Web UI for gallery-dl
with support for downloading videos via yt-dlp
.
This example uses the docker run
command to create the container to run the app.
docker run -d \
--name gallery-dl \
-p 9080:9080 \
-e UID=1000 \
-e GID=1000 \
-v "/path/to/config:/config" \
-v "/path/to/downloads:/gallery-dl" \
--restart on-failure \
qx6ghqkz/gallery-dl-server:latest
This is an example service definition that could be put in docker-compose.yaml
. This service uses a VPN client container for its networking.
Gluetun is recommended for VPN usage. See docs/docker-compose.yaml for a template.
services:
gallery-dl:
image: qx6ghqkz/gallery-dl-server:latest
container_name: gallery-dl
network_mode: container:vpn
# ports:
# - 9080:9080
environment:
- UID=1000
- GID=1000
volumes:
- "/path/to/config:/config"
- "/path/to/downloads:/gallery-dl"
restart: on-failure
-
Make sure you mount the directory containing the config file rather than the file itself. This ensures changes to the config file are propagated to the running Docker container and you will not need to restart the container. More information on this issue here.
-
The output download directory depends on the
base-directory
in your gallery-dl config file. Make sure it is the absolute path/gallery-dl/
instead of the relative path./gallery-dl/
or you will need to mount your download directory to/usr/src/app/gallery-dl
instead (not recommended). -
You can use the environment variables
UID
andGID
to change the user ID and group ID of the user running the server process. This is important because downloaded files will be owned by that user. Make sure the IDs match those of the user on your host system. The defaultUID:GID
is1000:1000
when left unspecified.
If you have Python 3.12 or later installed and on your PATH, you can simply run the server using the command line. Clone this repository and install the required dependencies located in requirements.txt
in a virtual environment.
Run the command below in the root folder while inside the virtual environment. On Windows, replace python3
with python
.
python3 -m uvicorn gallery_dl_server:app --host "0.0.0.0" --port "9080" --log-level "info" --no-access-log
The program can also be run as a package, and optional environment variable overrides can be provided inline. On Windows, this can be done using set "VAR=value" &&
in Command Prompt or $env:VAR="value";
in PowerShell.
HOST="0.0.0.0" PORT="9080" LOG_LEVEL="info" ACCESS_LOG="False" python3 -m gallery_dl_server
When running as a package, a random available port will be selected if PORT
is not set as an environment variable.
On Windows, the program can be run using the prebuilt executable (.exe) file, which includes a Python interpreter and the required Python packages. Prebuilt executables for each release can be found in Releases.
By default, any available port will be selected. To select a specific port, run the executable from the command line with PORT
set as an environment variable.
set "HOST=0.0.0.0" && set "PORT=9080" && gallery-dl-server.exe
The executable utilises the same environment variable overrides as the Python package.
By default, this service listens on port 9080. You can use any value for the host port, but if you would like to map to a different internal container port, you need to set the CONTAINER_PORT
environment variable. This can be done using the -e
flag with docker run
or in a Docker Compose file.
For example, if you need to run multiple instances of gallery-dl-server using Gluetun for the networking (every instance must use a different internal port), you can specify an internal port other than the default 9080.
All services defined in the same docker-compose.yaml
file:
services:
gallery-dl:
image: qx6ghqkz/gallery-dl-server:latest
container_name: gallery-dl
depends_on:
- gluetun
network_mode: service:gluetun
...
gallery-dl-2:
image: qx6ghqkz/gallery-dl-server:latest
container_name: gallery-dl-2
environment:
- CONTAINER_PORT=9090
depends_on:
- gluetun
network_mode: service:gluetun
...
gluetun:
image: qmcgaw/gluetun:latest
container_name: gluetun
ports:
# gallery-dl
- 9080:9080
# gallery-dl-2
- 9090:9090
...
Configuration of gallery-dl is as documented in the official documentation. A configuration file is required.
If run outside of Docker, the default locations will be used to search for a configuration file. If run as an executable, the current directory will also be searched for a valid configuration file.
Additionally, YAML and TOML configuration files are supported at any of the pre-defined locations.
When run with Docker, the configuration file must be mounted inside the /config
directory inside the container.
/config/gallery-dl.conf
/config/gallery-dl.{yaml, yml}
/config/gallery-dl.toml
/config/config.json
/config/config.{yaml, yml}
/config/config.toml
A default configuration file for use with gallery-dl-server has been provided and will automatically be placed in the directory mounted to /config
if no valid configuration file exists in that location.
For more information on configuration file options, see gallery-dl/docs/configuration.rst.
Any additional locations specified in the configuration file must also exist inside the Docker container. For example, if you specify a cookies file location, ensure that location is accessible from within the Docker container.
It is recommended you place any additional files such as archive, cache and cookies files inside the same directory mounted to /config
along with the configuration file.
Downloads can be triggered by supplying the {{url}}
of the requested video through the Web UI or through the REST interface via curl, etc.
Just navigate to http://{{host}}:9080/gallery-dl
and enter the requested {{url}}
.
curl -X POST --data-urlencode "url={{url}}" http://{{host}}:9080/gallery-dl/q
fetch(`http://${host}:9080/gallery-dl/q`, {
method: "POST",
body: new URLSearchParams({
url: url
}),
});
Add the following bookmarklet to your bookmark bar so you can conveniently send the current page URL to your gallery-dl-server instance.
javascript:!function(){fetch("http://${host}:9080/gallery-dl/q",{body:new URLSearchParams({url:window.location.href}),method:"POST"})}();
The server uses starlette
for the web framework with gallery-dl
and yt-dlp
to handle the downloading. The integration with gallery-dl uses Python as described here. For video downloads, gallery-dl imports yt-dlp.
The Docker image is based on python:alpine
and consequently alpine
.
- Documentation: gallery-dl/README.rst
- Config file outline: gallery-dl/wiki/config-file-outline
- Configuration options: gallery-dl/docs/configuration.rst
- List of supported sites: gallery-dl/docs/supportedsites.md
- Documentation: yt-dlp/README.md
- List of supported sites: yt-dlp/supportedsites.md
- List of extractors: yt-dlp/yt_dlp/extractor/_extractors.py