Any plans of adapting V3 Schema?

[legacyforker]

some of the comfyscripts are starting to have incompatibility issues when calling some of the new classes like TextEncodeQwenImageEditPlus

import node_helpers
import comfy.utils
import math
from typing_extensions import override
from comfy_api.latest import ComfyExtension, io


class TextEncodeQwenImageEdit(io.ComfyNode):
    @classmethod
    def define_schema(cls):
        return io.Schema(
            node_id="TextEncodeQwenImageEdit",
            category="advanced/conditioning",
            inputs=[
                io.Clip.Input("clip"),
                io.String.Input("prompt", multiline=True, dynamic_prompts=True),
                io.Vae.Input("vae", optional=True),
                io.Image.Input("image", optional=True),
            ],
            outputs=[
                io.Conditioning.Output(),
            ],
        )

    @classmethod
    def execute(cls, clip, prompt, vae=None, image=None) -> io.NodeOutput:
        ref_latent = None
        if image is None:
            images = []
        else:
            samples = image.movedim(-1, 1)
            total = int(1024 * 1024)

            scale_by = math.sqrt(total / (samples.shape[3] * samples.shape[2]))
            width = round(samples.shape[3] * scale_by)
            height = round(samples.shape[2] * scale_by)

            s = comfy.utils.common_upscale(samples, width, height, "area", "disabled")
            image = s.movedim(1, -1)
            images = [image[:, :, :, :3]]
            if vae is not None:
                ref_latent = vae.encode(image[:, :, :, :3])

        tokens = clip.tokenize(prompt, images=images)
        conditioning = clip.encode_from_tokens_scheduled(tokens)
        if ref_latent is not None:
            conditioning = node_helpers.conditioning_set_values(conditioning, {"reference_latents": [ref_latent]}, append=True)
        return io.NodeOutput(conditioning)


class TextEncodeQwenImageEditPlus(io.ComfyNode):
    @classmethod
    def define_schema(cls):
        return io.Schema(
            node_id="TextEncodeQwenImageEditPlus",
            category="advanced/conditioning",
            inputs=[
                io.Clip.Input("clip"),
                io.String.Input("prompt", multiline=True, dynamic_prompts=True),
                io.Vae.Input("vae", optional=True),
                io.Image.Input("image1", optional=True),
                io.Image.Input("image2", optional=True),
                io.Image.Input("image3", optional=True),
            ],
            outputs=[
                io.Conditioning.Output(),
            ],
        )

    @classmethod
    def execute(cls, clip, prompt, vae=None, image1=None, image2=None, image3=None) -> io.NodeOutput:
        ref_latents = []
        images = [image1, image2, image3]
        images_vl = []
        llama_template = "<|im_start|>system\nDescribe the key features of the input image (color, shape, size, texture, objects, background), then explain how the user's text instruction should alter or modify the image. Generate a new image that meets the user's requirements while maintaining consistency with the original input where appropriate.<|im_end|>\n<|im_start|>user\n{}<|im_end|>\n<|im_start|>assistant\n"
        image_prompt = ""

        for i, image in enumerate(images):
            if image is not None:
                samples = image.movedim(-1, 1)
                total = int(384 * 384)

                scale_by = math.sqrt(total / (samples.shape[3] * samples.shape[2]))
                width = round(samples.shape[3] * scale_by)
                height = round(samples.shape[2] * scale_by)

                s = comfy.utils.common_upscale(samples, width, height, "area", "disabled")
                images_vl.append(s.movedim(1, -1))
                if vae is not None:
                    total = int(1024 * 1024)
                    scale_by = math.sqrt(total / (samples.shape[3] * samples.shape[2]))
                    width = round(samples.shape[3] * scale_by / 8.0) * 8
                    height = round(samples.shape[2] * scale_by / 8.0) * 8

                    s = comfy.utils.common_upscale(samples, width, height, "area", "disabled")
                    ref_latents.append(vae.encode(s.movedim(1, -1)[:, :, :, :3]))

                image_prompt += "Picture {}: <|vision_start|><|image_pad|><|vision_end|>".format(i + 1)

        tokens = clip.tokenize(image_prompt + prompt, images=images_vl, llama_template=llama_template)
        conditioning = clip.encode_from_tokens_scheduled(tokens)
        if len(ref_latents) > 0:
            conditioning = node_helpers.conditioning_set_values(conditioning, {"reference_latents": ref_latents}, append=True)
        return io.NodeOutput(conditioning)


class QwenExtension(ComfyExtension):
    @override
    async def get_node_list(self) -> list[type[io.ComfyNode]]:
        return [
            TextEncodeQwenImageEdit,
            TextEncodeQwenImageEditPlus,
        ]


async def comfy_entrypoint() -> QwenExtension:
    return QwenExtension()

[Chaoses-Ib]

And specifically what incompatibility issues? Only real mode or both?

[legacyforker]

from comfy_script.runtime.real import load
load()

Others have the same issue too so like instead of returning a tensor, it is just comfy_api generator

Comfy-Org/ComfyUI@160698e

[Chaoses-Ib]

So the "incompatibility issues" you mean is some nodes returning comfy_api output instead of a torch tensor?

