danvics/ytptube
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1
ytptube/FAQ.md
arabcoders d9dbbb18ea feat: thumbnail generation
2026-05-14 23:22:27 +03:00

725 lines
34 KiB
Markdown
Raw Permalink Blame History

# Environment variables
Certain configuration values can be set via environment variables, using the `-e` parameter on the docker command line,
or the `environment:` section in `compose.yaml` file.
| Environment Variable | Description | Default |
| ------------------------------- | ------------------------------------------------------------------- | --------------------- |
| TZ | The timezone to use for the application | `(not_set)` |
| YTP_OUTPUT_TEMPLATE | The template for the filenames of the downloaded videos | `%(title)s.%(ext)s` |
| YTP_DEFAULT_PRESET | The default preset to use for the download | `default` |
| YTP_INSTANCE_TITLE | The title of the instance | `(not_set)` |
| YTP_FILE_LOGGING | Whether to log to file | `false` |
| YTP_DOWNLOAD_PATH | Path to where the downloads will be saved | `/downloads` |
| YTP_MAX_WORKERS | The maximum number of workers to use for downloading | `20` |
| YTP_MAX_WORKERS_PER_EXTRACTOR | The maximum number of concurrent downloads per extractor | `2` |
| YTP_AUTH_USERNAME | Username for basic authentication | `(not_set)` |
| YTP_AUTH_PASSWORD | Password for basic authentication | `(not_set)` |
| YTP_CONSOLE_ENABLED | Whether to enable the console | `false` |
| YTP_REMOVE_FILES | Remove the actual file when clicking the remove button | `false` |
| YTP_CONFIG_PATH | Path to where the config files will be stored. | `/config` |
| YTP_TEMP_PATH | Path to where tmp files are stored. | `/tmp` |
| YTP_TEMP_KEEP | Whether to keep the Individual video temp directory or remove it | `false` |
| YTP_HOST | Which IP address to bind to | `0.0.0.0` |
| YTP_PORT | Which port to bind to | `8081` |
| YTP_LOG_LEVEL | Log level | `info` |
| YTP_STREAMER_VCODEC | The video encoding codec, default to GPU and fallback to software | `""` |
| YTP_STREAMER_ACODEC | The audio codec to use for in-browser streaming | `aac` |
| YTP_VAAPI_DEVICE | The VAAPI device to use for hardware acceleration. | `/dev/dri/renderD128` |
| YTP_ACCESS_LOG | Whether to log access to the web server | `true` |
| YTP_DEBUG | Whether to turn on debug mode | `false` |
| YTP_DEBUGPY_PORT | The port to use for the debugpy debugger | `5678` |
| YTP_EXTRACT_INFO_TIMEOUT | The timeout for extracting video information | `70` |
| YTP_PIP_PACKAGES | A space separated list of pip packages to install | `(not_set)` |
| YTP_PIP_IGNORE_UPDATES | Do not update the custom pip packages | `false` |
| YTP_PICTURES_BACKENDS | A comma separated list of pictures urls to use | `(not_set)` |
| YTP_BROWSER_CONTROL_ENABLED | Whether to enable the file browser actions | `false` |
| YTP_YTDLP_AUTO_UPDATE | Whether to enable the auto update for yt-dlp | `true` |
| YTP_YTDLP_DEBUG | Whether to turn debug logging for the internal `yt-dlp` package | `false` |
| YTP_YTDLP_VERSION | The version of yt-dlp to use. Defaults to latest version | `(not_set)` |
| YTP_BROWSER_URL | Remote browser endpoint for the bundled `browser` extractor | `(not_set)` |
| YTP_FLARESOLVERR_URL | FlareSolverr endpoint URL. | `(not_set)` |
| YTP_FLARESOLVERR_MAX_TIMEOUT | Max FlareSolverr challenge timeout in seconds | `120` |
