wangbo/ComfyUI
1
0
Fork 0
mirror of https://github.com/comfyanonymous/ComfyUI.git synced 2026-03-05 17:27:42 +08:00
Code Issues Actions 18 Packages Projects Releases Wiki Activity

fix: resolve linting issues in SaveVideo codec options
Some checks failed
Python Linting / Run Ruff (push) Has been cancelled
Details
Python Linting / Run Pylint (push) Has been cancelled
Details

Browse Source 
- Add missing 'logging' import used in speed preset error handling
- Apply ruff format to all PR-changed files

Amp-Thread-ID: https://ampcode.com/threads/T-019ca1cb-0150-7549-8b1b-6713060d3408
This commit is contained in:
bymyself 2026-02-27 17:15:06 -08:00
parent 13a4008735
commit f2952d4634
5 changed files with 232 additions and 91 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
comfy_api/latest/__init__.py
View File

@ -7,7 +7,14 @@ from comfy_api.internal.singleton import ProxiedSingleton
from comfy_api.internal.async_to_sync import create_sync_class from comfy_api.internal.async_to_sync import create_sync_class
from ._input import ImageInput, AudioInput, MaskInput, LatentInput, VideoInput from ._input import ImageInput, AudioInput, MaskInput, LatentInput, VideoInput
from ._input_impl import VideoFromFile, VideoFromComponents from ._input_impl import VideoFromFile, VideoFromComponents
from ._util import VideoCodec, VideoContainer, VideoComponents, VideoSpeedPreset, MESH, VOXEL from ._util import (
VideoCodec,
VideoContainer,
VideoComponents,
VideoSpeedPreset,
MESH,
VOXEL,
)
from . import _io_public as io from . import _io_public as io
from . import _ui_public as ui from . import _ui_public as ui
from comfy_execution.utils import get_executing_context from comfy_execution.utils import get_executing_context
@ -45,7 +52,9 @@ class ComfyAPI_latest(ComfyAPIBase):
raise ValueError("node_id must be provided if not in executing context") raise ValueError("node_id must be provided if not in executing context")
# Convert preview_image to PreviewImageTuple if needed # Convert preview_image to PreviewImageTuple if needed
to_display: PreviewImageTuple | Image.Image | ImageInput | None = preview_image to_display: PreviewImageTuple | Image.Image | ImageInput | None = (
preview_image
)
if to_display is not None: if to_display is not None:
# First convert to PIL Image if needed # First convert to PIL Image if needed
if isinstance(to_display, ImageInput): if isinstance(to_display, ImageInput):
@ -75,6 +84,7 @@ class ComfyAPI_latest(ComfyAPIBase):
execution: Execution execution: Execution
class ComfyExtension(ABC): class ComfyExtension(ABC):
async def on_load(self) -> None: async def on_load(self) -> None:
""" """
@ -88,6 +98,7 @@ class ComfyExtension(ABC):
Returns a list of nodes that this extension provides. Returns a list of nodes that this extension provides.
""" """
class Input: class Input:
Image = ImageInput Image = ImageInput
Audio = AudioInput Audio = AudioInput
@ -95,10 +106,12 @@ class Input:
Latent = LatentInput Latent = LatentInput
Video = VideoInput Video = VideoInput
class InputImpl: class InputImpl:
VideoFromFile = VideoFromFile VideoFromFile = VideoFromFile
VideoFromComponents = VideoFromComponents VideoFromComponents = VideoFromComponents
class Types: class Types:
VideoCodec = VideoCodec VideoCodec = VideoCodec
VideoContainer = VideoContainer VideoContainer = VideoContainer
@ -107,6 +120,7 @@ class Types:
MESH = MESH MESH = MESH
VOXEL = VOXEL VOXEL = VOXEL
ComfyAPI = ComfyAPI_latest ComfyAPI = ComfyAPI_latest
# Create a synchronous version of the API # Create a synchronous version of the API

145
comfy_api/latest/_input_impl/video_types.py
View File

@ -10,7 +10,13 @@ import json
import numpy as np import numpy as np
import math import math
import torch import torch
from .._util import VideoContainer, VideoCodec, VideoComponents, VideoSpeedPreset, quality_to_crf from .._util import (
VideoContainer,
VideoCodec,
VideoComponents,
VideoSpeedPreset,
quality_to_crf,
)
def container_to_output_format(container_format: str | None) -> str | None: def container_to_output_format(container_format: str | None) -> str | None:
@ -82,9 +88,9 @@ class VideoFromFile(VideoInput):
""" """
if isinstance(self.__file, io.BytesIO): if isinstance(self.__file, io.BytesIO):
