Merge branch 'main' into fix-tiffNode/feature-outputNode

2026-01-29 15:23:43 -05:00
parent 7fee4c9575 938185cd42
commit faa6a499bf
15 changed files with 212 additions and 21 deletions
@@ -72,6 +72,7 @@ from .modules.settings import RunwareSettings
 from .modules.videoAdvancedFeatureInputs import videoAdvancedFeatureInputs
 from .modules.advancedFeatureSettings.wanAnimateAdvancedFeatureSettings import RunwareWanAnimateAdvancedFeatureSettings
 from .modules.saveImage import RunwareSaveImage
+from .modules.videoOutputs import RunwareVideoInferenceOutputs

 RUNWARE_COMFYUI_VERSION = "1.4.0 Beta"

@@ -158,6 +159,7 @@ NODE_CLASS_MAPPINGS = {
    "Runware Video Advanced Feature Inputs": videoAdvancedFeatureInputs,
    "Runware Wan Animate Advanced Feature Settings": RunwareWanAnimateAdvancedFeatureSettings,
    "Runware Save Image": RunwareSaveImage,
+    "Runware Video Inference Outputs": RunwareVideoInferenceOutputs,
 }

 NODE_DISPLAY_NAME_MAPPINGS = {
@@ -1,6 +1,6 @@
 import { app } from "../../scripts/app.js";
 import { api } from "../../scripts/api.js";
-import { promptEnhanceHandler, syncDimensionsNodeHandler, searchNodeHandler, APIKeyHandler, captionNodeHandler, mediaUUIDHandler, videoTranscriptionHandler, handleCustomErrors, videoInferenceDimensionsHandler, videoModelSearchFilterHandler, audioModelSearchFilterHandler, useParameterToggleHandler, imageInferenceToggleHandler, upscalerToggleHandler, videoUpscalerToggleHandler, audioInferenceToggleHandler, acceleratorOptionsToggleHandler, bytedanceProviderSettingsToggleHandler, openaiProviderSettingsToggleHandler, lightricksProviderSettingsToggleHandler, klingProviderSettingsToggleHandler, lumaProviderSettingsToggleHandler, briaProviderSettingsToggleHandler, pixverseProviderSettingsToggleHandler, alibabaProviderSettingsToggleHandler, mireloProviderSettingsToggleHandler, googleProviderSettingsToggleHandler, syncProviderSettingsToggleHandler, syncSegmentToggleHandler, settingsToggleHandler, audioInputToggleHandler, speechInputToggleHandler, briaProviderMaskToggleHandler, wanAnimateAdvancedFeatureSettingsToggleHandler, videoAdvancedFeatureInputsToggleHandler, audioInferenceInputsToggleHandler } from "./utils.js";
+import { promptEnhanceHandler, syncDimensionsNodeHandler, searchNodeHandler, APIKeyHandler, captionNodeHandler, mediaUUIDHandler, videoTranscriptionHandler, videoOutputsHandler, handleCustomErrors, videoInferenceDimensionsHandler, videoModelSearchFilterHandler, audioModelSearchFilterHandler, useParameterToggleHandler, imageInferenceToggleHandler, upscalerToggleHandler, videoUpscalerToggleHandler, audioInferenceToggleHandler, acceleratorOptionsToggleHandler, bytedanceProviderSettingsToggleHandler, openaiProviderSettingsToggleHandler, lightricksProviderSettingsToggleHandler, klingProviderSettingsToggleHandler, lumaProviderSettingsToggleHandler, briaProviderSettingsToggleHandler, pixverseProviderSettingsToggleHandler, alibabaProviderSettingsToggleHandler, mireloProviderSettingsToggleHandler, googleProviderSettingsToggleHandler, syncProviderSettingsToggleHandler, syncSegmentToggleHandler, settingsToggleHandler, audioInputToggleHandler, speechInputToggleHandler, briaProviderMaskToggleHandler, wanAnimateAdvancedFeatureSettingsToggleHandler, videoAdvancedFeatureInputsToggleHandler, audioInferenceInputsToggleHandler } from "./utils.js";
 import { RUNWARE_NODE_TYPES, RUNWARE_NODE_PROPS, SEARCH_TERMS } from "./types.js";

