Marco
793402f93f
This commit introduces several new performance-oriented custom nodes and utilities: - Added LTXV Multi Concat and LTXV Multi Concat (beta) for faster frame injection using latent inpainting logic. - Refactored LTXV index resolution logic into ltxv_utils.py. - Updated ImageListSampler to anchor edge frames properly and support output index normalization scaling via target_frames. - Added metadata-based prompt extraction image loaders (LoadImageAndExtractPrompt, FolderImageAndExtractPrompt, and FolderImageMetadataByName). - Included animation tracking and camera switcher nodes.
30 lines
1.2 KiB
Python
30 lines
1.2 KiB
Python
import json
|
|
|
|
def analyze_definitions(path):
|
|
with open(path, 'r', encoding='utf-8') as f:
|
|
data = json.load(f)
|
|
|
|
definitions = data.get('definitions', {})
|
|
print(f"Found {len(definitions)} definitions.")
|
|
|
|
for def_id, def_content in definitions.items():
|
|
print(f"\n--- Definition ID: {def_id} ---")
|
|
# Definitions usually contain their own 'nodes' and 'links'
|
|
inner_nodes = def_content.get('nodes', [])
|
|
print(f"Number of inner nodes: {len(inner_nodes)}")
|
|
|
|
node_types = {}
|
|
for node in inner_nodes:
|
|
ntype = node.get('type')
|
|
node_types[ntype] = node_types.get(ntype, 0) + 1
|
|
|
|
# Check for samplers or guide nodes
|
|
if ntype and ('Sampler' in ntype or 'Guide' in ntype or 'Inpaint' in ntype or 'Conditioning' in ntype):
|
|
print(f" Interesting Node found: {ntype}")
|
|
if node.get('widgets_values'):
|
|
print(f" Widgets: {node.get('widgets_values')}")
|
|
|
|
print(f"Types summary: {node_types}")
|
|
|
|
analyze_definitions('F:/AppsCrucial/ComfyUI_phoenix3/ComfyUI/custom_nodes/reference/multiframe/LTX23_MULTIFRAME.json')
|