Releases: Cyberdrop-DL/cyberdrop-dl
Releases · Cyberdrop-DL/cyberdrop-dl
Cyberdrop-DL V10.0 Start Files
This version is incompatible with any previous release. Config file options need to be manually migrated by user.
CDL will refuse to start if the current database schema is older than
v9.15.0. Runcyberdrop-dl database transferto upgrade older databases
Added
- New
--cookiesoption to read cookies from a file/folder - New
--hashesoption to control which hashes CDL computes for new downloads - New
--database-fileoption - New
--cache-fileoption - New
downloadCLI command (replaces--downloadargument) - New
retryCLI command (replaces--retry-failedand--retry-allarguments) - New
configCLI command - New
cacheCLI command - New
hashCLI command - New
reportCLI command
Changed
- The config file format has changed. All configs (
Auth,GlobalandSetttings) are now a single file. Several options have new names, new defaults and new groups - A default config file will no longer be created. You can manually create a default one from the
Edit configoption on the main menu or runningcyberdrop-dl config new - On Windows, all application files are now stored at
%AppData%/cyberdrop-dl - On Unix systems, application files are stored according to the XDG Base Directory Specification
- Validate config in strict mode. If a config has an unknown entry, CDL will refuse to run instead of ignoring it
- Apprise URLs are now part of the main config file instead of a dedicated
apprise.txtfile - Refuse to start if the current database schema is older than
v9.15.0 - Cookies will not be automatically loaded from
AppData/cookies. Path to cookies needs to be specified with--cookies - Retry options will no longer use the original path of the first download attempt. Download path will be generated based on current config options (like normal downloads)
- Compute
xxh128,md5andsha256hashes by default --deep-scrapewill no longer reset after a single run--input-fileis now a CLI only arg- If supplied,
--input-fileneeds to be a valid file that already exists - Refuse to run if both URLs and
--input-fileare passed as arguments - Integers in config options that expect a timedelta value will be parsed as seconds instead of days
- Detect and report BasedFlare anti-bot protection
- Limit max queued downloads of a single site to the config concurrency limit x10 (capped at 50). All scraping for a site will be paused if its queue is full
- Always remove generated id from filenames (Cyberdrop)
- Revert mandatory
Content-Lengthheader introduced in v9.15.0
The following options, which were CLI only arguments, now have dedicated config entries:
--ui(entry:ui.mode)--portrait(entry:ui.portrait)--stats
Several config options have new names:
--disable-crawlers->--crawlers.disabled--disable-file-timestamps->--mtime--maximum-audio-duration->--audio.duration.max--maximum-image-size->--image.size.max--maximum-number-of-children->--max-children--maximum-other-size->--non-media.size.max--maximum-thread-depth->--max-thread-depth--maximum-thread-folder-depth->--max-thread-folder-depth--maximum-video-duration->--video.duration.max--maximum-video-size->--video.size.max--minimum-audio-duration->--audio.duration.min--minimum-image-size->--image.size.min--minimum-other-size->--non-media.size.min--minimum-video-duration->--video.duration.min--minimum-video-size->--video.size.min--required-free-space->--min-free-space--print-stats->--stats,--no-stats--send-deleted-to-trash->--hashing.dedupe.use-trash-bin
Jdownloader
--jdownloader-autostart->--jdownloader.autostart--jdownloader-download-dir->--jdownloader.download-folder--jdownloader-whitelist->--jdownloader.whitelist--send-unsupported-to-jdownloader->--jdownloader/--no-jdownloader
Sorting
--sort-downloads->--sort/--no-sort--scan-folder->--sort.input-folder--sort-folder->--sort.output-folder--sort-incrementer-format->--sort.formats.incrementer--sorted-audio->--sort.formats.audio--sorted-image->--sort.formats.image--sorted-other->--sort.formats.non-media--sorted-video->--sort.formats.video
Ignore options (filters)
The behavior of --filename-regex has been reversed (files that DO NOT match the regex will be skipped)