self.__file.seek(0) # Reset the BytesIO object to the beginning self.__file.seek(0) # Reset the BytesIO object to the beginning
with av.open(self.__file, mode='r') as container: with av.open(self.__file, mode="r") as container:
for stream in container.streams: for stream in container.streams:
if stream.type == 'video': if stream.type == "video":
assert isinstance(stream, av.VideoStream) assert isinstance(stream, av.VideoStream)
return stream.width, stream.height return stream.width, stream.height
raise ValueError(f"No video stream found in file '{self.__file}'") raise ValueError(f"No video stream found in file '{self.__file}'")
@ -138,7 +144,9 @@ class VideoFromFile(VideoInput):
# 2. Try to estimate from duration and average_rate using only metadata # 2. Try to estimate from duration and average_rate using only metadata
if container.duration is not None and video_stream.average_rate: if container.duration is not None and video_stream.average_rate:
duration_seconds = float(container.duration / av.time_base) duration_seconds = float(container.duration / av.time_base)
estimated_frames = int(round(duration_seconds * float(video_stream.average_rate))) estimated_frames = int(
round(duration_seconds * float(video_stream.average_rate))
)
if estimated_frames > 0: if estimated_frames > 0:
return estimated_frames return estimated_frames
@ -148,7 +156,9 @@ class VideoFromFile(VideoInput):
and video_stream.average_rate and video_stream.average_rate
): ):
duration_seconds = float(video_stream.duration * video_stream.time_base) duration_seconds = float(video_stream.duration * video_stream.time_base)
estimated_frames = int(round(duration_seconds * float(video_stream.average_rate))) estimated_frames = int(
round(duration_seconds * float(video_stream.average_rate))
)
if estimated_frames > 0: if estimated_frames > 0:
return estimated_frames return estimated_frames
@ -160,7 +170,9 @@ class VideoFromFile(VideoInput):
frame_count += 1 frame_count += 1
if frame_count == 0: if frame_count == 0:
raise ValueError(f"Could not determine frame count for file '{self.__file}'") raise ValueError(
f"Could not determine frame count for file '{self.__file}'"
)
return frame_count return frame_count
def get_frame_rate(self) -> Fraction: def get_frame_rate(self) -> Fraction:
@ -181,7 +193,9 @@ class VideoFromFile(VideoInput):
if video_stream.frames and container.duration: if video_stream.frames and container.duration:
duration_seconds = float(container.duration / av.time_base) duration_seconds = float(container.duration / av.time_base)
if duration_seconds > 0: if duration_seconds > 0:
return Fraction(video_stream.frames / duration_seconds).limit_denominator() return Fraction(
video_stream.frames / duration_seconds
).limit_denominator()
# Last resort: match get_components_internal default # Last resort: match get_components_internal default
return Fraction(1) return Fraction(1)
@ -195,53 +209,69 @@ class VideoFromFile(VideoInput):
""" """
if isinstance(self.__file, io.BytesIO): if isinstance(self.__file, io.BytesIO):
self.__file.seek(0) self.__file.seek(0)
with av.open(self.__file, mode='r') as container: with av.open(self.__file, mode="r") as container:
return container.format.name return container.format.name
def get_components_internal(self, container: InputContainer) -> VideoComponents: def get_components_internal(self, container: InputContainer) -> VideoComponents:
# Get video frames # Get video frames
frames = [] frames = []
for frame in container.decode(video=0): for frame in container.decode(video=0):
img = frame.to_ndarray(format='rgb24') # shape: (H, W, 3) img = frame.to_ndarray(format="rgb24") # shape: (H, W, 3)
img = torch.from_numpy(img) / 255.0 # shape: (H, W, 3) img = torch.from_numpy(img) / 255.0 # shape: (H, W, 3)
frames.append(img) frames.append(img)
images = torch.stack(frames) if len(frames) > 0 else torch.zeros(0, 3, 0, 0) images = torch.stack(frames) if len(frames) > 0 else torch.zeros(0, 3, 0, 0)
# Get frame rate # Get frame rate
video_stream = next(s for s in container.streams if s.type == 'video') video_stream = next(s for s in container.streams if s.type == "video")
frame_rate = Fraction(video_stream.average_rate) if video_stream and video_stream.average_rate else Fraction(1) frame_rate = (
Fraction(video_stream.average_rate)
