Error Message while executing new image load

[gonzalu]

Description

Apologies for double posting

I am getting the following error when trying to parse any image:

Error occurred when executing SDPromptReader:

invalid literal for int() with base 10: 'None'

File "E:\ComfyUI\ComfyUI\execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\ComfyUI\ComfyUI\execution.py", line 82, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\ComfyUI\ComfyUI\execution.py", line 75, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\ComfyUI\ComfyUI\custom_nodes\comfyui-prompt-reader-node\nodes.py", line 161, in load_image
seed = int(
^^^^

and a sample image. Have tried a few and all fail. Was working earlier this week.

Thank you!

Reproduction steps

1. load the node
2. Drag+Drop image on to node
3. Press Queue Prompt

Image file

[gonzalu]

Looks like there was a change in ComfyUI main that broke a number of custom nodes. Maybe this is related.

https://github.com/comfyanonymous/ComfyUI/commit/c2cb8e889b6a6d45eba26a534bf57aff86b320eb

[receyuki]

Due to custom nodes and complex workflows potentially causing issues with SD Prompt Reader's ability to read image metadata correctly, it is recommended to embed the Prompt Saver Node within the workflow to ensure maximum compatibility.

Perhaps I should put this sentence in the most conspicuous place in the readme.

[gonzalu]

Ah ok, my bad if I did not RTFM :D

Let me go play and report back... so sorry if I missed an important step!

Thanks for the wonderful node.

[gonzalu]

OK, so using your example workflow, I get this error...

I deleted the custom node completely and reinstalled via Manager.

I then deleted it again and installed it manually. Ensured all requirements were met. Still not working.

Just let me know if it is me doing something wrong. Happy to correct anything I may be doing now correctly. All I know is that the SD Prompt Reader used to work up until a few days ago regardless of what image I loaded ;)

Thanks again,

[receyuki]

Can you post this ComfyUI01796(2).png here

[gonzalu]

[gonzalu]

I am not sure that is the name of the file I drag and dropped in the node but...

here is what it looks like after I hit the Queue Prompt button after closing the error

[gonzalu]

Every time I drop the file in the node, the node increases the number in parenthesis... same file.

[receyuki]

You might have misunderstood what I meant. ComfyUI doesn't store metadata but only the complete workflow. In order to make the prompt reader node correctly read the metadata, I need to support each custom node individually, which is a never-ending task. Therefore, I've created the prompt saver node.

To ensure the prompt reader or other tools can read the metadata of the image, you need to save the image through the prompt saver node. Therefore, you need to embed the prompt saver node and the parameter generator node into your workflow to generate images that can be read by any tool..

[gonzalu]

Ah gotcha... no worries.

My bad ... I literally thought the node would read ANY metadata and parse at a minimum the PROSITIVE/NEGATIVE and whatever it can find... Looks like the PNG that breaks for me is saved with some other custom node that places the entire workflow in a different spot in the PNG.

No worries... When your node does pick up the data from a file, it is FANTASTIC. I have come to rely on it :D hence why I was frustrated trying to figure out what I was doing wrong.

As always, MANY MANY thanks and I appreciate you... really wonderful nodes :D

[receyuki]

No worries. Feel free to ask me any questions :)

[gonzalu]

Is there a way the node can gracefully keep going if the image is not supported instead of erroring out? It currently freezes the workflow if a bad image is loaded. It would be great if it just collects the filename, dimensions and the image itself. I use the image output for img2image and currently did not want to use yet another image loader if this single one can do both for me :D I don;t always need the settings and positive/negative.

Thanks much!

Cheers,

[receyuki]

I'm currently planning to add more user-friendly warnings.