The behavior of --before and --after has been reversed and the --exclude prefix removed (they now include instead of excluding files)
--exclude-audio->--no-audio--exclude-images->--no-images--exclude-other->--no-non-media--exclude-videos->--no-videos--exclude-before->--before--exclude-after->--after--exclude-files-with-no-extension->--allow-files-with-no-extension--filename-regex-filter->--filename-regex--download-tiktok-src-quality-videos-> ---crawlers.tiktok.original
Logs
--log-level->--logs.level--console-log-level->--logs.console-level--log-folder->--logs.folder--logs-expire-after->--logs.expire-after--main-log->--logs.files.mainwith alias--log-file--download-error-urls->--logs.files.download-errors--rotate-logs->--logs-rotate--scrape-error-urls->--logs.files.scrape-errors--unsupported-urls->--logs.files.unsupported
DownloadOptions
--block-download-sub-folders->--subfolders,--no-subfolders--include-album-id-in-folder-name->--subfolders.include.album-id--include-thread-id-in-folder-name->--subfolders.include.thread-id--remove-domains-from-folder-names->--subfolders.include.domain--separate-posts-format->--subfolders.separate-posts.format--separate-posts->--subfolders.separate-posts--skip-download-mark-completed->--skip-and-mark-completed--max-simultaneous-downloads->--downloads--max-simultaneous-downloads-per-domain->--downloads.per-domain--slow-download-speed->--slow-speed--download-attempts->--attempts--download-delay->--delay--download-speed-limit->--speed-limit
Generic crawlers
--wordpress-media->--crawlers.generic.wordpress-media--wordpress-html->--crawlers.generic.wordpress-html--discourse->--crawlers.generic.discourse--chevereto->--crawlers.generic.chevereto
Removed
- Support for python 3.11
--log-leveland--console-log-level(--logs.leveland--logs.console-level) no longer accept integers. Only log level names as valid, ex:INFO,DEBUG,WARNING- Posts filtering by URL params (Wordpress)
- Kemono support
- Coomer support
- Coomerfans support
- Auto cookie extraction support
Several config options have been removed:
--appdata-folder--add-md5-hash--add-sha256-hash--auto-import--browser--completed-after--completed-before--disable-download-attempt-limit--download-tiktok-audios--last-forum-post--max-items-retry--remove-generated-id-from-filename--retry-all--retry-failed--retry-maintenance--save-pages-html--scrape-single-forum-post--sites--skip-check-for-partial-files--update-last-forum-post--ignore-coomer-post-content--ignore-coomer-ads
The following authentication entries has been removed:
Imgur.client_idKemono.sessionCoomer.session
Fixed
- Do not skip initialization segments (HLS)
- 403 errors on embeded videos (twing)
- Switch to v2 API (Filester)
Cyberdrop-DL V9 Start Files
[9.3.1] - 2026-04-21
Fixed
- Download errors showing as scrape errors
[9.3.0] - 2026-04-20
Added
- Patreon support
- OnePace support
- Giphy support
- XMegaDrive support
- Search, tag and category support (ThotHub)
Changed
- Hash multiple files (20) at once when using the UI menu option
- Default to HLS downloads (xHamster)
- Improve support for Flaresolverr forks using NoDriver/DrissionPage
Fixed
- Downloads panel not showing up in vertical layout
- Do no try to hash folders
- Do not try to normalize usernames (RedGifs)
- Search and user's pagination (Rule34Video, CamWhores)
- Do not download random unrelated videos from profiles (Rule34Video)
- Merging of m3u8 subtitles
- Subtitles sometimes being deleted after download
- Handle subdomains (Tranny.one)
KeyErrortrying to parse a response from Flaresolverr instance using Selenium as backend- Update query hashes (Twitch)
[9.2.0] - 2026-04-15
Added
- Tabootube support
Changed
- Check for DDoS-Guard even on successful responses
- New hashing, dedupe and scraping UI
- Show accurate download progress for HLS downloads
Removed
- Flugel anime support (No longer exists)
- Safe.soul support (No longer exists)
- Xbunkr support (No longer exists)
Fixed
- Switch to v2 API (nHentai)
- Download URL extraction (MixDrop)
- Do not try to parse relative dates (KVS)
- Use impersonation for downloads (MissAV)
[9.1.0] - 2026-04-08
This version logs raw HTTP requests and responses to the main log file. They may include credentials, IP, etc. Remove personal information before sharing them or just extract the relevant logs