if video_stream and video_stream.average_rate
else Fraction(1)
)
# Get audio if available # Get audio if available
audio = None audio = None
try: try:
container.seek(0) # Reset the container to the beginning container.seek(0) # Reset the container to the beginning
for stream in container.streams: for stream in container.streams:
if stream.type != 'audio': if stream.type != "audio":
continue continue
assert isinstance(stream, av.AudioStream) assert isinstance(stream, av.AudioStream)
audio_frames = [] audio_frames = []
for packet in container.demux(stream): for packet in container.demux(stream):
for frame in packet.decode(): for frame in packet.decode():
assert isinstance(frame, av.AudioFrame) assert isinstance(frame, av.AudioFrame)
audio_frames.append(frame.to_ndarray()) # shape: (channels, samples) audio_frames.append(
frame.to_ndarray()
) # shape: (channels, samples)
if len(audio_frames) > 0: if len(audio_frames) > 0:
audio_data = np.concatenate(audio_frames, axis=1) # shape: (channels, total_samples) audio_data = np.concatenate(
audio_tensor = torch.from_numpy(audio_data).unsqueeze(0) # shape: (1, channels, total_samples) audio_frames, axis=1
audio = AudioInput({ ) # shape: (channels, total_samples)
"waveform": audio_tensor, audio_tensor = torch.from_numpy(audio_data).unsqueeze(
"sample_rate": int(stream.sample_rate) if stream.sample_rate else 1, 0
}) ) # shape: (1, channels, total_samples)
audio = AudioInput(
{
"waveform": audio_tensor,
"sample_rate": int(stream.sample_rate)
if stream.sample_rate
else 1,
}
)
except StopIteration: except StopIteration:
pass # No audio stream pass # No audio stream
metadata = container.metadata metadata = container.metadata
return VideoComponents(images=images, audio=audio, frame_rate=frame_rate, metadata=metadata) return VideoComponents(
images=images, audio=audio, frame_rate=frame_rate, metadata=metadata
)
def get_components(self) -> VideoComponents: def get_components(self) -> VideoComponents:
if isinstance(self.__file, io.BytesIO): if isinstance(self.__file, io.BytesIO):
self.__file.seek(0) # Reset the BytesIO object to the beginning self.__file.seek(0) # Reset the BytesIO object to the beginning
with av.open(self.__file, mode='r') as container: with av.open(self.__file, mode="r") as container:
return self.get_components_internal(container) return self.get_components_internal(container)
raise ValueError(f"No video stream found in file '{self.__file}'") raise ValueError(f"No video stream found in file '{self.__file}'")
@ -260,13 +290,23 @@ class VideoFromFile(VideoInput):
): ):
if isinstance(self.__file, io.BytesIO): if isinstance(self.__file, io.BytesIO):
self.__file.seek(0) self.__file.seek(0)
with av.open(self.__file, mode='r') as container: with av.open(self.__file, mode="r") as container:
container_format = container.format.name container_format = container.format.name
video_encoding = container.streams.video[0].codec.name if len(container.streams.video) > 0 else None video_encoding = (
container.streams.video[0].codec.name
if len(container.streams.video) > 0
else None
)
reuse_streams = True reuse_streams = True
if format != VideoContainer.AUTO and format not in container_format.split(","): if format != VideoContainer.AUTO and format not in container_format.split(
","
):
reuse_streams = False reuse_streams = False
if codec != VideoCodec.AUTO and codec != video_encoding and video_encoding is not None: if (
codec != VideoCodec.AUTO
and codec != video_encoding
and video_encoding is not None
):
reuse_streams = False reuse_streams = False
if quality is not None or speed is not None: if quality is not None or speed is not None:
reuse_streams = False reuse_streams = False
@ -309,8 +349,12 @@ class VideoFromFile(VideoInput):
# Add streams to the new container # Add streams to the new container
stream_map = {} stream_map = {}
for stream in streams: for stream in streams:
if isinstance(stream, (av.VideoStream, av.AudioStream, SubtitleStream)): if isinstance(
out_stream = output_container.add_stream_from_template(template=stream, opaque=True) stream, (av.VideoStream, av.AudioStream, SubtitleStream)
):
out_stream = output_container.add_stream_from_template(
template=stream, opaque=True
)
stream_map[stream] = out_stream stream_map[stream] = out_stream
