": "CLIP Set Last Layer",
    "ConditioningCombine": "Conditioning (Combine)",
    "ConditioningAverage ": "Conditioning (Average)",
    "ConditioningSetArea": "Conditioning (Set Area)",
    "ConditioningSetMask": "Conditioning (Set Mask)",
    "ControlNetApply": "Apply ControlNet",
    # Latent
    "VAEEncodeForInpaint": "VAE Encode (for Inpainting)",
    "SetLatentNoiseMask": "Set Latent Noise Mask",
    "VAEDecode": "VAE Decode",
    "VAEEncode": "VAE Encode",
    "LatentRotate": "Rotate Latent",
    "LatentFlip": "Flip Latent",
    "LatentCrop": "Crop Latent",
    "EmptyLatentImage": "Empty Latent Image",
    "LatentUpscale": "Upscale Latent",
    "LatentUpscaleBy": "Upscale Latent By",
    "LatentComposite": "Latent Composite",
    "LatentFromBatch" : "Latent From Batch",
    "RepeatLatentBatch": "Repeat Latent Batch",
    # Image
    "SaveImage": "Save Image",
    "PreviewImage": "Preview Image",
    "LoadImage": "Load Image",
    "LoadImageMask": "Load Image (as Mask)",
    "ImageScale": "Upscale Image",
    "ImageUpscaleWithModel": "Upscale Image (using Model)",
    "ImageInvert": "Invert Image",
    "ImagePadForOutpaint": "Pad Image for Outpainting",
    # _for_testing
    "VAEDecodeTiled": "VAE Decode (Tiled)",
    "VAEEncodeTiled": "VAE Encode (Tiled)",
}

def load_custom_node(module_path):
    module_name = os.path.basename(module_path)
    if os.path.isfile(module_path):
        sp = os.path.splitext(module_path)
        module_name = sp[0]
    try:
        if os.path.isfile(module_path):
            module_spec = importlib.util.spec_from_file_location(module_name, module_path)
        else:
            module_spec = importlib.util.spec_from_file_location(module_name, os.path.join(module_path, "__init__.py"))
        module = importlib.util.module_from_spec(module_spec)
        sys.modules[module_name] = module
        module_spec.loader.exec_module(module)
        if hasattr(module, "NODE_CLASS_MAPPINGS") and getattr(module, "NODE_CLASS_MAPPINGS") is not None:
            NODE_CLASS_MAPPINGS.update(module.NODE_CLASS_MAPPINGS)
            if hasattr(module, "NODE_DISPLAY_NAME_MAPPINGS") and getattr(module, "NODE_DISPLAY_NAME_MAPPINGS") is not None:
                NODE_DISPLAY_NAME_MAPPINGS.update(module.NODE_DISPLAY_NAME_MAPPINGS)
            return True
        else:
            print(f"Skip {module_path} module for custom nodes due to the lack of NODE_CLASS_MAPPINGS.")
            return False
    except Exception as e:
        print(traceback.format_exc())
        print(f"Cannot import {module_path} module for custom nodes:", e)
        return False

def load_custom_nodes():
    node_paths = folder_paths.get_folder_paths("custom_nodes")
    node_import_times = []
    for custom_node_path in node_paths:
        possible_modules = os.listdir(custom_node_path)
        if "__pycache__" in possible_modules:
            possible_modules.remove("__pycache__")

        for possible_module in possible_modules:
            module_path = os.path.join(custom_node_path, possible_module)
            if os.path.isfile(module_path) and os.path.splitext(module_path)[1] != ".py": continue
            if module_path.endswith(".disabled"): continue
            time_before = time.perf_counter()
            success = load_custom_node(module_path)
            node_import_times.append((time.perf_counter() - time_before, module_path, success))

    if len(node_import_times) > 0:
        print("\nImport times for custom nodes:")
        for n in sorted(node_import_times):
            if n[2]:
                import_message = ""
            else:
                import_message = " (IMPORT FAILED)"
            print("{:6.1f} seconds{}:".format(n[0], import_message), n[1])
        print()

def init_custom_nodes():
    load_custom_node(os.path.join(os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy_extras"), "nodes_hypernetwork.py"))
    load_custom_node(os.path.join(os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy_extras"), "nodes_upscale_model.py"))
    load_custom_node(os.path.join(os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy_extras"), "nodes_post_processing.py"))
    load_custom_node(os.path.join(os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy_extras"), "nodes_mask.py"))
    load_custom_node(os.path.join(os.path.join(os.path.dirname(os.path.realpath(__file__)), "comfy_extras"), "nodes_rebatch.py"))
    load_custom_nodes()