diff --git a/__init__.py b/__init__.py index 021854b..739332d 100644 --- a/__init__.py +++ b/__init__.py @@ -342,6 +342,47 @@ class ModelLinkerExtension: # Get available models for existence checking available_models = get_model_files() available_paths = {m.get("path") for m in available_models} + # Create lookup for full paths by filename (with and without extension) + path_by_filename = {} + for m in available_models: + rel_path = m.get("relative_path", "") + if rel_path: + filename = rel_path.split("/")[-1].split("\\")[-1] + path_by_filename[filename] = m.get("path") + # Also add without extension for matching (simple approach) + if "." in filename: + filename_no_ext = filename.rsplit(".", 1)[0] + if filename_no_ext not in path_by_filename: + path_by_filename[filename_no_ext] = m.get("path") + # Add the full relative path as key too + path_by_filename[rel_path] = m.get("path") + + # Also use folder_paths.get_full_path() to get paths + import folder_paths + + for cat in [ + "loras", + "checkpoints", + "vae", + "controlnet", + "upscale_models", + ]: + try: + filenames = folder_paths.get_filename_list(cat) + for fn in filenames: + full_path = folder_paths.get_full_path(cat, fn) + if ( + full_path + and full_path not in path_by_filename.values() + ): + path_by_filename[fn] = full_path + fn_no_ext = ( + fn.rsplit(".", 1)[0] if "." in fn else fn + ) + if fn_no_ext not in path_by_filename: + path_by_filename[fn_no_ext] = full_path + except Exception: + pass # Analyze workflow to get all model references all_model_refs = analyze_workflow_models(workflow_json) @@ -416,6 +457,10 @@ class ModelLinkerExtension: "is_lora_v2": ref.get("is_lora_v2", False), "active": ref.get("active"), "connected": ref.get("connected", True), + "resolved_path": ( + path_by_filename.get(model_name) + or path_by_filename.get(original_path) + ), } ) @@ -468,6 +513,134 @@ class ModelLinkerExtension: ) return web.json_response({"error": str(e)}, status=500) + # ==================== CIVITAI SEARCH ROUTE ==================== + + @routes.post("/model_linker/civitai-search") + async def civitai_search(request): + """Search CivitAI for a model using file hash.""" + try: + data = await request.json() + filename = data.get("filename", "") + category = data.get("category", "") + resolved_path = data.get("resolved_path", "") + + if not filename: + return web.json_response( + {"error": "Filename is required"}, status=400 + ) + + # Clean filename for display + import os as _os + + clean_name = _os.path.splitext(filename)[0] + + # Get the file path to hash + file_path = resolved_path if resolved_path else None + + if not file_path and category: + # Try to find the file in the model directories using folder_paths + try: + import folder_paths + + # Map category to folder_paths type + category_map = { + "loras": "loras", + "checkpoints": "checkpoints", + "vae": "vae", + "controlnet": "controlnet", + "upscale_models": "upscale_models", + } + folder_type = category_map.get( + category.lower(), category.lower() + ) + file_path = folder_paths.get_full_path( + folder_type, filename + ) + except Exception: + pass + + # If not found, try scanner + if not file_path: + try: + from .core.scanner import get_model_files + + available_models = get_model_files() + for m in available_models: + if ( + m.get("relative_path", "").endswith(filename) + or m.get("filename", "") == filename + ): + file_path = m.get("path") + break + except Exception: + pass + + # Search CivitAI for the model using hash + if download_available and file_path and _os.path.exists(file_path): + try: + from .core.sources.civitai import ( + get_model_info_for_file, + ) + + result = get_model_info_for_file(file_path) + if result and result.get("url"): + return web.json_response( + { + "filename": filename, + "url": result.get("url"), + "version_url": result.get("version_url"), + "model_id": result.get("model_id"), + "model_name": result.get( + "model_name", clean_name + ), + "model_type": result.get("model_type", ""), + "version_id": result.get("version_id"), + "version_name": result.get("version_name", ""), + "sha256": result.get("sha256"), + "base_model": result.get("base_model"), + "tags": result.get("tags", []), + "trained_words": result.get( + "trained_words", [] + ), + "images": result.get("images", []), + "clip_skip": result.get("clip_skip"), + "description": result.get("description", ""), + "model_description": result.get( + "model_description", "" + ), + } + ) + except Exception as e: + self.logger.warning(f"CivitAI search error: {e}") + + # No result found - try fallback to filename search + if download_available: + try: + from .core.sources.civitai import ( + search_civitai_for_file, + ) + + result = search_civitai_for_file(filename) + if result and result.get("url"): + return web.json_response( + { + "url": result["url"], + "model_name": result.get("name", clean_name), + "version_id": result.get("version_id"), + } + ) + except Exception as e: + self.logger.warning(f"CivitAI fallback search error: {e}") + + # No result found + return web.json_response({"url": None}) + + except Exception as e: + self.logger.error( + f"Model Linker civitai-search error: {e}", exc_info=True + ) + return web.json_response({"error": str(e)}, status=500) + # ==================== DOWNLOAD ROUTES ==================== if download_available: diff --git a/core/sources/civitai.py b/core/sources/civitai.py index c34aaf7..9b9faea 100644 --- a/core/sources/civitai.py +++ b/core/sources/civitai.py @@ -6,7 +6,9 @@ Search and download models from CivitAI. import os import re +import json import logging +import hashlib import requests from typing import Dict, Any, Optional, List from urllib.parse import urlparse, parse_qs, quote @@ -18,6 +20,7 @@ CIVITAI_API_URL = "https://civitai.com/api/v1" # Cache for search results and URN resolutions _search_cache: Dict[str, Any] = {} _urn_cache: Dict[tuple[int, int], Dict[str, Any]] = {} +_hash_cache: Dict[str, Dict[str, Any]] = {} def parse_civitai_url(url: str) -> Optional[Dict[str, Any]]: @@ -25,22 +28,22 @@ def parse_civitai_url(url: str) -> Optional[Dict[str, Any]]: Parse a CivitAI URL to extract model/version info. """ parsed = urlparse(url) - if 'civitai.com' not in parsed.netloc: + if "civitai.com" not in parsed.netloc: return None - - if '/api/download/models/' in parsed.path: - match = re.search(r'/api/download/models/(\d+)', parsed.path) + + if "/api/download/models/" in parsed.path: + match = re.search(r"/api/download/models/(\d+)", parsed.path) if match: - return {'version_id': int(match.group(1))} - - match = re.search(r'/models/(\d+)', parsed.path) + return {"version_id": int(match.group(1))} + + match = re.search(r"/models/(\d+)", parsed.path) if match: - result = {'model_id': int(match.group(1))} + result = {"model_id": int(match.group(1))} query = parse_qs(parsed.query) - if 'modelVersionId' in query: - result['version_id'] = int(query['modelVersionId'][0]) + if "modelVersionId" in query: + result["version_id"] = int(query["modelVersionId"][0]) return result - + return None @@ -59,117 +62,122 @@ def clean_filename_for_search(filename: str) -> str: """ base = os.path.splitext(filename)[0] # Remove common precision/format suffixes - base = re.sub(r'[-_]?