# Write packets to the new container # Write packets to the new container
@ -338,7 +382,7 @@ class VideoFromComponents(VideoInput):
return VideoComponents( return VideoComponents(
images=self.__components.images, images=self.__components.images,
audio=self.__components.audio, audio=self.__components.audio,
frame_rate=self.__components.frame_rate frame_rate=self.__components.frame_rate,
) )
def save_to( def save_to(
@ -399,7 +443,9 @@ class VideoFromComponents(VideoInput):
if resolved_format == VideoContainer.MP4: if resolved_format == VideoContainer.MP4:
container_options["movflags"] = "use_metadata_tags" container_options["movflags"] = "use_metadata_tags"
with av.open(path, mode='w', options=container_options, **extra_kwargs) as output: with av.open(
path, mode="w", options=container_options, **extra_kwargs
) as output:
if metadata is not None: if metadata is not None:
for key, value in metadata.items(): for key, value in metadata.items():
output.metadata[key] = json.dumps(value) output.metadata[key] = json.dumps(value)
@ -409,48 +455,50 @@ class VideoFromComponents(VideoInput):
video_stream.width = self.__components.images.shape[2] video_stream.width = self.__components.images.shape[2]
video_stream.height = self.__components.images.shape[1] video_stream.height = self.__components.images.shape[1]
video_stream.pix_fmt = 'yuv420p' video_stream.pix_fmt = "yuv420p"
if resolved_codec == VideoCodec.VP9: if resolved_codec == VideoCodec.VP9:
video_stream.bit_rate = 0 video_stream.bit_rate = 0
if quality is not None: if quality is not None:
crf = quality_to_crf(quality, ffmpeg_codec) crf = quality_to_crf(quality, ffmpeg_codec)
video_stream.options['crf'] = str(crf) video_stream.options["crf"] = str(crf)
if speed is not None and speed != VideoSpeedPreset.AUTO: if speed is not None and speed != VideoSpeedPreset.AUTO:
if isinstance(speed, str): if isinstance(speed, str):
speed = VideoSpeedPreset(speed) speed = VideoSpeedPreset(speed)
preset = speed.to_ffmpeg_preset(ffmpeg_codec) preset = speed.to_ffmpeg_preset(ffmpeg_codec)
if resolved_codec == VideoCodec.VP9: if resolved_codec == VideoCodec.VP9:
video_stream.options['cpu-used'] = preset video_stream.options["cpu-used"] = preset
else: else:
video_stream.options['preset'] = preset video_stream.options["preset"] = preset
# H.264-specific options # H.264-specific options
if resolved_codec == VideoCodec.H264: if resolved_codec == VideoCodec.H264:
if profile is not None: if profile is not None:
video_stream.options['profile'] = profile video_stream.options["profile"] = profile
if tune is not None: if tune is not None:
video_stream.options['tune'] = tune video_stream.options["tune"] = tune
# VP9-specific options # VP9-specific options
if resolved_codec == VideoCodec.VP9: if resolved_codec == VideoCodec.VP9:
if row_mt: if row_mt:
video_stream.options['row-mt'] = '1' video_stream.options["row-mt"] = "1"
if tile_columns is not None: if tile_columns is not None:
video_stream.options['tile-columns'] = str(tile_columns) video_stream.options["tile-columns"] = str(tile_columns)
audio_sample_rate = 1 audio_sample_rate = 1
audio_stream: Optional[av.AudioStream] = None audio_stream: Optional[av.AudioStream] = None
if self.__components.audio: if self.__components.audio:
audio_sample_rate = int(self.__components.audio['sample_rate']) audio_sample_rate = int(self.__components.audio["sample_rate"])
audio_codec = 'libopus' if resolved_format == VideoContainer.WEBM else 'aac' audio_codec = (
"libopus" if resolved_format == VideoContainer.WEBM else "aac"
)
audio_stream = output.add_stream(audio_codec, rate=audio_sample_rate) audio_stream = output.add_stream(audio_codec, rate=audio_sample_rate)
for i, frame in enumerate(self.__components.images): for i, frame in enumerate(self.__components.images):
img = (frame * 255).clamp(0, 255).byte().cpu().numpy() img = (frame * 255).clamp(0, 255).byte().cpu().numpy()
video_frame = av.VideoFrame.from_ndarray(img, format='rgb24') video_frame = av.VideoFrame.from_ndarray(img, format="rgb24")
video_frame = video_frame.reformat(format='yuv420p') video_frame = video_frame.reformat(format="yuv420p")
packet = video_stream.encode(video_frame) packet = video_stream.encode(video_frame)
output.mux(packet) output.mux(packet)
