Python Runtime Info

It has no inputs, does no SVG work, and produces nothing you'd ever save. It's still one of the smartest things in this pack. Python Runtime Info reports which Python interpreter ComfyUI is actually running and whether key packages are importable - and if you've ever run pip install and watched ComfyUI still scream "ModuleNotFoundError," you know exactly why that matters.

Why this node exists

The pack's own README mentions CairoSVG as the upgrade path for fancy SVG rendering (gradients, filters - the stuff its built-in Pillow renderer can't do). Here's the catch that eats beginners alive: ComfyUI often runs on an embedded Python - the portable install ships its own interpreter inside the ComfyUI folder. pip install cairosvg from your system terminal installs into your system Python, and ComfyUI's embedded Python never sees it. This node's source docstring says it outright: it exists so you can determine "which interpreter is executing the custom nodes so that dependencies (e.g. cairosvg) can be installed in the correct environment."

So the fix loop goes: run this node → read sys_executable → that's the exact path to the interpreter you should be pip-installing into. No more guessing, no more installing into the void.

What it outputs

One output: info, a STRING containing a JSON blob with:

• sys_executable - the full path to the Python interpreter running your nodes. This is the one you care about.
• python_version, platform, implementation - sanity info.
• site_packages - where packages live for that interpreter.
• has_cairosvg / has_cairocffi / has_tinycss2 / has_cssselect2 / has_PIL / has_skimage / has_cv2 - a quick importability check for each, with the module's install location when present.

Run it once, wire info to a text preview node, and you've got a full picture of your environment in one glance. If has_cv2 or has_skimage is false, that's your "why is Line Region Detector failing" answer - the pack's requirements.txt wants opencv-python, opencv-contrib-python, and scikit-image, and any of them can silently be missing from the embedded Python.

The practical install fix

If this node tells you the interpreter is, say, ComfyUI_windows_portable/python_embeded/python.exe:

"ComfyUI_windows_portable/python_embeded/python.exe" -m pip install cairosvg

(or pip install -r requirements.txt the same way). If it tells you ComfyUI is on your system Python, then your terminal pip was probably fine and the problem is elsewhere. Either way, you now know instead of guessing. It's a 30-second diagnostic that saves the hour-long "why won't this import" hunt.

Install

It's part of the pack - no separate install:

cd ComfyUI/custom_nodes
git clone https://github.com/TJ16th/TJ_ComfyUI_Lineart2Vector.git
# restart ComfyUI

or ComfyUI Manager → search TJ_ComfyUI_Lineart2Vector. No models, no keys, and this node has effectively zero dependencies of its own - it only inspects sys, site, and importlib. It's the one node in the pack that's guaranteed to work even when everything else is broken, which is precisely the point.

Where people get burned

The classic failure is reading the output wrong: has_cairosvg: false with a non-empty cairosvg_location field can't happen - if it's importable the location is there; if it's not importable, that's your signal to install into the interpreter named in sys_executable, not the one you think you're using. And remember info is a JSON string - to read it, plug it into a text display; if you want a real object you'll need a string-to-JSON conversion node. Use this node the moment any dependency error appears, not after you've tried three wrong pip commands.