270 lines
11 KiB
Python
270 lines
11 KiB
Python
# SPDX-License-Identifier: Apache-2.0
|
|
# Copyright 2025-2026 Raykosan (RaykoStudio)
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
import os
|
|
import json
|
|
import numpy as np
|
|
import re
|
|
from datetime import datetime
|
|
from PIL import Image
|
|
from PIL.PngImagePlugin import PngInfo
|
|
import folder_paths
|
|
import torch
|
|
|
|
|
|
class RS_VAE_Decode_Save:
|
|
def __init__(self):
|
|
self.output_dir = folder_paths.get_output_directory()
|
|
|
|
@classmethod
|
|
def INPUT_TYPES(cls):
|
|
return {
|
|
"required": {
|
|
"samples": ("LATENT",),
|
|
"vae": ("VAE",),
|
|
"filename_prefix": ("STRING", {"default": "ComfyUI"}),
|
|
"format": (["png", "jpg", "webp"], {"default": "png"}),
|
|
},
|
|
"hidden": {
|
|
"prompt": "PROMPT",
|
|
"extra_pnginfo": "EXTRA_PNGINFO",
|
|
},
|
|
}
|
|
|
|
RETURN_TYPES = ("IMAGE",)
|
|
RETURN_NAMES = ("image",)
|
|
FUNCTION = "decode_and_save"
|
|
OUTPUT_NODE = True
|
|
CATEGORY = "🦊 RaykoStudio"
|
|
DESCRIPTION = "Combines native VAE Decode and Save Image into a single node."
|
|
|
|
PNG_COMPRESSION = 1
|
|
JPG_QUALITY = 90
|
|
WEBP_QUALITY = 90
|
|
EMBED_WORKFLOW = True
|
|
OVERWRITE_EXISTING = False
|
|
|
|
@staticmethod
|
|
def _sanitize_path_component(component: str) -> str:
|
|
"""Sanitizes a string to be safe for use in file paths."""
|
|
if not isinstance(component, str):
|
|
component = str(component)
|
|
component = component.replace("\\", "/")
|
|
while "../" in component or "./" in component:
|
|
component = component.replace("../", "").replace("./", "")
|
|
if os.path.isabs(component):
|
|
component = os.path.basename(component)
|
|
component = re.sub(r'[<>:"|?*]', '_', component)
|
|
if not component.strip():
|
|
component = "sanitized_empty"
|
|
return component.strip()
|
|
|
|
@staticmethod
|
|
def _normalize_images(images):
|
|
while images.dim() > 4 and images.shape[0] == 1:
|
|
images = images.squeeze(0)
|
|
if images.dim() == 2:
|
|
images = images.unsqueeze(0).unsqueeze(-1)
|
|
if images.dim() == 3:
|
|
images = images.unsqueeze(0)
|
|
if images.dim() != 4:
|
|
raise ValueError(f"[RS] Unexpected tensor dims: {images.shape}")
|
|
b, d1, d2, d3 = images.shape
|
|
second_is_ch = d1 in (1, 3, 4) and d3 not in (1, 3, 4)
|
|
last_is_ch = d3 in (1, 3, 4) and d1 not in (1, 3, 4)
|
|
if second_is_ch and not last_is_ch:
|
|
images = images.permute(0, 2, 3, 1)
|
|
elif d1 == 1 and d3 > 4:
|
|
images = images.squeeze(1).unsqueeze(-1)
|
|
if images.shape[-1] == 1:
|
|
images = images.repeat(1, 1, 1, 3)
|
|
if images.dtype != torch.float32:
|
|
images = images.float()
|
|
images = torch.clamp(images, 0.0, 1.0)
|
|
return images
|
|
|
|
def _get_next_filename(self, directory, base_name, extension, is_batch, batch_index=None):
|
|
try:
|
|
safe_base_name = self._sanitize_path_component(base_name)
|
|
safe_extension = self._sanitize_path_component(extension).lower()
|
|
|
|
test_path = os.path.join(directory, f"{safe_base_name}.{safe_extension}")
|
|
real_dir = os.path.realpath(directory)
|
|
real_path = os.path.realpath(test_path)
|
|
|
|
if not real_path.startswith(real_dir + os.sep) and real_path != real_dir:
|
|
print(f"[RS] Security Warning: Path traversal detected in '{base_name}'. Falling back to safe name.")