Options to change log level are ignored. It will be fixed on a future version
Added
- Gupload.xyz support
- Vidara support
- OwnCloud support
- Support premium URLs (MediaFire)
- Support subfolders (Filester)
- Add support for archives (.zip) files (ImagePond)
Changed
- Adjust default rate limit (RedGifs)
- Create crawlers only if they are going to be used
- Webhook notifications will now be sent as plain text
- Use original filenames (ImagePond)
- New UI while sorting files
- CDL is way more verbose now
- Several performance improvents
Deprecated
- Auto import cookies
Removed
- All settings and UI options about changing configs have been removed. CDL will always use the default config at
AppData/Configs/Default/setting.yaml. A different config can be specified only via CLI args (--config-file) - The UI Option to export cookies has been removed
Fixed
- Download all videos on paginated playlists (50+ videos) (PMVHaven )
- Handle deleted videos (TubeCorporate)
- 404 downloads (Bunkr)
- Update selectors for files with MD5 hashes instead of SHA256 (Filester)
- Profile, albums and images crawling (ImagePond)
- Android support (requires python 3.13)
- Player info parsing (ashemaletube)
[9.0.2] - 2026-03-25
Changed
- Support users with up to 20K posts (Redgifs)
Removed
- UI option to check updates
Fixed
- Database path (Filester)
- Requote download redirects (ePorner)
- Update website salt (Gofile)
[9.0.1] - 2026-03-17
Fixed
- Update website salt (Gofile)
[9.0.0] - 2026-03-09
You can only upgrade to this version from v8.10.
Added
- Upload.ee support
- Filester support
Changed
- Use original filename for turbo.cr
- Apprise is now an optional dependency (required for notifications)
Removed
- Reddit support
Several config options:
configdisable_cachelog_line_widthfile_host_cache_expire_afterforum_cache_expire_afterdownloading_item_limitscraping_item_limitvi_mode
Fixed
- Download of multipage profile albums (Chevereto)
- Handle
redirect/tolinks (Xenforo) - Download of files with broken thumbnails (bunkr)
- Send custom one-time-use headers (GoFile)
- Download subfolder from filesystem URLs (pixeldrain)
Cyberdrop-DL V8 Start Files
Added
- Allow downloading google docs files with a custom format by using a
formatquery param - Cloudflare stream support
- pCloud support
- MegaCloud support
- PimpBunny support
- HiAnime support
- Safe.Soul support
- Voe.sx support
- Support for subfolders (MediaFire)
- Support for short URL formats (MediaFire)
- Support for subfolders (Dropbox)
- Support for alternative pixeldrain domains:
pixeldrain.net,pixeldra.in - Support for
pd.cybar.xyzURLs to bypass pixeldrain daily limit - The
--separate-postand--separate-post-formatoptions now work for tiktok posts - CDL can now download private TikTok posts when using cookies
- New config option to download tiktok videos in original quality: https://script-ware.gitbook.io/cyberdrop-dl/reference/cli-arguments#download-tiktok-src-quality-videos
- New
--max-thread-folder-depthconfig option: https://script-ware.gitbook.io/cyberdrop-dl/reference/configuration-options/settings/download_options#maximum_thread_folder_depth - New
--ignore-coomer-post-contentconfig option: https://script-ware.gitbook.io/cyberdrop-dl/reference/configuration-options/settings/ignore_options#ignore_coomer_post_content - The
--ignore-coomer-adsconfig option now affects all Kemono based crawlers (Coomer, Kemono & Nekohouse)
Changed
- The recommended python version is now python 3.12+, to take advandage of eager tasks
- Start scripts will automatically install a compatible python version
- Start scripts no longer support 32bit systems
- Handle shared folders with single file as loose files (Yetishare)
- Increase default rate limit, from 10req/s to 25req/s
- Increase max concurrent bunkr downloads, from 1 to 3
- Parse full spec cookies from Flaresolverr responses
- CDL will no longer try to merge subtitles for HLS downloads. They will be downloaded as separate files
- Make requests with the same domain as the original input URL (saint2.su)
- Tiktok downloads use a new logic to create filenames and folders. Folder structure will be different from previous versions.