(fp16|fp32|fp8|bf16|e4m3fn|scaled|pruned|emaonly|q4|q8).*$', '', base, flags=re.IGNORECASE) + base = re.sub( + r"[-_]?(fp16|fp32|fp8|bf16|e4m3fn|scaled|pruned|emaonly|q4|q8).*$", + "", + base, + flags=re.IGNORECASE, + ) # Remove version numbers at end - base = re.sub(r'[-_]?v?\d+(\.\d+)*$', '', base, flags=re.IGNORECASE) + base = re.sub(r"[-_]?v?\d+(\.\d+)*$", "", base, flags=re.IGNORECASE) return base def search_civitai_for_file( - filename: str, - api_key: Optional[str] = None, - exact_only: bool = False + filename: str, api_key: Optional[str] = None, exact_only: bool = False ) -> Optional[Dict[str, Any]]: """ Search CivitAI for a specific model file. Returns the first model that actually has this exact filename. - + Args: filename: Exact filename to search for api_key: Optional API key exact_only: If True, only return exact filename matches (for downloads). If False, also try partial matching (for local file resolution). - + Returns: Dict with download info if found, None otherwise """ global _search_cache - + cache_key = f"civit_{filename}_exact{exact_only}" if cache_key in _search_cache: return _search_cache[cache_key] - + try: # Clean filename for search search_term = clean_filename_for_search(filename) - + headers = {} if api_key: - headers['Authorization'] = f'Bearer {api_key}' - + headers["Authorization"] = f"Bearer {api_key}" + search_url = f"{CIVITAI_API_URL}/models?query={quote(search_term)}&limit=10" - + response = requests.get(search_url, headers=headers, timeout=15) if response.status_code != 200: logger.debug(f"CivitAI search returned {response.status_code}") return None - + data = response.json() - items = data.get('items', []) - + items = data.get("items", []) + filename_base = os.path.splitext(filename.lower())[0] - + for item in items: - model_id = item.get('id') - model_name = item.get('name', '') - model_type = item.get('type', '') - - model_versions = item.get('modelVersions', []) + model_id = item.get("id") + model_name = item.get("name", "") + model_type = item.get("type", "") + + model_versions = item.get("modelVersions", []) for version in model_versions: - version_id = version.get('id') - files = version.get('files', []) - + version_id = version.get("id") + files = version.get("files", []) + for file_info in files: - file_name = file_info.get('name', '') + file_name = file_info.get("name", "") file_base = os.path.splitext(file_name.lower())[0] - + # Check for exact filename match (case-insensitive) - always try this if file_name.lower() == filename.lower(): - download_url = file_info.get('downloadUrl', '') + download_url = file_info.get("downloadUrl", "") if download_url: result = { - 'source': 'civitai', - 'model_id': model_id, - 'version_id': version_id, - 'name': model_name, - 'type': model_type, - 'filename': file_name, - 'url': f"https://civitai.com/models/{model_id}", - 'download_url': download_url, - 'size': file_info.get('sizeKB', 0) * 1024, - 'match_type': 'exact' + "source": "civitai", + "model_id": model_id, + "version_id": version_id, + "name": model_name, + "type": model_type, + "filename": file_name, + "url": f"https://civitai.com/models/{model_id}", + "download_url": download_url, + "size": file_info.get("sizeKB", 0) * 1024, + "match_type": "exact", } _search_cache[cache_key] = result logger.info(f"Found {filename} on CivitAI: {model_name}") return result - + # Check for partial match (filename_base in file_base or vice versa) # Skip partial matches if exact_only is True - prevents confusing # users with wrong model suggestions for downloads if not exact_only: if filename_base in file_base or file_base in filename_base: - download_url = file_info.get('downloadUrl', '') + download_url = file_info.get("downloadUrl", "") if download_url: result = { - 'source': 'civitai', - 'model_id': model_id, - 'version_id': version_id, - 'name': model_name, - 'type': model_type, - 'filename': file_name, - 'url': f"https://civitai.com/models/{model_id}", - 'download_url': download_url, - 'size': file_info.get('sizeKB', 0) * 1024, - 'match_type': 'partial' + "source": "civitai", + "model_id": model_id, + "version_id": version_id, + "name": model_name, + "type": model_type, + "filename": file_name, + "url": f"https://civitai.com/models/{model_id}", + "download_url": download_url, + "size": file_info.get("sizeKB", 0) * 1024, + "match_type": "partial", } _search_cache[cache_key] = result - logger.info(f"Found similar file for {filename} on CivitAI: {model_name}") + logger.info( + f"Found similar file for {filename} on CivitAI: {model_name}" + ) return result - + # Not found _search_cache[cache_key] = None return None - + except Exception as e: logger.error(f"CivitAI search error: {e}") return None @@ -179,221 +187,638 @@ def search_civitai( query: str, model_type: Optional[str] = None, limit: int = 10, - api_key: Optional[str] = None + api_key: Optional[str] = None, ) -> List[Dict[str, Any]]: """ Search CivitAI for models (general search). Returns models that might be relevant. """ results = [] - + type_map = { - 'checkpoint': 'Checkpoint', - 'checkpoints': 'Checkpoint', - 'lora': 'LORA', - 'loras': 'LORA', - 'vae': 'VAE', - 'controlnet': 'Controlnet', - 'embedding': 'TextualInversion', - 'embeddings': 'TextualInversion', - 'upscaler': 'Upscaler', - 'upscale_models': 'Upscaler', + "checkpoint": "Checkpoint", + "checkpoints": "Checkpoint", + "lora": "LORA", + "loras": "LORA", + "vae": "VAE", + "controlnet": "Controlnet", + "embedding": "TextualInversion", + "embeddings": "TextualInversion", + "upscaler": "Upscaler", + "upscale_models": "Upscaler", } - + try: - params = { - 'query': query, - 'limit': limit, - 'nsfw': 'false' - } - + params = {"query": query, "limit": limit, "nsfw": "false"} + if model_type: civitai_type = type_map.get(model_type.lower()) if civitai_type: - params['types'] = civitai_type - + params["types"] = civitai_type + headers = {} if api_key: - headers['Authorization'] = f'Bearer {api_key}' - + headers["Authorization"] = f"Bearer {api_key}" + response = requests.get( - f"{CIVITAI_API_URL}/models", - params=params, - headers=headers, - timeout=15 + f"{CIVITAI_API_URL}/models", params=params, headers=headers, timeout=15 ) - + if response.status_code == 200: data = response.json() - - for model in data.get('items', []): - model_id = model.get('id') - model_name = model.get('name', '') - model_type = model.get('type', '') - - versions = model.get('modelVersions', []) + + for model in data.get("items", []): + model_id = model.get("id") + model_name = model.get("name", "") + model_type = model.get("type", "") + + versions = model.get("modelVersions", []) if versions: latest = versions[0] - version_id = latest.get('id') - - files = latest.get('files', []) + version_id = latest.get("id") + + files = latest.get("files", []) primary_file = None for f in files: - if f.get('primary', False) or f.get('type') == 'Model': + if f.get("primary", False) or f.get("type") == "Model": primary_file = f break - + if not primary_file and files: primary_file = files[0] - + result = { - 'source': 'civitai', - 'model_id': model_id, - 'version_id': version_id, - 'name': model_name, - 'type': model_type, - 'url': f"https://civitai.com/models/{model_id}", - 'download_url': get_civitai_download_url(version_id, api_key), - 'downloads': model.get('stats', {}).get('downloadCount', 0), + "source": "civitai", + "model_id": model_id, + "version_id": version_id, + "name": model_name, + "type": model_type, + "url": f"https://civitai.com/models/{model_id}", + "download_url": get_civitai_download_url(version_id, api_key), + "downloads": model.get("stats", {}).get("downloadCount", 0), } - + if primary_file: - result['filename'] = primary_file.get('name', '') - result['size'] = primary_file.get('sizeKB', 0) * 1024 - + result["filename"] = primary_file.get("name", "") + result["size"] = primary_file.get("sizeKB", 0) * 1024 + results.append(result) - + except Exception as e: logger.error(f"CivitAI search error: {e}") - + return results def search_civitai_by_hash( - hash_value: str, - api_key: Optional[str] = None + hash_value: str, api_key: Optional[str] = None ) -> Optional[Dict[str, Any]]: """Look up a model by file hash on CivitAI.""" try: headers = {} if api_key: - headers['Authorization'] = f'Bearer {api_key}' - + headers["Authorization"] = f"Bearer {api_key}" + response = requests.get( f"{CIVITAI_API_URL}/model-versions/by-hash/{hash_value}", headers=headers, - timeout=15 + timeout=15, ) - + if response.status_code == 200: data = response.json() - - model_id = data.get('modelId') - version_id = data.get('id') - files = data.get('files', []) + + model_id = data.get("modelId") + version_id = data.get("id") + files = data.get("files", []) primary_file = files[0] if files else {} - + return { - 'source': 'civitai', - 'model_id': model_id, - 'version_id': version_id, - 'name': data.get('model', {}).get('name', ''), - 'url': f"https://civitai.com/models/{model_id}", - 'download_url': get_civitai_download_url(version_id, api_key), - 'filename': primary_file.get('name', ''), - 'size': primary_file.get('sizeKB', 0) * 1024 + "source": "civitai", + "model_id": model_id, + "version_id": version_id, + "name": data.get("model", {}).get("name", ""), + "url": f"https://civitai.com/models/{model_id}", + "download_url": get_civitai_download_url(version_id, api_key), + "filename": primary_file.get("name", ""), + "size": primary_file.get("sizeKB", 0) * 1024, } - + except Exception as e: logger.error(f"CivitAI hash lookup error: {e}") - + return None def resolve_urn( - model_id: int, - version_id: int, - api_key: Optional[str] = None + model_id: int, version_id: int, api_key: Optional[str] = None ) -> Optional[Dict[str, Any]]: """ Resolve URN model_id/version_id to model info and expected filename. - + Args: model_id: CivitAI model ID - version_id: CivitAI version ID + version_id: CivitAI version ID api_key: Optional API key - + Returns: Dict with model name and primary filename, or None """ cache_key = (model_id, version_id) if cache_key in _urn_cache: return _urn_cache[cache_key] - + try: headers = {} if api_key: - headers['Authorization'] = f'Bearer {api_key}' - + headers["Authorization"] = f"Bearer {api_key}" + url = f"{CIVITAI_API_URL}/models/{model_id}" - params = {'modelVersionId': version_id} - + params = {"modelVersionId": version_id} + response = requests.get(url, headers=headers, params=params, timeout=15) - + if response.status_code != 200: logger.warning(f"CivitAI URN resolve failed: {response.status_code}") _urn_cache[cache_key] = None return None - + data = response.json() - versions = data.get('modelVersions', []) - + versions = data.get("modelVersions", []) + if not versions: - logger.warning(f"No versions found for model {model_id}/version {version_id}") + logger.warning( + f"No versions found for model {model_id}/version {version_id}" + ) _urn_cache[cache_key] = None return None - + # Get specific version target_version = None for v in versions: - if v.get('id') == version_id: + if v.get("id") == version_id: target_version = v break - + if not target_version: logger.warning(f"Version {version_id} not found in model {model_id}") _urn_cache[cache_key] = None return None - - files = target_version.get('files', []) + + files = target_version.get("files", []) primary_file = None - + # Prefer primary file or type=='Model' for f in files: - if f.get('primary') or f.get('type') == 'Model': + if f.get("primary") or f.get("type") == "Model": primary_file = f break - + if not primary_file and files: primary_file = files[0] # Fallback to first - + if not primary_file: logger.warning(f"No files found for version {version_id}") _urn_cache[cache_key] = None return None - + result = { - 'model_name': data.get('name', 'Unknown'), - 'version_name': target_version.get('name', 'Unknown'), - 'expected_filename': primary_file.get('name', 'Unknown'), - 'files': [{'name': f.get('name'), 'size': f.get('sizeKB', 0)*1024} for f in files] + "model_name": data.get("name", "Unknown"), + "version_name": target_version.get("name", "Unknown"), + "expected_filename": primary_file.get("name", "Unknown"), + "files": [ + {"name": f.get("name"), "size": f.get("sizeKB", 0) * 1024} + for f in files + ], } - + _urn_cache[cache_key] = result - logger.info(f"Resolved URN model {model_id}@{version_id} → {result['expected_filename']}") + logger.info( + f"Resolved URN model {model_id}@{version_id} → {result['expected_filename']}" + ) return result - + except Exception as e: logger.error(f"CivitAI URN resolve error for {model_id}@{version_id}: {e}") _urn_cache[cache_key] = None return None + + +def _get_sha256_hash(file_path: str) -> Optional[str]: + """ + Compute sha256 hash of a file by reading it in chunks. + + Args: + file_path: Full path to the file + + Returns: + SHA256 hash as hex string, or None if file doesn't exist + """ + if not file_path or not os.path.exists(file_path): + return None + + BUF_SIZE = 1024 * 128 # 128KB chunks + sha256_hash = hashlib.sha256() + + try: + with open(file_path, "rb") as f: + for