| YTP_FLARESOLVERR_CLIENT_TIMEOUT | HTTP client timeout (seconds) when calling FlareSolverr | `120` |
| YTP_FLARESOLVERR_CACHE_TTL | The cache TTL (in seconds) for FlareSolverr solutions | `600` |
| YTP_BASE_PATH | Set this if you are serving YTPTube from sub-folder | `/` |
| YTP_PREVENT_LIVE_PREMIERE | Prevents the initial youtube premiere stream from being downloaded | `false` |
| YTP_LIVE_PREMIERE_BUFFER | buffer time in minutes to add to video duration | `5` |
| YTP_TASKS_HANDLER_TIMER | The cron expression for the tasks handler timer | `15 */1 * * *` |
| YTP_TEMP_DISABLED | Disable temp files handling. | `false` |
| YTP_DOWNLOAD_PATH_DEPTH | How many subdirectories to show in auto complete. | `1` |
| YTP_ALLOW_INTERNAL_URLS | Allow requests to internal URLs | `false` |
| YTP_SIMPLE_MODE | Switch default interface to Simple mode. | `false` |
| YTP_STATIC_UI_PATH | Path to custom static UI files. | `(not_set)` |
| YTP_AUTO_CLEAR_HISTORY_DAYS | Number of days after which completed download history is cleared. | `0` |
| YTP_DEFAULT_PAGINATION | The default number of items per page for history. | `50` |
| YTP_TASK_HANDLER_RANDOM_DELAY | The maximum random delay in seconds before starting a task handler. | `60` |
| YTP_IGNORE_ARCHIVED_ITEMS | Don't report archived items in the download history. | `false` |
| YTP_CHECK_FOR_UPDATES | Whether to check for application updates. | `true` |
| YTP_EXTRACT_INFO_CONCURRENCY | The number of concurrent extract info operations. | `4` |
| YTP_THUMB_CONCURRENCY | The number of concurrent ffmpeg thumbnail generations allowed. | `2` |
| YTP_THUMB_GENERATE | Enable ffmpeg thumbnail generation when no local thumbnail exists. | `true` |
| YTP_THUMB_SIDECAR | Save generated thumbnails next to media instead of temp cache. | `false` |
> [!NOTE]
> To raise the maximum workers for specific extractor, you need to add a ENV variable that follows the pattern `YTP_MAX_WORKERS_FOR_<EXTRACTOR_NAME>`.
> The extractor name must be in uppercase, to know the extractor name, check the log for the specific extractor used for the download.
> The limit should not exceed the `YTP_MAX_WORKERS` value as it will be ignored.
> [!IMPORTANT]
> The env variable `YTP_SIMPLE_MODE` only control what being displayed for first time visitor, the users can still switch between the two modes via the WebUI settings page.
## Notes about YTP_AUTO_CLEAR_HISTORY_DAYS
- `0` days means no automatic clearing of the download history. lowest value that will trigger the clearing is `1` day.
- This setting will **NOT** delete the downloaded files, it will only clear the history from the database.
# Browser extensions & bookmarklets
## Simple bookmarklet
```javascript
javascript:(() => { const url = "https://ytp.example.org"; const preset = "default"; const mUrl = new URL(url);mUrl.pathname="/api/history/add";mUrl.searchParams.set("url",document.location.href);mUrl.searchParams.set("preset",preset);fetch(mUrl,{method: "GET"}).then(j => j.json()).then(json =>alert(json.message)).catch(err =>alert(err)); })()
```
Change the the variable `url` and `preset` variables to match your YTPTube instance and preset name.
> [!NOTE]
> The bookmarklet should be served from https page, otherwise, some browsers will block the request. for mixed content.