- CDL will now iterate over each file and subfolder recursively, instead of downloading an entire folder as a zip (dropbox)
--read-timeoutnow acceptsNoneas value. This will make CDL keep the socket connection open indefinitely, even if the server is not sending data anymore--read-timeoutand--connection-timeoutnow accept fractions of a seconds as value (ex:0.3)--read-timeoutand--connection-timeoutnow affect all requests CDL will make (flaresolverr requests, scrape requests and download requests)- General performance improvements
Deprecated
- Support for v5 and v6 (config, database and auto-migration) is deprecated and will be removed on the next major release
Removed
--download-dropbox-folders-as-zipconfig option
Fixed
--dump-jsonoutput file missing some fields- CDL trying to download all HLS segments at once
- Download retries always re-starting from zero
- Global rate limiter being ignored while crawling
- Crashing when using any value for
--download-speed-limit - Handle b64 encoded URLs (MediaFire)
- Previously downloaded HLS fragments not being detected
- HLS subtitle downloads always failing
- Parsing of 1440p, 2K, 4K and 8K videos (KVS sites)
- Possible infinite loop when using the bunkrr.org domain (Bunkr)
- Crashing if a single error happens while crawling a Discord channel (Kemono)
- Nekohouse crawler
- Thumbnails being downloaded as is instead of full resolution (Coomer, Kemono, Nekohouse)
- Handle new rot13 encryption (xHamster)
- Handle new hex encryption (xHamster)
- Videos being skipped if the audio was downloaded before (TikTok)
- Do not crash if the
URL.txtfile has ipv6 URLs or invalid URLs ConnectionErrorafter the first attempt of a DNS lookup failed, without trying any DNS fallbacks
Cyberdrop-DL V7 Start Files
Added
- Camwhores.tv support
- Flugel-anime support
- Missav support
- Beeg.com support
- efukt support
- Vipergirls support
- Initial HLS support (requires ffmpeg)
- Support for any WordPress media site. See: https://script-ware.gitbook.io/cyberdrop-dl/reference/configuration-options/global-settings/generic_crawlers_instances
- Support for any Discourse site See: https://script-ware.gitbook.io/cyberdrop-dl/reference/configuration-options/global-settings/generic_crawlers_instances
- Add a config option to disable specific crawlers: https://script-ware.gitbook.io/cyberdrop-dl/reference/configuration-options/global-settings/general#disable_crawlers
- Add new
ssl_contextconfig option: https://script-ware.gitbook.io/cyberdrop-dl/reference/configuration-options/global-settings/general#ssl_context - Add a config option to map unknown URLs to generic crawlers: https://script-ware.gitbook.io/cyberdrop-dl/reference/configuration-options/global-settings/generic_crawlers_instances
Changed
- Date parsing errors will be logged as warnings, not errors
- Use an asynchronous DNS resolver
- Use native system certificates by default, with fallback to
certifi. See: https://script-ware.gitbook.io/cyberdrop-dl/reference/configuration-options/global-settings/general#ssl_context - Some crawlers have a new logic to create a filename. The name format for newly downloaded files may not match previous versions.