 const nodeInitList = [];
@@ -11,6 +11,7 @@ app.registerExtension({
        api.addEventListener('runwareImageCaption', captionNodeHandler);
        api.addEventListener('runwareMediaUUID', mediaUUIDHandler);
        api.addEventListener('runwareVideoTranscription', videoTranscriptionHandler);
+        api.addEventListener('runwareVideoOutputs', videoOutputsHandler);
    },

    async nodeCreated(node) {
@@ -39,6 +40,18 @@ app.registerExtension({
                mediaUUIDWidget.inputEl.style.fontSize = "11px";
                mediaUUIDWidget.inputEl.style.fontFamily = "monospace";
            }
+        } else if(nodeClass === RUNWARE_NODE_TYPES.VIDEOINFERENCEOUTPUTS) {
+            // Style draftId and videoId widgets to show full IDs (like Media Upload)
+            const draftIdWidget = node.widgets.find(widget => widget.name === "draftId");
+            const videoIdWidget = node.widgets.find(widget => widget.name === "videoId");
+            for(const w of [draftIdWidget, videoIdWidget]) {
+                if(w && w.inputEl) {
+                    w.inputEl.style.width = "100%";
+                    w.inputEl.style.minWidth = "300px";
+                    w.inputEl.style.fontSize = "11px";
+                    w.inputEl.style.fontFamily = "monospace";
+                }
+            }
        }

        const nodeWidgets = node.widgets;
@@ -122,6 +122,7 @@ const RUNWARE_NODE_TYPES = {
    VIDEOADVANCEDFEATUREINPUTS: "Runware Video Advanced Feature Inputs",
    WANANIMATEADVANCEDFEATURESETTINGS: "Runware Wan Animate Advanced Feature Settings",
    SAVEIMAGE: "Runware Save Image",
+    VIDEOINFERENCEOUTPUTS: "Runware Video Inference Outputs",
 };

 const RUNWARE_NODE_PROPS = {
@@ -425,6 +426,10 @@ const RUNWARE_NODE_PROPS = {
        bgColor: DEFAULT_BGCOLOR,
        colorModeOnly: true,
    },
+    [RUNWARE_NODE_TYPES.VIDEOINFERENCEOUTPUTS]: {
+        bgColor: DEFAULT_BGCOLOR,
+        colorModeOnly: true,
+    },
 };

 export {
@@ -106,6 +106,44 @@ function videoTranscriptionHandler(msgEvent) {
    return false;
 }