## Browser stores
- For Firefox via [Firefox Store](https://addons.mozilla.org/en-US/firefox/addon/ytptube-extension/)
- For Chrome/Chromium Browsers via [Chrome Store](https://chromewebstore.google.com/detail/ytptube-extension/kiepfnpeflemfokokgjiaelddchglfil)
## iOS Shortcuts
You can download [Add To YTPTube](https://www.icloud.com/shortcuts/6df61c97d97b4e539c9100999ba39dd4) shortcut and use it to send links to your YTPTube instance.
You have to edit the shortcut and replace the following:
- `https://ytp.example.org` with your YTPTube instance.
- `username:password` replace this with your own username & password or leave it empty if you don't have authentication enabled.
This shortcut is powerful, as it's allow you to select your preset on the fly pulled directly from your instance.
Combined with the new and powerful presets system, you could add presets for specific websites that need cookies,
and use that preset to download directly from your iOS device.
### Advanced iOS Shortcut
This shortcut [YTPTube To Media](https://www.icloud.com/shortcuts/4dc579382f254635ad5785424055f173) is more advanced, as it's parses
the `yt-dlp` output and attempt to download the media directly to your iOS device. It doesn't always work, but it's a good
starting point for those who want to download media directly to their iOS device. We provide no support for this use case
other than the shortcut itself. this shortcut missing support for parsing the http_headers, it's only parse the cookies.
# Authentication
To enable basic authentication, set the `YTP_AUTH_USERNAME` and `YTP_AUTH_PASSWORD` environment variables. And restart
the container. This will prompt the user to enter the username and password before accessing the web interface/API.
As this is a simple basic authentication, if your browser doesn't show the prompt, you can use the following URL
`http://username:password@your_ytptube_url:port`
# I cant download anything
If you are receiving errors like:
- "OSError: [Errno 5] I/O error"
- "OSError: [Errno 18] Cross-device link: '/tmp/random_id/name.webm' -> '/downloads/name.webm'
- "Operation not permitted: '/downloads/name.webm'
This indicates an error with your mounts and how they interact with the container. So, the basic solution is to do the following:
```yaml
services:
ytptube:
user: "${UID:-1000}:${UID:-1000}" # change this to your user id and group id, for example: "1000:1000"
image: ghcr.io/arabcoders/ytptube:latest
container_name: ytptube
restart: unless-stopped
environment:
- YTP_TEMP_PATH=/downloads/tmp
- YTP_DOWNLOAD_PATH=/downloads/files
ports:
- "8081:8081"
volumes:
- ./config:/config:rw
- ./downloads:/downloads:rw
```
Then run the following command to create the necessary directories and start the container:
```bash
mkdir -p ./config && mkdir -p ./downloads/{tmp,files} && docker compose -f compose.yaml up -d
```
Reference: [Issue #363](https://github.com/arabcoders/ytptube/issues/363)
# I want to use link with playlist but only download the video not all the videos in the playlist?
Simply create a preset, and in the `Command options for yt-dlp` field set `--no-playlist`. Then select the preset
whenever the link includes a playlist id.
> [!NOTE]
> You can also do the same via advanced options `Command options for yt-dlp` field, but presets are more convenient.
# Install specific yt-dlp version?
You can force specific version of `yt-dlp` by setting the `YTP_YTDLP_VERSION` environment variable for example
```env
YTP_YTDLP_VERSION=2025.07.21 or master or nightly
```
Then restart the container to apply the changes.
# Custom output template placeholders
YTPTube supports custom `ytp_*` placeholders in `yt-dlp` output template via the following syntax `%(ytp_*:<args>)s`.
## Currently available extra placeholders are:
- `ytp_random`: random mixed letters and digits,
- `N` A number is required to specify the length of the random string, for example `%(ytp_random:8)s` will generate a random string of 8 characters.
- if the args followed by `:s` it will generate random letters only, if followed by `:d` it will generate random digits only.
