Swarmui_Lightning/symb.py

# setup_civitai.py
import os
import subprocess
from pathlib import Path

# ---------- configuración ----------
TOKEN_CIVITAI = "4487c60c8a71cbfe100a7a03ad9e6ae9"
STORE_FILE    = Path.home() / ".civitai_token.pkl"   # sustituto de %store
# -----------------------------------

def _store(var: str, val: str) -> None:
    """Simula %store var val (persistente entre kernels)."""
    import pickle, builtins
    # escribe en el archivo
    data = {var: val}
    STORE_FILE.write_bytes(pickle.dumps(data))
    # opcional: también lo deja disponible en builtins para este kernel
    setattr(builtins, var, val)
    print(f"✅ Variable '{var}' guardada en {STORE_FILE} (persistente).")

def _run(cmd: str) -> None:
    """Ejecuta un comando shell y muestra la salida."""
    print(f"+ {cmd}")
    subprocess.run(cmd, shell=True, check=False)

def preparar():
    """Función principal: guarda el token y crea enlaces."""
    if not TOKEN_CIVITAI.strip():
        print("⚠️ No se ingresó token.")
        return
    _store("token_civitai", TOKEN_CIVITAI)

    # limpieza y enlaces
    cmds = [
        "rm -rf /teamspace/studios/this_studio/.cache",
        "rm -rf /teamspace/studios/this_studio/tmp ~/tmp",
        "ln -vs /tmp ~/tmp",
        "rm -rf /teamspace/studios/this_studio/SwarmUI/Models/Stable-Diffusion/tmp_models",
        "mkdir -p /tmp/models",
        "ln -vs /tmp/models /teamspace/studios/this_studio/SwarmUI/Models/Stable-Diffusion/tmp_models",
        "rm -rf /teamspace/studios/this_studio/SwarmUI/Models/Lora/tmp_lora",
        "mkdir -p /tmp/lora",
        "ln -vs /tmp/lora /teamspace/studios/this_studio/SwarmUI/Models/Lora/tmp_lora",
        "rm -rf /teamspace/studios/this_studio/SwarmUI/Models/diffusion_models",
        "mkdir -p /tmp/diffusion_models",
        "ln -vs /tmp/diffusion_models /teamspace/studios/this_studio/SwarmUI/Models/diffusion_models",
        "rm -rf /teamspace/studios/this_studio/SwarmUI/Models/text_encoders",
        "mkdir -p /tmp/text_encoders",
        "ln -vs /tmp/text_encoders /teamspace/studios/this_studio/SwarmUI/Models/text_encoders",
        "rm -rf /teamspace/studios/this_studio/SwarmUI/Models/unet",
        "mkdir -p /tmp/unet",
        "ln -vs /tmp/unet /teamspace/studios/this_studio/SwarmUI/Models/unet",
        "rm -rf /teamspace/studios/this_studio/SwarmUI/Models/controlnet",
        "mkdir -p /tmp/controlnet",
        "ln -vs /tmp/controlnet /teamspace/studios/this_studio/SwarmUI/Models/controlnet",
    ]
    for c in cmds:
        _run(c)
    print("🎉 Entorno listo.")

if __name__ == "__main__":
    preparar()