9001/copyparty

💾🎉 copyparty

turn almost any device into a file server with resumable uploads/downloads using any web browser

• server only needs Python (2 or 3), all dependencies optional
• 🔌 protocols: http(s) // webdav // sftp // ftp(s) // tftp // smb/cifs
• 📱 android app // iPhone shortcuts

👉 Get started! or visit the read-only demo server 👀 running on a nuc in my basement

📷 screenshots: browser // upload // unpost // thumbnails // search // fsearch // zip-DL // md-viewer

🎬 videos: upload // cli-upload // race-the-beam // 👉 feature-showcase (youtube)

built in Norway 🇳🇴 with contributions from not-norway

readme toc

• top
  • quickstart - just run copyparty-sfx.py -- that's it! 🎉
    • mirrors - other places to download copyparty from
    • at home - make it accessible over the internet
    • on servers - you may also want these, especially on servers
  • features - also see comparison to similar software
  • testimonials - small collection of user feedback
• motivations - project goals / philosophy
  • notes - general notes
• bugs - roughly sorted by chance of encounter
  • not my bugs - same order here too
• breaking changes - upgrade notes
• FAQ - "frequently" asked questions
• accounts and volumes - per-folder, per-user permissions
  • shadowing - hiding specific subfolders
  • dotfiles - unix-style hidden files/folders
• the browser - accessing a copyparty server using a web-browser
  • tabs - the main tabs in the ui
  • hotkeys - the browser has the following hotkeys
  • navpane - switching between breadcrumbs or navpane
  • thumbnails - press g or 田 to toggle grid-view instead of the file listing
  • zip downloads - download folders (or file selections) as zip or tar files
  • uploading - drag files/folders into the web-browser to upload
    • file-search - dropping files into the browser also lets you see if they exist on the server
    • unpost - undo/delete accidental uploads
    • self-destruct - uploads can be given a lifetime
    • race the beam - download files while they're still uploading (demo video)
    • incoming files - the control-panel shows the ETA for all incoming files
  • file manager - cut/paste, rename, and delete files/folders (if you have permission)
  • shares - share a file or folder by creating a temporary link
  • batch rename - select some files and press F2 to bring up the rename UI
  • rss feeds - monitor a folder with your RSS reader
  • opds feeds - browse and download files from your e-book reader
  • recent uploads - list all recent uploads
  • media player - plays almost every audio format there is
    • playlists - create and play m3u8 playlists
    • creating a playlist - with a standalone mediaplayer or copyparty
    • audio equalizer - and dynamic range compressor
    • fix unreliable playback on android - due to phone / app settings
  • textfile viewer - with realtime streaming of logfiles and such (demo)
  • markdown viewer - and there are two editors
    • markdown vars - dynamic docs with serverside variable expansion
  • other tricks
  • searching - search by size, date, path/name, mp3-tags, ...
• server config - using arguments or config files, or a mix of both
  • version-checker - sleep better at night
  • logging - serverlog is sent to stdout by default
  • zeroconf - announce enabled services on the LAN (pic)
    • mdns - LAN domain-name and feature announcer
    • ssdp - windows-explorer announcer
  • qr-code - print a qr-code (screenshot) for quick access
  • ftp server - an FTP server can be started using --ftp 3921
  • sftp server - goes roughly 700 MiB/s (slower than webdav and ftp)
  • webdav server - with read-write support
    • connecting to webdav from windows - using the GUI
  • tftp server - a TFTP server (read/write) can be started using --tftp 3969
  • smb server - unsafe, slow, not recommended for wan
  • wopi server - edit office documents in the web-ui
  • browser ux - tweaking the ui
  • opengraph - discord and social-media embeds
  • file deduplication - enable symlink-based upload deduplication
  • file indexing - enable music search, upload-undo, and better dedup
    • exclude-patterns - to save some time
    • filesystem guards - avoid traversing into other filesystems
    • periodic rescan - filesystem monitoring
  • u2sz - configure upload-chunking
  • upload rules - set upload rules using volflags
  • compress uploads - files can be autocompressed on upload
  • chmod and chown - per-volume filesystem-permissions and ownership
  • other flags
  • descript.ion - add a description to each file in a folder
  • dothidden - cosmetically hide specific files in a folder
  • thumbnail pregen - if you want to pre-generate everything on startup
  • database location - in-volume (.hist/up2k.db, default) or somewhere else
  • metadata from audio files - set -e2t to index tags on upload
    • metadata from xattrs - unix extended file attributes
  • file parser plugins - provide custom parsers to index additional tags
  • event hooks - trigger a program on uploads, renames etc (examples)
    • zeromq - event-hooks can send zeromq messages
    • upload events - the older, more powerful approach (examples)
  • handlers - redefine behavior with plugins (examples)
  • ip auth - autologin based on IP range (CIDR)
    • restrict to ip - limit a user to certain IP ranges (CIDR)
  • identity providers - replace copyparty passwords with oauth and such
    • generic header auth - other ways to auth by header
  • user-changeable passwords - if permitted, users can change their own passwords
  • using the cloud as storage - connecting to an aws s3 bucket and similar
  • hiding from google - tell search engines you don't wanna be indexed
  • themes
  • complete examples
  • listen on port 80 and 443 - become a real webserver
  • reverse-proxy - running copyparty next to other websites
    • real-ip - teaching copyparty how to see client IPs
    • reverse-proxy performance
  • permanent cloudflare tunnel - if you have a domain and want to get your copyparty online real quick
  • prometheus - metrics/stats can be enabled
  • other extremely specific features - you'll never find a use for these
    • custom mimetypes - change the association of a file extension
    • GDPR compliance - imagine using copyparty professionally...
    • feature chickenbits - buggy feature? rip it out
    • feature beefybits - force-enable features with known issues on your OS/env
• packages - the party might be closer than you think
  • arch package - pacman -S copyparty (in arch linux extra)
  • fedora package - does not exist yet
  • gentoo ::guru package - emerge www-servers/copyparty::guru (in ::guru)
  • homebrew formulae - brew install copyparty ffmpeg
  • nix package - nix profile install github:9001/copyparty
  • nixos module
• browser support - TLDR: yes
• server hall of fame - unexpected things that run copyparty
• client examples - interact with copyparty using non-browser clients
  • folder sync - sync folders to/from copyparty
  • mount as drive - a remote copyparty server as a local filesystem
• android app - upload to copyparty with one tap
• iOS shortcuts - there is no iPhone app, but
• performance - defaults are usually fine - expect 8 GiB/s download, 1 GiB/s upload
  • client-side - when uploading files
• security - there is a discord server with announcements
  • gotchas - behavior that might be unexpected
  • csp nonce - unauthorized javascript is blocked
  • cors - cross-site request config
  • filekeys - prevent filename bruteforcing
    • dirkeys - share specific folders in a volume
  • password hashing - you can hash passwords
  • https - both HTTP and HTTPS are accepted
• recovering from crashes
  • client crashes
    • firefox wsod - firefox 87 can crash during uploads
• HTTP API - see devnotes
• dependencies - mandatory deps
  • optional dependencies - enable bonus features
    • dependency chickenbits - prevent loading an optional dependency
    • dependency unvendoring - force use of system modules
  • optional gpl stuff
• sfx - the self-contained "binary" (recommended!)
  • copyparty.exe - download copyparty.exe (win8+) or copyparty32.exe (win7+)
  • zipapp - another emergency alternative, copyparty.pyz
• install on android
• install on iOS
• reporting bugs - ideas for context to include, and where to submit them
• devnotes - for build instructions etc, see ./docs/devnotes.md

quickstart

just run copyparty-sfx.py -- that's it! 🎉

ℹ️ the sfx is a self-extractor which unpacks an embedded tar.gz into $TEMP -- if this looks too scary, you can use the zipapp which has slightly worse performance

• or install through pypi: python3 -m pip install --user -U copyparty
• or if you cannot install python, you can use copyparty.exe instead
• or install on arch / homebrew ╱ on NixOS ╱ through nix
• or if you are on android, install copyparty in termux
• or maybe an iPhone or iPad? install in a-Shell on iOS
• or maybe you have a synology nas / dsm
• or if you have uv installed, run uv tool run copyparty
• or if your computer is messed up and nothing else works, try the pyz
• or if your OS is dead, give the bootable flashdrive / cd-rom a spin
• or if you don't trust copyparty yet and want to isolate it a little, then...
  • ...maybe prisonparty to create a tiny chroot (very portable),
  • ...or bubbleparty to wrap it in bubblewrap (much better)
• or if you prefer to use docker 🐋 you can do that too
  • docker has all deps built-in, so skip this step:

enable thumbnails (images/audio/video), media indexing, and audio transcoding by installing some recommended deps:

• Alpine: apk add py3-pillow ffmpeg
• Debian: apt install --no-install-recommends python3-pil ffmpeg
• Fedora: rpmfusion + dnf install python3-pillow ffmpeg --allowerasing
• FreeBSD: pkg install py311-sqlite3 py311-pillow ffmpeg
• MacOS: port install py-Pillow ffmpeg
• MacOS (alternative): brew install pillow ffmpeg
• Windows: python -m pip install --user -U Pillow
  • install python and ffmpeg manually; do not use winget or Microsoft Store (it breaks $PATH)
  • copyparty.exe comes with Pillow and only needs ffmpeg for mediatags/videothumbs
• see optional dependencies to enable even more features

running copyparty without arguments (for example doubleclicking it on Windows) will give everyone read/write access to the current folder; you may want accounts and volumes

or see some usage examples for inspiration, or the complete windows example

some recommended options:

• -e2dsa enables general file indexing
• -e2ts enables audio metadata indexing (needs either FFprobe or Mutagen)
• -v /mnt/music:/music:r:rw,foo -a foo:bar shares /mnt/music as /music, readable by anyone, and read-write for user foo, password bar
  • replace :r:rw,foo with :r,foo to only make the folder readable by foo and nobody else
  • see accounts and volumes (or --help-accounts) for the syntax and other permissions

mirrors

other places to download copyparty from (non-github links):

• https://copyparty.eu/ (hetzner, finland, official mirror):
  • https://copyparty.eu/py = https://copyparty.eu/copyparty-sfx.py = the sfx
  • https://copyparty.eu/en = https://copyparty.eu/copyparty-en.py = the english-only sfx
  • https://copyparty.eu/pyz = https://copyparty.eu/copyparty.pyz = the zipapp
  • https://copyparty.eu/enz = https://copyparty.eu/copyparty-en.pyz = the enterprise pyz
  • https://copyparty.eu/cli = online cli helptext

DO NOT download from sourceforge

at home

make it accessible over the internet by starting a cloudflare quicktunnel like so:

first download cloudflared and then start the tunnel with cloudflared tunnel --url http://127.0.0.1:3923

as the tunnel starts, it will show a URL which you can share to let anyone browse your stash or upload files to you

but if you have a domain, then you probably want to skip the random autogenerated URL and instead make a permanent cloudflare tunnel

since people will be connecting through cloudflare, run copyparty with --xff-hdr cf-connecting-ip to detect client IPs correctly

on servers

you may also want these, especially on servers:

• contrib/systemd/copyparty.service to run copyparty as a systemd service (see guide inside)
• contrib/systemd/prisonparty.service to run it in a chroot (for extra security)
• contrib/podman-systemd/ to run copyparty in a Podman container as a systemd service (see guide inside)
• contrib/openrc/copyparty to run copyparty on Alpine / Gentoo
• contrib/rc/copyparty to run copyparty on FreeBSD
• nixos module to run copyparty on NixOS hosts
• contrib/nginx/copyparty.conf to reverse-proxy behind nginx (for better https)

because the following environment variables are commonly used in service-scripts, they are understood by copyparty:

• NOTIFY_SOCKET as provided by systemd with service type=notify (see systemd/copyparty.service above)
• S6_NOTIFY_FD for dinit ready-notification = pipevar:S6_NOTIFY_FD and s6 supervision suite notification support

and remember to open the ports you want; here's a complete example including every feature copyparty has to offer:

firewall-cmd --permanent --add-port={80,443,3921,3922,3923,3945,3990}/tcp  # --zone=libvirt
firewall-cmd --permanent --add-port=12000-12099/tcp  # --zone=libvirt
firewall-cmd --permanent --add-port={69,1900,3969,5353}/udp  # --zone=libvirt
firewall-cmd --reload

(69:tftp, 1900:ssdp, 3921:ftp, 3922:sftp, 3923:http/https, 3945:smb, 3969:tftp, 3990:ftps, 5353:mdns, 12000:passive-ftp)

features

also see comparison to similar software

• backend stuff
  • ☑ IPv6 + unix-sockets
  • ☑ multiprocessing (actual multithreading)
  • ☑ volumes (mountpoints)
  • ☑ accounts
  • ☑ ftp server
  • ☑ tftp server
  • ☑ webdav server
  • ☑ smb/cifs server
  • ☑ qr-code for quick access
  • ☑ upnp / zeroconf / mdns / ssdp
  • ☑ event hooks / script runner
  • ☑ reverse-proxy support
  • ☑ cross-platform (Windows, Linux, Macos, Android, iOS, FreeBSD, arm32/arm64, ppc64le, s390x, risc-v/riscv64, SGI IRIX)
• upload
  • ☑ basic: plain multipart, ie6 support
  • ☑ up2k: js, resumable, multithreaded
    • no filesize limit! even on Cloudflare
  • ☑ stash: simple PUT filedropper
  • ☑ filename randomizer
  • ☑ write-only folders
  • ☑ unpost: undo/delete accidental uploads
  • ☑ self-destruct (specified server-side or client-side)
  • ☑ race the beam (almost like peer-to-peer)
  • ☑ symlink/discard duplicates (content-matching)
• download
  • ☑ single files in browser
  • ☑ folders as zip / tar files
  • ☑ FUSE client (read-only)
• browser
  • ☑ navpane (directory tree sidebar)
  • ☑ file manager (cut/paste, delete, batch-rename)
  • ☑ audio player (with OS media controls and opus/mp3 transcoding)
    • ☑ play video files as audio (converted on server)
    • ☑ create and play m3u8 playlists
  • ☑ image gallery with webm player
    • ☑ and cbz manga/comics reader
  • ☑ textfile browser with syntax highlighting
    • ☑ realtime streaming of growing files (logfiles and such)
  • ☑ thumbnails
    • ☑ ...of images using Pillow, pyvips, or FFmpeg
    • ☑ ...of RAW images using libraw-dcraw_emu or rawpy
    • ☑ ...of videos using FFmpeg
    • ☑ ...of audio (spectrograms) using FFmpeg
    • ☑ cache eviction (max-age; maybe max-size eventually)
  • ☑ multilingual UI (english, norwegian, chinese, add your own))
  • ☑ SPA (browse while uploading)
• server indexing
  • ☑ locate files by contents
  • ☑ search by name/path/date/size
  • ☑ search by ID3-tags etc.
• client support
  • ☑ folder sync (one-way only; full sync will never be supported)
  • ☑ curl-friendly
  • ☑ opengraph (discord embeds)
• markdown
  • ☑ viewer
  • ☑ editor (sure why not)
  • ☑ variables

PS: something missing? post any crazy ideas you've got as a feature request or discussion 🤙

testimonials

small collection of user feedback

good enough, surprisingly correct, certified good software, just works, why, wow this is better than nextcloud

• UI просто ужасно. Если буду описывать детально не смогу удержаться в рамках приличий

motivations

project goals / philosophy

• inverse unix philosophy -- do all the things, and do an okay job
  • quick drop-in service to get a lot of features in a pinch
  • some of the alternatives might be a better fit for you
• run anywhere, support everything
  • as many web-browsers and python versions as possible
    • every browser should at least be able to browse, download, upload files
    • be a good emergency solution for transferring stuff between ancient boxes
  • minimal dependencies
    • but optional dependencies adding bonus-features are ok
    • everything being plaintext makes it possible to proofread for malicious code
  • no preparations / setup necessary, just run the sfx (which is also plaintext)
• adaptable, malleable, hackable
  • no build steps; modify the js/python without needing node.js or anything like that

becoming rich is specifically not a motivation, but if you wanna donate then see my github profile regarding donations for my FOSS stuff in general (also THANKS!)

notes

general notes:

• paper-printing is affected by dark/light-mode! use lightmode for color, darkmode for grayscale
  • because no browsers currently implement the media-query to do this properly orz

browser-specific:

• iPhone/iPad: use Firefox to download files
• Android-Chrome: increase "parallel uploads" for higher speed (android bug)
• Android-Firefox: takes a while to select files (their fix for ☝️)
• Desktop-Firefox: may use gigabytes of RAM if your files are massive seems to be OK now
• Desktop-Firefox: may stop you from unplugging USB flashdrives until you visit about:memory and click Minimize memory usage

server-os-specific:

• RHEL8 / Rocky8: you can run copyparty using /usr/libexec/platform-python

server notes:

• pypy is supported but regular cpython is faster if you enable the database

bugs

roughly sorted by chance of encounter

• general:

  • --th-ff-jpg may fix video thumbnails on some FFmpeg versions (macos, some linux)
  • --th-ff-swr may fix audio thumbnails on some FFmpeg versions
  • if the up2k.db (filesystem index) is on a samba-share or network disk, you'll get unpredictable behavior if the share is disconnected for a bit
    • use --hist or the hist volflag (-v [...]:c,hist=/tmp/foo) to place the db and thumbnails on a local disk instead
    • or, if you only want to move the db (and not the thumbnails), then use --dbpath or the dbpath volflag
  • all volumes must exist / be available on startup; up2k (mtp especially) gets funky otherwise
  • probably more, pls let me know

• python 3.4 and older (including 2.7):

  • many rare and exciting edge-cases because python didn't handle EINTR yet
    • downloads from copyparty may suddenly fail, but uploads should be fine

• python 2.7 on Windows:

  • cannot index non-ascii filenames with -e2d
  • cannot handle filenames with mojibake

if you have a new exciting bug to share, see reporting bugs

not my bugs

same order here too

• Chrome issue 1317069 -- if you try to upload a folder which contains symlinks by dragging it into the browser, the symlinked files will not get uploaded

• Chrome issue 1352210 -- plaintext http may be faster at filehashing than https (but also extremely CPU-intensive)

• Chrome issue 383568268 -- filereaders in webworkers can OOM / crash the browser-tab

  • copyparty has a workaround which seems to work well enough

• Firefox issue 1790500 -- entire browser can crash after uploading ~4000 small files

• Windows: Uploading from a webbrowser may fail with "directory iterator got stuck" due to the max path length; try moving the files somewhere shorter before uploading

• Android: music playback randomly stops due to battery usage settings

• iPhones: the volume control doesn't work because apple doesn't want it to

  • AudioContext will probably never be a viable workaround as apple introduces new issues faster than they fix current ones

• iPhones: music volume goes on a rollercoaster during song changes

  • nothing I can do about it because AudioContext is still broken in safari

• iPhones: the preload feature (in the media-player-options tab) can cause a tiny audio glitch 20sec before the end of each song, but disabling it may cause worse iOS bugs to appear instead

  • just a hunch, but disabling preloading may cause playback to stop entirely, or possibly mess with bluetooth speakers
  • tried to add a tooltip regarding this but looks like apple broke my tooltips

• iPhones: preloaded awo files make safari log MEDIA_ERR_NETWORK errors as playback starts, but the song plays just fine so eh whatever

  • awo, opus-weba, is apple's new take on opus support, replacing opus-caf which was technically limited to cbr opus

• iPhones: preloading another awo file may cause playback to stop

  • can be somewhat mitigated with mp.au.play() in mp.onpreload but that can hit a race condition in safari that starts playing the same audio object twice in parallel...