## Examples of the custom placeholders in action:
- Template: `%(title)s [%(ytp_random:8)s].%(ext)s`
- Example result: `My Video [A7k2Pq9Z].mp4`
- Template: `%(uploader)s/%(ytp_random:6:d)s - %(title)s.%(ext)s`
- Example result: `MyChannel/483920 - My Video.mp4`
- Template: `%(playlist)s/%(ytp_random:10:s)s/%(title)s.%(ext)s`
- Example result: `Favorites/QwErTyUiOp/My Video.mp4`
> [!NOTE]
> `%(ytp_` placeholders are a YTPTube extension and not avaliable via console or directly via yt-dlp.
# How can I monitor sites without RSS feeds?
YTPTube includes a **generic task handler** that turns JSON definitions into site-specific scrapers. You can use it
to watch pages that do not expose RSS or public APIs and automatically enqueue new links into the download queue.
1. Create definition via the WebUI > tasks > Definitions.
2. Create task that reference same url click on inspect to see the results. Make sure it uses a preset that enables
a download archive (`--download-archive`).
3. When the task handler run, the handler scans the definitions, picks the first definition whose `match` rule covers
the task URL, fetches the page, extracts items, and queues the unseen ones.
### Definition schema
Each definition must contain a single JSON object with the following keys:
```json5
{
"name": "example", // Friendly identifier shown in logs
"match_url": [
"https://example.com/articles/*", // Glob strings
"https://example.com/post/[0-9]+" // Regex strings
],
"engine": { // Optional, defaults to HTTPX
"type": "httpx", // "httpx" (default) or "selenium"
"options": {
"url": "http://selenium:4444/wd/hub", // Selenium-only: remote hub URL
"arguments": ["--headless", "--disable-gpu"],
"wait_for": { "type": "css", "expression": ".article" },
"wait_timeout": 15,
"page_load_timeout": 60
}
},
"request": { // Optional HTTP settings
"method": "GET", // GET or POST
"url": "https://example.com/articles/latest", // Override the task URL if needed
"headers": { "User-Agent": "MyAgent/1.0" },
"params": { "page": 1 },
"data": null,
"json_data": null,
"timeout": 30
},
"response": { // Optional: how to interpret the body
"type": "html" // "html" (default) or "json"
},
"parse": {
"items": { // Optional container for per-item extraction
"selector": ".columns .card", // Defaults to CSS; set "type": "xpath" to use XPath
"fields": {
"link": { // Required inside fields: the per-item URL
"type": "css",
"expression": ".card-header a",
"attribute": "href"
},
"title": {
"type": "css",
"expression": ".card-header a",
"attribute": "text"
},
"poet": {
"type": "css",
"expression": "footer .card-footer-item:first-child a",
"attribute": "text"
}
}
},
"page_title": { // Optional global field outside the container
"type": "css",
"expression": "title",
"attribute": "text"
}
}
}
```
For JSON endpoints, switch the response format and use `jsonpath` selectors:
```json5
{
...
"response": { "type": "json" },
"parse": {
"items": {
"type": "jsonpath",
"selector": "items",
"fields": {
"link": { "type": "jsonpath", "expression": "url" },
"title": { "type": "jsonpath", "expression": "title" }
}
}
}
}
```
### Parsing rules
- Every definition must provide a `link` field either at the top level or inside `parse.items.fields`. Other fields are optional metadata attached to the queued item.
- CSS and XPath rules may specify `attribute`:
- `text` / `inner_text` applies `normalize-space()`.
- `html` / `outer_html` returns the raw HTML fragment.
- Any other value reads that attribute from the element. When omitted, the handler uses text and, for `link`, falls
back to `href` automatically.
- Regex rules scan the HTML fragment associated with the current scope (page-level or container). Set `attribute` to a named/numbered capture group or rely on the first group.
- `post_filter` lets you run a final regex on the extracted value and pick a named (`value`) group.
- When you declare `parse.items`, each matching container is processed independently so missing fields in one card do not shift values for the rest.