- Config options that take format strings will be validated at startup
- URLs will only be processed once, ignoring duplicates, even if they come from different crawlers. ex: If you have a dropbox URL in
URLs.txtand CDL finds that same URL while crawling a forum, it will silently ignore the URL from the forum. - General performance improvements
- New blocked domains:
gfycat
ko-fi.com
paypal.me
amazon.com
throne.com
youtu.be
youtube.com
linktr.ee
beacons.page
beacons.ai
allmylinks.com
Deprecated
- Support for 32-bit Python and 32-bit operating systems in general is deprecated. On a future version, the start scripts will explicitly fail when trying to install on 32-bit systems. This deprecation only applies to the start scripts. 32-bit versions can still be installed manually via
pip, but the user may need to compile any missing wheels.
Removed
no-uioptionallow_insecure_connectionsoption- Importing cookies from multi browsers is no longer supported. CDL will only try to extract cookies from a single browser. The
browsersoption was renamedbrowser
Fixed
ImportError: cannot import name 'MethodType'on startup- Handle errors while processing confirmation links (Xenforo)
- Redgifs not being downloaded (Xenforo)
- Attachments not being downloaded (Xenforo)
- Posts not being fully scraped if a single link in them failed (Xenforo)
- Download from user profiles (xHamster)
- Folder downloads (Cyberfile)
- Do not crash when using the
Create file hashesoption from the main menu - Do not download "related" or "favorites" videos from a profile (PornHub)
KeyErrorwhen using the--separate-postsoption- Using Real Debrid no longer causes CDL to crash
- Image downloads (HotPic)
- Do not try to download emojis (twitter_images)
- Fix albums downloads (saint)
- Handle all possible response errors (reddit)
- Do not use reddit's
user-agentfor every site - Fix
unclose client sessionerror at the end of a run - Fix parsing of dropbox folders URLS
- Handle embeded URLS (box.com)
- [Regression] Fixed free space check when using network drives (windows)
Cyberdrop-DL V6 Start Files
- Stricter config validation logic. CDL will verify that every config value and CLI argument is valid before running
- All dedupe and hashing functions has been replaced.
xxh128is now the default hashing algorithm, replacingmd5- By default, all v5 hashes will be ignored for deduplication matches. You can re-enable md5 with
--add-md5-hash- It's no longer possible to keep the newest file while deduplicating.
- Every file previously downloaded is taken into account for deduplication. Files no longer need to currently exists on disk
- Removed
--sort-cdl-onlyoption
- The following arguments are deprecated and will be removed in a future version:
--download-folder--download-all-configs--sort-all-configs--sort-all-downloads--main-log-filename--last-forum-post-filename--unsupported-urls-filename--download-error-urls-filename--scrape-error-urls-filename
This update introduces the following changes:
- Add Python 3.13 support
- Add requests caching to speed up future download sessions for forums, Coomer, and Kemono.
- Update Kemono & Coomer crawlers to use API calls
- Update GoFile crawler to the new site design
- Update Bunkr crawler to handle new UUID filename format
- Rework entire deduplication and hashing logic. Now multiple hashing algorithms are supported
- Use Pydantic for config and CLI values validation
- Allow downloading duplicate files on the same run with
--ignore-history - Fix sorting functionality
- Allow running all configs sequentially using
--config ALL - Better cookies management system
- Added option to clear cookies for selected websites
- General improvements and bug fixes
Cyberdrop-DL V5 Start Files
If you install and run the program with PIP yourself, you do not need anything here.
Download the Cyberdrop-DL zip file below, you do not need to download the source code zip files.
The zip file contains start files that will download and install Cyberdrop-DL and keep it up to date.
Cyberdrop-DL V5 Start Files
If you install and run the program with PIP yourself, you do not need anything here.
Download the Cyberdrop-DL zip file below, you do not need to download the source code zip files.
The zip file contains start files that will download and install Cyberdrop-DL and keep it up to date.