• Windows: folders cannot be accessed if the name ends with .

  • python or windows bug

• Windows: msys2-python 3.8.6 occasionally throws RuntimeError: release unlocked lock when leaving a scoped mutex in up2k

  • this is an msys2 bug, the regular windows edition of python is fine

• VirtualBox: sqlite throws Disk I/O Error when running in a VM and the up2k database is in a vboxsf

  • use --hist or the hist volflag (-v [...]:c,hist=/tmp/foo) to place the db and thumbnails inside the vm instead
    • or, if you only want to move the db (and not the thumbnails), then use --dbpath or the dbpath volflag
  • also happens on mergerfs, so put the db elsewhere

• Ubuntu: dragging files from certain folders into firefox or chrome is impossible

  • due to snap security policies -- see snap connections firefox for the allowlist, removable-media permits all of /mnt and /media apparently

breaking changes

upgrade notes

• 1.9.16 (2023-11-04):
  • --stats/prometheus: cpp_bans renamed to cpp_active_bans, and that + cpp_uptime are gauges
• 1.6.0 (2023-01-29):
  • http-api: delete/move is now POST instead of GET
  • everything other than GET and HEAD must pass cors validation
• 1.5.0 (2022-12-03): new chunksize formula for files larger than 128 GiB
  • users: upgrade to the latest cli uploader if you use that
  • devs: update third-party up2k clients (if those even exist)

FAQ

"frequently" asked questions

• CopyParty?

  • nope! the name is either copyparty (all-lowercase) or Copyparty -- it's one word after all :>

• what is a volflag?

  • per-volume configuration; many (not all) global-options can be set as volflags, and most (not all) volflags can be set as global-options; complete list of volflags

• what is a volume?

  • a mapping from a URL (/music/) to a folder on your server's local filesystem (C:\Users\ed\Music) which can then be accessed through copyparty, depending on the permissions and options you set on it -- see accounts and volumes

• can I change the 🌲 spinning pine-tree loading animation?

  • yeah... :-(

• is it possible to block read-access to folders unless you know the exact URL for a particular file inside?

  • yes, using the g permission, see the examples there
  • you can also do this with linux filesystem permissions; chmod 111 music will make it possible to access files and folders inside the music folder but not list the immediate contents -- also works with other software, not just copyparty

• can I link someone to a password-protected volume/file by including the password in the URL?

  • yes, by adding ?pw=hunter2 to the end; replace ? with & if there are parameters in the URL already, meaning it contains a ? near the end
    • if you have enabled --usernames then do ?pw=username:password instead
    • ?pw can be disabled with --pw-urlp=A but this breaks support for many clients

• how do I stop .hist folders from appearing everywhere on my HDD?

  • by default, a .hist folder is created inside each volume for the filesystem index, thumbnails, audio transcodes, and markdown document history. Use the --hist global-option or the hist volflag to move it somewhere else; see database location

• can I make copyparty download a file to my server if I give it a URL?

  • yes, using hooks

• firefox refuses to connect over https, saying "Secure Connection Failed" or "SEC_ERROR_BAD_SIGNATURE", but the usual button to "Accept the Risk and Continue" is not shown

  • firefox has corrupted its certstore; fix this by exiting firefox, then find and delete the file named cert9.db somewhere in your firefox profile folder

• the server keeps saying thank you for playing when I try to access the website

  • you've gotten banned for malicious traffic! if this happens by mistake, and you're running a reverse-proxy and/or something like cloudflare, see real-ip on how to fix this

• copyparty seems to think I am using http, even though the URL is https

  • your reverse-proxy is not sending the X-Forwarded-Proto: https header; this could be because your reverse-proxy itself is confused. Ensure that none of the intermediates (such as cloudflare) are terminating https before the traffic hits your entrypoint

• thumbnails are broken (you get a colorful square which says the filetype instead)

  • you need to install FFmpeg or Pillow; see thumbnails

• thumbnails are broken, specifically for photos and videos taken by iphones

  • the docker image and bootable flashdrive are not able to read heif/heic images and h265/HEVC video due to legal reasons

• thumbnails are broken (some images appear, but other files just get a blank box, and/or the broken-image placeholder)

  • probably due to a reverse-proxy messing with the request URLs and stripping the query parameters (?th=w), so check your URL rewrite rules
  • could also be due to incorrect caching settings in reverse-proxies and/or CDNs, so make sure that nothing is set to ignore the query string
  • could also be due to misbehaving privacy-related browser extensions, so try to disable those

• i want to learn python and/or programming and am considering looking at the copyparty source code in that occasion

  •  _|  _      __   _  _|_
    (_| (_)     | | (_)  |_
    

accounts and volumes

per-folder, per-user permissions - if your setup is getting complex, consider making a config file instead of using arguments

• much easier to manage, and you can modify the config at runtime with systemctl reload copyparty or more conveniently using the [reload cfg] button in the control-panel (if the user has a/admin in any volume)
  • changes to the [global] config section requires a restart to take effect
• when using config-files, ignore all the -a and -v stuff below, and CTRL-F config file example on this page instead

a quick summary can be seen using --help-accounts

configuring accounts/volumes with arguments:

• -a usr:pwd adds account usr with password pwd
• -v .::r adds current-folder . as the webroot, readable by anyone
  • the syntax is -v src:dst:perm:perm:... so local-path, url-path, and one or more permissions to set
  • granting the same permissions to multiple accounts:
    -v .::r,usr1,usr2:rw,usr3,usr4 = usr1/2 read-only, 3/4 read-write

permissions:

• r (read): browse folder contents, download files, download as zip/tar, see filekeys/dirkeys
• w (write): upload files, move/copy files into this folder
• m (move): move files/folders from this folder
• d (delete): delete files/folders
• . (dots): user can ask to show dotfiles in directory listings
• g (get): only download files, cannot see folder contents or zip/tar
• G (upget): same as g except uploaders get to see their own filekeys (see fk in examples below)
• h (html): same as g except folders return their index.html, and filekeys are not necessary for index.html
• a (admin): can see upload time, uploader IPs, config-reload
• A ("all"): same as rwmda. (read/write/move/delete/admin/dotfiles)

examples:

• add accounts named u1, u2, u3 with passwords p1, p2, p3: -a u1:p1 -a u2:p2 -a u3:p3
• make folder /srv the root of the filesystem, read-only by anyone: -v /srv::r
• make folder /mnt/music available at /music, read-only for u1 and u2, read-write for u3: -v /mnt/music:music:r,u1,u2:rw,u3
  • unauthorized users accessing the webroot can see that the music folder exists, but cannot open it
• make folder /mnt/incoming available at /inc, write-only for u1, read-move for u2: -v /mnt/incoming:inc:w,u1:rm,u2
  • unauthorized users accessing the webroot can see that the inc folder exists, but cannot open it
  • u1 can open the inc folder, but cannot see the contents, only upload new files to it
  • u2 can browse it and move files from /inc into any folder where u2 has write-access
• make folder /mnt/ss available at /i, read-write for u1, get-only for everyone else, and enable filekeys: -v /mnt/ss:i:rw,u1:g:c,fk=4
  • c,fk=4 sets the fk (filekey) volflag to 4, meaning each file gets a 4-character accesskey
  • u1 can upload files, browse the folder, and see the generated filekeys
  • other users cannot browse the folder, but can access the files if they have the full file URL with the filekey
  • replacing the g permission with wg would let anonymous users upload files, but not see the required filekey to access it
  • replacing the g permission with wG would let anonymous users upload files, receiving a working direct link in return

if you want to grant access to all users who are logged in, the group acct will always contain all known users, so for example -v /mnt/music:music:r,@acct

• to do the opposite, granting access to everyone who is NOT logged in. *,-@acct does the trick, for example -v /srv/welcome:welcome:r,*,-@acct
• single users can also be subtracted from a group: @admins,-james

anyone trying to bruteforce a password gets banned according to --ban-pw; default is 24h ban for 9 failed attempts in 1 hour

and if you want to use config files instead of commandline args (good!) then here's the same examples as a configfile; save it as foobar.conf and use it like this: python copyparty-sfx.py -c foobar.conf

• you can also PRTY_CONFIG=foobar.conf python copyparty-sfx.py (convenient in docker etc)

config file example:

[accounts]
  u1: p1  # create account "u1" with password "p1"
  u2: p2  #  (note that comments must have
  u3: p3  #   two spaces before the # sign)

[groups]
  g1: u1, u2  # create a group

[/]     # this URL will be mapped to...
  /srv  # ...this folder on the server filesystem
  accs:
    r: *  # read-only for everyone, no account necessary

[/music]       # create another volume at this URL,
  /mnt/music   # which is mapped to this folder
  accs:
    r: u1, u2  # only these accounts can read,
    r: @g1     # (exactly the same, just with a group instead)
    r: @acct   # (alternatively, ALL users who are logged in)
    rw: u3     # and only u3 can read-write

[/inc]
  /mnt/incoming
  accs:
    w: u1   # u1 can upload but not see/download any files,
    rm: u2  # u2 can browse + move files out of this volume

[/i]
  /mnt/ss
  accs:
    rw: u1  # u1 can read-write,
    g: *    # everyone can access files if they know the URL
  flags:
    fk: 4   # each file URL will have a 4-character password

shadowing

hiding specific subfolders by mounting another volume on top of them

for example -v /mnt::r -v /var/empty:web/certs: (note: no permissions) mounts the server folder /mnt as the webroot, but another volume is mounted at /web/certs -- so visitors can only see the contents of /mnt and /mnt/web (at URLs / and /web), but not /mnt/web/certs because URL /web/certs is mapped to /var/empty

to fully unmap it from the filesystem, specify //NULL instead of a real path such as /var/empty, so for example -v /mnt::r -v //NULL:web/certs: ensures /web/certs will never be accessible by anyone

the example config file right above this section may explain this better; the first volume / is mapped to /srv which means http://127.0.0.1:3923/music would try to read /srv/music on the server filesystem, but since there's another volume at /music mapped to /mnt/music then it'll go to /mnt/music instead

so, to shadow a file/folder, define a volume but leave out the accs: section

ℹ️ this also works for single files, because files can also be volumes

config file example for unmapping folders by shadowing:

[/drives]
  /mnt   # url "/drives" goes to "/mnt"
  accs:
    r: *  # everyone can read

[/drives/foo/bar]
  //NULL  # blocks access to "/mnt/foo/bar"

dotfiles

unix-style hidden files/folders by starting the name with a dot

anyone can access these if they know the name, but they normally don't appear in directory listings

a client can request to see dotfiles in directory listings if global option -ed is specified, or the volume has volflag dots, or the user has permission .

for shares, the dots volflag is ignored

dotfiles do not appear in search results unless one of the above is true, and the global option / volflag dotsrch is set

even if user has permission to see dotfiles, they are default-hidden unless --see-dots is set, and/or user has enabled the dotfiles option in the settings tab

config file example, where the same permission to see dotfiles is given in two different ways just for reference:

[/foo]
  /srv/foo
  accs:
    r.: ed   # user "ed" has read-access + dot-access in this volume;
             # dotfiles are visible in listings, but not in searches
  flags:
    dotsrch  # dotfiles will now appear in search results too
    dots     # another way to let everyone see dotfiles in this vol

the browser

accessing a copyparty server using a web-browser

tabs

the main tabs in the ui

• [🔎] search by size, date, path/name, mp3-tags ...
• [🧯] unpost: undo/delete accidental uploads
• [🚀] and [🎈] are the uploaders
• [📂] mkdir: create directories
• [📝] new-file: create a new textfile
• [📟] send-msg: either to server-log or into textfiles if --urlform save
• [🎺] audio-player config options
• [⚙️] general client config options

hotkeys

the browser has the following hotkeys (always qwerty)

• ? show hotkeys help
• B toggle breadcrumbs / navpane
• I/K prev/next folder
• M parent folder (or unexpand current)
• V toggle folders / textfiles in the navpane
• G toggle list / grid view -- same as 田 bottom-right
• T toggle thumbnails / icons
• ESC close various things
• ctrl-K delete selected files/folders
• ctrl-X cut selected files/folders
• ctrl-C copy selected files/folders to clipboard
• ctrl-V paste (move/copy)
• Y download selected files
• F2 rename selected file/folder
• when a file/folder is selected (in not-grid-view):
  • Up/Down move cursor
  • shift+Up/Down select and move cursor
  • ctrl+Up/Down move cursor and scroll viewport
  • Space toggle file selection
  • Ctrl-A toggle select all
• when a textfile is open:
  • I/K prev/next textfile
  • S toggle selection of open file
  • M close textfile
• when playing audio:
  • J/L prev/next song
  • U/O skip 10sec back/forward
  • 0..9 jump to 0%..90%
  • P play/pause (also starts playing the folder)
  • Y download file
• when viewing images / playing videos:
  • J/L, Left/Right prev/next file
  • Home/End first/last file
  • F toggle fullscreen
  • S toggle selection
  • R rotate clockwise (shift=ccw)
  • Y download file
  • Esc close viewer
  • videos:
    • U/O skip 10sec back/forward
    • 0..9 jump to 0%..90%
    • P/K/Space play/pause
    • M mute
    • C continue playing next video
    • V loop entire file
    • [ loop range (start)
    • ] loop range (end)
• when the navpane is open:
  • A/D adjust tree width
• in the grid view:
  • S toggle multiselect
  • shift+A/D zoom
• in the markdown editor:
  • ^s save
  • ^h header
  • ^k autoformat table
  • ^u jump to next unicode character
  • ^e toggle editor / preview
  • ^up, ^down jump paragraphs

navpane

switching between breadcrumbs or navpane

click the 🌲 or pressing the B hotkey to toggle between breadcrumbs path (default), or a navpane (tree-browser sidebar thing)

• [+] and [-] (or hotkeys A/D) adjust the size
• [🎯] jumps to the currently open folder
• [📃] toggles between showing folders and textfiles
• [📌] shows the name of all parent folders in a docked panel
• [a] toggles automatic widening as you go deeper
• [↵] toggles wordwrap
• [👀] show full name on hover (if wordwrap is off)

thumbnails

press g or 田 to toggle grid-view instead of the file listing and t toggles icons / thumbnails

• can be made default globally with --grid or per-volume with volflag grid
• enable by adding ?imgs to a link, or disable with ?imgs=0

it does static images with Pillow / pyvips / FFmpeg, and uses FFmpeg for video files, so you may want to --no-thumb or maybe just --no-vthumb depending on how dangerous your users are

• Pillow is 3x faster (and safer) than FFmpeg
• disable thumbnails for specific volumes with volflag dthumb for all, or dvthumb / dathumb / dithumb for video/audio/images only
• for installing FFmpeg on windows, see optional dependencies

audio files are converted into spectrograms using FFmpeg unless you --no-athumb (and some FFmpeg builds may need --th-ff-swr)

images with the following names (see --th-covers) become the thumbnail of the folder they're in: folder.png, folder.jpg, cover.png, cover.jpg

• the order is significant, so if both cover.png and folder.jpg exist in a folder, it will pick the first matching --th-covers entry (folder.jpg)
• and, if you enable file indexing, it will also try those names as dotfiles (.folder.jpg and so), and then fallback on the first picture in the folder (if it has any pictures at all)
• disable folderthumbs with --th-covers no

enabling multiselect lets you click files to select them, and then shift-click another file for range-select

• multiselect is mostly intended for phones/tablets, but the sel option in the [⚙️] settings tab is better suited for desktop use, allowing selection by CTRL-clicking and range-selection with SHIFT-click, all without affecting regular clicking
  • the sel option can be made default globally with --gsel or per-volume with volflag gsel

to show /icons/exe.png and /icons/elf.gif as the thumbnail for all .exe and .elf files respectively, do this: --ext-th=exe=/icons/exe.png --ext-th=elf=/icons/elf.gif

• optionally as separate volflags for each mapping; see config file example below
• the supported image formats are jpg, png, gif, webp, ico
  • be careful with svg; chrome will crash if you have too many unique svg files showing on the same page (the limit is 250 or so) -- showing the same handful of svg files thousands of times is ok however

note:

• heif/heifs/heic/heics images usually require the libvips optional dependency but this is not possible with the docker-images due to legal reasons
• if you do not want thumbnails to be generated on-the-fly, and instead wish to generate all of them on server startup, then see thumbnail pregen

config file example:

[global]
  no-thumb   # disable ALL thumbnails and audio transcoding
  no-vthumb  # only disable video thumbnails

[/music]
  /mnt/nas/music
  accs:
    r: *     # everyone can read
  flags:
    dthumb   # disable ALL thumbnails and audio transcoding
    dvthumb  # only disable video thumbnails
    ext-th:  exe=/ico/exe.png  # /ico/exe.png is the thumbnail of *.exe
    ext-th:  elf=/ico/elf.gif  # ...and /ico/elf.gif is used for *.elf
    th-covers:  folder.png,folder.jpg,cover.png,cover.jpg  # the default

zip downloads

download folders (or file selections) as zip or tar files

select which type of archive you want in the [⚙️] config tab:

• gzip default level is 3 (0=fast, 9=best), change with ?tar=gz:9
• xz default level is 1 (0=fast, 9=best), change with ?tar=xz:9
• bz2 default level is 2 (1=fast, 9=best), change with ?tar=bz2:9
• hidden files (dotfiles) are excluded unless account is allowed to list them
  • up2k.db and dir.txt is always excluded
• bsdtar supports streaming unzipping: curl foo?zip | bsdtar -xv
  • good, because copyparty's zip is faster than tar on small files
    • but ?tar is better for large files, especially if the total exceeds 4 GiB
• zip_crc will take longer to download since the server has to read each file twice
  • this is only to support MS-DOS PKZIP v2.04g (october 1993) and older
    • how are you accessing copyparty actually

you can also zip a selection of files or folders by clicking them in the browser, that brings up a selection editor and zip button in the bottom right

cool trick: download a folder by appending url-params ?tar&opus or ?tar&mp3 to transcode all audio files (except aac|m4a|mp3|ogg|opus|wma) to opus/mp3 before they're added to the archive

