File List to File Path

This node's job is to take a block of text - a list of file paths you've already got, however you got them - and turn it into a proper list type that the rest of this pack's tooling, particularly its loop nodes, can actually iterate over.

Why you'd reach for it

This pack ships a folder scanner (File-List-Loader) for when you want everything in a directory, but not every job is "process a whole folder." Sometimes you've got a specific, curated set of files - output from an earlier step, a list you pasted in from elsewhere - and you need it in the same shape a loop node expects. This is the adapter for that case: text in, iterable list out.

How it works

It takes the file_paths text and splits it into individual entries, producing both the list itself and a count. That count is worth noting - it's exactly the kind of number SimpleForLoopStart's total input wants, so this node pairs naturally with that pack's loop pair if you're processing the list one item at a time.

Inputs and outputs

• file_paths (STRING, required) - your block of paths as text.
• Outputs: file_path (STRING, list) - the individual paths, and count (INT) - how many there are.

Installing it

No models, no pip dependencies mentioned in the README - it's a plain text-processing node. Search ComfyUI Manager for the repo name first, but as a small personal pack it may not be indexed there. The manual route from the README:

cd ComfyUI/custom_nodes
git clone https://github.com/huyl3-cpu/comfyui-sortlist.git

Restart ComfyUI afterward.

Common issues and troubleshooting

There's no delimiter setting exposed. The schema doesn't give you a comma-vs-newline toggle, so if your paths aren't splitting the way you expect, check how you're separating them - one path per line is the safe bet here, and it matches how this pack's own File-List-Loader returns its results, so staying consistent with that format is a reasonable default.

It doesn't validate that the paths exist. This is a string-splitting utility, not a filesystem check - a typo'd or nonexistent path won't error here. It'll pass through fine and only fail later, when something actually tries to open it. If a downstream loader is failing mysteriously, drop a text-preview node right after this one and eyeball the list before it goes any further.

Pair it with the loop nodes, not a plain single-file node. The whole point of getting a count back alongside the list is to drive iteration - if you're only ever processing one file, you don't need this node at all; just use the path directly.