+function videoOutputsHandler(msgEvent) {
+    const outputData = msgEvent.detail;
+    const draftId = outputData.draftId || "";
+    const videoId = outputData.videoId || "";
+    const nodeID = outputData.nodeID;
+    
+    if(outputData.success && nodeID !== undefined && nodeID !== null) {
+        const nodeIdInt = typeof nodeID === 'string' ? parseInt(nodeID) : nodeID;
+        const outputNode = app.graph.getNodeById(nodeIdInt);
+        
+        if(outputNode !== null && outputNode !== undefined) {
+            const draftIdWidget = outputNode.widgets.find(widget => widget.name === "draftId");
+            const videoIdWidget = outputNode.widgets.find(widget => widget.name === "videoId");
+            
+            if(draftIdWidget) {
+                if(draftIdWidget.value !== undefined) draftIdWidget.value = draftId;
+                if(draftIdWidget.inputEl) {
+                    draftIdWidget.inputEl.value = draftId;
+                    draftIdWidget.inputEl.dispatchEvent(new Event('input', { bubbles: true }));
+                    draftIdWidget.inputEl.dispatchEvent(new Event('change', { bubbles: true }));
+                }
+            }
+            
+            if(videoIdWidget) {
+                if(videoIdWidget.value !== undefined) videoIdWidget.value = videoId;
+                if(videoIdWidget.inputEl) {
+                    videoIdWidget.inputEl.value = videoId;
+                    videoIdWidget.inputEl.dispatchEvent(new Event('input', { bubbles: true }));
+                    videoIdWidget.inputEl.dispatchEvent(new Event('change', { bubbles: true }));
+                }
+            }
+            
+            outputNode.setDirtyCanvas(true);
+        }
+    }
+    return false;
+}
+
 function notifyUser(message, type="info", title = "Runware", life = 4.5) {
    app.extensionManager.toast.add({
        severity: type, // 'info', 'success', 'warn', 'error' \\
@@ -1131,6 +1169,8 @@ function bytedanceProviderSettingsToggleHandler(bytedanceNode) {
    const fastModeWidget = bytedanceNode.widgets.find(w => w.name === "fastMode");
    const useAudioWidget = bytedanceNode.widgets.find(w => w.name === "useAudio");
    const audioWidget = bytedanceNode.widgets.find(w => w.name === "audio");
+    const useDraftWidget = bytedanceNode.widgets.find(w => w.name === "useDraft");
+    const draftWidget = bytedanceNode.widgets.find(w => w.name === "draft");
    
    // Helper function to toggle widget enabled state (exact same pattern)
    function toggleWidgetState(useWidget, paramWidget, paramName) {
@@ -1189,6 +1229,10 @@ function bytedanceProviderSettingsToggleHandler(bytedanceNode) {
    if (useAudioWidget && audioWidget) {
        toggleWidgetState(useAudioWidget, audioWidget, "audio");
    }
+
+    if (useDraftWidget && draftWidget) {
+        toggleWidgetState(useDraftWidget, draftWidget, "draft");
+    }
 }

 function openaiProviderSettingsToggleHandler(openaiNode) {
@@ -1715,7 +1759,7 @@ function videoModelSearchFilterHandler(videoModelSearchNode) {
        ],
        "PixVerse": [
            "pixverse:1@1 (PixVerse v3.5)", "pixverse:1@2 (PixVerse v4)",
-            "pixverse:1@3 (PixVerse v4.5)", "pixverse:1@5-fast (PixVerse v5 Fast)", "pixverse:1@6 (PixVerse v5.5)", "pixverse:lipsync@1 (PixVerse LipSync)",
+            "pixverse:1@3 (PixVerse v4.5)", "pixverse:1@5-fast (PixVerse v5 Fast)", "pixverse:1@6 (PixVerse v5.5)", "pixverse:1@7 (PixVerse v5.6)", "pixverse:lipsync@1 (PixVerse LipSync)",
        ],
        "Vidu": [
            "vidu:1@0 (Vidu Q1 Classic)", "vidu:1@1 (Vidu Q1)",
@@ -1801,6 +1845,7 @@ function videoModelSearchFilterHandler(videoModelSearchNode) {
        "pixverse:1@2": {"width": 640, "height": 360},
        "pixverse:1@3": {"width": 640, "height": 360},
        "pixverse:1@6": {"width": 640, "height": 360},
+        "pixverse:1@7": {"width": 640, "height": 360},
        "pixverse:lipsync@1": {"width": 640, "height": 360},
        "vidu:1@0": {"width": 1920, "height": 1080},
        "vidu:1@1": {"width": 1920, "height": 1080},
@@ -1869,6 +1914,7 @@ function videoModelSearchFilterHandler(videoModelSearchNode) {
        "pixverse:1@2": "360p",
        "pixverse:1@3": "360p",
        "pixverse:1@6": "360p",
+        "pixverse:1@7": "360p",
        "pixverse:lipsync@1": "360p",
        "vidu:1@0": "1080p",
        "vidu:1@1": "1080p",
@@ -3149,6 +3195,7 @@ export {
    mediaUUIDHandler,
    captionNodeHandler,
    videoTranscriptionHandler,
+    videoOutputsHandler,
    handleCustomErrors,
    APIKeyHandler,
    videoInferenceDimensionsHandler,
@@ -47,6 +47,14 @@ class RunwareBytedanceProviderSettings:
                    "tooltip": "Enable audio generation support when available.",
                    "default": False,
                }),
+                "useDraft": ("BOOLEAN", {
+                    "tooltip": "Enable to include draft parameter in provider settings",
+                    "default": False,
+                }),
+                "draft": ("BOOLEAN", {
+                    "tooltip": "Enable draft mode for Bytedance provider",
+                    "default": False,
+                }),
            }
        }
    