• super useful if you're 5 minutes away from takeoff and realize you don't have any music on your phone but your server only has flac files and downloading those will burn through all your data + there wouldn't be enough time anyways
• and url-param &name=foo changes the name of the toplevel folder in the archive to foo, and just &name removes the folder entirely
• and url-param &nodot skips dotfiles/dotfolders; they are included by default if your account has permission to see them
• and url-params &j / &w produce jpeg/webm thumbnails/spectrograms instead of the original audio/video/images (&p for audio waveforms)
  • can also be used to pregenerate thumbnails; combine with --th-maxage=9999999 or --th-clean=0
    • but now there is also a real thumbnail pregen so just use that

uploading

drag files/folders into the web-browser to upload

dragdrop is the recommended way, but you may also:

• select some files (not folders) in your file explorer and press CTRL-V inside the browser window
• use the command-line uploader
• upload using curl, sharex, ishare, ...

when uploading files through dragdrop or CTRL-V, this initiates an upload using up2k; there are two browser-based uploaders available:

• [🎈] bup, the basic uploader, supports almost every browser since netscape 4.0
• [🚀] up2k, the good / fancy one

NB: you can undo/delete your own uploads with [🧯] unpost (and this is also where you abort unfinished uploads, but you have to refresh the page first)

up2k has several advantages:

• you can drop folders into the browser (files are added recursively)
• files are processed in chunks, and each chunk is checksummed
  • uploads autoresume if they are interrupted by network issues
  • uploads resume if you reboot your browser or pc, just upload the same files again
  • server detects any corruption; the client reuploads affected chunks
  • the client doesn't upload anything that already exists on the server
  • no filesize limit, even when a proxy limits the request size (for example Cloudflare)
• much higher speeds than ftp/scp/tarpipe on some internet connections (mainly american ones) thanks to parallel connections
• the last-modified timestamp of the file is preserved

it is perfectly safe to restart / upgrade copyparty while someone is uploading to it!
all known up2k clients will resume just fine 💪

see up2k for details on how it works, or watch a demo video

protip: you can avoid scaring away users with contrib/plugins/minimal-up2k.js which makes it look much simpler

protip: if you enable favicon in the [⚙️] settings tab (by typing something into the textbox), the icon in the browser tab will indicate upload progress -- also, the [🔔] and/or [🔊] switches enable visible and/or audible notifications on upload completion

the up2k UI is the epitome of polished intuitive experiences:

• "parallel uploads" specifies how many chunks to upload at the same time
• [🏃] analysis of other files should continue while one is uploading
• [🥔] shows a simpler UI for faster uploads from slow devices
• [🛡️] decides when to overwrite existing files on the server
  • 🛡️ = never (generate a new filename instead)
  • 🕒 = overwrite if the server-file is older
  • ♻️ = always overwrite if the files are different
• [🎲] generate random filenames during upload
• [🔎] switch between upload and file-search mode
  • ignore [🔎] if you add files by dragging them into the browser

and then there's the tabs below it,

• [ok] is the files which completed successfully
• [ng] is the ones that failed / got rejected (already exists, ...)
• [done] shows a combined list of [ok] and [ng], chronological order
• [busy] files which are currently hashing, pending-upload, or uploading
  • plus up to 3 entries each from [done] and [que] for context
• [que] is all the files that are still queued

note that since up2k has to read each file twice, [🎈] bup can theoretically be up to 2x faster in some extreme cases (files bigger than your ram, combined with an internet connection faster than the read-speed of your HDD, or if you're uploading from a cuo2duo)

if you are resuming a massive upload and want to skip hashing the files which already finished, you can enable turbo in the [⚙️] config tab, but please read the tooltip on that button

if the server is behind a proxy which imposes a request-size limit, you can configure up2k to sneak below the limit with server-option --u2sz

if you want to replace existing files on the server with new uploads by default, run with --u2ow 2 (only works if users have the delete-permission, and can still be disabled with 🛡️ in the UI)

file-search

dropping files into the browser also lets you see if they exist on the server

when you drag/drop files into the browser, you will see two dropzones: Upload and Search

on a phone? toggle the [🔎] switch green before tapping the big yellow Search button to select your files

the files will be hashed on the client-side, and each hash is sent to the server, which checks if that file exists somewhere

files go into [ok] if they exist (and you get a link to where it is), otherwise they land in [ng]

• the main reason filesearch is combined with the uploader is cause the code was too spaghetti to separate it out somewhere else, this is no longer the case but now i've warmed up to the idea too much

if you have a "wark" (file-identifier/checksum) then you can also search for that in the [🔎] tab by putting w = kFpDiztbZc8Z1Lzi in the raw field

unpost

undo/delete accidental uploads using the [🧯] tab in the UI

you can unpost even if you don't have regular move/delete access, however only for files uploaded within the past --unpost seconds (default 12 hours) and the server must be running with -e2d

config file example:

[global]
  e2d            # enable up2k database (remember uploads)
  unpost: 43200  # 12 hours (default)

self-destruct

uploads can be given a lifetime, after which they expire / self-destruct

the feature must be enabled per-volume with the lifetime upload rule which sets the upper limit for how long a file gets to stay on the server

clients can specify a shorter expiration time using the up2k ui -- the relevant options become visible upon navigating into a folder with lifetimes enabled -- or by using the life upload modifier

specifying a custom expiration time client-side will affect the timespan in which unposts are permitted, so keep an eye on the estimates in the up2k ui

race the beam

download files while they're still uploading (demo video) -- it's almost like peer-to-peer

requires the file to be uploaded using up2k (which is the default drag-and-drop uploader), alternatively the command-line program

incoming files

the control-panel shows the ETA for all incoming files , but only for files being uploaded into volumes where you have read-access

file manager

cut/paste, rename, and delete files/folders (if you have permission)

file selection: click somewhere on the line (not the link itself), then:

• space to toggle

• up/down to move

• shift-up/down to move-and-select

• ctrl-shift-up/down to also scroll

• shift-click another line for range-select

• cut: select some files and ctrl-x

• copy: select some files and ctrl-c

• paste: ctrl-v in another folder

• rename: F2

you can copy/move files across browser tabs (cut/copy in one tab, paste in another)

shares

share a file or folder by creating a temporary link

when enabled in the server settings (--shr), click the bottom-right share button to share the folder you're currently in, or alternatively:

• select a folder first to share that folder instead
• select one or more files to share only those files

this feature was made with identity providers in mind -- configure your reverseproxy to skip the IdP's access-control for a given URL prefix and use that to safely share specific files/folders sans the usual auth checks

when creating a share, the creator can choose any of the following options:

• password-protection
• expire after a certain time; 0 or blank means infinite
• allow visitors to upload (if the user who creates the share has write-access)

semi-intentional limitations:

• cleanup of expired shares only works when global option e2d is set, and/or at least one volume on the server has volflag e2d
• only folders from the same volume are shared; if you are sharing a folder which contains other volumes, then the contents of those volumes will not be available
• if you change password hashing settings after creating a password-protected share, then that share will stop working
• related to IdP volumes being forgotten on shutdown, any shares pointing into a user's IdP volume will be unavailable until that user makes their first request after a restart
• no option to "delete after first access" because tricky
  • when linking something to discord (for example) it'll get accessed by their scraper and that would count as a hit
  • browsers wouldn't be able to resume a broken download unless the requester's IP gets allowlisted for X minutes (ref. tricky)

specify --shr /foobar to enable this feature; a toplevel virtual folder named foobar is then created, and that's where all the shares will be served from

• you can name it whatever, foobar is just an example
• if you're using config files, put shr: /foobar inside the [global] section instead

users can delete their own shares in the controlpanel, and a list of privileged users (--shr-adm) are allowed to see and/or delet any share on the server

the volflag --shr-who lets you control who can create a share from that volume, either no (nobody), a (people with admin permission), or auth (people who are logged in)

after a share has expired, it remains visible in the controlpanel for --shr-rt minutes (default is 1 day), and the owner can revive it by extending the expiration time there

security note: using this feature does not mean that you can skip the accounts and volumes section -- you still need to restrict access to volumes that you do not intend to share with unauthenticated users! it is not sufficient to use rules in the reverseproxy to restrict access to just the /share folder.

batch rename

select some files and press F2 to bring up the rename UI

quick explanation of the buttons,

• [✅ apply rename] confirms and begins renaming
• [❌ cancel] aborts and closes the rename window
• [↺ reset] reverts any filename changes back to the original name
• [decode] does a URL-decode on the filename, fixing stuff like & and %20
• [advanced] toggles advanced mode

advanced mode: rename files based on rules to decide the new names, based on the original name (regex), or based on the tags collected from the file (artist/title/...), or a mix of both

in advanced mode,

• [case] toggles case-sensitive regex
• regex is the regex pattern to apply to the original filename; any files which don't match will be skipped
• format is the new filename, taking values from regex capturing groups and/or from file tags
  • very loosely based on foobar2000 syntax
• presets lets you save rename rules for later

available functions:

• $lpad(text, length, pad_char)
• $rpad(text, length, pad_char)

two counters are available; .n.s is the nth file in the selection, and .n.d the nth file in the folder, for example rename-output file(.n.d).(ext) gives file5.bin, and beach-$lpad((.n.s),3,0).(ext) is beach-017.jpg and the initial value of each counter can be set in the textboxes underneath the preset dropdown

so,

say you have a file named meganeko - Eclipse - 07 Sirius A.mp3 (absolutely fantastic album btw) and the tags are: Album:Eclipse, Artist:meganeko, Title:Sirius A, tn:7

you could use just regex to rename it:

• regex = (.*) - (.*) - ([0-9]{2}) (.*)
• format = (3). (1) - (4)
• output = 07. meganeko - Sirius A.mp3

or you could use just tags:

• format = $lpad((tn),2,0). (artist) - (title).(ext)
• output = 7. meganeko - Sirius A.mp3

or a mix of both:

• regex = - ([0-9]{2})
• format = (1). (artist) - (title).(ext)
• output = 07. meganeko - Sirius A.mp3

the metadata keys you can use in the format field are the ones in the file-browser table header (whatever is collected with -mte and -mtp)

rss feeds

monitor a folder with your RSS reader , optionally recursive

must be enabled per-volume with volflag rss or globally with --rss

the feed includes itunes metadata for use with podcast readers such as AntennaPod

a feed example: https://cd.ocv.me/a/d2/d22/?rss&fext=mp3

url parameters:

• pw=hunter2 for password auth
  • if you enabled --usernames then do pw=username:password instead
• nopw disables embedding the password (if provided) into item-URLs in the feed
• nopw=a disables mentioning the password anywhere at all in the feed; may break some readers
• recursive to also include subfolders
• title=foo changes the feed title (default: folder name)
• fext=mp3,opus only include mp3 and opus files (default: all)
• nf=30 only show the first 30 results (default: 250)
• sort=m sort by mtime (file last-modified), newest first (default)
  • u = upload-time; NOTE: non-uploaded files have upload-time 0
  • n = filename
  • a = filesize
  • uppercase = reverse-sort; M = oldest file first

opds feeds

browse and download files from your e-book reader

enabled with the opds volflag or --opds global option

add ?opds to the end of the url you would like to browse, then input that in your opds client. for example: https://copyparty.example/books/?opds.

to log in with a password, enter it into either of the username or password fields in your client.

• if you've enabled --usernames, then you need to enter both username and password .

note: some clients (e.g. Moon+ Reader) will not send the password when downloading cover images, which will cause your ip to be banned by copyparty. to work around this, you can grant the g permission to unauthenticated requests and enable filekeys to prevent guessing filenames. for example: -vbooks:books:r,ed:g:c,fk,opds

by default, not all file types will be listed in opds feeds. to change this, add the extension to --opds-exts (volflag: opds_exts), or empty the list to list everything

recent uploads

list all recent uploads by clicking "show recent uploads" in the controlpanel

will show uploader IP and upload-time if the visitor has the admin permission

• global-option --ups-when makes upload-time visible to all users, and not just admins

• global-option --ups-who (volflag ups_who) specifies who gets access (0=nobody, 1=admins, 2=everyone), default=2

note that the 🧯 unpost feature is better suited for viewing your own recent uploads, as it includes the option to undo/delete them

config file example:

[global]
  ups-when    # everyone can see upload times
  ups-who: 1  # but only admins can see the list,
              # so ups-when doesn't take effect

media player

plays almost every audio format there is (if the server has FFmpeg installed for on-demand transcoding)

the following audio formats are usually always playable, even without FFmpeg: aac|flac|m4a|mp3|ogg|opus|wav

some highlights:

• OS integration; control playback from your phone's lockscreen (windows // iOS // android)
• shows the audio waveform in the seekbar
• not perfectly gapless but can get really close (see settings + eq below); good enough to enjoy gapless albums as intended
• videos can be played as audio, without wasting bandwidth on the video
• adding ?v to the end of an audio/video/image link will make it open in the mediaplayer

click the play link next to an audio file, or copy the link target to share it (optionally with a timestamp to start playing from, like that example does)

open the [🎺] media-player-settings tab to configure it,

• "switches":
  • [🔁] repeats one single song forever
  • [🔀] shuffles the files inside each folder
  • [preload] starts loading the next track when it's about to end, reduces the silence between songs
  • [full] does a full preload by downloading the entire next file; good for unreliable connections, bad for slow connections
  • [~s] toggles the seekbar waveform display
  • [/np] enables buttons to copy the now-playing info as an irc message
  • [📻] enables buttons to create an m3u playlist with the selected songs
  • [os-ctl] makes it possible to control audio playback from the lockscreen of your device (enables mediasession)
  • [seek] allows seeking with lockscreen controls (buggy on some devices)
  • [art] shows album art on the lockscreen
  • [🎯] keeps the playing song scrolled into view (good when using the player as a taskbar dock)
  • [⟎] shrinks the playback controls
• "buttons":
  • [uncache] may fix songs that won't play correctly due to bad files in browser cache
• "at end of folder":
  • [loop] keeps looping the folder
  • [next] plays into the next folder
• "transcode":
  • [flac] converts flac and wav files into opus (if supported by browser) or mp3
  • [aac] converts aac and m4a files into opus (if supported by browser) or mp3
  • [oth] converts all other known formats into opus (if supported by browser) or mp3
    • aac|ac3|aif|aiff|alac|alaw|amr|ape|au|dfpwm|dts|flac|gsm|it|m4a|m4b|m4r|mka|mo3|mod|mp2|mp3|mpc|mptm|mt2|mulaw|ogg|okt|opus|ra|s3m|tak|tta|ulaw|wav|wma|wv|xm|xpk
• "transcode to":
  • [opus] produces an opus whenever transcoding is necessary (the best choice on Android and PCs)
  • [awo] is opus in a weba file, good for iPhones (iOS 17.5 and newer) but Apple is still fixing some state-confusion bugs as of iOS 18.2.1
  • [caf] is opus in a caf file, good for iPhones (iOS 11 through 17), technically unsupported by Apple but works for the most part
  • [mp3] -- the myth, the legend, the undying master of mediocre sound quality that definitely works everywhere
  • [flac] -- lossless but compressed, for LAN and/or fiber playback on electrostatic headphones
  • [wav] -- lossless and uncompressed, for LAN and/or fiber playback on electrostatic headphones connected to very old equipment
    • flac and wav must be enabled with --allow-flac / --allow-wav to allow spending the disk space
• "tint" reduces the contrast of the playback bar

playlists

create and play m3u8 playlists -- see example text and player

click a file with the extension m3u or m3u8 (for example mixtape.m3u or touhou.m3u8 ) and you get two choices: Play / Edit

playlists can include songs across folders anywhere on the server, but filekeys/dirkeys are NOT supported, so the listener must have read-access or get-access to the files

creating a playlist

with a standalone mediaplayer or copyparty

you can use foobar2000, deadbeef, just about any standalone player should work -- but you might need to edit the filepaths in the playlist so they fit with the server-URLs

alternatively, you can create the playlist using copyparty itself:

• open the [🎺] media-player-settings tab and enable the [📻] create-playlist feature -- this adds two new buttons in the bottom-right tray, [📻add] and [📻copy] which appear when you listen to music, or when you select a few audiofiles

• click the 📻add button while a song is playing (or when you've selected some songs) and they'll be added to "the list" (you can't see it yet)

• at any time, click 📻copy to send the playlist to your clipboard

  • you can then continue adding more songs if you'd like
  • if you want to wipe the playlist and start from scratch, just refresh the page

• create a new textfile, name it something.m3u and paste the playlist there

audio equalizer

and dynamic range compressor

can also boost the volume in general, or increase/decrease stereo width (like crossfeed just worse)

has the convenient side-effect of reducing the pause between songs, so gapless albums play better with the eq enabled (just make it flat)

not available on iPhones / iPads because AudioContext currently breaks background audio playback on iOS (15.7.8)

fix unreliable playback on android

due to phone / app settings, android phones may randomly stop playing music when the power saver kicks in, especially at the end of an album -- you can fix it by disabling power saving in the app settings of the browser you use for music streaming (preferably a dedicated one)

textfile viewer

with realtime streaming of logfiles and such (demo) , and terminal colors work too

click -txt- next to a textfile to open the viewer, which has the following toolbar buttons:

• ✏️ edit opens the textfile editor
• 📡 follow starts monitoring the file for changes, streaming new lines in realtime
  • similar to tail -f
  • link directly to a file with tailing enabled by adding &tail to the textviewer URL

markdown viewer

and there are two editors

there is a built-in extension for inline clickable thumbnails;

• enable it by adding <!-- th --> somewhere in the doc
• add thumbnails with !th[l](your.jpg) where l means left-align (r = right-align)
• a single line with --- clears the float / inlining
• in the case of README.md being displayed below a file listing, thumbnails will open in the gallery viewer

other notes,

• the document preview has a max-width which is the same as an A4 paper when printed

markdown vars

dynamic docs with serverside variable expansion to replace stuff like {{self.ip}} with the client's IP, or {{srv.htime}} with the current time on the server

see ./srv/expand/ for usage and examples

other tricks

• you can link a particular timestamp in an audio file by adding it to the URL, such as &20 / &20s / &1m20 / &t=1:20 after the .../#af-c8960dab

• enabling the audio equalizer can help make gapless albums fully gapless in some browsers (chrome), so consider leaving it on with all the values at zero

• get a plaintext file listing by adding ?ls=t to a URL, or a compact colored one with ?ls=v (for unix terminals)

• if you are using media hotkeys to switch songs and are getting tired of seeing the OSD popup which Windows doesn't let you disable, consider ./contrib/media-osd-bgone.ps1

• click the bottom-left π to open a javascript prompt for debugging

• files named .prologue.html / .epilogue.html will be rendered before/after directory listings unless --no-logues

• files named descript.ion / DESCRIPT.ION are parsed and displayed in the file listing, or as the epilogue if nonstandard

• files named README.md / readme.md will be rendered after directory listings unless --no-readme (but .epilogue.html takes precedence)

  • and PREADME.md / preadme.md is shown above directory listings unless --no-readme or .prologue.html

• README.md and *logue.html can contain placeholder values which are replaced server-side before embedding into directory listings; see --help-exp

searching

search by size, date, path/name, mp3-tags, ...

when started with -e2dsa copyparty will scan/index all your files. This avoids duplicates on upload, and also makes the volumes searchable through the web-ui:

• make search queries by size/date/directory-path/filename, or...
• drag/drop a local file to see if the same contents exist somewhere on the server, see file-search

path/name queries are space-separated, AND'ed together, and words are negated with a - prefix, so for example:

• path: shibayan -bossa finds all files where one of the folders contain shibayan but filters out any results where bossa exists somewhere in the path
• name: demetori styx gives you good stuff

the raw field allows for more complex stuff such as ( tags like *nhato* or tags like *taishi* ) and ( not tags like *nhato* or not tags like *taishi* ) which finds all songs by either nhato or taishi, excluding collabs (terrible example, why would you do that)

for the above example to work, add the commandline argument -e2ts to also scan/index tags from music files, which brings us over to:

server config

using arguments or config files, or a mix of both:

• using a config-file (-c some.conf) is best, easier to read/maintain; see ./docs/example.conf
  • for complex and intricate setups, see how ./docs/example2.conf includes additional config-files/folders with % copyparty.d which loads ./docs/copyparty.d/some.conf
• kill -s USR1 (same as systemctl reload copyparty) to reload accounts and volumes from config files without restarting
  • or click the [reload cfg] button in the control-panel if the user has a/admin in any volume
  • changes to the [global] config section requires a restart to take effect

NB: as humongous as this readme is, there is also a lot of undocumented features. Run copyparty with --help (or click that link) to see all available global options; all of those can be used in the [global] section of config files, and everything listed in --help-flags can be used in volumes as volflags (per-volume configuration).