byte_block in iter(lambda: f.read(BUF_SIZE), b""): + sha256_hash.update(byte_block) + return sha256_hash.hexdigest() + except Exception as e: + logger.error(f"Error computing hash for {file_path}: {e}") + return None + + +def get_model_info_by_hash( + file_hash: str, api_key: Optional[str] = None, use_cache: bool = True +) -> Optional[Dict[str, Any]]: + """ + Look up a model on CivitAI using its sha256 hash. + Uses the CivitAI API endpoint: /api/v1/model-versions/by-hash/{hash} + + Args: + file_hash: SHA256 hash of the model file + api_key: Optional CivitAI API key + use_cache: Whether to use cached results + + Returns: + Dict with model info from CivitAI, or None if not found + """ + global _hash_cache + + if not file_hash: + return None + + cache_key = f"hash_{file_hash}" + + if use_cache and cache_key in _hash_cache: + return _hash_cache[cache_key] + + api_url = f"{CIVITAI_API_URL}/model-versions/by-hash/{file_hash}" + + try: + headers = {} + if api_key: + headers["Authorization"] = f"Bearer {api_key}" + + response = requests.get(api_url, headers=headers, timeout=15) + + if response.status_code == 200: + data = response.json() + + # Extract useful info from the response + model_info = data.get("model", {}) + version_info = data + + # Get model_id from top level or nested model object + model_id = data.get("modelId") or model_info.get("id") + + # Extract images with metadata + images = _extract_model_images(version_info) + + result = { + "source": "civitai", + "model_id": model_id, + "model_name": model_info.get("name"), + "model_type": model_info.get("type"), + "version_id": version_info.get("id"), + "version_name": version_info.get("name"), + "sha256": file_hash, + "url": f"https://civitai.com/models/{model_id}" if model_id else None, + "version_url": f"https://civitai.com/models/{model_id}?modelVersionId={version_info.get('id')}" + if model_id + else None, + "download_url": version_info.get("downloadUrl"), + "base_model": version_info.get("baseModel"), + "tags": model_info.get("tags", []), + "trained_words": _extract_trained_words(version_info), + "images": images, + "clip_skip": version_info.get("clipSkip"), + "description": version_info.get("description", ""), + "model_description": model_info.get("description", ""), + } + + _hash_cache[cache_key] = result + logger.info(f"Found model by hash {file_hash}: {result.get('model_name')}") + return result + + elif response.status_code == 404: + logger.info(f"Model not found on CivitAI for hash {file_hash}") + _hash_cache[cache_key] = None + return None + else: + logger.warning( + f"CivitAI hash lookup returned {response.status_code} for {file_hash}" + ) + return None + + except Exception as e: + logger.error(f"Error looking up model by hash {file_hash}: {e}") + return None + + cache_key = f"hash_{file_hash}" + + if use_cache and cache_key in _hash_cache: + return _hash_cache[cache_key] + + api_url = f"{CIVITAI_API_URL}/model-versions/by-hash/{file_hash}" + + try: + headers = {} + if api_key: + headers["Authorization"] = f"Bearer {api_key}" + + response = requests.get(api_url, headers=headers, timeout=15) + + if response.status_code == 200: + data = response.json() + + # Extract useful info from the response + model_info = data.get("model", {}) + version_info = data + + # Get model_id from top level or nested model object + model_id = data.get("modelId") or model_info.get("id") + + # Extract images with metadata + images = _extract_model_images(version_info) + + result = { + "source": "civitai", + "model_id": model_id, + "model_name": model_info.get("name"), + "model_type": model_info.get("type"), + "version_id": version_info.get("id"), + "version_name": version_info.get("name"), + "sha256": file_hash, + "url": f"https://civitai.com/models/{model_id}" if model_id else None, + "version_url": f"https://civitai.com/models/{model_id}?modelVersionId={version_info.get('id')}" + if model_id + else None, + "download_url": version_info.get("downloadUrl"), + "base_model": version_info.get("baseModel"), + "tags": model_info.get("tags", []), + "trained_words": _extract_trained_words(version_info), + "images": images, + "clip_skip": version_info.get("clipSkip"), + "description": version_info.get("description", ""), + "model_description": model_info.get("description", ""), + } + + _hash_cache[cache_key] = result + logger.info(f"Found model by hash {file_hash}: {result.get('model_name')}") + return result + + elif response.status_code == 404: + logger.info(f"Model not found on CivitAI for hash {file_hash}") + _hash_cache[cache_key] = None + return None + else: + logger.warning( + f"CivitAI hash lookup returned {response.status_code} for {file_hash}" + ) + return None + + except Exception as e: + logger.error(f"Error looking up model by hash {file_hash}: {e}") + return None + + +def _extract_trained_words(version_info: Dict[str, Any]) -> List[str]: + """ + Extract trained words/phrases from model version info. + """ + trained_words = [] + + # Try to get from metadata + metadata = version_info.get("trainedWords", []) + if isinstance(metadata, list): + trained_words.extend(metadata) + elif isinstance(metadata, str) and metadata: + trained_words.append(metadata) + + # Also check metadata field + model = version_info.get("model", {}) + if isinstance(model, dict): + model_tags = model.get("tags", []) + if isinstance(model_tags, list): + for tag in model_tags: + if tag not in trained_words: + trained_words.append(tag) + + return trained_words + + +def _extract_model_images(version_info: Dict[str, Any]) -> List[Dict[str, Any]]: + """ + Extract images with metadata from model version info. + Each image can have: url, civitaiUrl, seed, steps, cfg, sampler, model, positive, negative + """ + images = [] + + # Get images from modelVersions or direct images field + images_data = version_info.get("images", []) + if not images_data: + # Check nested modelVersions + model_versions = version_info.get("modelVersions", []) + if model_versions and len(model_versions) > 0: + images_data = model_versions[0].get("images", []) + + for img in images_data: + if not isinstance(img, dict): + continue + + # Get the image URL + img_url = img.get("url", "") + if not img_url: + continue + + # Get metadata from image info (may be nested in 'meta' object) + meta = img.get("meta", {}) + + img_info = { + "url": img_url, + "civitaiUrl": img.get("civitaiUrl", ""), + "seed": img.get("seed") or meta.get("seed"), + "steps": img.get("steps") or meta.get("steps"), + "cfg": img.get("cfg") or meta.get("cfg"), + "sampler": img.get("sampler") or meta.get("sampler"), + "model": img.get("model") or meta.get("model"), + "positive": img.get("positive") or meta.get("prompt"), + "negative": img.get("negative") or meta.get("negative_prompt"), + } + + # Only