@@ -63,12 +71,14 @@ class RunwareBytedanceProviderSettings:
        useMaxSequentialImages = kwargs.get("useMaxSequentialImages", False)
        useFastMode = kwargs.get("useFastMode", False)
        useAudio = kwargs.get("useAudio", False)
+        useDraft = kwargs.get("useDraft", False)
        
        # Get value parameters
        cameraFixed = kwargs.get("cameraFixed", False)
        maxSequentialImages = kwargs.get("maxSequentialImages", 1)
        fastMode = kwargs.get("fastMode", False)
        audio = kwargs.get("audio", False)
+        draft = kwargs.get("draft", False)
        
        # Build settings dictionary - only include what is enabled
        settings = {}
@@ -88,6 +98,10 @@ class RunwareBytedanceProviderSettings:
        # Add audio only if useAudio is enabled
        if useAudio:
            settings["audio"] = audio
+
+        # Add draft only if useDraft is enabled
+        if useDraft:
+            settings["draft"] = draft
        
        # Clean up None values
        settings = {k: v for k, v in settings.items() if v is not None}
@@ -317,6 +317,19 @@ def sendVideoTranscription(transcriptionText, nodeID):
        },
    )

+
+def sendVideoOutputs(draftId, videoId, nodeID):
+    """Send draftId and videoId to frontend for display in Video Inference Outputs node textboxes."""
+    PromptServer.instance.send_sync(
+        "runwareVideoOutputs",
+        {
+            "success": True,
+            "draftId": draftId or "",
+            "videoId": videoId or "",
+            "nodeID": nodeID,
+        },
+    )
+
 def inferenecRequest(genConfig):
    global RUNWARE_API_KEY, RUNWARE_API_BASE_URL, SESSION_TIMEOUT
    RUNWARE_API_KEY = os.getenv("RUNWARE_API_KEY")
@@ -125,7 +125,8 @@ class videoBgRemoval:
                    if status == "success":
                        if "videoURL" in videoData or "videoBase64Data" in videoData:
                            videos = rwUtils.convertVideoB64List(pollResult, 1920, 1080)
-                            return videos
+                            # SaveVideo expects a single video object, not a tuple
+                            return (videos[0],) if len(videos) > 0 else (None,)
            
            comfy.model_management.throw_exception_if_processing_interrupted()
            
@@ -141,8 +141,8 @@ class txt2vid:

    DESCRIPTION = "Generates Videos Lightning Fast With Runware Video Inference Engine."
    FUNCTION = "generateVideo"
-    RETURN_TYPES = ("VIDEO", "RUNWARETASK")
-    RETURN_NAMES = ("VIDEO", "RW-Task")
+    RETURN_TYPES = ("VIDEO", "RUNWARETASK", "RUNWARETASK")
+    RETURN_NAMES = ("VIDEO", "RW-Task", "OUTPUT")
    CATEGORY = "Runware"

    def generateVideo(self, **kwargs):
@@ -325,7 +325,7 @@ class txt2vid:
            genConfig[0]["acceleration"] = acceleration

        if (multiInferenceMode):
-            return (None, genConfig)
+            return (None, genConfig, None)
        else:
            try:
                # Debug: Print the request being sent
@@ -388,7 +388,9 @@ class txt2vid:
                        if status == "success":
                            if "videoURL" in video_data or "videoBase64Data" in video_data:
                                videos = rwUtils.convertVideoB64List(pollResult, width, height)
-                                return videos
+                                # SaveVideo expects a single video object, not a tuple
+                                video_output = videos[0] if len(videos) > 0 else None
+                                return (video_output, genConfig, pollResult)
                        