|
|
safe_base_name = "unsafe_input_sanitized"
|
|
|
|
if self.OVERWRITE_EXISTING:
|
|
if is_batch and batch_index is not None:
|
|
return os.path.join(directory, f"{safe_base_name}_{batch_index:03d}.{safe_extension}")
|
|
return os.path.join(directory, f"{safe_base_name}_001.{safe_extension}")
|
|
|
|
if is_batch and batch_index is not None:
|
|
base_with_index = f"{safe_base_name}_{batch_index:03d}"
|
|
pattern = re.compile(rf'^{re.escape(base_with_index)}_(\d+)\.{re.escape(safe_extension)}$')
|
|
else:
|
|
pattern = re.compile(rf'^{re.escape(safe_base_name)}_(\d+)\.{re.escape(safe_extension)}$')
|
|
|
|
max_num = 0
|
|
if os.path.exists(directory):
|
|
for f in os.listdir(directory):
|
|
match = pattern.match(f)
|
|
if match:
|
|
num = int(match.group(1))
|
|
if num > max_num:
|
|
max_num = num
|
|
|
|
next_num = max_num + 1
|
|
if is_batch and batch_index is not None:
|
|
filename = f"{safe_base_name}_{batch_index:03d}_{next_num:03d}.{safe_extension}"
|
|
else:
|
|
filename = f"{safe_base_name}_{next_num:03d}.{safe_extension}"
|
|
|
|
return os.path.join(directory, filename)
|
|
except Exception as e:
|
|
print(f"[RS] Filename error: {e}")
|
|
ts = datetime.now().strftime("%Y%m%d_%H%M%S")
|
|
return os.path.join(directory, f"error_fallback_{ts}.png")
|
|
|
|
def _build_workflow_json(self, prompt, extra_pnginfo):
|
|
try:
|
|
if extra_pnginfo and 'workflow' in extra_pnginfo:
|
|
wf = extra_pnginfo['workflow']
|
|
if 'prompt' not in wf:
|
|
wf['prompt'] = prompt
|
|
return wf
|
|
return {"prompt": prompt or {}, "workflow_info": "Generated by RS Decode Save"}
|
|
except Exception as e:
|
|
print(f"[RS] Workflow JSON error: {e}")
|
|
return {"prompt": prompt or {}}
|
|
|
|
def _save_workflow_json(self, image_path, workflow_data):
|
|
try:
|
|
json_path = os.path.splitext(image_path)[0] + ".json"
|
|
with open(json_path, 'w', encoding='utf-8') as f:
|
|
json.dump(workflow_data, f, ensure_ascii=False, indent=2)
|
|
except Exception as e:
|
|
print(f"[RS] Workflow JSON save error: {e}")
|
|
|
|
def _save_single(self, image_tensor, save_dir, base_filename, fmt,
|
|
prompt, extra_pnginfo, is_batch, batch_index=None):
|
|
try:
|
|
ext_map = {"png": "png", "jpg": "jpg", "webp": "webp"}
|
|
extension = ext_map.get(fmt, "png")
|
|
filepath = self._get_next_filename(save_dir, base_filename, extension, is_batch, batch_index)
|
|
img_array = np.clip(255.0 * image_tensor.cpu().numpy(), 0, 255).astype(np.uint8)
|
|
img = Image.fromarray(img_array)
|
|
save_kwargs = {}
|
|
if fmt == "png":
|
|
save_kwargs["compress_level"] = self.PNG_COMPRESSION
|
|
if self.EMBED_WORKFLOW and prompt:
|
|
metadata = PngInfo()
|
|
metadata.add_text("prompt", json.dumps(prompt))
|
|
if extra_pnginfo:
|
|
for key, value in extra_pnginfo.items():
|
|
try:
|
|
metadata.add_text(key, json.dumps(value))
|
|
except Exception:
|
|
pass
|
|
save_kwargs["pnginfo"] = metadata
|
|
elif fmt == "jpg":
|
|
save_kwargs["quality"] = self.JPG_QUALITY