• if running in docker/podman, try this: docker run --rm -it copyparty/ac --help
• or if you prefer plaintext, https://copyparty.eu/helptext.txt

version-checker

sleep better at night by telling copyparty to periodically check whether your version has a known vulnerability

this feature can be enabled by setting the global-option --vc-url to one of the following URLs; choose what severity level you want to be notified for:

• https://api.copyparty.eu/advisories-panic -- only really bad stuff, the "UPGRADE NOW" kind
• https://api.copyparty.eu/advisories -- everything important / noteworthy, "upgrade when you can"
• https://api.copyparty.eu/advisories-all -- everything, including stuff that's unlikely to affect anyone
• https://api.github.com/repos/9001/copyparty/security-advisories?per_page=9 -- same as advisories-all

note that https://api.copyparty.eu/advisories may (for example) skip some advisories rated High but include some Low; that's because an easily-reachable Low in a default-enabled feature is more severe than a High which is a theoretical bug in a contrived use of a fringe feature, but the CVE calculator would still classify that as High

if you want to use the github advisory feed but only care about advisories rated medium/moderate or higher, then global-option --vc-sev medium does that, but see previous paragraph

to see what happens when a bad version is detected, try --vc-url https://api.copyparty.eu/advisories-test

also consider the following options:

• global-option --vc-age is how often (in hours) to check that URL; default is 3
• global-option --vc-exit can be enabled to panic and immediately exit if a vulnerability is indicated
  • if --vc-exit is not enabled, it just shows a warning on the controlpanel for all users with permission a or A

config file example:

[global]
  vc-url: https://api.copyparty.eu/advisories
  vc-age: 3  # how many hours to wait between each check
  vc-exit    # emergency-exit if current version is vulnerable
  vc-sev: medium  # only care about severity 'Medium'/'Moderate' or higher (github-only; don't use this with api.copyparty.eu)

logging

serverlog is sent to stdout by default (but logging to a file is also possible)

"stdout" usually means either the terminal, or journalctl, or whatever is collecting logs from your docker containers, so that depends on your setup

• -q disables logging to stdout, and may improve performance a little bit
  • combine it with -lo logfolder/cpp-%Y-%m-%d.txt to log to a file instead
  • the %Y-%m-%d makes it create a new logfile every day, with the date as filename
  • global-option --rlo decides what happens if the filename is taken
• -lo whatever.txt can be used without -q to log to both at the same time
  • by default, the logfile will have colors if the terminal does (usually the case)
  • use the textfile-viewer or less -R in a terminal to see colors correctly
• if you want no colors:
  • --flo 2 disables colors for just the logfile
  • --no-ansi disables colors for both the terminal and logfile

config file example:

[global]
  log-date: %Y-%m-%d  # show dates on stdout too
  lo: /var/log/cpp/%Y-%m-%d.txt  # logfile path
  flo: 2  # just text (no colors) in logfile
  q       # disable stdout; use logfile only

zeroconf

announce enabled services on the LAN (pic) -- -z enables both mdns and ssdp

• --z-on / --z-off limits the feature to certain networks

config file example:

[global]
  z      # enable all zeroconf features (mdns, ssdp)
  zm     # only enables mdns (does nothing since we already have z)
  z-on: 192.168.0.0/16, 10.1.2.0/24  # restrict to certain subnets

mdns

LAN domain-name and feature announcer

uses multicast dns to give copyparty a domain which any machine on the LAN can use to access it

all enabled services (webdav, ftp, smb) will appear in mDNS-aware file managers (KDE, gnome, macOS, ...)

the domain will be partybox.local if the machine's hostname is partybox unless --name specifies something else

and the web-UI will be available at http://partybox.local:3923/

• if you want to get rid of the :3923 so you can use http://partybox.local/ instead then see listen on port 80 and 443

ssdp

windows-explorer announcer

uses ssdp to make copyparty appear in the windows file explorer on all machines on the LAN

doubleclicking the icon opens the "connect" page which explains how to mount copyparty as a local filesystem

if copyparty does not appear in windows explorer, use --zsv to see why:

• maybe the discovery multicast was sent from an IP which does not intersect with the server subnets

qr-code

print a qr-code (screenshot) for quick access, great between phones on android hotspots which keep changing the subnet

• --qr enables it
• --qrs does https instead of http
• --qrl lootbox/?pw=hunter2 appends to the url, linking to the lootbox folder with password hunter2
• --qrz 1 forces 1x zoom instead of autoscaling to fit the terminal size
  • 1x may render incorrectly on some terminals/fonts, but 2x should always work
• --qr-pin 1 makes the qr-code stick to the bottom of the console (never scrolls away)
• --qr-file qr.txt:1:2 writes a small qr-code to qr.txt
• --qr-file qr.txt:2:2 writes a big qr-code to qr.txt
• --qr-file qr.svg:1:2 writes a vector-graphics qr-code to qr.svg
• --qr-file qr.png:8:4:333333:ffcc55 writes an 8x-magnified yellow-on-gray qr.png
• --qr-file qr.png:8:4::ffffff writes an 8x-magnified white-on-transparent qr.png

it uses the server hostname if mdns is enabled, otherwise it'll use your external ip (default route) unless --qri specifies a specific ip-prefix or domain

ftp server

an FTP server can be started using --ftp 3921, and/or --ftps for explicit TLS (ftpes)

• based on pyftpdlib
• needs a dedicated port (cannot share with the HTTP/HTTPS API)
• uploads are not resumable -- delete and restart if necessary
• runs in active mode by default, you probably want --ftp-pr 12000-13000
  • if you enable both ftp and ftps, the port-range will be divided in half
  • some older software (filezilla on debian-stable) cannot passive-mode with TLS
• login with any username + your password, or put your password in the username field
  • unless you enabled --usernames

some recommended FTP / FTPS clients; wark = example password:

• https://winscp.net/eng/download.php
• https://filezilla-project.org/ struggles a bit with ftps in active-mode, but is fine otherwise
• https://rclone.org/ does FTPS with tls=false explicit_tls=true
• lftp -u k,wark -p 3921 127.0.0.1 -e ls
• lftp -u k,wark -p 3990 127.0.0.1 -e 'set ssl:verify-certificate no; ls'
• curl ftp://127.0.0.1:3921/ (plaintext ftp)
• curl --ssl-reqd ftp://127.0.0.1:3990/ (encrypted ftps)

config file example, which restricts FTP to only use ports 3921 and 12000-12099 so all of those ports must be opened in your firewall:

[global]
  ftp: 3921
  ftp-pr: 12000-12099

sftp server

goes roughly 700 MiB/s (slower than webdav and ftp)

this is not ftps (which copyparty also supports); ftps is ftp-tls (think http/https), while sftp is ssh-based and (preferably) uses ssh-keys for authentication

the sftp-server requires the optional dependency paramiko;

• if you are not using docker, then install paramiko somehow
• if you are using docker, then use one of the following image variants: ac / im / iv / dj

enable sftpd with --sftp 3922 to listen on port 3922;

• use global-option sftp-key to associate an ssh-key with a user;
  • commandline: --sftp-key 'david ssh-ed25519 AAAAC3NzaC...'
  • config-file: sftp-key: david ssh-ed25519 AAAAC3NzaC...
• --sftp-pw enables login with passwords (default is ssh-keys only)
• --sftp-anon foo enables login with username foo and no password; gives the same access/permissions as the website does when not logged in

see the sftp section in --help for the other options

webdav server

with read-write support, supports winXP and later, macos, nautilus/gvfs ... a great way to access copyparty straight from the file explorer in your OS

click the connect button in the control-panel to see connection instructions for windows, linux, macos

general usage:

• login with any username + your password, or put your password in the username field (password field can be empty/whatever)
  • unless you enabled --usernames

on macos, connect from finder:

• [Go] -> [Connect to Server...] -> http://192.168.123.1:3923/

to be able to edit existing files, the client must have the Delete-permission, and some webdav clients will also require the daw volflag or global-option (not necessary if the client sends the x-oc-mtime header). Without daw, those clients will fail to modify existing files and instead create new copies with names like notes.txt-1771978661.726032-3i9GPghL.txt. NOTE: Enabling daw will also make all PUT-uploads overwrite existing files if the user has delete-access, so use with caution. Another alternative is the dav-port option

note: if you have enabled IdP authentication then that may cause issues for some/most webdav clients; see the webdav section in the IdP docs

connecting to webdav from windows

using the GUI (winXP or later):

• rightclick [my computer] -> [map network drive] -> Folder: http://192.168.123.1:3923/
  • on winXP only, click the Sign up for online storage hyperlink instead and put the URL there
  • providing your password as the username is recommended; the password field can be anything or empty
    • unless you enabled --usernames

the webdav client that's built into windows has the following list of bugs; you can avoid all of these by connecting with rclone instead:

• win7+ doesn't actually send the password to the server when reauthenticating after a reboot unless you first try to login with an incorrect password and then switch to the correct password
  • or just type your password into the username field instead to get around it entirely
• connecting to a folder which allows anonymous read will make writing impossible, as windows has decided it doesn't need to login
  • workaround: connect twice; first to a folder which requires auth, then to the folder you actually want, and leave both of those mounted
  • or set the server-option --dav-auth to force password-auth for all webdav clients
• win7+ may open a new tcp connection for every file and sometimes forgets to close them, eventually needing a reboot
  • maybe NIC-related (??), happens with win10-ltsc on e1000e but not virtio
• windows cannot access folders which contain filenames with invalid unicode or forbidden characters (<>:"/\|?*), or names ending with .
• winxp cannot show unicode characters outside of some range
  • latin-1 is fine, hiragana is not (not even as shift-jis on japanese xp)

tftp server

a TFTP server (read/write) can be started using --tftp 3969 (you probably want ftp instead unless you are actually communicating with hardware from the 90s (in which case we should definitely hang some time))

that makes this the first RTX DECT Base that has been updated using copyparty 🎉

• based on partftpy
• no accounts; read from world-readable folders, write to world-writable, overwrite in world-deletable
• needs a dedicated port (cannot share with the HTTP/HTTPS API)
  • run as root (or see below) to use the spec-recommended port 69 (nice)
• can reply from a predefined portrange (good for firewalls)
• only supports the binary/octet/image transfer mode (no netascii)
• RFC 7440 is not supported, so will be extremely slow over WAN
  • assuming default blksize (512), expect 1100 KiB/s over 100BASE-T, 400-500 KiB/s over wifi, 200 on bad wifi

most clients expect to find TFTP on port 69, but on linux and macos you need to be root to listen on that. Alternatively, listen on 3969 and use NAT on the server to forward 69 to that port;

• on linux: iptables -t nat -A PREROUTING -i eth0 -p udp --dport 69 -j REDIRECT --to-port 3969

some recommended TFTP clients:

• curl (cross-platform, read/write)
  • get: curl --tftp-blksize 1428 tftp://127.0.0.1:3969/firmware.bin
  • put: curl --tftp-blksize 1428 -T firmware.bin tftp://127.0.0.1:3969/
• windows: tftp.exe (you probably already have it)
  • tftp -i 127.0.0.1 put firmware.bin
• linux: tftp-hpa, atftp
  • atftp --option "blksize 1428" 127.0.0.1 3969 -p -l firmware.bin -r firmware.bin
  • tftp -v -m binary 127.0.0.1 3969 -c put firmware.bin

smb server

unsafe, slow, not recommended for wan, enable with --smb for read-only or --smbw for read-write

click the connect button in the control-panel to see connection instructions for windows, linux, macos

dependencies: python3 -m pip install --user -U impacket==0.13.0

• newer versions of impacket will hopefully work just fine but there is monkeypatching so maybe not

some BIG WARNINGS specific to SMB/CIFS, in decreasing importance:

• not entirely confident that read-only is read-only
• the smb backend is not fully integrated with vfs, meaning there could be security issues (path traversal). Please use --smb-port (see below) and prisonparty or bubbleparty
  • account passwords work per-volume as expected, and so does account permissions (read/write/move/delete), but --smbw must be given to allow write-access from smb
  • shadowing probably works as expected but no guarantees
• not compatible with pw-hashing or --usernames

and some minor issues,

• clients only see the first ~400 files in big folders;
  • this was originally due to impacket#1433 which was fixed in impacket-0.12, so you can disable the workaround with --smb-nwa-1 but then you get unacceptably poor performance instead
• hot-reload of server config (/?reload=cfg) does not include the [global] section (commandline args)
• listens on the first IPv4 -i interface only (default = :: = 0.0.0.0 = all)
• login doesn't work on winxp, but anonymous access is ok -- remove all accounts from copyparty config for that to work
  • win10 onwards does not allow connecting anonymously / without accounts
• python3 only
• slow (the builtin webdav support in windows is 5x faster, and rclone-webdav is 30x faster)
  • those numbers are specifically for copyparty's smb-server (because it sucks); other smb-servers should be similar to webdav

known client bugs:

• on win7 only, --smb1 is much faster than smb2 (default) because it keeps rescanning folders on smb2
  • however smb1 is buggy and is not enabled by default on win10 onwards
• windows cannot access folders which contain filenames with invalid unicode or forbidden characters (<>:"/\|?*), or names ending with .

the smb protocol listens on TCP port 445, which is a privileged port on linux and macos, which would require running copyparty as root. However, this can be avoided by listening on another port using --smb-port 3945 and then using NAT on the server to forward the traffic from 445 to there;

• on linux: iptables -t nat -A PREROUTING -i eth0 -p tcp --dport 445 -j REDIRECT --to-port 3945

authenticate with one of the following:

• username $username, password $password
• username $password, password k

wopi server

edit office documents in the web-ui

needs a wopi client to work, and the only supported wopi client is a selfhosted collabora online server; eventually onlyoffice will probably also be supported: issue #1574

no further documentation or examples yet, issue #1575

browser ux

tweaking the ui

• set default sort order globally with --sort or per-volume with the sort volflag; specify one or more comma-separated columns to sort by, and prefix the column name with - for reverse sort
  • the column names you can use are visible as tooltips when hovering over the column headers in the directory listing, for example href ext sz ts tags/.up_at tags/Circle tags/.tn tags/Artist tags/Title
  • to sort in music order (album, track, artist, title) with filename as fallback, you could --sort tags/Circle,tags/.tn,tags/Artist,tags/Title,href
  • to sort by upload date, first enable showing the upload date in the listing with -e2d -mte +.up_at and then --sort tags/.up_at

see ./docs/rice for more, including:

• how to hide ui-elements
• custom fonts
• custom loading-spinner
• adding stuff (css/<meta>/...) to the html <head> tag
• adding your own translation

opengraph

discord and social-media embeds

can be enabled globally with --og or per-volume with volflag og

note that this disables hotlinking because the opengraph spec demands it; to sneak past this intentional limitation, you can enable opengraph selectively by user-agent, for example --og-ua '(Discord|Twitter|Slack)bot' (or volflag og_ua)

you can also hotlink files regardless by appending ?raw to the url

WARNING: if you plan to use WebDAV, then --og-ua / og_ua must be configured

if you want to entirely replace the copyparty response with your own jinja2 template, give the template filepath to --og-tpl or volflag og_tpl (all members of HttpCli are available through the this object)

file deduplication

enable symlink-based upload deduplication globally with --dedup or per-volume with volflag dedup

by default, when someone tries to upload a file that already exists on the server, the upload will be politely declined, and the server will copy the existing file over to where the upload would have gone

if you enable deduplication with --dedup then it'll create a symlink instead of a full copy, thus reducing disk space usage

• on the contrary, if your server is hooked up to s3-glacier or similar storage where reading is expensive, and you cannot use --safe-dedup=1 because you have other software tampering with your files, so you want to entirely disable detection of duplicate data instead, then you can specify --no-clone globally or noclone as a volflag

warning: when enabling dedup, you should also:

• enable indexing with -e2dsa or volflag e2dsa (see file indexing section below); strongly recommended
• ...and/or --hardlink-only to use hardlink-based deduplication instead of symlinks; see explanation below
• ...and/or --reflink to use CoW/reflink-based dedup (much safer than hardlink, but OS/FS-dependent)

it will not be safe to rename/delete files if you only enable dedup and none of the above; if you enable indexing then it is not necessary to also do hardlinks (but you may still want to)

by default, deduplication is done based on symlinks (symbolic links); these are tiny files which are pointers to the nearest full copy of the file

you can choose to use hardlinks instead of softlinks, globally with --hardlink-only or volflag hardlinkonly, and you can choose to use reflinks with --reflink or volflag reflink

advantages of using reflinks (CoW, copy-on-write):

• entirely safe (when your filesystem supports it correctly); either file can be edited or deleted without affecting other copies
• only linux 5.3 or newer, only python 3.14 or newer, only some filesystems (btrfs probably ok, maybe xfs too, but zfs had bugs)

advantages of using hardlinks:

• hardlinks are more compatible with other software; they behave entirely like regular files
• you can safely move and rename files using other file managers
  • symlinks need to be managed by copyparty to ensure the destinations remain correct

advantages of using symlinks (default):

• each symlink can have its own last-modified timestamp, but a single timestamp is shared by all hardlinks
• symlinks make it more obvious to other software that the file is not a regular file, so this can be less dangerous
  • hardlinks look like regular files, so other software may assume they are safe to edit without affecting the other copies

warning: if you edit the contents of a deduplicated file, then you will also edit all other copies of that file! This is especially surprising with hardlinks, because they look like regular files, but that same file exists in multiple locations

global-option --xlink / volflag xlink additionally enables deduplication across volumes, but this is probably buggy and not recommended

config file example:

[global]
  e2dsa  # scan and index filesystem on startup
  dedup  # symlink-based deduplication for all volumes

[/media]
  /mnt/nas/media
  flags:
    hardlinkonly  # this vol does hardlinks instead of symlinks

file indexing

enable music search, upload-undo, and better dedup

file indexing relies on two database tables, the up2k filetree (-e2d) and the metadata tags (-e2t), stored in .hist/up2k.db. Configuration can be done through arguments, volflags, or a mix of both.

through arguments:

• -e2d enables file indexing on upload
• -e2ds also scans writable folders for new files on startup
• -e2dsa also scans all mounted volumes (including readonly ones)
• -e2t enables metadata indexing on upload
• -e2ts also scans for tags in all files that don't have tags yet
• -e2tsr also deletes all existing tags, doing a full reindex
• -e2v verifies file integrity at startup, comparing hashes from the db
• -e2vu patches the database with the new hashes from the filesystem
• -e2vp panics and kills copyparty instead

the same arguments can be set as volflags, in addition to d2d, d2ds, d2t, d2ts, d2v for disabling:

• -v ~/music::r:c,e2ds,e2tsr does a full reindex of everything on startup
• -v ~/music::r:c,d2d disables all indexing, even if any -e2* are on
• -v ~/music::r:c,d2t disables all -e2t* (tags), does not affect -e2d*
• -v ~/music::r:c,d2ds disables on-boot scans; only index new uploads
• -v ~/music::r:c,d2ts same except only affecting tags

note:

• upload-times can be displayed in the file listing by enabling the .up_at metadata key, either globally with -e2d -mte +.up_at or per-volume with volflags e2d,mte=+.up_at (will have a ~17% performance impact on directory listings)
  • and file checksums can be shown with global-option -e2d -mte +w or volflag e2d,mte=+w (always active for users with permission a)
• e2tsr is probably always overkill, since e2ds/e2dsa would pick up any file modifications and e2ts would then reindex those, unless there is a new copyparty version with new parsers and the release note says otherwise

config file example (these options are recommended btw):

[global]
  e2dsa  # scan and index all files in all volumes on startup
  e2ts   # check newly-discovered or uploaded files for media tags

exclude-patterns

to save some time, you can provide a regex pattern for filepaths to only index by filename/path/size/last-modified (and not the hash of the file contents) by setting --no-hash '\.iso$' or the volflag :c,nohash=\.iso$, this has the following consequences:

• initial indexing is way faster, especially when the volume is on a network disk
• makes it impossible to file-search
• if someone uploads the same file contents, the upload will not be detected as a dupe, so it will not get symlinked or rejected

similarly, you can fully ignore files/folders using --no-idx [...] and :c,noidx=\.iso$

NOTE: no-idx and/or no-hash prevents deduplication of those files

• when running on macos, all the usual apple metadata files are excluded by default

if you set --no-hash [...] globally, you can enable hashing for specific volumes using flag :c,nohash=

to exclude certain filepaths from search-results, use --srch-excl or volflag srch_excl instead of --no-idx, for example --srch-excl 'password|logs/[0-9]'

config file example:

[/games]
  /mnt/nas/games
  flags:
    noidx: \.iso$  # skip indexing iso-files
    srch_excl: password|logs/[0-9]  # filter search results

filesystem guards

avoid traversing into other filesystems using --xdev / volflag :c,xdev, skipping any symlinks or bind-mounts to another HDD for example

and/or you can --xvol / :c,xvol to ignore all symlinks leaving the volume's top directory, but still allow bind-mounts pointing elsewhere

• symlinks are permitted with xvol if they point into another volume where the user also has some sort of access, keeping permissions from outer/initial volume

these options will reduce performance; unlikely worst-case estimates are 14% reduction for directory listings, 35% for download-as-tar

as of copyparty v1.7.0 these options also prevent file access at runtime -- in previous versions it was just hints for the indexer

periodic rescan

filesystem monitoring; if copyparty is not the only software doing stuff on your filesystem, you may want to enable periodic rescans to keep the index up to date

argument --re-maxage 60 will rescan all volumes every 60 sec, same as volflag :c,scan=60 to specify it per-volume

uploads are disabled while a rescan is happening, so rescans will be delayed by --db-act (default 10 sec) when there is write-activity going on (uploads, renames, ...)

note: folder-thumbnails are selected during filesystem indexing, so periodic rescans can be used to keep them accurate as images are uploaded/deleted (or manually do a rescan with the reload button in the controlpanel)

config file example:

[global]
  re-maxage: 3600

[/pics]
  /mnt/nas/pics
  flags:
    scan: 900

u2sz

configure upload-chunking; the default 64,96 is optimal for nearly every scenario, however:

• if you are seeing the message bad server config; see "u2sz" in readme then you have a proxy/CDN/whatever with a max-request-body-size which is too small, and you must either increase the limit in the reverseproxy config to at least 96 MiB (recommended) or reduce the target chunksize in copyparty (continue reading)

the global-option --u2sz is two numbers, target,max MiB of each chunk to send, aiming for target MiB and allowing up-to-and-including max MiB depending on filesize, and if you only specify one number then it sets that for both

cloudflare max is 96 MiB, but if you have a reverseproxy (or CDN/WAF/whatever) which is more restrictive, then try some of the following:

• u2sz: 32 is still acceptable performance in most cases
• u2sz: 8 is the lowest recommended value; anything lower will reduce performance over long physical distances and/or shitty routes
• u2sz: 1 is the absolute minimum, and will be slow in many cases

the setting can also be changed in the web-UI; the textbox next to date-chk in the settings tab

upload rules

set upload rules using volflags, some examples:

• :c,sz=1k-3m sets allowed filesize between 1 KiB and 3 MiB inclusive (suffixes: b, k, m, g)
• :c,df=4g block uploads if there would be less than 4 GiB free disk space afterwards
• :c,vmaxb=1g block uploads if total volume size would exceed 1 GiB afterwards
• :c,vmaxn=4k block uploads if volume would contain more than 4096 files afterwards
• :c,nosub disallow uploading into subdirectories; goes well with rotn and rotf:
• :c,rotn=1000,2 moves uploads into subfolders, up to 1000 files in each folder before making a new one, two levels deep (must be at least 1)
• :c,rotf=%Y/%m/%d/%H enforces files to be uploaded into a structure of subfolders according to that date format
  • :c,rotf_tz=Europe/Oslo sets the timezone (default is UTC unless global-option rotf-tz is changed)
  • if someone uploads to /foo/bar the path would be rewritten to /foo/bar/2021/08/06/23 for example
  • but the actual value is not verified, just the structure, so the uploader can choose any values which conform to the format string
    • just to avoid additional complexity in up2k which is enough of a mess already
• :c,lifetime=300 delete uploaded files when they become 5 minutes old

you can also set transaction limits which apply per-IP and per-volume, but these assume -j 1 (default) otherwise the limits will be messed up, for example -j 4 would allow anywhere between 1x and 4x the limits you set depending on which processing node the client gets routed to

• :c,maxn=250,3600 allows 250 files over 1 hour from each IP (tracked per-volume)
• :c,maxb=1g,300 allows 1 GiB total over 5 minutes from each IP (tracked per-volume)

notes:

• vmaxb and vmaxn requires either the e2ds volflag or -e2dsa global-option

config file example:

[/inc]
  /mnt/nas/uploads
  accs:
    w: *    # anyone can upload here
    rw: ed  # only user "ed" can read-write
  flags:
    e2ds       # filesystem indexing is required for many of these:
    sz: 1k-3m  # accept upload only if filesize in this range
    df: 4g     # free disk space cannot go lower than this
    vmaxb: 1g  # volume can never exceed 1 GiB
    vmaxn: 4k  # ...or 4000 files, whichever comes first
    nosub      # must upload to toplevel folder
    lifetime: 300   # uploads are deleted after 5min
    maxn: 250,3600  # each IP can upload 250 files in 1 hour
    maxb: 1g,300    # each IP can upload 1 GiB over 5 minutes

compress uploads

files can be autocompressed on upload, either on user-request (if config allows) or forced by server-config

• volflag gz allows gz compression
• volflag xz allows lzma compression
• volflag pk forces compression on all files
• url parameter pk requests compression with server-default algorithm
• url parameter gz or xz requests compression with a specific algorithm
• url parameter xz requests xz compression

things to note,

• the gz and xz arguments take a single optional argument, the compression level (range 0 to 9)
• the pk volflag takes the optional argument ALGORITHM,LEVEL which will then be forced for all uploads, for example gz,9 or xz,0
• default compression is gzip level 9
• all upload methods except up2k are supported
• the files will be indexed after compression, so dupe-detection and file-search will not work as expected

some examples,

• -v inc:inc:w:c,pk=xz,0
  folder named inc, shared at inc, write-only for everyone, forces xz compression at level 0
• -v inc:inc:w:c,pk
  same write-only inc, but forces gz compression (default) instead of xz
• -v inc:inc:w:c,gz
  allows (but does not force) gz compression if client uploads to /inc?pk or /inc?gz or /inc?gz=4

chmod and chown

per-volume filesystem-permissions and ownership

by default:

• all folders are chmod 755
• files are usually chmod 644 (umask-defined)
• user/group is whatever copyparty is running as

this can be configured per-volume:

• volflag chmod_f sets file permissions; default=644 (usually)
• volflag chmod_d sets directory permissions; default=755
• volflag uid sets the owner user-id
• volflag gid sets the owner group-id

notes:

• gid can only be set to one of the groups which the copyparty process is a member of
• uid can only be set if copyparty is running as root (i appreciate your faith)

other flags

• :c,magic enables filetype detection for nameless uploads, same as --magic
  • needs https://pypi.org/project/python-magic/ python3 -m pip install --user -U python-magic
  • on windows grab this instead python3 -m pip install --user -U python-magic-bin
• cachectl changes how webbrowser will cache responses (the Cache-Control response-header); default is no-cache which will prevent repeated downloading of the same file unless necessary (browser will ask copyparty if the file has changed)
  • adding ?cache to a link will override this with "fully cache this for 69 seconds"; ?cache=321 is 321 seconds, and ?cache=i is 7 days

descript.ion

add a description to each file in a folder by adding them to a textfile named descript.ion

see https://copyparty.eu/beta/ for an example -- here's a basic descript.ion file:

bookmark.mp3 Taishi feat. Rita - Bookmark Memories
slowstep.mp3 Taishi feat. 向日葵 - Slow Step -F.L.C.A-
prsnlzr.mp3 Taishi feat. みとせのりこ - Personalizer
cosmos.mp3 Taishi feat. Rita - Into the cosmos

dothidden

cosmetically hide specific files in a folder by adding them to a textfile named .hidden

this option is default-disabled; enable the volflag and/or global-option dothidden

this is cosmetic only! the files are still easily accessible in many ways, for example with download-as-zip/tar, so do not rely on this for security.

also see the --unlist option which is somewhat similar -- unlist applies to the whole volume instead of just one folder; however, while dothidden also affects sftp and ftp, the unlist option is http/https-only

thumbnail pregen

if you want to pre-generate everything on startup (usually a bad idea);

by default, thumbnails are created on-the-fly when a client needs it, and then cached on the server for --th-maxage seconds (default is one week), so most thumbnails only need to be created once, and are then eventually deleted from the cache to preserver disk space

but if you need every thumbnail instantly available when a folder is viewed, then first increase the thumbnail expiration time to something really big, and then set global-option th-pregen and volflag th_pregen to a comma-separated list of thumbnail formats to automatically generate on server startup;

the full list of all possible formats is: j,jf,jf3,j3,w,wf,wf3,w3,x,xf,xf3,x3,opus,mp3,flac,wav and I'll explain what those mean soon

• j = jpeg cropped, jf = jpeg uncropped, jf3 = jpeg uncropped triplesize, j3 jpeg cropped triplesize
• w = webm cropped, wf = webm uncropped, ..., x = jxl cropped, xf = jxl uncropped, ...
• and yes, audio-transcodes are technically thumbnails according to copyparty -- don't think too much about it ( ﾟ ヮﾟ)
  • unlike thumbnails, the expiry time for audio-transcodes is configured with --ac-maxage

anyways, obviously you do not want to pregenerate flac/wav because they're HUGE, and everything else also gets pretty big because it all adds up;

• each regular thumbnail ( j, jf, w, wf, x, xf ) takes about 16 KiB of disk space
• each triplesize thumb ( j3, jf3, w3, wf3, x3, xf3 ) takes about 96 KiB
• each opus / mp3 audiotranscode takes... idk, 6 MiB? depends on song length

so a thousand pictures converted to every possible regular-size image format (j,jf,w,wf,x,xf) takes 96 MiB, and every possible 3x-size (jf3,j3,wf3,w3,xf3,x3) takes 562 MiB, alternatively 658 MiB in total for all, so that's why the default is to not pregenerate on startup, but instead do on-demand with a cache

database location

in-volume (.hist/up2k.db, default) or somewhere else

copyparty creates a subfolder named .hist inside each volume where it stores the database, thumbnails, and some other stuff

this can instead be kept in a single place using the --hist argument, or the hist= volflag, or a mix of both:

• --hist ~/.cache/copyparty -v ~/music::r:c,hist=- sets ~/.cache/copyparty as the default place to put volume info, but ~/music gets the regular .hist subfolder (- restores default behavior)

by default, the per-volume up2k.db sqlite3-database for -e2d and -e2t is stored next to the thumbnails according to the --hist option, but the global-option --dbpath and/or volflag dbpath can be used to put the database somewhere else

if your storage backend is unreliable (NFS or bad HDDs), you can specify one or more "landmarks" to look for before doing anything database-related. A landmark is a file which is always expected to exist inside the volume. This avoids spurious filesystem rescans in the event of an outage. One line per landmark (see example below)

note:

• putting the hist-folders on an SSD is strongly recommended for performance
• markdown edits are always stored in a local .hist subdirectory
• on windows the volflag path is cyglike, so /c/temp means C:\temp but use regular paths for --hist
  • you can use cygpaths for volumes too, -v C:\Users::r and -v /c/users::r both work

config file example:

[global]
  hist: ~/.cache/copyparty  # put db/thumbs/etc. here by default

[/pics]
  /mnt/nas/pics
  flags:
    hist: -  # restore the default (/mnt/nas/pics/.hist/)
    hist: /mnt/nas/cache/pics/  # can be absolute path
    landmark: me.jpg  # /mnt/nas/pics/me.jpg must be readable to enable db
    landmark: info/a.txt^=ok  # and this textfile must start with "ok"

metadata from audio files

set -e2t to index tags on upload

-mte decides which tags to index and display in the browser (and also the display order), this can be changed per-volume:

• -v ~/music::r:c,mte=title,artist indexes and displays title followed by artist

if you add/remove a tag from mte you will need to run with -e2tsr once to rebuild the database, otherwise only new files will be affected

but instead of using -mte, -mth is a better way to hide tags in the browser: these tags will not be displayed by default, but they still get indexed and become searchable, and users can choose to unhide them in the [⚙️] config pane

-mtm can be used to add or redefine a metadata mapping, say you have media files with foo and bar tags and you want them to display as qux in the browser (preferring foo if both are present), then do -mtm qux=foo,bar and now you can -mte artist,title,qux

tags that start with a . such as .bpm and .dur(ation) indicate numeric value

see the beautiful mess of a dictionary in mtag.py for the default mappings (should cover mp3,opus,flac,m4a,wav,aif,)

--no-mutagen disables Mutagen and uses FFprobe instead, which...