- For JSON responses (`response.type = "json"`), set the container `type` and field `type` to `jsonpath` and supply [JMESPath](https://jmespath.org/) expressions. Relative values are resolved against each container object and converted to strings automatically.
### Fetch engines
- **httpx (default)**: supports custom headers, query params, JSON/body payloads, proxy inherited from the task preset,
and optional timeout.
- **selenium**: uses a remote Chrome session. Provide the hub URL under `engine.options.url`; only Chrome is supported at
the moment. Optional keys: `arguments` (list or string), `wait_for` (type `css`/`xpath` + `expression`), `wait_timeout`,
and `page_load_timeout`.
> [!NOTE]
> A machine-readable schema is available at `app/schema/task_definition.json` if you want to validate your JSON with editors or CI tools.
# How to generate POT tokens?
You need a pot provider server we already have the extractor `bgutil-ytdlp-pot-provider` pre-installed in the container.
You can simply do the following to enable the support for it.
```yaml
services:
ytptube:
user: "${UID:-1000}:${UID:-1000}" # change this to your user id and group id, for example: "1000:1000"
image: ghcr.io/arabcoders/ytptube:latest
container_name: ytptube
restart: unless-stopped
ports:
- "8081:8081"
volumes:
- ./config:/config:rw
- ./downloads:/downloads:rw
tmpfs:
- /tmp
depends_on:
- bgutil_provider
bgutil_provider:
init: true
image: brainicism/bgutil-ytdlp-pot-provider:latest
container_name: bgutil_provider
restart: unless-stopped
```
Then simply create a new preset, and in the `Command options for yt-dlp` field set the following:
```bash
--extractor-args "youtubepot-bgutilhttp:base_url=http://bgutil_provider:4416"
```
For more information please visit [bgutil-ytdlp-pot-provider](https://github.com/Brainicism/bgutil-ytdlp-pot-provider) project.
# Troubleshooting and submitting issues
Before asking a question or submitting an issue for YTPTube, please remember that YTPTube is only a wrapper for
[yt-dlp](https://github.com/yt-dlp/yt-dlp). Any issues you might be experiencing with authentication to video websites,
postprocessing, permissions, other `yt-dlp options` configurations which seem not to work, or anything else that
concerns the workings of the underlying yt-dlp library, need not be opened on the YTPTube project.
In order to debug and troubleshoot them, it's advised to try using the yt-dlp binary directly first, bypassing the UI,
and once that is working, importing the options that worked for you into a new `preset`.
## Via HTTP
If you have enabled the web terminal via `YTP_CONSOLE_ENABLED` environment variable, simply go to `Other > Terminal` use
the yt-dlp command, the interface is jailed to the `yt-dlp` binary you can't access anything else. Or from download form
by clicking `advanced options` button than the yellow terminal icon `Run directly in console`.
## Via CLI
Assuming your YTPTube container is called `ytptube`, run the following on your docker host to get a shell inside the container:
```bash
docker exec -ti ytptube bash
cd /downloads
yt-dlp ....
```
Once there, you can use the yt-dlp command freely.
# Run behind reverse proxy.
It's advisable to run YTPTube behind a reverse proxy, if better authentication and/or HTTPS support are required.
## Caddy http server
The following example Caddyfile gets a reverse proxy going behind [caddy](https://caddyserver.com).
```caddyfile
# If you are using sub-domain.
# make sure to change "ytptube:8081" to the actual name of your YTPTube container/port.
ytp.example.org {
reverse_proxy ytptube:8081
}
# If you are using sub-folder, for example: https://example.org/ytptube/
# Also make sure to set the `YTP_BASE_PATH` environment variable to `/ytptube/`
# make sure to change "ytptube:8081" to the actual name of your YTPTube container/port.
example.org {
redir /ytptube /ytptube/
route /ytptube/* {
reverse_proxy ytptube:8081
}
}
```
# How to autoload yt-dlp plugins?
Loading yt-dlp plugins in YTPTube is quite simple, we already have everything setup for you. simply, create a folder
inside the `/config` directory named `yt-dlp` so, the path will be `/config/yt-dlp`. then follow
[yt-dlp plugins docs](https://github.com/yt-dlp/yt-dlp?tab=readme-ov-file#plugins) to know how to install the plugins.