|
|
save_kwargs["optimize"] = True
|
|
save_kwargs["progressive"] = True
|
|
if self.EMBED_WORKFLOW and (prompt or extra_pnginfo):
|
|
self._save_workflow_json(filepath, self._build_workflow_json(prompt, extra_pnginfo))
|
|
elif fmt == "webp":
|
|
save_kwargs["quality"] = self.WEBP_QUALITY
|
|
save_kwargs["method"] = 4
|
|
if self.EMBED_WORKFLOW and (prompt or extra_pnginfo):
|
|
self._save_workflow_json(filepath, self._build_workflow_json(prompt, extra_pnginfo))
|
|
img.save(filepath, **save_kwargs)
|
|
return filepath
|
|
except Exception as e:
|
|
print(f"[RS] Save error: {e}")
|
|
return ""
|
|
|
|
def decode_and_save(self, samples, vae, filename_prefix, format,
|
|
prompt=None, extra_pnginfo=None):
|
|
images = vae.decode(samples["samples"])
|
|
images = self._normalize_images(images)
|
|
|
|
safe_prefix = self._sanitize_path_component(filename_prefix)
|
|
|
|
full_output_folder, fn_prefix, counter, subfolder, _ = \
|
|
folder_paths.get_save_image_path(safe_prefix, self.output_dir,
|
|
images[0].shape[1], images[0].shape[0])
|
|
|
|
batch_size = images.shape[0]
|
|
is_batch = batch_size > 1
|
|
ext_map = {"png": "png", "jpg": "jpg", "webp": "webp"}
|
|
extension = ext_map.get(format, "png")
|
|
|
|
saved_files = []
|
|
for i in range(batch_size):
|
|
batch_index = i + 1 if is_batch else None
|
|
|
|
file_counter = counter + i
|
|
filename = f"{fn_prefix}_{file_counter:05}.{extension}"
|
|
filepath = os.path.join(full_output_folder, filename)
|
|
|
|
img_array = np.clip(255.0 * images[i].cpu().numpy(), 0, 255).astype(np.uint8)
|
|
img = Image.fromarray(img_array)
|
|
|
|
save_kwargs = {}
|
|
if format == "png":
|
|
save_kwargs["compress_level"] = self.PNG_COMPRESSION
|
|
if self.EMBED_WORKFLOW and prompt:
|
|
metadata = PngInfo()
|
|
metadata.add_text("prompt", json.dumps(prompt))
|
|
if extra_pnginfo:
|
|
for key, value in extra_pnginfo.items():
|
|
try:
|
|
metadata.add_text(key, json.dumps(value))
|
|
except Exception:
|
|
pass
|
|
save_kwargs["pnginfo"] = metadata
|
|
elif format == "jpg":
|
|
save_kwargs["quality"] = self.JPG_QUALITY
|
|
save_kwargs["optimize"] = True
|
|
save_kwargs["progressive"] = True
|
|
if self.EMBED_WORKFLOW and (prompt or extra_pnginfo):
|
|
self._save_workflow_json(filepath, self._build_workflow_json(prompt, extra_pnginfo))
|
|
elif format == "webp":
|
|
save_kwargs["quality"] = self.WEBP_QUALITY
|
|
save_kwargs["method"] = 4
|
|
if self.EMBED_WORKFLOW and (prompt or extra_pnginfo):
|
|
self._save_workflow_json(filepath, self._build_workflow_json(prompt, extra_pnginfo))
|
|
|
|
img.save(filepath, **save_kwargs)
|
|
|
|
saved_files.append({
|
|
"filename": filename,
|
|
"subfolder": subfolder,
|
|
"type": "output"
|
|
})
|
|
|
|
return {"ui": {"images": saved_files}, "result": (images,)}
|
|
|
|
|
|
NODE_CLASS_MAPPINGS = {
|
|
"RS_VAE_Decode_Save": RS_VAE_Decode_Save,
|
|
}
|
|
|
|
NODE_DISPLAY_NAME_MAPPINGS = {
|
|
"RS_VAE_Decode_Save": "🦊 RS Decode Save",
|
|
} |