Prompt Adapter v2

Documentation

• Class name: SeargePromptAdapterV2
• Category: Searge/UI/Prompting
• Output node: False

The SeargePromptAdapterV2 node is designed to adapt input prompts for further processing or generation tasks, enabling customization and refinement of text inputs to fit specific requirements or formats.

Input types

Required

Optional

• data
  • This input represents a data stream that can include various types of information, serving as a flexible container for inputs that the node can process or transform.
  • Comfy dtype: SRG_DATA_STREAM
  • Python dtype: Dict[str, Any]
• main_prompt
  • Captures the primary text prompt intended for guiding the generation process, focusing on the main subject or theme.
  • Comfy dtype: SRG_PROMPT_TEXT
  • Python dtype: str
• secondary_prompt
  • This input captures additional text prompts that complement the main prompt, offering further context or detail.
  • Comfy dtype: SRG_PROMPT_TEXT
  • Python dtype: str
• style_prompt
  • Specifies a style or tone for the text generation, influencing the mood or character of the output.
  • Comfy dtype: SRG_PROMPT_TEXT
  • Python dtype: str
• negative_main_prompt
  • Captures negative text prompts to avoid certain topics or attributes in the generation process.
  • Comfy dtype: SRG_PROMPT_TEXT
  • Python dtype: str
• negative_secondary_prompt
  • Additional negative prompts that help refine the filtering process by excluding undesired elements.
  • Comfy dtype: SRG_PROMPT_TEXT
  • Python dtype: str
• negative_style_prompt
  • Defines a style or tone to be avoided in the text generation, ensuring the output does not carry undesired characteristics.
  • Comfy dtype: SRG_PROMPT_TEXT
  • Python dtype: str

Output types

• data
  • Comfy dtype: SRG_DATA_STREAM
  • A data stream that includes the processed or transformed inputs, ready for further use or generation tasks.
  • Python dtype: Dict[str, Any]
• prompts
  • Comfy dtype: SRG_DATA_STREAM
  • A collection of prompts, both positive and negative, formatted and ready for use in generation tasks.
  • Python dtype: Dict[str, Any]

Usage tips

• Infra type: CPU
• Common nodes: unknown

Source code

class SeargePromptAdapterV2:
    @classmethod
    def INPUT_TYPES(s):
        return {
            "required": {
            },
            "optional": {
                "data": ("SRG_DATA_STREAM",),
                "main_prompt": ("SRG_PROMPT_TEXT",),
                "secondary_prompt": ("SRG_PROMPT_TEXT",),
                "style_prompt": ("SRG_PROMPT_TEXT",),
                "negative_main_prompt": ("SRG_PROMPT_TEXT",),
                "negative_secondary_prompt": ("SRG_PROMPT_TEXT",),
                "negative_style_prompt": ("SRG_PROMPT_TEXT",),
            },
        }

    RETURN_TYPES = ("SRG_DATA_STREAM", "SRG_DATA_STREAM",)
    RETURN_NAMES = ("data", UI.S_PROMPTS,)
    FUNCTION = "get_value"

    CATEGORY = UI.CATEGORY_UI_PROMPTING

    @staticmethod
    def create_dict(main_prompt=None, secondary_prompt=None, style_prompt=None,
                    negative_main_prompt=None, negative_secondary_prompt=None, negative_style_prompt=None):
        return {
            UI.F_MAIN_PROMPT: main_prompt,
            UI.F_SECONDARY_PROMPT: secondary_prompt,
            UI.F_STYLE_PROMPT: style_prompt,
            UI.F_NEGATIVE_MAIN_PROMPT: negative_main_prompt,
            UI.F_NEGATIVE_SECONDARY_PROMPT: negative_secondary_prompt,
            UI.F_NEGATIVE_STYLE_PROMPT: negative_style_prompt,
        }

    def get_value(self, main_prompt=None, secondary_prompt=None, style_prompt=None,
                  negative_main_prompt=None, negative_secondary_prompt=None, negative_style_prompt=None, data=None):
        if data is None:
            data = {}

        data[UI.S_PROMPTS] = self.create_dict(
            main_prompt,
            secondary_prompt,
            style_prompt,
            negative_main_prompt,
            negative_secondary_prompt,
            negative_style_prompt
        )

        return (data, data[UI.S_PROMPTS],)