• is about 20x slower than Mutagen
• catches a few tags that Mutagen doesn't
  • melodic key, video resolution, framerate, pixfmt
• avoids pulling any GPL code into copyparty
• more importantly runs FFprobe on incoming files which is bad if your FFmpeg has a cve

--mtag-to sets the tag-scan timeout; very high default (60 sec) to cater for zfs and other randomly-freezing filesystems. Lower values like 10 are usually safe, allowing for faster processing of tricky files

metadata from xattrs

unix extended file attributes (Linux-only) can be indexed into the db and made searchable;

• --db-xattr user.foo,user.bar will index the xattrs user.foo and user.bar,
• --db-xattr user.foo=foo,user.bar=bar will index them with the names foo and bar,
• --db-xattr ~~user.foo,user.bar will index everything except user.foo and user.bar,
• --db-xattr ~~ will index everything

however note that the tags must also be enabled with -mte so here are some complete examples:

• -e2ts --db-xattr user.foo,user.bar -mte +user.foo,user.bar
• -e2ts --db-xattr user.foo=foo,user.bar=bar -mte +foo,bar

as for actually adding the xattr user.foo to a file in the first place,

• setfattr -n user.foo -v 'utsikt fra fløytoppen' photo.jpg

file parser plugins

provide custom parsers to index additional tags, also see ./bin/mtag/README.md

copyparty can invoke external programs to collect additional metadata for files using mtp (either as argument or volflag), there is a default timeout of 60sec, and only files which contain audio get analyzed by default (see ay/an/ad below)

• -mtp .bpm=~/bin/audio-bpm.py will execute ~/bin/audio-bpm.py with the audio file as argument 1 to provide the .bpm tag, if that does not exist in the audio metadata
• -mtp key=f,t5,~/bin/audio-key.py uses ~/bin/audio-key.py to get the key tag, replacing any existing metadata tag (f,), aborting if it takes longer than 5sec (t5,)
• -v ~/music::r:c,mtp=.bpm=~/bin/audio-bpm.py:c,mtp=key=f,t5,~/bin/audio-key.py both as a per-volume config wow this is getting ugly

but wait, there's more! -mtp can be used for non-audio files as well using the a flag: ay only do audio files (default), an only do non-audio files, or ad do all files (d as in dontcare)

• "audio file" also means videos btw, as long as there is an audio stream
• -mtp ext=an,~/bin/file-ext.py runs ~/bin/file-ext.py to get the ext tag only if file is not audio (an)
• -mtp arch,built,ver,orig=an,eexe,edll,~/bin/exe.py runs ~/bin/exe.py to get properties about windows-binaries only if file is not audio (an) and file extension is exe or dll
• if you want to daisychain parsers, use the p flag to set processing order
  • -mtp foo=p1,~/a.py runs before -mtp foo=p2,~/b.py and will forward all the tags detected so far as json to the stdin of b.py
• option c0 disables capturing of stdout/stderr, so copyparty will not receive any tags from the process at all -- instead the invoked program is free to print whatever to the console, just using copyparty as a launcher
  • c1 captures stdout only, c2 only stderr, and c3 (default) captures both
• you can control how the parser is killed if it times out with option kt killing the entire process tree (default), km just the main process, or kn let it continue running until copyparty is terminated

if something doesn't work, try --mtag-v for verbose error messages

config file example; note that mtp is an additive option so all of the mtp options will take effect:

[/music]
  /mnt/nas/music
  flags:
    mtp: .bpm=~/bin/audio-bpm.py  # assign ".bpm" (numeric) with script
    mtp: key=f,t5,~/bin/audio-key.py  # force/overwrite, 5sec timeout
    mtp: ext=an,~/bin/file-ext.py  # will only run on non-audio files
    mtp: arch,built,ver,orig=an,eexe,edll,~/bin/exe.py  # only exe/dll

event hooks

trigger a program on uploads, renames etc (examples)

you can set hooks before and/or after an event happens, and currently you can hook uploads, moves/renames, and deletes

there's a bunch of flags and stuff, see --help-hooks

if you want to write your own hooks, see devnotes

zeromq

event-hooks can send zeromq messages instead of running programs

to send a 0mq message every time a file is uploaded,

• --xau zmq:pub:tcp://*:5556 sends a PUB to any/all connected SUB clients
• --xau t3,zmq:push:tcp://*:5557 sends a PUSH to exactly one connected PULL client
• --xau t3,j,zmq:req:tcp://localhost:5555 sends a REQ to the connected REP client

the PUSH and REQ examples have t3 (timeout after 3 seconds) because they block if there's no clients to talk to

• the REQ example does t3,j to send extended upload-info as json instead of just the filesystem-path

see zmq-recv.py if you need something to receive the messages with

config file example; note that the hooks are additive options, so all of the xau options will take effect:

[global]
  xau: zmq:pub:tcp://*:5556`  # send a PUB to any/all connected SUB clients
  xau: t3,zmq:push:tcp://*:5557`  # send PUSH to exactly one connected PULL cli
  xau: t3,j,zmq:req:tcp://localhost:5555`  # send REQ to the connected REP cli

upload events

the older, more powerful approach (examples):

-v /mnt/inc:inc:w:c,e2d,e2t,mte=+x1:c,mtp=x1=ad,kn,/usr/bin/notify-send

that was the commandline example; here's the config file example:

[/inc]
  /mnt/inc
  accs:
    w: *
  flags:
    e2d, e2t  # enable indexing of uploaded files and their tags
    mte: +x1
    mtp: x1=ad,kn,/usr/bin/notify-send

so filesystem location /mnt/inc shared at /inc, write-only for everyone, appending x1 to the list of tags to index (mte), and using /usr/bin/notify-send to "provide" tag x1 for any filetype (ad) with kill-on-timeout disabled (kn)

that'll run the command notify-send with the path to the uploaded file as the first and only argument (so on linux it'll show a notification on-screen)

note that this is way more complicated than the new event hooks but this approach has the following advantages:

• non-blocking and multithreaded; doesn't hold other uploads back
• you get access to tags from FFmpeg and other mtp parsers
• only trigger on new unique files, not dupes

note that it will occupy the parsing threads, so fork anything expensive (or set kn to have copyparty fork it for you) -- otoh if you want to intentionally queue/singlethread you can combine it with --mtag-mt 1

for reference, if you were to do this using event hooks instead, it would be like this: -e2d --xau notify-send,hello,--

handlers

redefine behavior with plugins (examples)

replace 404 and 403 errors with something completely different (that's it for now)

as for client-side stuff, there is plugins for modifying UI/UX

ip auth

autologin based on IP range (CIDR) , using the global-option --ipu

for example, if everyone with an IP that starts with 192.168.123 should automatically log in as the user spartacus, then you can either specify --ipu=192.168.123.0/24=spartacus as a commandline option, same as this config file example:

[global]
  ipu: 192.168.123.0/24=spartacus

repeat the option to map additional subnets

be careful with this one! if you have a reverseproxy, then you definitely want to make sure you have real-ip configured correctly, and it's probably a good idea to nullmap the reverseproxy's IP just in case; so if your reverseproxy is sending requests from 172.24.27.9 then that would be --ipu=172.24.27.9/32=

restrict to ip

limit a user to certain IP ranges (CIDR) , using the global-option --ipr

for example, if the user spartacus should get rejected if they're not connecting from an IP that starts with 192.168.123 or 172.16, then you can either specify --ipr=192.168.123.0/24,172.16.0.0/16=spartacus as a commandline option, same as this config file example:

[global]
  ipr: 192.168.123.0/24,172.16.0.0/16=spartacus

repeat the option to map additional users

identity providers

replace copyparty passwords with oauth and such

you can disable the built-in password-based login system, and instead replace it with a separate piece of software (an identity provider) which will then handle authenticating / authorizing of users; this makes it possible to login with passkeys / fido2 / webauthn / yubikey / ldap / active directory / oauth / many other single-sign-on contraptions

• the regular config-defined users will be used as a fallback for requests which don't include a valid (trusted) IdP username header

  • --auth-ord configured auth precedence, for example to allow overriding the IdP with a copyparty password

• the login/logout links/buttons can be replaced with links to your IdP with --idp-login and --idp-logout , for example --idp-login /idp/login/?redir={dst} will expand {dst} to the page the user was on when clicking Login

• if your IdP-server is slow, consider --idp-cookie and let requests with the cookie cppws bypass the IdP; experimental sessions-based feature added for a party

some popular identity providers are Authelia (config-file based) and authentik (GUI-based, more complex)

there is a docker-compose example which is hopefully a good starting point (alternatively see ./docs/idp.md if you're the DIY type)

a more complete example of the copyparty configuration options look like this

but if you just want to let users change their own passwords, then you probably want user-changeable passwords instead

generic header auth

other ways to auth by header

if you have a middleware which adds a header with a user identifier, for example tailscale's Tailscale-User-Login: [email protected] then you can automatically auth as alice by defining that mapping with --idp-hm-usr '^Tailscale-User-Login^[email protected]^alice' or the following config file example:

[global]
  idp-hm-usr: ^Tailscale-User-Login^[email protected]^alice

repeat the whole idp-hm-usr option to add more mappings

user-changeable passwords

if permitted, users can change their own passwords in the control-panel

• not compatible with identity providers

• must be enabled with --chpw because account-sharing is a popular usecase

  • if you want to enable the feature but deny password-changing for a specific list of accounts, you can do that with --chpw-no name1,name2,name3,...

• to perform a password reset, edit the server config and give the user another password there, then do a config reload or server restart

• the custom passwords are kept in a textfile at filesystem-path --chpw-db, by default chpw.json in the copyparty config folder

  • if you run multiple copyparty instances with different users you almost definitely want to specify separate DBs for each instance

  • if password hashing is enabled, the passwords in the db are also hashed

    • ...which means that all user-defined passwords will be forgotten if you change password-hashing settings

using the cloud as storage

connecting to an aws s3 bucket and similar

there is no built-in support for this, but you can use FUSE-software such as rclone / geesefs / JuiceFS to first mount your cloud storage as a local disk, and then let copyparty use (a folder in) that disk as a volume

if copyparty is unable to access the local folder that rclone/geesefs/JuiceFS provides (for example if it looks invisible) then you may need to run rclone with --allow-other and/or enable user_allow_other in /etc/fuse.conf

you will probably get decent speeds with the default config, however most likely restricted to using one TCP connection per file, so the upload-client won't be able to send multiple chunks in parallel

before v1.13.5 it was recommended to use the volflag sparse to force-allow multiple chunks in parallel; this would improve the upload-speed from 1.5 MiB/s to over 80 MiB/s at the risk of provoking latent bugs in S3 or JuiceFS. But v1.13.5 added chunk-stitching, so this is now probably much less important. On the contrary, nosparse may now increase performance in some cases. Please try all three options (default, sparse, nosparse) as the optimal choice depends on your network conditions and software stack (both the FUSE-driver and cloud-server)

someone has also tested geesefs in combination with gocryptfs with surprisingly good results, getting 60 MiB/s upload speeds on a gbit line, but JuiceFS won with 80 MiB/s using its built-in encryption

you may improve performance by specifying larger values for --iobuf / --s-rd-sz / --s-wr-sz

if you've experimented with this and made interesting observations, please share your findings so we can add a section with specific recommendations :-)

hiding from google

tell search engines you don't wanna be indexed, either using the good old robots.txt or through copyparty settings:

• --no-robots adds HTTP (X-Robots-Tag) and HTML (<meta>) headers with noindex, nofollow globally
• volflag [...]:c,norobots does the same thing for that single volume
• volflag [...]:c,robots ALLOWS search-engine crawling for that volume, even if --no-robots is set globally

also, --force-js disables the plain HTML folder listing, making things harder to parse for some search engines -- note that crawlers which understand javascript (such as google) will not be affected

themes

you can change the default theme with --theme 2, and add your own themes by modifying browser.css or providing your own css to --css-browser, then telling copyparty they exist by increasing --themes

0. classic dark	2. flat pm-monokai	4. vice
1. classic light	3. flat light	5. hotdog stand

the classname of the HTML tag is set according to the selected theme, which is used to set colors as css variables ++

• each theme generally has a dark theme (even numbers) and a light theme (odd numbers), showing in pairs
• the first theme (theme 0 and 1) is html.a, second theme (2 and 3) is html.b
• if a light theme is selected, html.y is set, otherwise html.z is
• so if the dark edition of the 2nd theme is selected, you use any of html.b, html.z, html.bz to specify rules

see the top of ./copyparty/web/browser.css where the color variables are set, and there's layout-specific stuff near the bottom

if you want to change the fonts, see ./docs/rice/

complete examples

• see running on windows for a fancy windows setup

  • or use any of the examples below, just replace python copyparty-sfx.py with copyparty.exe if you're using the exe edition

• allow anyone to download or upload files into the current folder:
  python copyparty-sfx.py

  • enable searching and music indexing with -e2dsa -e2ts

  • start an FTP server on port 3921 with --ftp 3921

  • announce it on your LAN with -z so it appears in windows/Linux file managers

• anyone can upload, but nobody can see any files (even the uploader):
  python copyparty-sfx.py -e2dsa -v .::w

  • block uploads if there's less than 4 GiB free disk space with --df 4

  • show a popup on new uploads with --xau bin/hooks/notify.py

• anyone can upload, and receive "secret" links for each upload they do:
  python copyparty-sfx.py -e2dsa -v .::wG:c,fk=8

• anyone can browse (r), only kevin (password okgo) can upload/move/delete (A) files:
  python copyparty-sfx.py -e2dsa -a kevin:okgo -v .::r:A,kevin

• read-only music server:
  python copyparty-sfx.py -v /mnt/nas/music:/music:r -e2dsa -e2ts --no-robots --force-js --theme 2

  • ...with bpm and key scanning
    -mtp .bpm=f,audio-bpm.py -mtp key=f,audio-key.py

  • ...with a read-write folder for kevin whose password is okgo
-a kevin:okgo -v /mnt/nas/inc:/inc:rw,kevin

  • ...with logging to disk
    -lo log/cpp-%Y-%m%d-%H%M%S.txt.xz

listen on port 80 and 443

become a real webserver which people can access by just going to your IP or domain without specifying a port

if you're on windows or macos, then you just need to add the commandline argument -p 80,443 and you're done! nice

if you're on Linux, you have the following 4 options:

• option 1: set up a reverse-proxy -- this one makes a lot of sense if you're running on a proper headless server, because that way you get real HTTPS too

• option 2: NAT to port 3923 -- this is cumbersome since you'll need to do it every time you reboot, and the exact command may depend on your linux distribution:

  iptables -t nat -A PREROUTING -p tcp --dport 80 -j REDIRECT --to-port 3923
  iptables -t nat -A PREROUTING -p tcp --dport 443 -j REDIRECT --to-port 3923
  

• option 3: disable the security policy which prevents the use of 80 and 443; this is probably fine:

  setcap CAP_NET_BIND_SERVICE=+eip $(realpath $(which python))
  python copyparty-sfx.py -p 80,443
  

• option 4: run copyparty as root (please don't)

reverse-proxy

running copyparty next to other websites hosted on an existing webserver such as nginx, caddy, or apache

you can either:

• give copyparty its own domain or subdomain (recommended)
• or do location-based proxying, using --rp-loc=/stuff to tell copyparty where it is mounted -- has a slight performance cost and higher chance of bugs
  • if copyparty says incorrect --rp-loc or webserver config; expected vpath starting with [...] it's likely because the webserver is stripping away the proxy location from the request URLs -- see the ProxyPass in the apache example below

when running behind a reverse-proxy (this includes services like cloudflare), it is important to configure real-ip correctly, as many features rely on knowing the client's IP. The best/safest approach is to configure your reverse-proxy so it gives copyparty a header which only contains the client's true/real IP-address, and then setting --xff-hdr theHeaderName --rproxy 1 but alternatively, if you want/need to let copyparty handle this, look out for red and yellow log messages which explain how to do that. Basically, the log will say this:

set --xff-hdr to the name of the http-header to read the IP from (usually x-forwarded-for, but cloudflare uses cf-connecting-ip), and then --xff-src to the IP of the reverse-proxy so copyparty will trust the xff-hdr. You will also need to configure --rproxy to 1 if the header only contains one IP (the correct one) or to a negative value if it contains multiple; -1 being the rightmost and most trusted IP (the nearest proxy, so usually not the correct one), -2 being the second-closest hop, and so on

Note that --rp-loc in particular will not work at all unless you configure the above correctly

some reverse proxies (such as Caddy) can automatically obtain a valid https/tls certificate for you, and some support HTTP/2 and QUIC which could be a nice speed boost, depending on a lot of factors

• warning: nginx-QUIC (HTTP/3) is still experimental and can make uploads much slower, so HTTP/1.1 is recommended for now
• depending on server/client, HTTP/1.1 can also be 5x faster than HTTP/2

for improved security (and a 10% performance boost) consider listening on a unix-socket with -i unix:770:www:/dev/shm/party.sock (permission 770 means only members of group www can access it)

example webserver / reverse-proxy configs:

• apache config
• caddy uds: caddy reverse-proxy --from :8080 --to unix///dev/shm/party.sock
• caddy tcp: caddy reverse-proxy --from :8081 --to http://127.0.0.1:3923
• haproxy config
• lighttpd subdomain -- entire domain/subdomain
• lighttpd subpath -- location-based (not optimal, but in case you need it)
• nginx config -- recommended
• traefik config -- only use v3.6.7 or newer due to CVE-2025-66490

real-ip

teaching copyparty how to see client IPs when running behind a reverse-proxy, or a WAF, or another protection service such as cloudflare

if you (and maybe everybody else) keep getting a message that says thank you for playing, then you've gotten banned for malicious traffic. This ban applies to the IP address that copyparty thinks identifies the shady client -- so, depending on your setup, you might have to tell copyparty where to find the correct IP

for most common setups, there should be a helpful message in the server-log explaining what to do, but see docs/xff.md if you want to learn more, including a quick hack to just make it work (which is not recommended, but hey...)

reverse-proxy performance

most reverse-proxies support connecting to copyparty either using uds/unix-sockets (/dev/shm/party.sock, faster/recommended) or using tcp (127.0.0.1)

with copyparty listening on a uds / unix-socket / unix-domain-socket and the reverse-proxy connecting to that:

when connecting the reverse-proxy to 127.0.0.1 instead (the basic and/or old-fasioned way), speeds are a bit worse:

in summary, haproxy > caddy > traefik > nginx > apache > lighttpd, and use uds when possible (traefik does not support it yet)

• if these results are bullshit because my config examples are bad, please submit corrections!

permanent cloudflare tunnel

if you have a domain and want to get your copyparty online real quick, either from your home-PC behind a CGNAT or from a server without an existing reverse-proxy setup, one approach is to create a Cloudflare Tunnel (formerly "Argo Tunnel")

I'd recommend making a Locally-managed tunnel for more control, but if you prefer to make a Remotely-managed tunnel then this is currently how:

• cloudflare dashboard » zero trust » networks » tunnels » create a tunnel » cloudflared » choose a cool subdomain and leave the path blank, and use service type = http and URL = 127.0.0.1:3923

• and if you want to just run the tunnel without installing it, skip the cloudflared service install BASE64 step and instead do cloudflared --no-autoupdate tunnel run --token BASE64

NOTE: since people will be connecting through cloudflare, as mentioned in real-ip you should run copyparty with --xff-hdr cf-connecting-ip to detect client IPs correctly

config file example:

[global]
  xff-hdr: cf-connecting-ip

prometheus

metrics/stats can be enabled at URL /.cpr/metrics for grafana / prometheus / etc (openmetrics 1.0.0)

must be enabled with --stats since it reduces startup time a tiny bit, and you probably want -e2dsa too

the endpoint is only accessible by admin accounts, meaning the a in rwmda in the following example commandline: python3 -m copyparty -a ed:wark -v /mnt/nas::rwmda,ed --stats -e2dsa

follow a guide for setting up node_exporter except have it read from copyparty instead; example /etc/prometheus/prometheus.yml below

scrape_configs:
  - job_name: copyparty
    metrics_path: /.cpr/metrics
    basic_auth:
      password: wark
    static_configs:
      - targets: ['192.168.123.1:3923']

currently the following metrics are available,

• cpp_uptime_seconds time since last copyparty restart
• cpp_boot_unixtime_seconds same but as an absolute timestamp
• cpp_active_dl number of active downloads
• cpp_http_conns number of open http(s) connections
• cpp_http_reqs number of http(s) requests handled
• cpp_sus_reqs number of 403/422/malicious requests
• cpp_active_bans number of currently banned IPs
• cpp_total_bans number of IPs banned since last restart

these are available unless --nos-vst is specified:

• cpp_db_idle_seconds time since last database activity (upload/rename/delete)
• cpp_db_act_seconds same but as an absolute timestamp
• cpp_idle_vols number of volumes which are idle / ready
• cpp_busy_vols number of volumes which are busy / indexing
• cpp_offline_vols number of volumes which are offline / unavailable
• cpp_hashing_files number of files queued for hashing / indexing
• cpp_tagq_files number of files queued for metadata scanning
• cpp_mtpq_files number of files queued for plugin-based analysis

and these are available per-volume only:

• cpp_disk_size_bytes total HDD size
• cpp_disk_free_bytes free HDD space

and these are per-volume and total:

• cpp_vol_bytes size of all files in volume
• cpp_vol_files number of files
• cpp_dupe_bytes disk space presumably saved by deduplication
• cpp_dupe_files number of dupe files
• cpp_unf_bytes currently unfinished / incoming uploads

some of the metrics have additional requirements to function correctly,

• cpp_vol_* requires either the e2ds volflag or -e2dsa global-option

the following options are available to disable some of the metrics:

• --nos-hdd disables cpp_disk_* which can prevent spinning up HDDs
• --nos-vol disables cpp_vol_* which reduces server startup time
• --nos-vst disables volume state, reducing the worst-case prometheus query time by 0.5 sec
• --nos-dup disables cpp_dupe_* which reduces the server load caused by prometheus queries
• --nos-unf disables cpp_unf_* for no particular purpose

note: the following metrics are counted incorrectly if multiprocessing is enabled with -j: cpp_http_conns, cpp_http_reqs, cpp_sus_reqs, cpp_active_bans, cpp_total_bans

other extremely specific features

you'll never find a use for these:

custom mimetypes

change the association of a file extension

using commandline args, you can do something like --mime gif=image/jif and --mime ts=text/x.typescript (can be specified multiple times)

that's the same as this config file example:

[global]
  mime: gif=image/jif
  mime: ts=text/x.typescript

run copyparty with --mimes to list all the default mappings

GDPR compliance

imagine using copyparty professionally... TINLA/IANAL; EU laws are hella confusing

• remember to disable logging, or configure logrotation to an acceptable timeframe with -lo cpp-%Y-%m%d.txt.xz or similar

• if running with the database enabled (recommended), then have it forget uploader-IPs after some time using --forget-ip 43200

  • don't set it too low; unposting a file is no longer possible after this takes effect

• if you actually are a lawyer then I'm open for feedback, would be fun

feature chickenbits

buggy feature? rip it out by setting any of the following environment variables to disable its associated bell or whistle,

example: PRTY_NO_IFADDR=1 python3 copyparty-sfx.py

feature beefybits

force-enable features with known issues on your OS/env by setting any of the following environment variables, also affectionately known as fuckitbits or hail-mary-bits

packages

the party might be closer than you think

if your distro/OS is not mentioned below, there might be some hints in the «on servers» section

arch package

pacman -S copyparty (in arch linux extra)

it comes with a systemd service as well as a user service, and expects to find a config file in /etc/copyparty.conf or ~/.config/copyparty/copyparty.conf

after installing, start either the system service or the user service and navigate to http://127.0.0.1:3923 for further instructions (unless you already edited the config files, in which case you are good to go, probably)

to start the systemd service, either do systemctl start --user copyparty to start it as your own user, or systemctl start copyparty@bob to use unix-user bob

fedora package

does not exist yet; there are rumours that it is being packaged! keep an eye on this space...

gentoo ::guru package

emerge www-servers/copyparty::guru (in ::guru)

but first enable the ::guru repo;

emerge -an app-eselect/eselect-repository
eselect repository enable guru
emerge --sync guru

to start the service as a user:

• OpenRC: rc-service -U copyparty start && rc-update -U add copyparty default
• systemd: [todo]

homebrew formulae

brew install copyparty ffmpeg -- https://formulae.brew.sh/formula/copyparty

should work on all macs (both intel and apple silicon) and all relevant macos versions

the homebrew package is maintained by the homebrew team (thanks!)

nix package

nix profile install github:9001/copyparty

requires a flake-enabled installation of nix

some recommended dependencies are enabled by default; override the package if you want to add/remove some features/deps

nixos module

for flake-enabled installations of NixOS:

{
  # add copyparty flake to your inputs
  inputs.copyparty.url = "github:9001/copyparty";

  # ensure that copyparty is an allowed argument to the outputs function
  outputs = { self, nixpkgs, copyparty }: {
    nixosConfigurations.yourHostName = nixpkgs.lib.nixosSystem {
      modules = [
        # load the copyparty NixOS module
        copyparty.nixosModules.default
        ({ pkgs, ... }: {
          # add the copyparty overlay to expose the package to the module
          nixpkgs.overlays = [ copyparty.overlays.default ];
          # (optional) install the package globally
          environment.systemPackages = [ pkgs.copyparty ];
          # configure the copyparty module
          services.copyparty.enable = true;
        })
      ];
    };
  };
}

if you don't use a flake in your configuration, you can use other dependency management tools like npins, niv, or even plain fetchTarball, like so:

{ pkgs, ... }:

let
  # npins example, adjust for your setup. copyparty should be a path to the downloaded repo
  # for niv, just replace the npins folder import with the sources.nix file
  copyparty = (import ./npins).copyparty;

  # or with fetchTarball:
  copyparty = fetchTarball "https://github.com/9001/copyparty/archive/hovudstraum.tar.gz";
in

{
  # load the copyparty NixOS module
  imports = [ "${copyparty}/contrib/nixos/modules/copyparty.nix" ];

  # add the copyparty overlay to expose the package to the module
  nixpkgs.overlays = [ (import "${copyparty}/contrib/package/nix/overlay.nix") ];
  # (optional) install the package globally
  environment.systemPackages = [ pkgs.copyparty ];
  # configure the copyparty module
  services.copyparty.enable = true;
}

copyparty on NixOS is configured via services.copyparty options, for example:

services.copyparty = {
  enable = true;
  # the user to run the service as
  user = "copyparty"; 
  # the group to run the service as
  group = "copyparty"; 
  # directly maps to values in the [global] section of the copyparty config.
  # see `copyparty --help` for available options
  settings = {
    i = "0.0.0.0";
    # use lists to set multiple values
    p = [ 3210 3211 ];
    # use booleans to set binary flags
    no-reload = true;
    # using 'false' will do nothing and omit the value when generating a config
    ignored-flag = false;
  };

  # create users
  accounts = {
    # specify the account name as the key
    ed = {
      # provide the path to a file containing the password, keeping it out of /nix/store
      # must be readable by the copyparty service user
      passwordFile = "/run/keys/copyparty/ed_password";
    };
    # or do both in one go
    k.passwordFile = "/run/keys/copyparty/k_password";
  };

  # create a group
  groups = {
    # users "ed" and "k" are part of the group g1
    g1 = [ "ed" "k" ];
  };

  # create a volume
  volumes = {
    # create a volume at "/" (the webroot), which will
    "/" = {
      # share the contents of "/srv/copyparty"
      path = "/srv/copyparty";
      # see `copyparty --help-accounts` for available options
      access = {
        # everyone gets read-access, but
        r = "*";
        # users "ed" and "k" get read-write
        rw = [ "ed" "k" ];
      };
      # see `copyparty --help-flags` for available options
      flags = {
        # "fk" enables filekeys (necessary for upget permission) (4 chars long)
        fk = 4;
        # scan for new files every 60sec
        scan = 60;
        # volflag "e2d" enables the uploads database
        e2d = true;
        # "d2t" disables multimedia parsers (in case the uploads are malicious)
        d2t = true;
        # skips hashing file contents if path matches *.iso
        nohash = "\.iso$";
      };
    };
  };
  # you may increase the open file limit for the process
  openFilesLimit = 8192;
  
  # override the package used by the module to add dependencies, e.g. for hooks
  package = pkgs.copyparty.override {
    # provides exiftool for bin/hooks/image-noexif.py
    extraPackages = [ pkgs.exiftool ];
  };
};

the passwordFile at /run/keys/copyparty/ could for example be generated by agenix, or you could just dump it in the nix store instead if that's acceptable

browser support

TLDR: yes

ie = internet-explorer, ff = firefox, c = chrome, iOS = iPhone/iPad, Andr = Android

• internet explorer 6 through 8 behave the same
• firefox 52 and chrome 49 are the final winxp versions
• *1 yes, but extremely slow (ie10: 1 MiB/s, ie11: 270 KiB/s)
• *2 only able to do plaintext documents (no markdown rendering)
• *3 iOS 11 and newer, opus only, and requires FFmpeg on the server

quick summary of more eccentric web-browsers trying to view a directory index:

server hall of fame

unexpected things that run copyparty:

• an old allwinner android tv-box (ziptie-strapped to an HDD) running a firmware which flips the CPU into Big-Endian mode early during boot
  • copyparty is certified BE ready -- thanks, Øl Telecom!
• an SGI O2 (photo) with a grand total of 64 MiB RAM running SGI IRIX; screenshot
  • thanks again to the wonderful people at Øl Telecom
• a wristwatch

client examples

interact with copyparty using non-browser clients

• javascript: dump some state into a file (two separate examples)

  • await fetch('//127.0.0.1:3923/', {method:"PUT", body: JSON.stringify(foo)});
  • var xhr = new XMLHttpRequest(); xhr.open('POST', '//127.0.0.1:3923/msgs?raw'); xhr.send('foo');

• curl/wget: upload some files (post=file, chunk=stdin)

  • post(){ curl -F f=@"$1" http://127.0.0.1:3923/?pw=wark;}
post movie.mkv (gives HTML in return)
  • post(){ curl -F f=@"$1" 'http://127.0.0.1:3923/?want=url&pw=wark';}
post movie.mkv (gives hotlink in return)
  • post(){ curl -H pw:wark -H rand:8 -T "$1" http://127.0.0.1:3923/;}
post movie.mkv (randomized filename)
  • post(){ wget --header='pw: wark' --post-file="$1" -O- http://127.0.0.1:3923/?raw;}
post movie.mkv
  • chunk(){ curl -H pw:wark -T- http://127.0.0.1:3923/;}
chunk <movie.mkv

• curl: append to existing file with ?apnd

  • log(){ curl -H pw:wark -T- http://127.0.0.1:3923/logfile.txt?apnd;}
echo hey | log

• bash: when curl and wget is not available or too boring

  • (printf 'PUT /junk?pw=wark HTTP/1.1\r\n\r\n'; cat movie.mkv) | nc 127.0.0.1 3923
  • (printf 'PUT / HTTP/1.1\r\n\r\n'; cat movie.mkv) >/dev/tcp/127.0.0.1/3923

• python: u2c.py is a command-line up2k client (webm)

  • file uploads, file-search, folder sync, autoresume of aborted/broken uploads
  • can be downloaded from copyparty: controlpanel -> connect -> u2c.py
  • see ./bin/README.md#u2cpy

• FUSE: mount a copyparty server as a local filesystem

  • cross-platform python client available in ./bin/
  • able to mount nginx and iis directory listings too, not just copyparty
  • can be downloaded from copyparty: controlpanel -> connect -> partyfuse.py
  • rclone as client can give ~5x performance, see ./docs/rclone.md

• sharex (screenshot utility): see ./contrib/sharex.sxcu

  • and for screenshots on macos, see ./contrib/ishare.iscu
  • and for screenshots on linux, see ./contrib/flameshot.sh

• Custom Uploader (an Android app) as an alternative to copyparty's own PartyUP!

  • works if you set UploadURL to https://your.com/foo/?want=url&pw=hunter2 and FormDataName f

• contextlet (web browser integration); see contrib contextlet

• igloo irc: Method: post Host: https://you.com/up/?want=url&pw=hunter2 Multipart: yes File parameter: f

copyparty returns a truncated sha512sum of your PUT/POST as base64; you can generate the same checksum locally to verify uploads:

b512(){ printf "$((sha512sum||shasum -a512)|sed -E 's/ .*//;s/(..)/\\x\1/g')"|base64|tr '+/' '-_'|head -c44;}
b512 <movie.mkv

you can provide passwords using header PW: hunter2, cookie cppwd=hunter2, url-param ?pw=hunter2, or with basic-authentication (either as the username or password)

for basic-authentication, all of the following are accepted: password / whatever:password / password:whatever (the username is ignored)

• unless you've enabled --usernames, then it's PW: usr:pwd, cookie cppwd=usr:pwd, url-param ?pw=usr:pwd

NOTE: curl will not send the original filename if you use -T combined with url-params! Also, make sure to always leave a trailing slash in URLs unless you want to override the filename

folder sync

sync folders to/from copyparty

NOTE: full bidirectional sync, like what nextcloud and syncthing does, will never be supported! Only single-direction sync (server-to-client, or client-to-server) is possible with copyparty

• if you want bidirectional sync, then copyparty and syncthing should be entirely safe to combine; they should be able to collaborate on the same folders without causing any trouble for eachother. Many people do this, and there have been no issues so far. But, if you do encounter any problems, please file a copyparty bug and I'll try to help -- just keep in mind I've never used syncthing before :-)

the commandline uploader u2c.py with --dr is the best way to sync a folder to copyparty; verifies checksums and does files in parallel, and deletes unexpected files on the server after upload has finished which makes file-renames really cheap (it'll rename serverside and skip uploading)

if you want to sync with u2c.py then:

• the e2dsa option (either globally or volflag) must be enabled on the server for the volumes you're syncing into
• ...but DON'T enable global-options no-hash or no-idx (or volflags nohash / noidx), or at least make sure they are configured so they do not affect anything you are syncing into
• ...and u2c needs the delete-permission, so either rwd at minimum, or just A which is the same as rwmd.a
  • quick reminder that a and A are different permissions, and . is very useful for sync

alternatively there is rclone which allows for bidirectional sync and is way more flexible (stream files straight from sftp/s3/gcs to copyparty, ...), although there is no integrity check and it won't work with files over 100 MiB if copyparty is behind cloudflare

• starting from rclone v1.63, rclone is faster than u2c.py on low-latency connections
  • but this is only true for the initial upload; u2c will be faster for periodic syncing

mount as drive

a remote copyparty server as a local filesystem; go to the control-panel and click connect to see a list of commands to do that

alternatively, some alternatives roughly sorted by speed (unreproducible benchmark), best first:

• rclone-webdav (25s), read/WRITE (rclone v1.63 or later)
• rclone-http (26s), read-only
• partyfuse.py (26s), read-only
• rclone-ftp (47s), read/WRITE
• davfs2 (103s), read/WRITE
• win10-webdav (138s), read/WRITE
• win10-smb2 (387s), read/WRITE

most clients will fail to mount the root of a copyparty server unless there is a root volume (so you get the admin-panel instead of a browser when accessing it) -- in that case, mount a specific volume instead

if you have volumes that are accessible without a password, then some webdav clients (such as davfs2) require the global-option --dav-auth to access any password-protected areas

android app

upload to copyparty with one tap

'' ''

the app is NOT the full copyparty server! just a basic upload client, nothing fancy yet

if you want to run the copyparty server on your android device, see install on android

iOS shortcuts

there is no iPhone app, but the following shortcuts are almost as good:

• upload to copyparty (offline) (png) based on the original by Daedren (thx!)
  • can strip exif, upload files, pics, vids, links, clipboard
  • can download links and rehost the target file on copyparty (see first comment inside the shortcut)
  • pics become lowres if you share from gallery to shortcut, so better to launch the shortcut and pick stuff from there

if you want to run the copyparty server on your iPhone or iPad, see install on iOS

performance

defaults are usually fine - expect 8 GiB/s download, 1 GiB/s upload

below are some tweaks roughly ordered by usefulness:

• disabling HTTP/2 and HTTP/3 can make uploads 5x faster, depending on server/client software

• -q disables logging and can help a bunch, even when combined with -lo to redirect logs to file

• --hist pointing to a fast location (ssd) will make directory listings and searches faster when -e2d or -e2t is set

  • and also makes thumbnails load faster, regardless of e2d/e2t

• --dedup enables deduplication and thus avoids writing to the HDD if someone uploads a dupe

• --safe-dedup 1 makes deduplication much faster during upload by skipping verification of file contents; safe if there is no other software editing/moving the files in the volumes

• --no-dirsz shows the size of folder inodes instead of the total size of the contents, giving about 30% faster folder listings

• --no-hash . when indexing a network-disk if you don't care about the actual filehashes and only want the names/tags searchable

• if your volumes are on a network-disk such as NFS / SMB / s3, specifying larger values for --iobuf and/or --s-rd-sz and/or --s-wr-sz may help; try setting all of them to 524288 or 1048576 or 4194304

• --no-htp --hash-mt=0 --mtag-mt=1 --th-mt=1 minimizes the number of threads; can help in some eccentric environments (like the vscode debugger)

• when running on AlpineLinux or other musl-based distro, try mimalloc for higher performance (and twice as much RAM usage); apk add mimalloc2 and run copyparty with env-var LD_PRELOAD=/usr/lib/libmimalloc-secure.so.2

  • note that mimalloc requires special care when combined with prisonparty and/or bubbleparty/bubblewrap; you must give it access to /proc and /sys otherwise you'll encounter issues with FFmpeg (audio transcoding, thumbnails)

• -j0 (usually not recommended) enables multiprocessing (actual multithreading), can reduce latency to 20+80/numCores percent and generally improve performance in cpu-intensive workloads, for example:

  • lots of connections (many users or heavy clients)
  • simultaneous downloads and uploads saturating a 20gbps connection
  • if -e2d is enabled, -j2 gives 4x performance for directory listings; -j4 gives 16x

  ...however it will probably reduce performance in most cases, since it also increases the server/filesystem/HDD load during uploads, and adds an overhead to internal communication, so keeping the default is generally best

• using pypy instead of cpython can be 70% faster for some workloads, but slower for many others

  • and pypy can sometimes crash on startup with -j0 (TODO make issue)

• if you are running the copyparty server on Windows or Macos:

  • --casechk=n makes it much faster, but also awakens the usual surprises you expect from a case-insensitive filesystem
    • this is the same as casechk: n in a config-file

client-side

when uploading files,

• when uploading from very fast storage (NVMe SSD) with chrome/firefox, enable [wasm] in the [⚙️] settings tab to more effectively use all CPU-cores for hashing