                        # If status is "processing", continue polling
                
@@ -43,6 +43,16 @@ class videoInferenceInputs:
                "Reference Voices": ("RUNWAREREFERENCEVOICES", {
                    "tooltip": "Connect the Reference Voices node to provide reference voice URLs or mediaUUIDs."
                }),
+                "draftId": ("STRING", {
+                    "default": "",
+                    "multiline": False,
+                    "tooltip": "Draft ID from a previous Bytedance video generation. Used when continuing or extending a draft output",
+                }),
+                "videoId": ("STRING", {
+                    "default": "",
+                    "multiline": False,
+                    "tooltip": "Video ID from a previous generation. Used when extending or editing an existing OpenAI Sora video",
+                }),
            }
        }

@@ -65,6 +75,8 @@ class videoInferenceInputs:
        references = kwargs.get("References", None)
        referenceVideos = kwargs.get("Reference Videos", None)
        referenceVoices = kwargs.get("Reference Voices", None)
+        draftId = kwargs.get("draftId", "")
+        videoId = kwargs.get("videoId", "")

        inputs = {}

@@ -112,4 +124,10 @@ class videoInferenceInputs:
        if referenceVoices is not None and len(referenceVoices) > 0:
            inputs["referenceVoices"] = referenceVoices

+        if draftId is not None and draftId.strip() != "":
+            inputs["draftId"] = draftId.strip()
+
+        if videoId is not None and videoId.strip() != "":
+            inputs["videoId"] = videoId.strip()
+
        return (inputs,)
@@ -47,6 +47,7 @@ class videoModelSearch:
            "pixverse:1@3 (PixVerse v4.5)",
            "pixverse:1@5-fast (PixVerse v5 Fast)",
            "pixverse:1@6 (PixVerse v5.5)",
+            "pixverse:1@7 (PixVerse v5.6)",
            "pixverse:lipsync@1 (PixVerse LipSync)",
        ],
        "Vidu": [
@@ -148,6 +149,7 @@ class videoModelSearch:
        "pixverse:1@3": {"width": 640, "height": 360},
        "pixverse:1@5-fast": {"width": 640, "height": 360},
        "pixverse:1@6": {"width": 640, "height": 360},
+        "pixverse:1@7": {"width": 640, "height": 360},
        "pixverse:lipsync@1": {"width": 640, "height": 360},
        
        # Vidu Models
@@ -248,6 +250,7 @@ class videoModelSearch:
        "pixverse:1@3": "360p",
        "pixverse:1@5-fast": "360p",
        "pixverse:1@6": "360p",
+        "pixverse:1@7": "360p",
        "pixverse:lipsync@1": "360p",
        
        # Vidu Models
@@ -390,8 +393,8 @@ class videoModelSearch:

    DESCRIPTION = "Directly Search and Connect Video Models to Runware Video Inference Nodes In ComfyUI."
    FUNCTION = "videoModelSearch"
-    RETURN_TYPES = ("RUNWAREVIDEOMODEL", "INT", "INT")
-    RETURN_NAMES = ("Runware Video Model", "Width", "Height")
+    RETURN_TYPES = ("RUNWAREVIDEOMODEL",)
+    RETURN_NAMES = ("Runware Video Model",)
    CATEGORY = "Runware"

