Meisoftcoltd
4d18b4e69d
- Replaced existing SequentialLoopTrigger logic. - Added dictionary scanning to intercept the queue POST JSON payload. - Injects a new 32-bit random integer (`0xffffffff` max) into any node possessing `seed` or `noise_seed` in its inputs. - Preserves the sequential auto-queue logic but now guarantees the sampler processes a new generation instead of instantly returning cached results in subsequent loops.
108 lines
3.9 KiB
Python
108 lines
3.9 KiB
Python
import random
|
|
import time
|
|
import urllib.request
|
|
import json
|
|
from . import register_node
|
|
|
|
# Variable global para mantener el estado del bucle entre ejecuciones
|
|
global_loop_index = 0
|
|
|
|
@register_node
|
|
class SequentialLoopStart:
|
|
"""Inicia el bucle y provee el índice actual a los nodos de imagen y vídeo."""
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {
|
|
"required": {
|
|
"reset_loop": ("BOOLEAN", {"default": False}),
|
|
}
|
|
}
|
|
|
|
RETURN_TYPES = ("INT",)
|
|
RETURN_NAMES = ("current_loop_index",)
|
|
FUNCTION = "get_index"
|
|
CATEGORY = "🔁 Sequential Batcher/Loop"
|
|
|
|
@classmethod
|
|
def IS_CHANGED(cls, **kwargs):
|
|
return time.time() # Garantiza 100% una ruptura de caché
|
|
|
|
def get_index(self, reset_loop):
|
|
global global_loop_index
|
|
|
|
# Saneamiento de booleano por si viene de un text input
|
|
is_reset = False
|
|
if isinstance(reset_loop, str):
|
|
is_reset = reset_loop.strip().lower() in ['true', '1', 't', 'y', 'yes']
|
|
else:
|
|
is_reset = bool(reset_loop)
|
|
|
|
if is_reset:
|
|
global_loop_index = 0
|
|
print("🔄 [Sequential Batcher] Loop Start: Bucle reiniciado a 0 manualmente.")
|
|
|
|
print(f"🔄 [Sequential Batcher] Loop Start: Ejecutando ciclo {global_loop_index}.")
|
|
return (global_loop_index, )
|
|
|
|
@register_node
|
|
class SequentialLoopTrigger:
|
|
"""Se ejecuta al final del flujo. Incrementa el contador, muta las semillas y auto-encola el siguiente ciclo."""
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {
|
|
"required": {
|
|
"trigger_dependency": ("*", ),
|
|
"target_loops": ("INT", {"default": 4, "min": 1, "max": 1000, "step": 1}),
|
|
"port": ("INT", {"default": 8188, "min": 1000, "max": 9999, "step": 1}),
|
|
},
|
|
"hidden": {
|
|
"prompt": "PROMPT",
|
|
"extra_pnginfo": "EXTRA_PNGINFO"
|
|
}
|
|
}
|
|
|
|
RETURN_TYPES = ()
|
|
RETURN_NAMES = ()
|
|
OUTPUT_NODE = True
|
|
FUNCTION = "trigger_next"
|
|
CATEGORY = "🔁 Sequential Batcher/Loop"
|
|
|
|
@classmethod
|
|
def IS_CHANGED(cls, **kwargs):
|
|
return time.time() # Garantiza 100% una ruptura de caché en nuestro nodo
|
|
|
|
def trigger_next(self, trigger_dependency, target_loops, port, prompt=None, extra_pnginfo=None):
|
|
global global_loop_index
|
|
|
|
global_loop_index += 1
|
|
|
|
if global_loop_index < target_loops:
|
|
print(f"🚀 [Sequential Batcher] Loop Trigger: Ciclo {global_loop_index}/{target_loops}. Encolando automáticamente el siguiente lote...")
|
|
|
|
# 💥 MAGIA ANTI-CACHÉ: Muta todas las semillas en el lienzo para obligar al Sampler a renderizar
|
|
if prompt is not None:
|
|
for node_id, node_data in prompt.items():
|
|
inputs = node_data.get("inputs", {})
|
|
for key in ["seed", "noise_seed"]:
|
|
if key in inputs and isinstance(inputs[key], (int, float)):
|
|
# Generamos una semilla de 32 bits compatible con Samplers antiguos
|
|
inputs[key] = random.randint(1, 0xffffffff)
|
|
|
|
p = {"prompt": prompt}
|
|
if extra_pnginfo:
|
|
p["extra_data"] = {"extra_pnginfo": extra_pnginfo}
|
|
|
|
data = json.dumps(p).encode('utf-8')
|
|
req = urllib.request.Request(f"http://127.0.0.1:{port}/prompt", data=data, headers={'Content-Type': 'application/json'})
|
|
|
|
try:
|
|
urllib.request.urlopen(req, timeout=5)
|
|
print("✅ [Sequential Batcher] Loop Trigger: Siguiente lote mutado e inyectado con éxito.")
|
|
except Exception as e:
|
|
print(f"❌ [Sequential Batcher] Error al auto-encolar: {e}")
|
|
else:
|
|
print(f"🏁 [Sequential Batcher] Loop Trigger: ¡Generación finalizada! Se han completado los {target_loops} ciclos.")
|
|
global_loop_index = 0
|
|
|
|
return ()
|