@ -458,12 +506,19 @@ class VideoFromComponents(VideoInput):
output.mux(packet) output.mux(packet)
if audio_stream and self.__components.audio: if audio_stream and self.__components.audio:
waveform = self.__components.audio['waveform'] waveform = self.__components.audio["waveform"]
waveform = waveform[:, :, :math.ceil((audio_sample_rate / frame_rate) * self.__components.images.shape[0])] waveform = waveform[
:,
:,
: math.ceil(
(audio_sample_rate / frame_rate)
* self.__components.images.shape[0]
),
]
audio_frame = av.AudioFrame.from_ndarray( audio_frame = av.AudioFrame.from_ndarray(
waveform.movedim(2, 1).reshape(1, -1).float().numpy(), waveform.movedim(2, 1).reshape(1, -1).float().numpy(),
format='flt', format="flt",
layout='mono' if waveform.shape[1] == 1 else 'stereo' layout="mono" if waveform.shape[1] == 1 else "stereo",
) )
audio_frame.sample_rate = audio_sample_rate audio_frame.sample_rate = audio_sample_rate
audio_frame.pts = 0 audio_frame.pts = 0

8
comfy_api/latest/_util/__init__.py
View File

@ -1,4 +1,10 @@
from .video_types import VideoContainer, VideoCodec, VideoComponents, VideoSpeedPreset, quality_to_crf from .video_types import (
VideoContainer,
VideoCodec,
VideoComponents,
VideoSpeedPreset,
quality_to_crf,
)
from .geometry_types import VOXEL, MESH from .geometry_types import VOXEL, MESH
from .image_types import SVG from .image_types import SVG

5
comfy_api/latest/_util/video_types.py
View File

@ -5,6 +5,7 @@ from fractions import Fraction
from typing import Optional from typing import Optional
from .._input import ImageInput, AudioInput from .._input import ImageInput, AudioInput
class VideoCodec(str, Enum): class VideoCodec(str, Enum):
AUTO = "auto" AUTO = "auto"
H264 = "h264" H264 = "h264"
@ -46,6 +47,7 @@ class VideoContainer(str, Enum):
class VideoSpeedPreset(str, Enum): class VideoSpeedPreset(str, Enum):
"""Encoding speed presets - slower = better compression at same quality.""" """Encoding speed presets - slower = better compression at same quality."""
AUTO = "auto" AUTO = "auto"
FASTEST = "Fastest" FASTEST = "Fastest"
FAST = "Fast" FAST = "Fast"
@ -104,6 +106,7 @@ def quality_to_crf(quality: int, codec: str = "h264") -> int:
# Default fallback # Default fallback
return 23 return 23
@dataclass @dataclass
class VideoComponents: class VideoComponents:
""" """
@ -114,5 +117,3 @@ class VideoComponents:
frame_rate: Fraction frame_rate: Fraction
audio: Optional[AudioInput] = None audio: Optional[AudioInput] = None
metadata: Optional[dict] = None metadata: Optional[dict] = None

147
comfy_extras/nodes_video.py
View File

@ -1,16 +1,19 @@
from __future__ import annotations from __future__ import annotations
import os
import av
import torch
import folder_paths
import json import json
import logging
import os
from typing import Optional from typing import Optional
import av
import folder_paths
import torch
from typing_extensions import override from typing_extensions import override
from fractions import Fraction from fractions import Fraction
from comfy_api.latest import ComfyExtension, io, ui, Input, InputImpl, Types from comfy_api.latest import ComfyExtension, io, ui, Input, InputImpl, Types
from comfy.cli_args import args from comfy.cli_args import args
class SaveWEBM(io.ComfyNode): class SaveWEBM(io.ComfyNode):
@classmethod @classmethod
def define_schema(cls): def define_schema(cls):
@ -23,7 +26,14 @@ class SaveWEBM(io.ComfyNode):
io.String.Input("filename_prefix", default="ComfyUI"), io.String.Input("filename_prefix", default="ComfyUI"),
io.Combo.Input("codec", options=["vp9", "av1"]), io.Combo.Input("codec", options=["vp9", "av1"]),
io.Float.Input("fps", default=24.0, min=0.01, max=1000.0, step=0.01), io.Float.Input("fps", default=24.0, min=0.01, max=1000.0, step=0.01),
io.Float.Input("crf", default=32.0, min=0, max=63.0, step=1, tooltip="Higher crf means lower quality with a smaller file size, lower crf means higher quality higher filesize."), io.Float.Input(
"crf",
default=32.0,
min=0,
max=63.0,
step=1,
tooltip="Higher crf means lower quality with a smaller file size, lower crf means higher quality higher filesize.",
),
], ],
hidden=[io.Hidden.prompt, io.Hidden.extra_pnginfo], hidden=[io.Hidden.prompt, io.Hidden.extra_pnginfo],
is_output_node=True, is_output_node=True,