Once you have installed the plugins, restart the container and the plugins will be auto-loaded on demand.
# How to load random backgrounds from WatchState or any other source?
YTPTube can be configured to pull random background images from different sources, including `WatchState` which is another
project of mine, simply change the `YTP_PICTURES_BACKENDS` environment variable to the following url
```env
YTP_PICTURES_BACKENDS=https://watchstate.ip/v1/api/system/images/background?apikey=[api_key]
```
Where `[api_key]` is the api key you get from your WatchState instance.
# How to use share folder or external storage as download target?
You can simply mount the share folder as target for `/downloads` path, but in some cases you might face issues with permissions or
cross-device link errors. To avoid these issues, you can mount the share folder as a named volume, and then mount the
named volume to `/downloads/smb` or `/downloads/nfs`.
```yaml
services:
ytptube:
user: "${UID:-1000}:${UID:-1000}" # change this to your user id and group id, for example: "1000:1000"
image: ghcr.io/arabcoders/ytptube:latest
container_name: ytptube
restart: unless-stopped
ports:
- "8081:8081"
volumes:
# Config must be mounted locally as read-write sqlite doesn't support network mounts.
- ./config:/config:rw
# Mount a local directory
- ./downloads:/downloads/local:rw
# Mount the NFS share
- nfs-data:/downloads/nfs:rw
# Mount the SMB share
- smb-data:/downloads/smb:rw
tmpfs:
- /tmp
volumes:
nfs-data:
driver: local
driver_opts:
type: nfs
o: addr=10.0.0.3,rw,nfsvers=4 # <--- Change server IP and options
device: ":/exported/path" # <--- Remote NFS path
smb-data:
driver: local
driver_opts:
type: cifs
o: username=my_username,password=my_password,vers=3.0,uid=1000,gid=1000,file_mode=0777,dir_mode=0777 # <--- Change options to fit your needs
device: "//10.0.0.3/public" # <--- Remote SMB path
```
If you prefer, you can bypass YTPTube `download_path` and set it to `/` and completely manage your own mounts. However,
please be aware that the file browser feature will expose whatever `download_path` is set to. **So, if you set it to `/`,
the file browser will expose the entire container filesystem.**
# The origin of the project.
The project first started as a fork [meTube](https://github.com/alexta69/metube), since then it has been completely
rewritten and redesigned. The original project was a great starting point, but it didn't align with my vision for the
project and what i wanted to achieve with it.
# How to use hardware acceleration for video transcoding?
As the container is rootless, we cannot do the necessary changes to the container to enable hardware acceleration.
However, We do have the drivers and ffmpeg already installed and the CPU transcoding should work regardless. To enable
hardware acceleration You need to alter your `compose.yaml` file to mount the necessary devices to the container. Here
is an example of how to do it for debian based systems.
```yaml
services:
ytptube:
........ # see above for the rest of the configuration
devices:
# mount the dri devices to the container if you only have one gpu you can simply do the following
- /dev/dri:/dev/dri
# Otherwise, selectively mount the devices you need.
- /dev/dri/card0 # Intel GPU device
- /dev/dri/renderD128 # Intel GPU render node
group_add:
# Add the necessary groups to the container to access the gpu devices.
- 44 # it might be different on your system.
- 105 # it might be different on your system.
```
This setup should work for at VAAPI encoding in `x86_64` containers.
> [!NOTE]
> Your `video`, `render` group id might be different from mine, you can run the follow command in docker host server to get the group ids for both groups.