    @classmethod
@@ -421,33 +424,27 @@ class videoModelSearch:
            # Use custom dimensions from Width/Height widgets
            width = customWidth
            height = customHeight
-            returnWidth = width
-            returnHeight = height
        elif useCustomDimensions == "Disabled":
-            # Don't pass width/height (use None in dict, but return 0 for INT output)
+            # Don't pass width/height (use None in dict)
            width = None
            height = None
-            returnWidth = 0
-            returnHeight = 0
        else:  # "Model Default"
            # Use model default dimensions
            width = dimensions["width"]
            height = dimensions["height"]
-            returnWidth = width
-            returnHeight = height
-        
+
        resultDict = {
            "model": modelAirCode,
            "useCustomDimensions": useCustomDimensions,
            "width": width,
            "height": height,
        }
-        
+
        # Add resolution if enabled and model supports it
        if useResolution and modelResolution is not None:
            resultDict["resolution"] = resolution
            resultDict["useResolution"] = True
        else:
            resultDict["useResolution"] = False
-        
-        return (resultDict, returnWidth, returnHeight)
+
+        return (resultDict,)
@@ -0,0 +1,75 @@
+from .utils import runwareUtils as rwUtils
+
+
+class RunwareVideoInferenceOutputs:
+    @classmethod
+    def INPUT_TYPES(cls):
+        return {
+            "required": {
+                "Output": (
+                    "RUNWARETASK",
+                    {
+                        "tooltip": "Connect the OUTPUT from Runware Video Inference.",
+                    },
+                ),
+            },
+            "optional": {
+                "draftId": ("STRING", {
+                    "placeholder": "Draft ID will appear here after generation.",
+                    "tooltip": "This field will be automatically populated with the draft ID from the video inference output.",
+                }),
+                "videoId": ("STRING", {
+                    "placeholder": "Video ID will appear here after generation.",
+                    "tooltip": "This field will be automatically populated with the video ID from the video inference output.",
+                }),
+            },
+            "hidden": {"node_id": "UNIQUE_ID"},
+        }
+
+    DESCRIPTION = "Extracts Runware video inference outputs such as draftId and videoId."
+    FUNCTION = "get_outputs"
+    RETURN_TYPES = ("STRING", "STRING")
+    RETURN_NAMES = ("draftID", "videoID")
+    CATEGORY = "Runware"
+    OUTPUT_NODE = True  # Ensure node executes even if outputs aren't used
+
+    def get_outputs(self, Output, draftId=None, videoId=None, **kwargs):
+        draft_id = ""
+        video_id = ""
+
+        try:
+            data_list = None
+
+            # Handle different possible response shapes defensively
+            if isinstance(Output, dict):
+                if "data" in Output:
+                    data_list = Output.get("data", [])
+                elif "response" in Output and isinstance(Output["response"], dict):
+                    data_list = Output["response"].get("data", [])
+
+            if data_list and len(data_list) > 0:
+                first = data_list[0]
+                outputs = first.get("outputs") or {}
+                if isinstance(outputs, dict):
+                    # draftId is used by some providers
+                    draft_id = outputs.get("draftId", "") or outputs.get("draftID", "")
+                    # videoId / videoID used by other providers
+                    video_id = outputs.get("videoId", "") or outputs.get("videoID", "")
+
+            # Send to frontend so values appear in textboxes (like Media Upload)
+            rwUtils.sendVideoOutputs(draft_id, video_id, kwargs.get("node_id"))
+
+        except Exception as e:
+            print(f"[Runware Video Outputs] Failed to parse outputs: {e}")
+
+        return (draft_id, video_id)
+
+
+NODE_CLASS_MAPPINGS = {
+    "RunwareVideoInferenceOutputs": RunwareVideoInferenceOutputs,
+}
+
+NODE_DISPLAY_NAME_MAPPINGS = {
+    "RunwareVideoInferenceOutputs": "Runware Video Inference Outputs",
+}
+
@@ -121,7 +121,8 @@ class videoUpscaler:
                        if status == "success":
                            if "mediaURL" in video_data or "videoURL" in video_data or "videoBase64Data" in video_data:
                                videos = rwUtils.convertVideoB64List(pollResult, 1920, 1080)
-                                return videos
+                                # SaveVideo expects a single video object, not a tuple
+                                return (videos[0],) if len(videos) > 0 else (None,)
                        
                        # If status is "processing", continue polling