@ -31,8 +41,13 @@ class SaveWEBM(io.ComfyNode):
@classmethod @classmethod
def execute(cls, images, codec, fps, filename_prefix, crf) -> io.NodeOutput: def execute(cls, images, codec, fps, filename_prefix, crf) -> io.NodeOutput:
full_output_folder, filename, counter, subfolder, filename_prefix = folder_paths.get_save_image_path( full_output_folder, filename, counter, subfolder, filename_prefix = (
filename_prefix, folder_paths.get_output_directory(), images[0].shape[1], images[0].shape[0] folder_paths.get_save_image_path(
filename_prefix,
folder_paths.get_output_directory(),
images[0].shape[1],
images[0].shape[0],
)
) )
file = f"{filename}_{counter:05}_.webm" file = f"{filename}_{counter:05}_.webm"
@ -46,23 +61,33 @@ class SaveWEBM(io.ComfyNode):
container.metadata[x] = json.dumps(cls.hidden.extra_pnginfo[x]) container.metadata[x] = json.dumps(cls.hidden.extra_pnginfo[x])
codec_map = {"vp9": "libvpx-vp9", "av1": "libsvtav1"} codec_map = {"vp9": "libvpx-vp9", "av1": "libsvtav1"}
stream = container.add_stream(codec_map[codec], rate=Fraction(round(fps * 1000), 1000)) stream = container.add_stream(
codec_map[codec], rate=Fraction(round(fps * 1000), 1000)
)
stream.width = images.shape[-2] stream.width = images.shape[-2]
stream.height = images.shape[-3] stream.height = images.shape[-3]
stream.pix_fmt = "yuv420p10le" if codec == "av1" else "yuv420p" stream.pix_fmt = "yuv420p10le" if codec == "av1" else "yuv420p"
stream.bit_rate = 0 stream.bit_rate = 0
stream.options = {'crf': str(crf)} stream.options = {"crf": str(crf)}
if codec == "av1": if codec == "av1":
stream.options["preset"] = "6" stream.options["preset"] = "6"
for frame in images: for frame in images:
frame = av.VideoFrame.from_ndarray(torch.clamp(frame[..., :3] * 255, min=0, max=255).to(device=torch.device("cpu"), dtype=torch.uint8).numpy(), format="rgb24") frame = av.VideoFrame.from_ndarray(
torch.clamp(frame[..., :3] * 255, min=0, max=255)
.to(device=torch.device("cpu"), dtype=torch.uint8)
.numpy(),
format="rgb24",
)
for packet in stream.encode(frame): for packet in stream.encode(frame):
container.mux(packet) container.mux(packet)
container.mux(stream.encode()) container.mux(stream.encode())
container.close() container.close()
return io.NodeOutput(ui=ui.PreviewVideo([ui.SavedResult(file, subfolder, io.FolderType.output)])) return io.NodeOutput(
ui=ui.PreviewVideo([ui.SavedResult(file, subfolder, io.FolderType.output)])
)
class SaveVideo(io.ComfyNode): class SaveVideo(io.ComfyNode):
@classmethod @classmethod
@ -76,7 +101,7 @@ class SaveVideo(io.ComfyNode):
step=1, step=1,
display_name="Quality", display_name="Quality",
tooltip="Output quality (0-100). Higher = better quality, larger files. " tooltip="Output quality (0-100). Higher = better quality, larger files. "
"Internally maps to CRF: 100→CRF 12, 50→CRF 23, 0→CRF 40.", "Internally maps to CRF: 100→CRF 12, 50→CRF 23, 0→CRF 40.",
) )
h264_speed = io.Combo.Input( h264_speed = io.Combo.Input(
"speed", "speed",
@ -84,7 +109,7 @@ class SaveVideo(io.ComfyNode):
default="auto", default="auto",
display_name="Encoding Speed", display_name="Encoding Speed",
tooltip="Encoding speed preset. Slower = better compression at same quality. " tooltip="Encoding speed preset. Slower = better compression at same quality. "
"Maps to FFmpeg presets: Fastest=ultrafast, Balanced=medium, Best=veryslow.", "Maps to FFmpeg presets: Fastest=ultrafast, Balanced=medium, Best=veryslow.",
) )
h264_profile = io.Combo.Input( h264_profile = io.Combo.Input(
"profile", "profile",
@ -92,16 +117,24 @@ class SaveVideo(io.ComfyNode):
default="auto", default="auto",
display_name="Profile", display_name="Profile",
tooltip="H.264 profile. 'baseline' for max compatibility (older devices), " tooltip="H.264 profile. 'baseline' for max compatibility (older devices), "
"'main' for standard use, 'high' for best quality/compression.", "'main' for standard use, 'high' for best quality/compression.",
advanced=True, advanced=True,
) )
h264_tune = io.Combo.Input( h264_tune = io.Combo.Input(
"tune", "tune",
options=["auto", "film", "animation", "grain", "stillimage", "fastdecode", "zerolatency"], options=[
"auto",
"film",
"animation",
"grain",
"stillimage",
"fastdecode",
"zerolatency",
],
default="auto", default="auto",