There is no guarantee that nodes will return tensors from the beginning. Real mode just returns what the node returns. So this change is actually the intended behavior.

You can get the tensor via the result prop:

[Chaoses-Ib]

Oh, sorry, I forgot about the input part. These outputs should be unpacked when passing in real mode nodes, otherwise it's both imcompatible and cumbersome to fix.

[Chaoses-Ib]

Should be fixed now. You can run git pull to update.

[svjack]

@Chaoses-Ib Hi when I use transpiler i get

from comfy_script.runtime import *
load()
from comfy_script.runtime.nodes import *

with Workflow():
    camera_embedding, width, height, length = WanCameraEmbedding('Pan Up', 768, 1024, 81, 1, 0.5, 0.5, 0.5, 0.5)
    

It seems like due to v3 schema issue, this can not be run, and yield the following

Nodes: 495
Traceback (most recent call last):
  File "/home/featurize/run_test.py", line 10, in <module>
    camera_embedding, width, height, length = WanCameraEmbedding('Pan Up', 768, 1024, 81, 1, 0.5, 0.5, 0.5, 0.5)
                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: NodeOutput.__init__() takes 4 positional arguments but 10 were given

How can I solve this ?

[Chaoses-Ib]

this can not be run, and yield the following

Probably not related to v3 schema but a name conflict. Have you tried WanCameraEmbedding_()?

[svjack]

this can not be run, and yield the following

Probably not related to v3 schema but a name conflict. Have you tried WanCameraEmbedding_()?

I use WanCameraEmbedding('Pan Up', 768, 1024) this not yield error in python shell

in https://github.com/comfyanonymous/ComfyUI/blob/v0.3.62/comfy_extras/nodes_camera_trajectory.py
have

class WanCameraEmbedding(io.ComfyNode):
    @classmethod
    def define_schema(cls):
        return io.Schema(
            node_id="WanCameraEmbedding",
            category="camera",
            inputs=[
                io.Combo.Input(
                    "camera_pose",
                    options=[
                        "Static",
                        "Pan Up",
                        "Pan Down",
                        "Pan Left",
                        "Pan Right",
                        "Zoom In",
                        "Zoom Out",
                        "Anti Clockwise (ACW)",
                        "ClockWise (CW)",
                    ],
                    default="Static",
                ),
                io.Int.Input("width", default=832, min=16, max=nodes.MAX_RESOLUTION, step=16),
                io.Int.Input("height", default=480, min=16, max=nodes.MAX_RESOLUTION, step=16),
                io.Int.Input("length", default=81, min=1, max=nodes.MAX_RESOLUTION, step=4),
                io.Float.Input("speed", default=1.0, min=0, max=10.0, step=0.1, optional=True),
                io.Float.Input("fx", default=0.5, min=0, max=1, step=0.000000001, optional=True),
                io.Float.Input("fy", default=0.5, min=0, max=1, step=0.000000001, optional=True),
                io.Float.Input("cx", default=0.5, min=0, max=1, step=0.01, optional=True),
                io.Float.Input("cy", default=0.5, min=0, max=1, step=0.01, optional=True),
            ],
            outputs=[
                io.WanCameraEmbedding.Output(display_name="camera_embedding"),
                io.Int.Output(display_name="width"),
                io.Int.Output(display_name="height"),
                io.Int.Output(display_name="length"),
            ],
        )

in https://github.com/comfyanonymous/ComfyUI/blob/v0.3.59/comfy_extras/nodes_camera_trajectory.py
have

class WanCameraEmbedding:
    @classmethod
    def INPUT_TYPES(cls):
        return {
            "required": {
                "camera_pose":(["Static","Pan Up","Pan Down","Pan Left","Pan Right","Zoom In","Zoom Out","Anti Clockwise (ACW)", "ClockWise (CW)"],{"default":"Static"}),
                "width": ("INT", {"default": 832, "min": 16, "max": MAX_RESOLUTION, "step": 16}),
                "height": ("INT", {"default": 480, "min": 16, "max": MAX_RESOLUTION, "step": 16}),
                "length": ("INT", {"default": 81, "min": 1, "max": MAX_RESOLUTION, "step": 4}),
            },
            "optional":{
                "speed":("FLOAT",{"default":1.0, "min": 0, "max": 10.0, "step": 0.1}),
                "fx":("FLOAT",{"default":0.5, "min": 0, "max": 1, "step": 0.000000001}),
                "fy":("FLOAT",{"default":0.5, "min": 0, "max": 1, "step": 0.000000001}),
                "cx":("FLOAT",{"default":0.5, "min": 0, "max": 1, "step": 0.01}),
                "cy":("FLOAT",{"default":0.5, "min": 0, "max": 1, "step": 0.01}),
            }

        }

    RETURN_TYPES = ("WAN_CAMERA_EMBEDDING","INT","INT","INT")
    RETURN_NAMES = ("camera_embedding","width","height","length")
    FUNCTION = "run"
    CATEGORY = "camera"

does this related with your code change ?

[Chaoses-Ib]

So have you tried WanCameraEmbedding_() or not?