add if we have at least a URL + if img_info["url"]: + images.append(img_info) + + return images + + +def _get_metadata_file_path(model_path: str) -> str: + """ + Get the path to the metadata file for a model. + For example, for 'model.safetensors', it returns 'model.metadata.json' + Also checks for variations without extension or with different extensions. + """ + if not model_path: + return "" + + # Get directory and filename + directory = os.path.dirname(model_path) + filename = os.path.basename(model_path) + + # Try different variations of the metadata file name + base_name = filename.rsplit(".", 1)[0] if "." in filename else filename + + possible_names = [ + base_name + ".metadata.json", + filename + ".metadata.json", + base_name + ".json", + filename.replace("_", " ").split()[0] + ".metadata.json" + if "_" in base_name + else None, + ] + + for name in possible_names: + if name: + path = os.path.join(directory, name) + if os.path.exists(path): + logger.info(f"Found metadata file: {path}") + return path + + return "" + + +def _read_model_metadata(metadata_path: str) -> Optional[Dict[str, Any]]: + """ + Read model metadata from a JSON file. + Returns the metadata if found and valid, None otherwise. + """ + if not metadata_path or not os.path.exists(metadata_path): + return None + + try: + with open(metadata_path, "r", encoding="utf-8") as f: + data = json.load(f) + + # Validate it's a valid metadata file with CivitAI info + if not isinstance(data, dict): + return None + + # Check if it has the needed info + if not data.get("sha256") and not data.get("civitai"): + return None + + logger.info(f"Successfully read metadata from: {metadata_path}") + return data + except Exception as e: + logger.debug(f"Error reading metadata file {metadata_path}: {e}") + return None + + +def _metadata_to_model_info(metadata: Dict[str, Any]) -> Dict[str, Any]: + """ + Convert metadata file format to model info format used by our extension. + """ + civitai_data = metadata.get("civitai") or {} + + # Extract images with metadata + images = [] + for img in civitai_data.get("images") or []: + if isinstance(img, dict) and img.get("url"): + img_meta = img.get("meta") or {} + images.append( + { + "url": img.get("url", ""), + "civitaiUrl": img.get("civitaiUrl", "") + or f"https://civitai.com/images/{img.get('id')}", + "seed": img_meta.get("seed"), + "steps": img_meta.get("steps"), + "cfg": img_meta.get("cfg"), + "sampler": img_meta.get("sampler"), + "positive": img_meta.get("prompt"), + } + ) + + # Get trained words from CivitAI data + trained_words = civitai_data.get("trainedWords") or [] + if isinstance(trained_words, str): + trained_words = [trained_words] if trained_words else [] + + # Get model info + model_info = civitai_data.get("model") or {} + version_info = civitai_data + + # Build model_id from CivitAI data + model_id = civitai_data.get("modelId") or civitai_data.get("id") + + return { + "source": "metadata", + "model_id": model_id, + "model_name": metadata.get("model_name") or metadata.get("file_name", ""), + "model_type": model_info.get("type", "") or civitai_data.get("type", ""), + "version_id": civitai_data.get("id"), + "version_name": civitai_data.get("name", ""), + "sha256": metadata.get("sha256", ""), + "url": f"https://civitai.com/models/{civitai_data.get('modelId')}" + if civitai_data.get("modelId") + else None, + "version_url": f"https://civitai.com/models/{civitai_data.get('modelId')}?modelVersionId={civitai_data.get('id')}" + if model_id + else None, + "download_url": civitai_data.get("downloadUrl"), + "base_model": (metadata.get("base_model") or civitai_data.get("baseModel", "")), + "tags": metadata.get("tags") or [], + "trained_words": trained_words, + "images": images, + "clip_skip": civitai_data.get("clipSkip"), + "description": metadata.get("model_description", "") + or civitai_data.get("description", ""), + "model_description": metadata.get("model_description", ""), + "from_metadata": True, + } + + +def get_model_info_for_file( + file_path: str, api_key: Optional[str] = None +) -> Optional[Dict[str, Any]]: + """ + Get model info from CivitAI by computing file hash and looking up by hash. + First checks if there's a metadata file in the same folder. + + Args: + file_path: Full path to the model file + api_key: Optional CivitAI API key + + Returns: + Dict with model info from CivitAI, or None if not found + """ + logger.info(f"get_model_info_for_file called with: {file_path}") + + # First check for metadata file + metadata_path = _get_metadata_file_path(file_path) + logger.info(f"Looking for metadata file, checked: {metadata_path}") + + if metadata_path: + metadata = _read_model_metadata(metadata_path) + if metadata: + logger.info(f"Using metadata file for {file_path}") + return _metadata_to_model_info(metadata) + else: + # Debug: list all files in the same directory + directory = os.path.dirname(file_path) + if directory and os.path.exists(directory): + files = os.listdir(directory) + metadata_files = [ + f for f in files if "metadata" in f.lower() or f.endswith(".json") + ] + logger.info(f"Files in {directory}: {metadata_files}") + + # If no metadata file, compute hash and look up on CivitAI + file_hash = _get_sha256_hash(file_path) + if not file_hash: + return None + + return get_model_info_by_hash(file_hash, api_key) diff --git a/web/linker.js b/web/linker.js index a1ae8d8..e16a0c2 100644 --- a/web/linker.js +++ b/web/linker.js @@ -43,6 +43,35 @@ class LinkerManagerDialog extends ComfyDialog { } }); + // Create context menu for model chips + this.contextMenu = $el("div.ml-context-menu", { + parent: document.body, + style: { + position: "fixed", + display: "none", + zIndex: "100001" + } + }, [ + $el("div.ml-context-menu-item", { + onclick: () => this.handleContextMenuAction('showInfo'), + style: { cursor: "pointer" } + }, [ + $el("span.ml-context-menu-item-icon", { textContent: "ℹ" }), + $el("span", { textContent: "Show Info" }) + ]), + $el("div.ml-context-menu-divider"), + $el("div.ml-context-menu-item", { + onclick: () => this.handleContextMenuAction('civitai'), + style: { cursor: "pointer" } + }, [ + $el("span.ml-context-menu-item-icon", { textContent: "🌐" }), + $el("span", { textContent: "Open in CivitAI" }) + ]) + ]); + + // Selected model for context menu + this._contextMenuModel = null; + // Create dialog element using $el this.element = $el("div.comfy-modal", { id: "model-linker-modal", @@ -78,6 +107,466 @@ class LinkerManagerDialog extends ComfyDialog { // Add click listener to backdrop this.backdrop.addEventListener('click', () => this.close()); + + // Add click listener to hide context menu when clicking outside + this.boundHandleContextMenuClick = (e) => this.handleContextMenuOutsideClick(e); + document.addEventListener('click', this.boundHandleContextMenuClick); + } + + /** + * Handle click outside context menu to hide it + */ + handleContextMenuOutsideClick(e) { + if (!this.contextMenu) return; + if (this.contextMenu.style.display === 'none') return; + + // Check if click is outside the context menu + if (!this.contextMenu.contains(e.target)) { + this.hideContextMenu(); + } + } + + /** + * Show context menu at the specified position + */ + showContextMenu(x, y, model) { + if (!this.contextMenu) return; + + this._contextMenuModel = model; + + // Position the menu + this.contextMenu.style.left = `${x}px`; + this.contextMenu.style.top = `${y}px`; + this.contextMenu.style.display = 'block'; + + // Adjust position if menu would go off screen + const rect = this.contextMenu.getBoundingClientRect(); + if (rect.right > window.innerWidth) { + this.contextMenu.style.left = `${window.innerWidth - rect.width - 10}px`; + } + if (rect.bottom > window.innerHeight) { + this.contextMenu.style.top = `${window.innerHeight - rect.height - 10}px`; + } + } + + /** + * Hide context menu + */ + hideContextMenu() { + if (!this.contextMenu) return; + this.contextMenu.style.display = 'none'; + this._contextMenuModel = null; + } + + /** + * Handle context menu item click + */ + handleContextMenuAction(action) { + const model = this._contextMenuModel; + this.hideContextMenu(); + + if (!model) return; + + if (action === 'civitai') { + this.openInCivitAI(model); + } else if (action === 'showInfo') { + this.showModelInfo(model); + } + } + + /** + * Open model in CivitAI + */ + async openInCivitAI(model) { + if (!model) return; + + const name = model.name || model.original_path?.split(/[\/\\]/).pop() || ''; + if (!name) return; + + try { + // Search CivitAI for this model using hash (pass resolved_path for hash lookup) + const response = await api.fetchApi('/model_linker/civitai-search', { + method: 'POST', + headers: { 'Content-Type': 'application/json' }, + body: JSON.stringify({ + filename: name, + category: model.category, + resolved_path: model.resolved_path || '' + }) + }); + + if (!response.ok) { + this.showNotification('Nie znaleziono modelu na CivitAI', 'error'); + return; + } + + const data = await response.json(); + if (data.url) { + window.open(data.url, '_blank'); + } else { + // Try direct search on CivitAI + const searchName = name.replace(/\.(safetensors|ckpt|pt|pth|bin|pkl|sft|onnx|gguf)$/i, ''); + const searchUrl = `https://civitai.com/search?q=${encodeURIComponent(searchName)}`; + window.open(searchUrl, '_blank'); + } + } catch (e) { + console.error('Model Linker: Error searching CivitAI:', e); + // Fall back to direct search + const searchName = name.replace(/\.(safetensors|ckpt|pt|pth|bin|pkl|sft|onnx|gguf)$/i, ''); + const searchUrl = `https://civitai.com/search?q=${encodeURIComponent(searchName)}`; + window.open(searchUrl, '_blank'); + } + } + + /** + * Show model info dialog (similar to rgthree's RgthreeLoraInfoDialog) + */ + async showModelInfo(model) { + if (!model) return; + + const name = model.name || model.original_path?.split(/[\/\\]/).pop() || ''; + if (!name) return; + + // Create and show the info dialog + this.showModelInfoDialog(name, model); + } + + /** + * Show the model info dialog + */ + showModelInfoDialog(loraName, modelData) { + // Create info dialog element + const dialog = this.createInfoDialog(loraName, modelData); + + // Show the dialog + document.body.appendChild(dialog); + + // Add close handlers + const closeBtn = dialog.querySelector('.ml-info-dialog-close'); + const footerCloseBtn = dialog.querySelector('.ml-info-dialog-close-btn'); + const backdrop = dialog.querySelector('.ml-info-dialog-backdrop'); + + if (closeBtn) { + closeBtn.addEventListener('click', () => this.closeInfoDialog(dialog)); + } + if (footerCloseBtn) { + footerCloseBtn.addEventListener('click', () => this.closeInfoDialog(dialog)); + } + if (backdrop) { + backdrop.addEventListener('click', (e) => { + // Only close if clicking backdrop itself, not its children + if (e.target === backdrop) { + this.closeInfoDialog(dialog); + } + }); + } + + // Fetch CivitAI info + this.fetchModelInfoForDialog(loraName, modelData, dialog); + } + + /** + * Create the info dialog element + */ + createInfoDialog(loraName, modelData) { + const loraDisplayName = loraName.replace(/\.(safetensors|ckpt|pt|pth|bin|pkl|sft|onnx|gguf)$/i, ''); + + const dialog = document.createElement('div'); + dialog.className = 'ml-info-dialog-backdrop'; + dialog.innerHTML = ` +
+
+

${loraDisplayName}

+ +
+
+
Loading...
+ +
+
+ +
+
+ `; + + return dialog; + } + + /** + * Fetch model info and update the dialog + */ + async fetchModelInfoForDialog(loraName, modelData, dialog) { + try { + const response = await api.fetchApi('/model_linker/civitai-search', { + method: 'POST', + headers: { 'Content-Type': 'application/json' }, + body: JSON.stringify({ + filename: loraName, + category: modelData?.category || '', + resolved_path: modelData?.resolved_path || '' + }) + }); + + if (response.ok) { + const data = await response.json(); + this.updateInfoDialogWithData(dialog, data); + } else { + this.updateInfoDialogError(dialog, 'Model not found on CivitAI'); + } + } catch (e) { + console.error('Model Linker: Error fetching model info:', e); + this.updateInfoDialogError(dialog, 'Error fetching info'); + } + } + + /** + * Update the info dialog with data + */ + updateInfoDialogWithData(dialog, data) { + const loadingDiv = dialog.querySelector('.ml-info-dialog-loading'); + const bodyDiv = dialog.querySelector('.ml-info-dialog-body'); + + if (loadingDiv) loadingDiv.style.display = 'none'; + if (bodyDiv) bodyDiv.style.display = 'block'; + + if (!data) { + this.updateInfoDialogError(dialog, 'No data received'); + return; + } + + // Update title + const titleEl = dialog.querySelector('.ml-info-dialog-title'); + if (titleEl) { + const modelName = data.model_name || data.modelName || 'Unknown Model'; + const versionName = data.version_name || data.versionName || ''; + titleEl.textContent = versionName ? `${modelName} - ${versionName}` : modelName; + } + + // Update type tag + const typeTag = dialog.querySelector('.ml-info-type'); + if (typeTag) { + const modelType = data.model_type || data.modelType || ''; + typeTag.textContent = modelType.toUpperCase(); + typeTag.className = `ml-info-tag ml-info-type -type-${modelType.toLowerCase()}`; + } + + // Update base model tag + const baseModelTag = dialog.querySelector('.ml-info-basemodel'); + if (baseModelTag) { + const baseModel = data.base_model || data.baseModel || ''; + baseModelTag.textContent = baseModel || ''; + if (baseModel) { + baseModelTag.style.display = ''; + baseModelTag.className = `ml-info-tag ml-info-basemodel -basemodel-${baseModel.toLowerCase().replace(/\s+/g, '-')}`; + } else { + baseModelTag.style.display = 'none'; + } + } + + // Update file + const fileEl = dialog.querySelector('.ml-info-file'); + if (fileEl && data.filename) { + fileEl.textContent = data.filename; + } + + // Update hash + const hashEl = dialog.querySelector('.ml-info-hash'); + if (hashEl) { + hashEl.textContent = data.sha256 || data.hash || ''; + } + + // Update CivitAI link + const civitaiLinkEl = dialog.querySelector('.ml-info-civitai-link'); + if (civitaiLinkEl) { + if (data.url || data.version_url) { + const url = data.version_url || data.url; + civitaiLinkEl.innerHTML = ` + + + View on Civitai + + `; + } else { + const searchName = data.model_name || data.modelName || 'Unknown'; + civitaiLinkEl.innerHTML = ` + Model not found + + 🔍 Search on CivitAI + + `; + } + } + + // Update name + const nameEl = dialog.querySelector('.ml-info-name'); + if (nameEl) { + nameEl.textContent = data.model_name || data.modelName || ''; + } + + // Update base model row + const baseModelRowEl = dialog.querySelector('.ml-info-base-model'); + if (baseModelRowEl) { + const baseModel = data.base_model || data.baseModel || ''; + baseModelRowEl.textContent = baseModel; + const row = baseModelRowEl.closest('tr'); + if (row && baseModel) { + row.style.display = ''; + } else if (row) { + row.style.display = 'none'; + } + } + + // Update trained words + const trainedWordsEl = dialog.querySelector('.ml-info-trained-words'); + if (trainedWordsEl) { + const words = data.trained_words || data.trainedWords || []; + if (Array.isArray(words) && words.length > 0) { + trainedWordsEl.textContent = words.join(', '); + const row = trainedWordsEl.closest('tr'); + if (row) row.style.display = ''; + } else { + const row = trainedWordsEl.closest('tr'); + if (row) row.style.display = 'none'; + } + } + + // Update clip skip + const clipSkipEl = dialog.querySelector('.ml-info-clip-skip'); + if (clipSkipEl) { + const clipSkip = data.clip_skip || data.clipSkip; + if (clipSkip && clipSkip !== 'None') { + clipSkipEl.textContent = clipSkip; + const row = clipSkipEl.closest('tr'); + if (row) row.style.display = ''; + } else { + const row = clipSkipEl.closest('tr'); + if (row) row.style.display = 'none'; + } + } + + // Update description + const descEl = dialog.querySelector('.ml-info-description'); + if (descEl) { + const desc = data.description || data.model_description || data.modelDescription || ''; + if (desc) { + descEl.textContent = desc.substring(0, 200) + (desc.length > 200 ? '...' : ''); + const row = descEl.closest('tr'); + if (row) row.style.display = ''; + } else { + const row = descEl.closest('tr'); + if (row) row.style.display = 'none'; + } + } + + // Update images + this.updateInfoDialogImages(dialog, data.images || data.modelImages || []); + } + + /** + * Update images in the info dialog + */ + updateInfoDialogImages(dialog, images) { + const imagesContainer = dialog.querySelector('.ml-info-images'); + if (!imagesContainer || !images.length) return; + + let imagesHtml = '
Example Images
'; + + for (const img of images.slice(0, 6)) { + if (!img.url) continue; + + let caption = ''; + if (img.civitaiUrl) { + caption += `civitai`; + } + if (img.seed) caption += ` ${img.seed}`; + if (img.steps) caption += ` ${img.steps}`; + if (img.cfg) caption += ` ${img.cfg}`; + if (img.sampler) caption += ` ${img.sampler}`; + if (img.positive) caption += ` ${img.positive.substring(0, 100)}${img.positive.length > 100 ? '...' : ''}`; + + imagesHtml += ` +
+
+ Example +
${caption}
+
+
+ `; + } + + imagesHtml += '
'; + imagesContainer.innerHTML = imagesHtml; + } + + /** + * Update the info dialog with error + */ + updateInfoDialogError(dialog, message) { + const civitaiLink = dialog.querySelector('.ml-info-civitai-link'); + if (civitaiLink) { + civitaiLink.innerHTML = `${message}`; + } + } + + /** + * Close the info dialog + */ + closeInfoDialog(dialog) { + if (dialog && dialog.parentNode) { + dialog.parentNode.removeChild(dialog); + } } /** @@ -529,6 +1018,309 @@ class LinkerManagerDialog extends ComfyDialog { .ml-link:hover { text-decoration: underline; } + + /* Context menu */ + .ml-context-menu { + position: fixed; + background: var(--comfy-menu-bg, #202020); + border: 1px solid var(--border-color, #555555); + border-radius: 6px; + box-shadow: 0 4px 16px rgba(0,0,0,0.5); + z-index: 100001; + min-width: 160px; + overflow: hidden; + } + .ml-context-menu-item { + display: flex; + align-items: center; + gap: 8px; + padding: 8px 12px; + font-size: 12px; + color: var(--input-text, #e0e0e0); + cursor: pointer; + transition: background 0.15s ease; + } + .ml-context-menu-item:hover { + background: rgba(255,255,255,0.1); + } + .ml-context-menu-item-icon { + font-size: 14px; + width: 20px; + text-align: center; + } + .ml-context-menu-divider { + height: 1px; + background: var(--border-color, #555555); + margin: 4px 0; + } + + /* Model Info Dialog (similar to rgthree's RgthreeLoraInfoDialog) */ + .ml-info-dialog-backdrop { + position: fixed; + top: 0; + left: 0; + width: 100vw; + height: 100vh; + background: rgba(0, 0, 0, 0.6); + z-index: 100002; + display: flex; + align-items: center; + justify-content: center; + } + .ml-info-dialog { + background: var(--comfy-menu-bg, #202020); + border: 2px solid var(--border-color, #555555); + border-radius: 8px; + width: 500px; + max-width: 90vw; + max-height: 80vh; + overflow: hidden; + display: flex; + flex-direction: column; + box-shadow: 0 8px 32px rgba(0,0,0,0.5); + } + .ml-info-dialog-header { + display: flex; + justify-content: space-between; + align-items: center; + padding: 16px 20px; + border-bottom: 1px solid var(--border-color, #555555); + background: rgba(0,0,0,0.2); + } + .ml-info-dialog-header h3 { + margin: 0; + font-size: 16px; + color: var(--input-text, #e0e0e0); + } + .ml-info-dialog-close { + background: none; + border: none; + font-size: 24px; + color: var(--input-text, #e0e0e0); + cursor: pointer; + padding: 0; + width: 30px; + height: 30px; + display: flex; + align-items: center; + justify-content: center; + border-radius: 4px; + } + .ml-info-dialog-close:hover { + background: rgba(255,255,255,0.1); + } + .ml-info-dialog-content { + padding: 16px 20px; + overflow-y: auto; + flex: 1; + } + .ml-info-dialog-loading { + text-align: center; + color: var(--ml-text-muted, #888); + padding: 20px; + } + .ml-info-section { + margin-bottom: 16px; + padding-bottom: 16px; + border-bottom: 1px solid var(--border-color, #555555); + } + .ml-info-section-header { + font-size: 12px; + font-weight: 600; + color: var(--ml-text-muted, #888); + text-transform: uppercase; + letter-spacing: 0.5px; + margin-bottom: 12px; + } + .ml-info-row { + display: flex; + justify-content: space-between; + align-items: center; + padding: 6px 0; + } + .ml-info-label { + color: var(--ml-text-muted, #888); + font-size: 13px; + } + .ml-info-value { + color: var(--input-text, #e0e0e0); + font-size: 13px; + font-weight: 500; + } + .ml-info-value.ml-info-active { + color: #4CAF50; + } + .ml-info-value.ml-info-inactive { + color: #888; + } + .ml-info-section-civitai { + border-bottom: none; + margin-bottom: 0; + padding-bottom: 0; + } + .ml-info-civitai-link { + display: flex; + flex-direction: column; + gap: 8px; + } + .ml-info-link { + display: inline-flex; + align-items: center; + gap: 6px; + padding: 8px 12px; + background: #2196F3; + color: white; + border-radius: 6px; + text-decoration: none; + font-size: 13px; + } + .ml-info-link:hover { + background: #1976D2; + } + .ml-info-loading-small { + color: var(--ml-text-muted, #888); + font-size: 12px; + font-style: italic; + } + .ml-info-not-found { + color: #888; + font-size: 12px; + } + + /* Info Area (type/base model tags) */ + .ml-info-area { + display: flex; + gap: 8px; + margin-bottom: 16px; + } + .ml-info-tag { + display: inline-flex; + align-items: center; + padding: 4px 10px; + background: rgba(255,255,255,0.1); + border-radius: 4px; + font-size: 11px; + font-weight: 600; + color: var(--input-text, #e0e0e0); + } + .ml-info-tag.-type-lora { background: rgba(33,150,243,0.2); color: #64B5F6; } + .ml-info-tag.-type-checkpoint { background: rgba(76,175,80,0.2); color: #81C784; } + .ml-info-tag.-type-vae { background: rgba(156,39,176,0.2); color: #BA68C8; } + .ml-info-tag.-basemodel-sd1 { background: rgba(244,67,54,0.2); color: #EF9A9A; } + .ml-info-tag.-basemodel-sdxl { background: rgba(33,150,243,0.2); color: #64B5F6; } + .ml-info-tag.-basemodel-sdxl-turbo { background: rgba(33,150,243,0.2); color: #64B5F6; } + .ml-info-tag.-basemodel-flux.1 { background: rgba(255,152,0,0.2); color: #FFB74D; } + .ml-info-tag.-basemodel-flux.1-d { background: rgba(255,152,0,0.2); color: #FFB74D; } + + /* Info Table */ + .ml-info-table { + width: 100%; + border-collapse: collapse; + margin-bottom: 16px; + } + .ml-info-table tr { + border-bottom: 1px solid rgba(255,255,255,0.05); + } + .ml-info-table tr:last-child { + border-bottom: none; + } + .ml-info-table td { + padding: 8px 4px; + font-size: 13px; + vertical-align: top; + } + .ml-info-table td:first-child { + color: var(--ml-text-muted, #888); + white-space: nowrap; + width: 140px; + } + .ml-info-table td:last-child { + color: var(--input-text, #e0e0e0); + word-break: break-word; + } + .ml-info-help { + display: inline-block; + width: 14px; + height: 14px; + background: rgba(255,255,255,0.1); + border-radius: 50%; + font-size: 10px; + text-align: center; + line-height: 14px; + cursor: help; + margin-left: 4px; + vertical-align: middle; + } + + /* Images Section */ + .ml-info-images { + margin-top: 16px; + padding-top: 16px; + border-top: 1px solid var(--border-color, #555555); + } + .ml-info-images-header { + font-size: 12px; + font-weight: 600; + color: var(--ml-text-muted, #888); + text-transform: uppercase; + letter-spacing: 0.5px; + margin-bottom: 12px; + } + .ml-info-images-grid { + display: grid; + grid-template-columns: repeat(auto-fill, minmax(150px, 1fr)); + gap: 12px; + } + .ml-info-image-item { + border-radius: 6px; + overflow: hidden; + background: rgba(0,0,0,0.2); + } + .ml-info-image-item img { + width: 100%; + height: auto; + display: block; + } + .ml-info-image-item figcaption { + padding: 8px; + font-size: 11px; + color: var(--ml-text-muted, #888); + } + .ml-info-image-item figcaption span { + display: block; + margin-bottom: 2px; + } + .ml-info-image-item figcaption span label { + font-weight: 600; + } + .ml-info-image-link { + color: #2196F3; + text-decoration: none; + display: block; + margin-bottom: 4px; + } + + /* CivitAI Logo */ + .ml-info-civitai-logo { + width: 16px; + height: 16px; + } + + /* Footer */ + .ml-info-dialog-footer { + padding: 12px 20px; + border-top: 1px solid var(--border-color, #555555); + display: flex; + justify-content: flex-end; + gap: 8px; + } + .ml-info-dialog-close-btn { + padding: 8px 16px; + } + .ml-info-error { + color: #f44336; + font-size: 12px; + } `; document.head.appendChild(styles); @@ -1139,7 +1931,8 @@ class LinkerManagerDialog extends ComfyDialog { const fullName = model.name || model.original_path?.split(/[\/\\]/).pop() || 'Unknown'; const name = fullName.replace(/\.(safetensors|ckpt|pt|pth|bin|pkl|sft|onnx|gguf)$/i, ''); const strength = model.strength !== null && model.strength !== undefined ? model.strength.toFixed(2) : null; - html += `${name}${strength !== null ? `${strength}` : ''}`; + const modelData = encodeURIComponent(JSON.stringify(model)); + html += `${name}${strength !== null ? `${strength}` : ''}`; } html += ``; } @@ -1163,7 +1956,8 @@ class LinkerManagerDialog extends ComfyDialog { const fullName = model.name || model.original_path?.split(/[\/\\]/).pop() || 'Unknown'; const name = fullName.replace(/\.(safetensors|ckpt|pt|pth|bin|pkl|sft|onnx|gguf)$/i, ''); const strength = model.strength !== null && model.strength !== undefined ? model.strength.toFixed(2) : null; - html += `${name}${strength !== null ? `${strength}` : ''}`; + const modelData = encodeURIComponent(JSON.stringify(model)); + html += `${name}${strength !== null ? `${strength}` : ''}`; } html += ``; } @@ -3529,6 +4323,7 @@ class ModelLinker { // Create dialog instance if (!this.dialog) { this.dialog = new LinkerManagerDialog(); + window.modelLinkerDialog = this.dialog; } // Register keyboard shortcut (Ctrl+Shift+L) @@ -3727,11 +4522,12 @@ class ModelLinker { // Small delay to let popup close await new Promise(r => setTimeout(r, 200)); - // Create dialog if needed +// Create dialog if needed if (!this.dialog) { this.dialog = new LinkerManagerDialog(); + window.modelLinkerDialog = this.dialog; } - + // Run auto-resolve for 100% matches - returns the updated workflow const updatedWorkflow = await this.dialog.autoResolve100Percent(); @@ -4041,6 +4837,7 @@ class ModelLinker { try { if (!this.dialog) { this.dialog = new LinkerManagerDialog(); + window.modelLinkerDialog = this.dialog; } this.dialog.show(); } catch (error) { @@ -4181,3 +4978,23 @@ window.MLCopyCode = function(sectionId, btn) { }); }; +window.MLOpenContextMenu = function(event, element) { + event.preventDefault(); + event.stopPropagation(); + + try { + const modelData = element.getAttribute('data-model'); + if (!modelData) return; + + const model = JSON.parse(decodeURIComponent(modelData)); + + // Get dialog instance + const dialog = window.modelLinkerDialog; + if (dialog && dialog.showContextMenu) { + dialog.showContextMenu(event.clientX, event.clientY, model); + } + } catch (e) { + console.error('Model Linker: Error opening context menu:', e); + } +}; +