display_name="Tune", display_name="Tune",
tooltip="Optimize encoding for specific content types. " tooltip="Optimize encoding for specific content types. "
"'film' for live action, 'animation' for cartoons/anime, 'grain' to preserve film grain.", "'film' for live action, 'animation' for cartoons/anime, 'grain' to preserve film grain.",
advanced=True, advanced=True,
) )
@ -114,7 +147,7 @@ class SaveVideo(io.ComfyNode):
step=1, step=1,
display_name="Quality", display_name="Quality",
tooltip="Output quality (0-100). Higher = better quality, larger files. " tooltip="Output quality (0-100). Higher = better quality, larger files. "
"Internally maps to CRF: 100→CRF 15, 50→CRF 33, 0→CRF 50.", "Internally maps to CRF: 100→CRF 15, 50→CRF 33, 0→CRF 50.",
) )
vp9_speed = io.Combo.Input( vp9_speed = io.Combo.Input(
"speed", "speed",
@ -122,7 +155,7 @@ class SaveVideo(io.ComfyNode):
default="auto", default="auto",
display_name="Encoding Speed", display_name="Encoding Speed",
tooltip="Encoding speed. Slower = better compression. " tooltip="Encoding speed. Slower = better compression. "
"Maps to VP9 cpu-used: Fastest=0, Balanced=2, Best=4.", "Maps to VP9 cpu-used: Fastest=0, Balanced=2, Best=4.",
) )
vp9_row_mt = io.Boolean.Input( vp9_row_mt = io.Boolean.Input(
"row_mt", "row_mt",
@ -137,7 +170,7 @@ class SaveVideo(io.ComfyNode):
default="auto", default="auto",
display_name="Tile Columns", display_name="Tile Columns",
tooltip="Number of tile columns (as power of 2). More tiles = faster encoding " tooltip="Number of tile columns (as power of 2). More tiles = faster encoding "
"but slightly worse compression. 'auto' picks based on resolution.", "but slightly worse compression. 'auto' picks based on resolution.",
advanced=True, advanced=True,
) )
@ -146,28 +179,39 @@ class SaveVideo(io.ComfyNode):
display_name="Save Video", display_name="Save Video",
category="image/video", category="image/video",
description="Saves video to the output directory. " description="Saves video to the output directory. "
"When format/codec/quality differ from source, the video is re-encoded.", "When format/codec/quality differ from source, the video is re-encoded.",
inputs=[ inputs=[
io.Video.Input("video", tooltip="The video to save."), io.Video.Input("video", tooltip="The video to save."),
io.String.Input( io.String.Input(
"filename_prefix", "filename_prefix",
default="video/ComfyUI", default="video/ComfyUI",
tooltip="The prefix for the file to save. " tooltip="The prefix for the file to save. "
"Supports formatting like %date:yyyy-MM-dd%.", "Supports formatting like %date:yyyy-MM-dd%.",
),
io.DynamicCombo.Input(
"codec",
options=[
io.DynamicCombo.Option("auto", []),
io.DynamicCombo.Option(
"h264", [h264_quality, h264_speed, h264_profile, h264_tune]
),
io.DynamicCombo.Option(
"vp9",
[vp9_quality, vp9_speed, vp9_row_mt, vp9_tile_columns],
),
],
tooltip="Video codec. 'auto' preserves source when possible. "
"h264 outputs MP4, vp9 outputs WebM.",
), ),
io.DynamicCombo.Input("codec", options=[
io.DynamicCombo.Option("auto", []),
io.DynamicCombo.Option("h264", [h264_quality, h264_speed, h264_profile, h264_tune]),
io.DynamicCombo.Option("vp9", [vp9_quality, vp9_speed, vp9_row_mt, vp9_tile_columns]),
], tooltip="Video codec. 'auto' preserves source when possible. "
"h264 outputs MP4, vp9 outputs WebM."),
], ],
hidden=[io.Hidden.prompt, io.Hidden.extra_pnginfo], hidden=[io.Hidden.prompt, io.Hidden.extra_pnginfo],
is_output_node=True, is_output_node=True,
) )
@classmethod @classmethod
def execute(cls, video: Input.Video, filename_prefix: str, codec: dict) -> io.NodeOutput: def execute(
cls, video: Input.Video, filename_prefix: str, codec: dict
) -> io.NodeOutput:
selected_codec = codec.get("codec", "auto") selected_codec = codec.get("codec", "auto")
quality = codec.get("quality") quality = codec.get("quality")
speed_str = codec.get("speed", "auto") speed_str = codec.get("speed", "auto")
@ -201,11 +245,10 @@ class SaveVideo(io.ComfyNode):
logging.warning(f"Invalid speed preset '{speed_str}', using default") logging.warning(f"Invalid speed preset '{speed_str}', using default")
width, height = video.get_dimensions() width, height = video.get_dimensions()
