smZ CLIPTextEncode seem easy to get error when the prompt too long or get Other encoded characters or punctuation

[xueqing0622]

smZ CLIPTextEncode seem easy to get error when the prompt too long or get Other encoded characters or punctuation

Error occurred when executing smZ CLIPTextEncode:

tuple index out of range

File "F:\ComfyUI\ComfyUI\execution.py", line 154, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "F:\ComfyUI\ComfyUI\execution.py", line 84, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "F:\ComfyUI\ComfyUI\execution.py", line 77, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "F:\ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\nodes.py", line 87, in encode
result = run(**params)
File "F:\ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 693, in run
cond, pooled = clip_clone.encode_from_tokens(tokens, True)
File "F:\ComfyUI\ComfyUI\comfy\sd.py", line 131, in encode_from_tokens
cond, pooled = self.cond_stage_model.encode_token_weights(tokens)
File "F:\ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 416, in encode_token_weights
if multi: schedules = prompt_parser.get_multicond_learned_conditioning(model_hijack.cond_stage_model, texts, steps, None, opts.use_old_scheduling)
File "F:\ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\modules\prompt_parser.py", line 270, in get_multicond_learned_conditioning
learned_conditioning = get_learned_conditioning(model, prompt_flat_list, steps, hires_steps, use_old_scheduling)
File "F:\ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\modules\prompt_parser.py", line 198, in get_learned_conditioning
conds = model.forward(texts)
File "F:\ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\modules\sd_hijack_clip.py", line 207, in forward
z = self.process_tokens(tokens, multipliers)
File "F:\ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\modules\sd_hijack_clip.py", line 237, in process_tokens
z = self.encode_with_transformers(tokens)
File "F:\ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 223, in encode_with_transformers
return self.encode_with_transformers_comfy_(tokens, return_pooled)
File "F:\ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 124, in encode_with_transformers_comfy_
z, pooled = ClipTextEncoderCustom._forward(self.wrapped, tokens_orig)
File "F:\ComfyUI\ComfyUI\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 113, in _forward
z, pooled_output = self.forward(tokens)
File "F:\ComfyUI\ComfyUI\comfy\sd1_clip.py", line 160, in forward
tokens = self.set_up_textual_embeddings(tokens, backup_embeds)
File "F:\ComfyUI\ComfyUI\comfy\sd1_clip.py", line 131, in set_up_textual_embeddings
if y.shape[0] == current_embeds.weight.shape[1]:

[xueqing0622]

And these are error prompts samplers: And I use in A1111 mode

1:
1gir,black hair,Night,Standing in the meadow,Blue stars,looking at viewers,
(extremely detailed 8k illustration), (extremely detailed and beautiful background:1.2), ultra detailed painting, professional illustrasion, Ultra-precise depiction, Ultra-detailed depiction, (beautiful and aesthetic:1.2), vivid,intricate, nice hands, perfect hands,

2:
highly detailed black sumi-e painting of . in-depth study of perfection, created by a master. best quality, high resolution

[shiimizu]

It seems like an embedding was used? Do you have any embeddings that have the same name as one of the words you used?

[xueqing0622]

shiimizu, you are right! after I remove all the embedding , It works,
How do avoid it happen but I have to keep my embeddings?

[shiimizu]

Make sure your embeddings are renamed to words you don't often use.