File List Sorter
Paste a list of file names or paths and put it in the order a file manager would use — natural numbering (img2 before img10), by extension, folder depth or size, with folders on top. Nothing leaves your browser.
About this tool
File List Sorter takes a pasted list of file names or paths and puts it in
the order a file manager would use, not the order a plain text sort gives.
A plain sort compares text character by character, so img10.png lands before
img2.png and README.md jumps ahead of every lowercase name. This tool
understands that a path has structure: a folder, a file name, an extension, a
depth, and sometimes a size column.
Worked example
Paste this list:
img10.png
img2.png
img1.png
With Sort by = Natural order, the result is:
img1.png
img2.png
img10.png
The digit runs are compared as numbers, so 2 sorts before 10 instead of
after it. Switch Sort by to Alphabetical and you get the machine order
(img1.png, img10.png, img2.png) — useful when you need to match what
sort or a Git tree does.
The options
- Sort by — Natural is human numbering over the whole path. Alphabetical
is plain codepoint order. File name only ignores the folders above each
file, so
zzz/apple.txtsorts beforeaaa/banana.txt. Extension groups by file type (entries with no extension come first), then naturally by path. Folder depth lists the shallowest paths first. Size reads a size column off each line. - Order — ascending or descending. Folders-first is not reversed: folders stay on top in both directions, the way Explorer and Finder behave.
- Ignore case — on by default, so
README.mdsits next toreadme.txt. Turn it off for the case-sensitive order ofls,sortor a Git tree. - Folders first — on by default. An entry counts as a folder when it ends in
a slash (
src/) or when another pasted entry lives underneath it (srcis a folder ifsrc/main.rsis also in the list). - Keep each folder's contents together — sorts by parent folder first, then
applies the chosen key inside each folder. Handy for sorting a deep
finddump by file name or size while keeping each folder's files side by side. - Remove duplicate paths — keeps the first spelling of each path.
./src/app.js,src/app.jsandSRC/APP.JScount as one entry when Ignore case is on. - Ignore surrounding whitespace — on by default, because indented
treeorlsoutput otherwise sorts by its indentation. - Output shape — a plain list ready to paste back into a script, a numbered list, a table with type/extension/depth/size columns, or JSON.
Handy for
- Ordering exported screenshots, scans, episodes or
page-1 … page-10files so they concatenate in the right sequence. - Tidying
ls -1,find .,git ls-filesordu -houtput before pasting it into a script, ticket or README. - Grouping a messy download folder by file type, or finding the deepest paths in a repository.
- Turning a
du -hdump into a biggest-files-first list.
Everything runs locally in your browser via WebAssembly — your file list is never uploaded, and nothing is read from your disk. Up to 20,000 paths per run; split a longer list and sort it in batches.
FAQ
What exactly is "natural" sorting?
Natural sorting compares runs of digits as numbers instead of as text, so
img2.png comes before img10.png and v1.9 before v1.10. The letters
around the digits are still compared as text. Leading zeros are handled too:
file01 and file1 compare as the same number, with the shorter spelling
first. Choose Alphabetical if you specifically want the raw codepoint order.
How does it know which entries are folders?
Two ways, since a pasted list carries no filesystem information. An entry that
ends in a slash (src/, assets/img/) is treated as a folder. So is any entry
that another pasted entry sits underneath — if the list contains both src and
src/main.rs, then src is a folder. Everything else is treated as a file. If
your listing has no trailing slashes and no parent entries, turn Folders
first off, since there is nothing for it to lift.
How do I sort by size?
Include a size on each line and pick Sort by = Size. A size is recognised
when it carries a unit — 4.0K src/app.js, src/app.js 1.2MB, 512B,
3GiB — either before or after the path, which is exactly what du -h and
ls -lh produce. K/M/G/T are 1024-based. A bare byte count is only read when a
TAB separates it from the path, so a name like 2024 report.txt keeps its year
instead of being mistaken for a size. Entries with no size always sort last, and
sorting by size with no size column anywhere is reported as an error rather than
silently ignored.
Does it handle Windows paths and mixed separators?
Yes. Both / and \ count as folder separators, so docs\report.docx and
docs/report.docx are understood the same way for depth, folder and extension
purposes. A leading ./ is ignored. Your entries are printed back in exactly
the spelling you pasted — the normalisation is only used for comparing.
What counts as the extension?
The text after the last dot in the file name. So archive.tar.gz has the
extension gz, README has none, and dotfiles like .gitignore have none
either (a leading dot starts the name, it does not start an extension). Folders
never carry an extension, even if their name contains a dot.
Is there a limit, and does my list leave my device?
A run is capped at 20,000 paths; past that you get a clear error asking you to split the list. Blank lines are skipped and do not count. The sorting itself runs in WebAssembly inside your browser, so the list never leaves your device and no file contents are read — only the names you paste.
Developer & Automation Access
Run it from the terminal
Same engine as this page, headless — via the gizza CLI:
gizza tool file-list-sorter "img10.png
img2.png
src/main.rs
docs/report.docx
README.md"New to the CLI? Get gizza →
Open it by URL
Pre-fill and auto-run this tool with query parameters — the names match the API/CLI:
https://gizza.ai/tools/file-list-sorter/?paths=img10.png%0Aimg2.png%0Asrc%2Fmain.rs%0Adocs%2Freport.docx%0AREADME.md&sort_by=natural&order=asc&ignore_case=true&dirs_first=true&group_by_dir=true&unique=true&trim=true&format=listMachine-readable descriptor: tool.json — title + parameters JSON Schema for agents.