full_output_folder, filename, counter, subfolder, filename_prefix = folder_paths.get_save_image_path( full_output_folder, filename, counter, subfolder, filename_prefix = (
filename_prefix, folder_paths.get_save_image_path(
folder_paths.get_output_directory(), filename_prefix, folder_paths.get_output_directory(), width, height
width, )
height
) )
saved_metadata = None saved_metadata = None
@ -233,7 +276,9 @@ class SaveVideo(io.ComfyNode):
tile_columns=int(tile_columns) if tile_columns != "auto" else None, tile_columns=int(tile_columns) if tile_columns != "auto" else None,
) )
return io.NodeOutput(ui=ui.PreviewVideo([ui.SavedResult(file, subfolder, io.FolderType.output)])) return io.NodeOutput(
ui=ui.PreviewVideo([ui.SavedResult(file, subfolder, io.FolderType.output)])
)
class CreateVideo(io.ComfyNode): class CreateVideo(io.ComfyNode):
@ -247,7 +292,9 @@ class CreateVideo(io.ComfyNode):
inputs=[ inputs=[
io.Image.Input("images", tooltip="The images to create a video from."), io.Image.Input("images", tooltip="The images to create a video from."),
io.Float.Input("fps", default=30.0, min=1.0, max=120.0, step=1.0), io.Float.Input("fps", default=30.0, min=1.0, max=120.0, step=1.0),
io.Audio.Input("audio", optional=True, tooltip="The audio to add to the video."), io.Audio.Input(
"audio", optional=True, tooltip="The audio to add to the video."
),
], ],
outputs=[ outputs=[
io.Video.Output(), io.Video.Output(),
@ -255,11 +302,18 @@ class CreateVideo(io.ComfyNode):
) )
@classmethod @classmethod
def execute(cls, images: Input.Image, fps: float, audio: Optional[Input.Audio] = None) -> io.NodeOutput: def execute(
cls, images: Input.Image, fps: float, audio: Optional[Input.Audio] = None
) -> io.NodeOutput:
return io.NodeOutput( return io.NodeOutput(
InputImpl.VideoFromComponents(Types.VideoComponents(images=images, audio=audio, frame_rate=Fraction(fps))) InputImpl.VideoFromComponents(
Types.VideoComponents(
images=images, audio=audio, frame_rate=Fraction(fps)
)
)
) )
class GetVideoComponents(io.ComfyNode): class GetVideoComponents(io.ComfyNode):
@classmethod @classmethod
def define_schema(cls): def define_schema(cls):
@ -269,7 +323,9 @@ class GetVideoComponents(io.ComfyNode):
category="image/video", category="image/video",
description="Extracts all components from a video: frames, audio, and framerate.", description="Extracts all components from a video: frames, audio, and framerate.",
inputs=[ inputs=[
io.Video.Input("video", tooltip="The video to extract components from."), io.Video.Input(
"video", tooltip="The video to extract components from."
),
], ],
outputs=[ outputs=[
io.Image.Output(display_name="images"), io.Image.Output(display_name="images"),
@ -281,21 +337,29 @@ class GetVideoComponents(io.ComfyNode):
@classmethod @classmethod
def execute(cls, video: Input.Video) -> io.NodeOutput: def execute(cls, video: Input.Video) -> io.NodeOutput:
components = video.get_components() components = video.get_components()
return io.NodeOutput(components.images, components.audio, float(components.frame_rate)) return io.NodeOutput(
components.images, components.audio, float(components.frame_rate)
)
class LoadVideo(io.ComfyNode): class LoadVideo(io.ComfyNode):
@classmethod @classmethod
def define_schema(cls): def define_schema(cls):
input_dir = folder_paths.get_input_directory() input_dir = folder_paths.get_input_directory()
files = [f for f in os.listdir(input_dir) if os.path.isfile(os.path.join(input_dir, f))] files = [
f
for f in os.listdir(input_dir)
if os.path.isfile(os.path.join(input_dir, f))
]
files = folder_paths.filter_files_content_types(files, ["video"]) files = folder_paths.filter_files_content_types(files, ["video"])
return io.Schema( return io.Schema(
node_id="LoadVideo", node_id="LoadVideo",
display_name="Load Video", display_name="Load Video",
category="image/video", category="image/video",
inputs=[ inputs=[
io.Combo.Input("file", options=sorted(files), upload=io.UploadType.video), io.Combo.Input(
"file", options=sorted(files), upload=io.UploadType.video
),
], ],
outputs=[ outputs=[
io.Video.Output(), io.Video.Output(),
@ -334,5 +398,6 @@ class VideoExtension(ComfyExtension):
LoadVideo, LoadVideo,
] ]
async def comfy_entrypoint() -> VideoExtension: async def comfy_entrypoint() -> VideoExtension:
return VideoExtension() return VideoExtension()