pipeKSampler: varying results on the same workflow

[diaopal]

• Happens on a highresfix workflow (1 pipeLoader, 2 pipeKSamplers)
• I think it's related to #9
• I think there's also a bug when you set the first pipeKSampler to hold along with your seed set to random, the 2nd pipeKSampler will use a random latent input.

Reproduction

1. Use a fixed seed. Generate an image and its upscaled image. On the 2nd pipeKSampler change its latent upscale to something else like bilinear. Immediately cancel and change it back to your previous latent upscaler, e.g. nearest-exact.
2. The results will be different from your first time generating the upscaled image
3. You can keep repeating this process, constantly going back to nearest-exact or your upscaler of your choice, and it will be different each time.

[TinyTerra]

Please let me know if 0bba92f0942a781f56374b31e8adc1ca10e36794 fixes both of those issues for you

[diaopal]

Thanks! The issue seems to be fixed now, but I get this error if I use pipeEdit:

!!! Exception during processing !!!
Traceback (most recent call last):
  File "C:\Users\anon\Downloads\Programs\ComfyUI_windows_portable\ComfyUI\execution.py", line 151, in recursive_execute
    output_data, output_ui = get_output_data(obj, input_data_all)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\anon\Downloads\Programs\ComfyUI_windows_portable\ComfyUI\execution.py", line 81, in get_output_data
    return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\anon\Downloads\Programs\ComfyUI_windows_portable\ComfyUI\execution.py", line 74, in map_node_over_list
    results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\anon\Downloads\Programs\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_tinyterraNodes\tinyterraNodes.py", line 1533, in flush
    new_model, new_pos, new_neg, new_latent, new_vae, new_clip, new_image, new_seed = pipe['orig'].values()
                                                                                      ~~~~^^^^^^^^
KeyError: 'orig'

[TinyTerra]

Oh of course.. should be fixed in 19fe90db224c0ceaf4ed553a6f6137f700ad2cc0 :)

[diaopal]

Thanks! It's fixed now.