```bash
cat /etc/group | grep -E 'render|video'
video:x:44:your_docker_username
render:x:105:your_docker_username
```
In my docker host the group id for `video` is `44` and for `render` is `105`. change what needed in the `compose.yaml`
file to match your setup.
If for some reason the initial test for GPU encoding fails, YTPTube will fallback to software encoding. You can force
software encoding by setting the `YTP_STREAMER_VCODEC` environment variable to `libx264`. If you want to force GPU encoding, set the
`YTP_STREAMER_VCODEC` environment variable to one of the supported GPU codecs, for example `h264_vaapi` or `h264_nvenc` depending on your GPU.
For more information about the supported codecs, please refer to the [SegmentEncoders.py](app/library/SegmentEncoders.py) file.
If GPU encoding fails and software encoding is used, you will have to restart the container to try GPU encoding again.
as we only test for GPU encoding once on first video stream.
# Allowing internal URLs requests
By default, YTPTube prevents requests to internal resources. However, if you want to allow requests to internal URLs, you can set the `YTP_ALLOW_INTERNAL_URLS` environment variable to `true`. This will allow requests to internal URLs.
We do not recommend enabling this option unless you know what you are doing, as it can expose your internal network to
potential security risks. This should only be used if it's truly needed.
# How to setup CI on Gitea?
The docker container builder already support self-hosted repositories like Gitea, you simply need to define two things at your repository settings.
1. Create a secret named `GIT_TOKEN` and set it to your Gitea personal access token.
2. Create a variable named `REGISTRY` and set it to your docker registry, for example `gitea.domain.org`.
Thats it, the `main.yml` will now disable the docker/github container registries, and use your Gitea repository instead. It will follow the usual
naming, your container name will be named `REGISTRY/ytptube` and the tags will be the same as the ones used in the github registry.
Unfortunately, the `native-builder.yml` workflow doesn't support self-hosted repositories at the moment.
# Getting No space left on device error
If you encounter this error: `OSError: [Errno 28] No space left on device` This indicates that either
the `/tmp` or `/downloads` directory has run out of available space.
This issue commonly occurs when:
- `/tmp` is mounted as `tmpfs` (memory-based storage)
- Your system has limited RAM
- You're downloading large video files
Since videos are temporarily stored in `/tmp` before being moved to the final download location, memory-based storage
may be insufficient for large downloads.
To fix the issue, modify your `compose.yaml` to use a disk-based directory for temporary files:
```yaml
services:
ytptube:
user: "${UID:-1000}:${UID:-1000}"
image: ghcr.io/arabcoders/ytptube:latest
container_name: ytptube
restart: unless-stopped
ports:
- "8081:8081"
volumes:
- ./config:/config:rw
- ./downloads:/downloads/local:rw
- ./temp:/tmp:rw
```
> [!NOTE]
> Replace the `tmpfs` mount with a local directory volume (`./temp:/tmp:rw`). This allows temporary files to use disk space instead of RAM.
After making the changes, restart your container. This should resolve the "No space left on device"
error during download.
# How to prevent loading screen during YouTube premieres?
Depending on how you look at it, YTPTube live download implementation is rather great and fast. However, during YouTube
premieres, usually streams will contain a loading screen of say, 1-5 minutes before the actual video content starts
playing. By default we wait for 5min + the duration of the video before starting the download to ensure we get the full video without
the loading screen. However, you can override the behavior by setting the following environment variable:
```env
YTP_LIVE_PREMIERE_BUFFER=10
```
Where `YTP_LIVE_PREMIERE_BUFFER` is the buffer time in minutes to add to the video duration before the download starts.
This will help in case the premiere has a longer loading screen than usual.
