Index Of Batch Ovum

Same trick as the pack's other IndexOf nodes, but this is the version you should actually use when your data arrives as a Comfy-style list: feed it a list and a search value, get back the first matching index, or -1 if there's no match. It's the modern replacement for the deprecated plain IndexOf, and it lives in the ovum/lists/comfy category to signal that it's built to sit directly on Comfy list sockets.

Think of it as JavaScript's Array.prototype.indexOf with Python equality underneath. If your list is ["sunset", "hdr", "sunset"] and you search for "hdr", you get 1. Search "sunset" and you get 0 - first match only.

The inputs that matter

The schema is minimal, and that's the appeal:

• py_list - the list to search. Must be an actual Python list; the node raises ValueError otherwise.
• search_element - the value to find. Any type, so wire in a string, an int, even a small dict.
• start - optional from-index. This is a STRING field (the pack's quirk): blank means 0, a negative value counts from the end and gets clamped to [0, n].

Output is a single INT. Wire it into a comparison to test "is this in the list?" - anything >= 0 means yes - or into an indexing node to grab the found entry from a parallel list.

Why "batch" and which one do I pick

The Batch suffix here mostly tells you which socket semantics the node expects. IndexOfBatchOvum treats the incoming value as the list itself (batch-style, one element per list item), while IndexOfOvum in ovum/lists/python is the flavor that takes a Python list object as a single value. In practice: if you built your list with a list-construction node or it came off a fan-out, use this one; if a node hands you a single py_list object, the python flavor matches better. If you're ever unsure, they're cheap to swap - same inputs, same output, same -1 contract.

One thing to know about every node in this family: they inherit the pack's NewPointer base, which forces IS_CHANGED to NaN, so they never cache and always re-execute. Great for loops where you need fresh answers every tick; slightly annoying if you were hoping ComfyUI would skip a no-op run.

Install

It ships in the comfy-ovum pack, so install once and you're done:

cd ComfyUI/custom_nodes
git clone https://github.com/sfinktah/comfy-ovum

or search comfy-ovum in ComfyUI Manager, then restart. No models, no keys, no heavy dependencies for this node specifically - the pack's Python deps (requests, pillow, numpy, pymediainfo, aiohttp, etc.) are handled by Manager automatically.

Troubleshooting

• ValueError: Input must be a Python list - you connected something that isn't a list, or you expected it to search a Comfy batch tensor. Wire a real list in.
• Getting -1 when you're sure the value is there - check the start field. It's a string; a stray space or leftover text can shift where the scan begins. Also remember it uses strict Python equality, so 2 (int) won't match "2" (string).
• Workflow wants IndexOf (no suffix) - that's the deprecated class. Swapping to this node is a straight replacement: same inputs, same output.