• Fork Features
• Installation
  • Docker
  • How to Build
  • Dependencies
• Usage
  • Options
  • Login
  • Cookies

AI-Assisted Development: This fork was developed with AI assistance. No warranties or guarantees. Use at your own risk.

Not Affiliated: This project is not affiliated with, endorsed by, or connected to Dropout, College Humor, or the original dropout-dl maintainers.

SUPPORT DROPOUT: We strongly encourage you to subscribe to Dropout. They create amazing content and deserve your support. This tool exists solely for personal convenience (offline viewing, backups, Plex organization) for paying subscribers.

NO PIRACY: This tool is intended for personal use by paying subscribers who want offline access or Plex organization. Redistribution of downloaded content is illegal. I'm a README, not a cop, but please don't be that person who ruins it for everyone.

This fork adds several performance improvements, stability fixes, and new features on top of the original dropout-dl:

• ⚡ Concurrent Segment Buffering - Downloads multiple segments simultaneously using a sliding window algorithm for significantly faster downloads
  • Controlled via --concurrent-segments flag (default: 5 concurrent segments)
  • Maintains sequential write order while downloading in parallel
  • Tested with 240p through 1080p content
• 🎯 Improved Audio Quality Matching - Audio quality now intelligently matches video quality tier instead of defaulting to medium

• 📦 MKV Container Format - Support for outputting to Matroska (MKV) format via --format mkv flag
• 🎬 Plex-Compatible Naming - Follows Plex naming conventions for proper library detection
  • Format: Series Name - S##E## - Episode Name.ext
  • Special episode support (S00Exx for specials, including "Last Looks" and "Behind the Scenes" content)
• 🔍 Preview Downloads - --list-urls flag to see what would be downloaded without actually downloading
• 🐛 Docker Debugging Support - Build with DEBUG_BUILD=1 for gdb integration and crash analysis
• 💾 Atomic File Operations & Auto-Resume - All files are downloaded to .tmp extensions and only renamed to final format upon successful completion
  • Automatically resumes interrupted downloads by detecting and completing partial .tmp files
  • Ensures interrupted downloads are easily identifiable
  • Prevents partially downloaded files from appearing as complete
  • No manual intervention needed to resume - just re-run the same command

• Multiple memory safety fixes preventing segmentation faults
• Bounds checking for HTML parsing loops
• Null pointer checks throughout codebase
• Fixed episode numbering bugs in series/season downloads

• Added TESTING.md with Docker testing workflows and debugging instructions

For the original project, see: mosswg/dropout-dl

A docker image was created that makes it easier to build and use dropout-dl. You can simply build the docker image without worrying about installing any system dependencies:

docker build -t dropout-dl:latest .

After its done building, you can use it by adding your arguments to the end of the docker run command:

docker run --rm -it -v $PWD/login:/app/login -v $PWD/out:/Downloads dropout-dl:latest --output-directory /Downloads --captions -e https://www.dropout.tv/dimension-20/season:10/videos/the-chosen-ones

Note: The docker image expects the login file to be at /app/login.
You must specify an output directory and mount that directory to the host so that you can retrieve the files from the docker container. In the above command I tell dropout-dl to output everything in /Downloads inside the container, which is mounted to a folder named out inside the current directory ($PWD is current directory).

This repository uses a json library. Either clone the repository with the --recurse-submodules flag or after cloning run:

git submodule update --init --recursive

cmake -S <source-dir> -B <build-dir>
cmake --build <build-dir>

• cURL - Required for downloading pages and videos.

• SQLite - Required for retrieving cookies from browsers.
• libgcrypt - Used for decrypting chrome cookies retrieved from the sqlite database.

sudo apt install libcurl4-gnutls-dev

To install the optional dependencies, run:

sudo apt install libsqlite3-dev libgcrypt-dev

sudo xbps-install -S libcurl

To install the optional dependencies, run:

sudo xbps-install -S sqlite-devel libgcrypt-devel

./dropout-dl [options] <url>

By default, dropout-dl will download episodes in a season with the format <series>/<season>/<series> - S<season-num>E<episode-num> - <episode-name>.mp4 and single episodes with the format <series>/<season>/<series> - <season> - <episode-name>.mp4.

--help              -h   Display this message
--quality           -q   Set the quality of the downloaded video. Quality can be set to 'all' which
                            will download all qualities and place them into separate folders
--output            -o   Set the output filename. Only works for single episode downloads
--output-directory  -d   Set the directory where files are output
--verbose           -v   Display debug information while running
--browser-cookies   -bc  Use cookies from the browser placed in 'firefox_profile' or 'chrome_profile'
--force-cookies          Interpret the next to argument as the session cookie
--series            -S   Interpret the url as a link to a series and download all episodes from all seasons
--season            -s   Interpret the url as a link to a season and download all episodes from all seasons
--episode           -e   Interpret the url as a link to a single episode
--captions          -c   Download the captions along with the episode. Overridden by --captions-only if set.
--captions-only     -co  Download the captions only, without the episode.
--format              -f   Set the output container format (mp4 or mkv). Defaults to mp4
--list-urls           -l   List the URLs of episodes that would be downloaded without downloading them
--concurrent-segments -cs  Set the number of segments to download concurrently per stream.
                            Higher values = faster downloads but more network connections. Defaults to 5

If series, season, or episode is not used, the type will be inferred based on the link format.

Login in information must be placed in a file called login in the same directory as the executable. The file must be email then on a new line password. For example if your email is [email protected] and password password123 the file would be:

[email protected]
password123

If you would like to avoid logging in for any reason, cookies can be used. The option browser-cookies must be provided.

Create a file named firefox_profile in the build directory and paste in your firefox profile folder path

Install libgcrypt and create a file named chrome_profile in the build directory and paste in your chrome profile folder path (found on chrome://version)

Use the --force-cookies program option to manually input cookies.

If you have any issues or would like a feature to be added please don't hesitate to submit an issue after checking to make sure it hasn't already been submitted. Using the templates is a good place to start, but sometimes they're overkill. For example, if the program segfaults for you, you don't need to state that the intended behaviour is to not segfault.

If you'd like to contribute a good place to start is looking at open issues and trying to fix one with a pull request. \

• mosswg
• SeanOMik - Docker support
• Hello-User - Skip downloading if file exists
• MisterSheeple - Optional dependency installation instructions
• BradMJustice - Caption improvements