# How to bypass CF challenges?
You need to setup [FlareSolverr](https://github.com/FlareSolverr/FlareSolverr) and then set the `YTP_FLARESOLVERR_URL`
environment variable to point to your FlareSolverr instance. For example:
```yaml
services:
ytptube:
user: "${UID:-1000}:${UID:-1000}" # change this to your user id and group id, for example: "1000:1000"
image: ghcr.io/arabcoders/ytptube:latest
container_name: ytptube
restart: unless-stopped
environment:
- YTP_FLARESOLVERR_URL=http://flaresolverr:8191/v1
ports:
- "8081:8081"
volumes:
- ./config:/config:rw
- ./downloads:/downloads:rw
tmpfs:
- /tmp
depends_on:
- flaresolverr
flaresolverr:
image: flaresolverr/flaresolverr:latest
container_name: flaresolverr
restart: unless-stopped
```
For more information please visit [FlareSolverr](https://github.com/FlareSolverr/FlareSolverr) project.
# How to use the browser extractor?
YTPTube ships with a bundled `browser` extractor plugin for `yt-dlp`. It opens the target page in a remote browser,
waits for the page network activity, then tries to extract media urls from the captured requests and media elements.
This is mainly useful for sites where the regular extractor or generic page parsing does not expose the final media url,
but a real browser session does.
## Environment variables
You can define the remote browser endpoint globally with the following environment variable.
### Selenium example
```env
YTP_BROWSER_URL=selenium+http://selenium:4444/wd/hub
```
### Playwright example
For a native Playwright server, point to the Playwright websocket endpoint:
```env
YTP_BROWSER_URL=playwright+ws://playwright:3000/
```
### Playwright CDP example
To connect Playwright over the Chrome DevTools Protocol, use either the shorthand `playwright+cdp://` form or an explicit transport form:
```env
YTP_BROWSER_URL=playwright+cdp://chrome:9222/
# same as playwright+cdp+http://chrome:9222/
```
- `YTP_BROWSER_URL` is the full backend selector and endpoint in one value.
- Supported forms are:
- `selenium+http://...`
- `selenium+https://...`
- `playwright+ws://...`
- `playwright+wss://...`
- `playwright+cdp://...` which is treated as `playwright+cdp+http://...`
- `playwright+cdp+http://...`
- `playwright+cdp+https://...`
- `playwright+cdp+ws://...`
- `playwright+cdp+wss://...`
## yt-dlp usage
If you want to set the browser extractor options directly on the yt-dlp side, you can also use `--extractor-args` with `generic:url=...`:
### Selenium example
```bash
--use-extractors "generic" --extractor-args "generic:url=selenium+http://selenium:4444/wd/hub"
```
### Playwright example
```bash
--use-extractors "generic" --extractor-args "generic:url=playwright+ws://playwright:3000/"
```
### Playwright CDP example
```bash
--use-extractors "generic" --extractor-args "generic:url=playwright+cdp://chrome:9222/"
```
The explicit `--extractor-args` value takes priority over `YTP_BROWSER_URL`.
## Example compose setup
```yaml
services:
ytptube:
user: "${UID:-1000}:${UID:-1000}"
image: ghcr.io/arabcoders/ytptube:latest
container_name: ytptube
restart: unless-stopped
environment:
- YTP_BROWSER_URL=selenium+http://selenium:4444/wd/hub # or playwright+ws://playwright:3000/ or playwright+cdp://chrome:9222/
ports:
- "8081:8081"
volumes:
- ./config:/config:rw
- ./downloads:/downloads:rw
selenium:
image: selenium/standalone-chrome:latest
container_name: selenium
restart: unless-stopped
shm_size: 2gb
playwright:
image: mcr.microsoft.com/playwright:v1.59.0-noble
container_name: playwright
restart: unless-stopped
command: /bin/sh -c "npx -y playwright@1.59.0 run-server --port 3000 --host 0.0.0.0"
```
> [!NOTE]
> The browser extractor is slower than the normal extractor flow and should only be used when a site actually needs a real browser session.
> playwright require same version for both the server and the client, so make sure to use the same version in the container and in your local environment if you want to test it locally.
Reference in a new issue View git blame Copy permalink