lerc / canvas_tab Goto Github PK

Running this on a 1366X768 monitor breaks the interface in two places.
1- the color swatches go out of position and do not register as the displayed colors.
2- when a scroll bar appears because of the stretching interface zooming in and out becomes unreliable and confusing at best.

Now, its not an issue for me, I can use another monitor but this is going to be an issue for anyone with smaller laptops or runnning SD over a local network on netbooks, phones, or other kinds of computers which is more common than you would think.

edit- running on OperaGX

-Love the node by the way, fantastic job, even made a tutorial about using it.
https://youtu.be/DROM8vfIYUY :D

Can you add the ability to load an image back into the Canvas so it can be touched up?

Could you please add an option for any image-type node to "Send to Editor Tab"? This option should be available in the contextual menu of any image node.

Currently, the "Send to Editor Tab" node works during execution, but having this option during workflow editing would be also convenient.

• Implementation of all tools inside comfy not an independent app similar to Alekpec Painter node

• make the UI more pleasant "beauty"

• add pressure sensitive to line art as reference see MyPaint brush system are open source

• wen sketch fast the line get polygonised

• it would be nice to have all SD1.5 and SDXL aspect ratio as canvas size

• Your canvas tab have some really nice tool like the mirror, layers and blend mode

• separate the brush size independent from the eraser like this the user don't have to tweak every time

• MyPaint brush engine system: https://github.com/mypaint/libmypaint

• MyPaint App: https://github.com/mypaint/mypaint

• MyPaint Brush: https://github.com/mypaint/mypaint-brushes

• Here is Alekpec custom node : https://github.com/AlekPet/ComfyUI_Custom_Nodes_AlekPet

• Alekpec Painter VS Canvas Tab
  

I can only use computer mouse to draw, can't use my Wacom Tablet because it only allows single dot

Thank you for the canvas node. I am trying to use your workflow and am getting an error. The ImageCompositeMasked is complaining that there is no image. How did you get the onioin in your canvas node? Not sure where to start with your workflow to duplicate your example.

What are those inputs? they don't accept latent image or image, or mask or anything. What do they need? Please update your description BEFORE you update a node

Do you have a video showing how you can send images to different named images on your canvas? I thought I remember there was a tutorial on that previously but I can't find it. Thanks.

Any chance you can add text in this with different fonts? It's good for Depth Leres to make custom text art and etc. as well as QR Code monster, so it would get a ton of use being inside Comfy.

But for the reason I want to use it... lol, I use remove background and a layer node to layer images on a new background and make multiple subjects. I'd really like to be able to generate text and use it within my work flow for my layer setup. If this is possible it would be awesome!

I'm using Canvas_tab in this Workflow currently. https://civitai.com/models/388864/geekyghost-aio-all-in-one-workflow

Is there a way you can create a node or a switch that will prompt/launch the queue to run when the canvas image is updated?

That would truly make it as close to "real time" as possible.

Could you please provide instructions how to get these nodes?
I'm trying to run the Onion workflow, but these two node types are missing.
Thanks.

I already download it but I couldn't find the node. If I use the workflow include Edit in Another Tab, it shows red and tell me install the missing node, but when I go to the missing node page of manager, it turned out and show that I already download it. So strange.

Hi thanks for this plugin, what would make it even better is to have shortcuts for the tools used, specifically the eyedropper tool to quickly sample a color

Hello,

the title is self explanitory, would be nice to have a way of manipulating the pixels quickly, move them around using a move tool kinda like photoshop, i now its a big feature request and would be quite hard to implement. but i think that would make canvas tab even greater, or an easier idea would be to have only a selection tool, that allows you to put in a cache chunk of the images then have a cached item list ui somewhere, where you would be able to apply that cached chunk back to any image canvas via a simple click on that list. both idea have some pro and cond, but both idea would improve your ability to manipulate pixels to work in an iterative kind of workflow.

At the moment, if we have multiple "edit in another tab" nodes, all of them get updated with the same image, if we change something in the canvas editor tab.

It would be much more flexible if each node would send some unique identifier when clicking on "edit", so the canvas tab can only update/change that specific "edit in another tab" node.

For example I would need up to 5 different masks in my workflow, but at the moment all of them get updated at once.

The canvas tab was perfect last time when i did the same for a demo . Now after all the new updates. I am not getting the canvas edits updating from the new canvas window to the edit window , neither the masks nor the base is coming thru.
This results in workflow getting stuck. Is there some way this can be updated as no error message .. it just loops through again and again only. There needs to be a way to ensure the canvas edit coming in to the main window. Also looking forward to a better edit tab where one can load an image as input for canvas tab as dropping it in the tab does not bring it into the canvas editor.

Also the larger workflows are failing due to unknown buffer or some other error cannot get.. can you recheck same and update. Thanks a ton.

This is what happens whenever you do an edit to the image, but it doesn't happen when the image changes come in from the "Send to Editor Tab" control.
The result can be when I'm using this control to make changes to an image, the ComfyUI "Edit in another tab" control can lag behind.
Bonus points if you can have a switch to only trigger the Queue Prompt when this button is pressed.

But even without this feature this is an amazing node.

Or did I not figure out how to do it? Anyway, if it exists, it's not intuitive.

Love this node, by the way!

I was using layers to drop transparent pngs onto a background for compositing - when you resize a layer it loses the image element and cant get back other than through the undo button

i can't seem to get the image to send a new size to canvas , or edit a different size. here's a simple workflow i'm just testing. note how the final image is stuck at 512x512

I downloaded the module, added it to custom_nodes, then added the node, and click "Edit".
It launches "http://127.0.0.1:8188/extensions/canvas_tab/page/index.html" which gives "404: Not Found"
any idea what I'm doing wrong?
Thanks.

I'm glad someone gave this project some attention. I hope it'll energize you to come back & add a few items to the Canvas tab for the use that https://www.youtube.com/watch?v=fotVrleRgi4 highlighted.

Can you add an Aspect Ratio option with some common presets?
1:2 (250x500)
1:3 (166x498)
2:3 (332x498)
3:4 (375x500)
9:16 (279x496)

With a Horiz v. Vert toggle? Then let us reinit the Canvas space?