  • don't do this on Safari (runs faster without)
  • don't do this on older browsers; likely to provoke browser-bugs (browser eats all RAM and crashes)
  • can be made default-enabled serverside with --nosubtle 137 (chrome v137+) or --nosubtle 2 (chrome+firefox)

• chrome is recommended (unfortunately), at least compared to firefox:

  • up to 90% faster when hashing, especially on SSDs
  • up to 40% faster when uploading over extremely fast internets
  • but u2c.py can be 40% faster than chrome again

• if you're cpu-bottlenecked, or the browser is maxing a cpu core:

  • up to 30% faster uploads if you hide the upload status list by switching away from the [🚀] up2k ui-tab (or closing it)
    • optionally you can switch to the lightweight potato ui by clicking the [🥔]
    • switching to another browser-tab also works, the favicon will update every 10 seconds in that case
  • unlikely to be a problem, but can happen when uploading many small files, or your internet is too fast, or PC too slow

security

there is a discord server with announcements ; an @everyone for all important updates (at the lack of better ideas)

some notes on hardening

• set --rproxy 0 if and only if your copyparty is directly facing the internet (not through a reverse-proxy)
  • cors doesn't work right otherwise
• if you allow anonymous uploads or otherwise don't trust the contents of a volume, you can prevent XSS with volflag nohtml
  • this returns html documents and svg images as plaintext, and also disables markdown rendering
  • the nohtml volflag also enables noscript which, on its own, prevents most javascript from running; enabling just noscript without nohtml makes it probably-safe (see below) to view html and svg files, but nohtml is necessary to block javascript in markdown documents
    • "probably-safe" because it relies on Content-Security-Policy so it depends on the reverseproxy to forward it, and the browser to understand it, but nohtml (the nuclear option) always works
• when running behind a reverse-proxy, listen on a unix-socket for tighter access control (and more performance); see reverse-proxy or --help-bind
• put ffmpeg into a sandbox (helps against future ffmpeg vulns) by configuring --th-bwrap and enable it with --use-bwrap y
  • usually impossible when running in docker or podman
  • the default value of --th-bwrap (see --help on your own server) is autogenerated best-effort guess based on your setup; installing bwrap and configuring this is up to you (good luck)

safety profiles:

• option -s is a shortcut to set the following options:

  • --no-thumb disables thumbnails and audio transcoding to stop copyparty from running FFmpeg/Pillow/VIPS on uploaded files, which is a good idea if anonymous upload is enabled
  • --no-mtag-ff uses mutagen to grab music tags instead of FFmpeg, which is safer and faster but less accurate
  • --dotpart hides uploads from directory listings while they're still incoming
  • --no-robots and --force-js makes life harder for crawlers, see hiding from google

• option -ss is a shortcut for the above plus:

  • --unpost 0, --no-del, --no-mv disables all move/delete support
  • --hardlink creates hardlinks instead of symlinks when deduplicating uploads, which is less maintenance
    • however note if you edit one file it will also affect the other copies
  • --vague-403 returns a "404 not found" instead of "401 unauthorized" which is a common enterprise meme
  • -nih removes the server hostname from directory listings

• option -sss is a shortcut for the above plus:

  • --no-dav disables webdav support
  • --no-logues and --no-readme disables support for readme's and prologues / epilogues in directory listings, which otherwise lets people upload arbitrary (but sandboxed) <script> tags
  • -lo cpp-%Y-%m%d-%H%M%S.txt.xz enables logging to disk
  • -ls **,*,ln,p,r does a scan on startup for any dangerous symlinks

other misc notes:

• you can disable directory listings by giving permission g instead of r, only accepting direct URLs to files
  • you may want filekeys to prevent filename bruteforcing
  • permission h instead of r makes copyparty behave like a traditional webserver with directory listing/index disabled, returning index.html instead
    • compatibility with filekeys: index.html itself can be retrieved without the correct filekey, but all other files are protected

gotchas

behavior that might be unexpected

• users without read-access to a folder can still see the .prologue.html / .epilogue.html / PREADME.md / README.md contents, for the purpose of showing a description on how to use the uploader for example
• users can submit <script>s which autorun (in a sandbox) for other visitors in a few ways;
  • uploading a README.md -- avoid with --no-readme
  • renaming some.html to .epilogue.html -- avoid with either --no-logues or --no-dot-ren
  • the directory-listing embed is sandboxed (so any malicious scripts can't do any damage) but the markdown editor is not 100% safe, see below
• markdown documents can contain html and <script>s; attempts are made to prevent scripts from executing (unless -emp is specified) but this is not 100% bulletproof, so setting the nohtml volflag is still the safest choice
  • or eliminate the problem entirely by only giving write-access to trustworthy people :^)

csp nonce

unauthorized javascript is blocked by means of the default content-security-policy which enforces "script nonces", to help prevent XSS and similar

usually this Just Works and is correctly configured by default; you can stop reading this section and move on, nothing to see here

but if you are the unlucky 0.1% with a hosting-setup funky enough for this to not work as intended, then you will also be seeing this message: Looks like javascript is broken; see "csp nonce" in the readme

there is two possible reasons why this might happen:

1. your browser is simply too old to understand the javascript, which is not likely, because just about every browser is supported, even internet explorer 11

2. something in your hosting-setup is tampering with the responses from copyparty and destroying this security feature; the culprit is probably a misconfigured reverseproxy, a CDN, a WAF, or some other mitm, but a popular offender is "cloudflare rocket loader"

you can try to temporarily set global-options --csp-ui no and --csp-dl no and restart copyparty to check whether it is due to reason #2. Remember to refresh the webbrowser with CTRL-SHIFT-R to check properly. If that makes the problem go away, then turn them back off (never keep them enabled), restart copyparty, open the Network tab in the webbrowser, and refresh the website; you should see the content-security-policy header including a nonce, for example nonce-BxcJbvdu1z3S/ycPBjVWuA== which should also appear inside the HTML of that response like this: <script nonce="BxcJbvdu1z3S/ycPBjVWuA=="> -- if it doesn't then that's the problem; something in your hosting stack is dangerously misconfigured

cors

cross-site request config

by default, except for GET and HEAD operations, all requests must either:

• not contain an Origin header at all
• or have an Origin matching the server domain
• or the header PW with your password as value

cors can be configured with --acao and --acam, or the protections entirely disabled with --allow-csrf

filekeys

prevent filename bruteforcing

volflag fk generates filekeys (per-file accesskeys) for all files; users which have full read-access (permission r) will then see URLs with the correct filekey ?k=... appended to the end, and g users must provide that URL including the correct key to avoid a 404

by default, filekeys are generated based on salt (--fk-salt) + filesystem-path + file-size + inode (if not windows); add volflag fka to generate slightly weaker filekeys which will not be invalidated if the file is edited (only salt + path)

permissions wG (write + upget) lets users upload files and receive their own filekeys, still without being able to see other uploads

dirkeys

share specific folders in a volume without giving away full read-access to the rest -- the visitor only needs the g (get) permission to view the link

ℹ️ a better alternative to this is shares

volflag dk generates dirkeys (per-directory accesskeys) for all folders, granting read-access to that folder; by default only that folder itself, no subfolders

volflag dky disables the actual key-check, meaning anyone can see the contents of a folder where they have g access, but not its subdirectories

• dk + dky gives the same behavior as if all users with g access have full read-access, but subfolders are hidden files (as if their names start with a dot), so dky is an alternative to renaming all the folders for that purpose, maybe just for some users

volflag dks lets people enter subfolders as well, and also enables download-as-zip/tar

if you enable dirkeys, it is probably a good idea to enable filekeys too, otherwise it will be impossible to hotlink files from a folder which was accessed using a dirkey

dirkeys are generated based on another salt (--dk-salt) + filesystem-path and have a few limitations:

• the key does not change if the contents of the folder is modified
  • if you need a new dirkey, either change the salt or rename the folder
• linking to a textfile (so it opens in the textfile viewer) is not possible if recipient doesn't have read-access

password hashing

you can hash passwords before putting them into config files / providing them as arguments; see --help-pwhash for all the details

--ah-alg argon2 enables it, and if you have any plaintext passwords then it'll print the hashed versions on startup so you can replace them

optionally also specify --ah-cli to enter an interactive mode where it will hash passwords without ever writing the plaintext ones to disk

the default configs take about 0.4 sec and 256 MiB RAM to process a new password on a decent laptop

when generating hashes using --ah-cli for docker or systemd services, make sure it is using the same --ah-salt by:

• inspecting the generated salt using --show-ah-salt in copyparty service configuration
• setting the same --ah-salt in both environments

⚠️ if you have enabled --usernames then provide the password as username:password when hashing it, for example ed:hunter2

https

both HTTP and HTTPS are accepted by default, but please ignore copyparty's built-in https/tls support and instead use a reverse proxy to handle https/tls/ssl

• reverseproxies do a better job following best practices meaning they are more secure, and probably also have higher performance
• also, copyparty doesn't speak HTTP/2 or QUIC, so using a reverse proxy would solve that as well -- but note that HTTP/1 is usually faster than both HTTP/2 and HTTP/3

if cfssl is installed, copyparty will automatically create a CA and server-cert on startup

• the certs are written to --crt-dir for distribution, see --help for the other --crt options
• this will be a self-signed certificate so you must install your ca.pem into all your browsers/devices
• if you want to avoid the hassle of distributing certs manually, please consider using a reverse proxy

to install cfssl on windows:

• download cfssl_windows_amd64.exe, cfssljson_windows_amd64.exe, cfssl-certinfo_windows_amd64.exe
• rename them to cfssl.exe, cfssljson.exe, cfssl-certinfo.exe
• put them in PATH, for example inside c:\windows\system32

if you really wanna give copyparty an existing TLS certificate then do one of the following:

• --no-crt --cert server.pem where server.pem is a concatenation of key + cert + chain (in that order), or...
• --no-crt --cert server.crt --certkey server.key where server.key is the key, and server.crt is a concatenation of cert + chain (in that order)
• file-extensions don't matter, but all files are expected to be PEM-style

recovering from crashes

client crashes

firefox wsod

firefox 87 can crash during uploads -- the entire browser goes, including all other browser tabs, everything turns white

however you can hit F12 in the up2k tab and use the devtools to see how far you got in the uploads:

• get a complete list of all uploads, organized by status (ok / no-good / busy / queued):
  var tabs = { ok:[], ng:[], bz:[], q:[] }; for (var a of up2k.ui.tab) tabs[a.in].push(a); tabs

• list of filenames which failed:
  ​var ng = []; for (var a of up2k.ui.tab) if (a.in != 'ok') ng.push(a.hn.split('<a href=\"').slice(-1)[0].split('\">')[0]); ng

• send the list of filenames to copyparty for safekeeping:
  await fetch('/inc', {method:'PUT', body:JSON.stringify(ng,null,1)})

HTTP API

see devnotes

dependencies

mandatory deps:

• jinja2 (is built into the SFX)

optional dependencies

enable bonus features by installing these python-packages from pypi or so:

enable hashed passwords in config: argon2-cffi

enable ftp-server:

• for just plaintext FTP, pyftpdlib (is built into the SFX)
• with TLS encryption, pyftpdlib pyopenssl

enable sftp-server: paramiko

enable music tags:

• either mutagen (fast, pure-python, skips a few tags, makes copyparty GPL? idk)
• or ffprobe (20x slower, more accurate, possibly dangerous depending on your distro and users)

enable thumbnails of...

• images: Pillow and/or pyvips and/or ffmpeg (requires py2.7 or py3.5+)
• videos/audio: ffmpeg and ffprobe somewhere in $PATH
• HEIF pictures: pyvips or ffmpeg or pillow-heif
• AVIF pictures: pyvips or ffmpeg or pillow-avif-plugin or pillow v11.3+
• JPEG XL pictures: pyvips or ffmpeg
• RAW photos: either libraw dcraw_emu or rawpy, plus either pyvips or Pillow

enable sending zeromq messages from event-hooks: pyzmq

enable smb support (not recommended): impacket==0.13.0

pyvips adds support for some additional image formats, but can use extreme amounts of RAM

• to install pyvips on Linux: sudo apt install libvips42 && python3 -m pip install --user -U pyvips
• to install pyvips on windows: pip install --user -U "pyvips[binary]"

to install FFmpeg on Windows, grab a recent build -- you need ffmpeg.exe and ffprobe.exe from inside the bin folder; copy them into C:\Windows\System32 or any other folder that's in your %PATH%

if your ffmpeg/ffprobe binaries have nonstandard names -- such as ffmpeg8 (macports) -- set environment variables PRTY_FFMPEG_BIN and PRTY_FFPROBE_BIN to the corret name (or full path)

dependency chickenbits

prevent loading an optional dependency , for example if:

• you have an incompatible version installed and it causes problems
• you just don't want copyparty to use it, maybe to save ram

set any of the following environment variables to disable its associated optional feature,

example: PRTY_NO_PIL=1 python3 copyparty-sfx.py

• PRTY_NO_PIL saves ram
• PRTY_NO_VIPS saves ram and startup time
• python2.7 on windows: PRTY_NO_FFMPEG + PRTY_NO_FFPROBE saves startup time

dependency unvendoring

force use of system modules instead of the vendored versions:

to debug, run copyparty with PRTY_MODSPEC=1 to see where it's getting each module from

optional gpl stuff

some bundled tools have copyleft dependencies, see ./bin/#mtag

these are standalone programs and will never be imported / evaluated by copyparty, and must be enabled through -mtp configs

sfx

the self-contained "binary" (recommended!) copyparty-sfx.py will unpack itself and run copyparty, assuming you have python installed of course

if you only need english, copyparty-en.py is the same thing but smaller

you can reduce the sfx size by repacking it; see ./docs/devnotes.md#sfx-repack

copyparty.exe

download copyparty.exe (win8+) or copyparty32.exe (win7+)

can be convenient on machines where installing python is problematic, however is not recommended -- if possible, please use copyparty-sfx.py instead

• copyparty.exe runs on win8 or newer, was compiled on win10, does thumbnails + media tags, and is currently safe to use, but any future python/expat/pillow CVEs can only be remedied by downloading a newer version of the exe

  • on win8 it needs vc redist 2015, on win10 it just works
  • some antivirus may freak out (false-positive), possibly Avast, AVG, and McAfee

• dangerous: copyparty32.exe is compatible with windows7, which means it uses an ancient copy of python (3.7.9) which cannot be upgraded and should never be exposed to the internet (LAN is fine)

• dangerous and deprecated: copyparty-winpe64.exe lets you run copyparty in WinPE and is otherwise completely useless

meanwhile copyparty-sfx.py instead relies on your system python which gives better performance and will stay safe as long as you keep your python install up-to-date

then again, if you are already into downloading shady binaries from the internet, you may also want my minimal builds of ffmpeg and ffprobe which enables copyparty to extract multimedia-info, do audio-transcoding, and thumbnails/spectrograms/waveforms, however it's much better to instead grab a recent official build every once ina while if you can afford the size

zipapp

another emergency alternative, copyparty.pyz has less features, is slow, requires python 3.7 or newer, worse compression, and more importantly is unable to benefit from more recent versions of jinja2 and such (which makes it less secure)... lots of drawbacks with this one really -- but, unlike the sfx, it is a completely normal zipfile which does not unpack any temporary files to disk, so it may just work if the regular sfx fails to start because the computer is messed up in certain funky ways, so it's worth a shot if all else fails

run it by doubleclicking it, or try typing python copyparty.pyz in your terminal/console/commandline/telex if that fails

it is a python zipapp meaning it doesn't have to unpack its own python code anywhere to run, so if the filesystem is busted it has a better chance of getting somewhere

there is also copyparty-en.pyz, english-only and without smb support (enterprise-friendly)

install on android

install Termux + its companion app Termux:API (see ocv.me/termux) and then copy-paste this into Termux (long-tap) all at once:

yes | pkg upgrade && termux-setup-storage && yes | pkg install python termux-api && python -m ensurepip && python -m pip install --user -U copyparty && { grep -qE 'PATH=.*\.local/bin' ~/.bashrc 2>/dev/null || { echo 'PATH="$HOME/.local/bin:$PATH"' >> ~/.bashrc && . ~/.bashrc; }; }
echo $?

after the initial setup, you can launch copyparty at any time by running copyparty anywhere in Termux -- and if you run it with --qr you'll get a neat qr-code pointing to your external ip

if you want thumbnails (photos+videos) and you're okay with spending another 132 MiB of storage, pkg install ffmpeg && python3 -m pip install --user -U pillow

• or if you want to use vips for photo-thumbs instead, pkg install libvips && python -m pip install --user -U wheel && python -m pip install --user -U pyvips && (cd /data/data/com.termux/files/usr/lib/; ln -s libgobject-2.0.so{,.0}; ln -s libvips.so{,.42})

if you are suddenly unable to access storage (permission issues), try forcequitting termux, revoke all of its permissions in android settings, and run the command termux-setup-storage

install on iOS

first install one of the following:

• a-Shell mini gives you the essential features
• a-Shell also enables audio transcoding and better thubmnails

and then copypaste the following command into a-Shell:

curl -L https://github.com/9001/copyparty/raw/refs/heads/hovudstraum/contrib/setup-ashell.sh | sh

if you want the latest copyparty beta, then do this instead:
curl -L https://copyparty.eu/beta/setup-ashell.sh | sh

what this does:

• creates a basic config file named cpc which you can edit with vim cpc
• adds the command cpp to launch copyparty with that config file

known issues:

• cannot run in the background; it needs to be on-screen to accept connections / uploads / downloads
• the best way to exit copyparty is to swipe away the app

reporting bugs

ideas for context to include, and where to submit them

please get in touch using any of the following URLs:

• https://github.com/9001/copyparty/ (primary)
• https://gitlab.com/9001/copyparty/ (mirror)
• https://codeberg.org/9001/copyparty (mirror)

in general, commandline arguments (and config file if any)

if something broke during an upload (replacing FILENAME with a part of the filename that broke):

journalctl -aS '48 hour ago' -u copyparty | grep -C10 FILENAME | tee bug.log

if there's a wall of base64 in the log (thread stacks) then please include that, especially if you run into something freezing up or getting stuck, for example OperationalError('database is locked') -- alternatively you can visit /?stack to see the stacks live, so http://127.0.0.1:3923/?stack for example

devnotes

for build instructions etc, see ./docs/devnotes.md

specifically you may want to build the sfx or build from scratch

see ./docs/TODO.md for planned features / fixes / changes