nixos-config/modules/nixos/tasks/multimedia-archive/README.adoc

107 lines
4.4 KiB
Plaintext
Raw Normal View History

= Multimedia archiving
:toc:
More like offline delivery, really.
Just wait for the task to complete and you have your videos, pictures, music, and whatever questionable files you want to download.
It's a nice offline repository for it especially that internet usually randomly disconnects so that's nice while I still have something working, yeah?
== Project structure
The following listing block shows the files and folders that this project should have.
[source, tree]
----
./modules/nixos/tasks/multimedia-archive/
├── data/
├── scripts/
├── default.nix
└── README.adoc
----
Some points of interests include...
* link:./data/[`./data/`] contains a bunch of data (i.e., jobs for various services).
This is used to easily import them from a process that is used to externally generate them.
Some of the data may have a schema available (e.g., link:./data/jobs.schema.json[`./data/jobs.schema.json`] for the jobs database).
* link:./scripts/[`./scripts/`] contains a bunch of scripts specific for this tasks.
2022-10-03 08:43:26 +00:00
Mainly, it contains scripts to generate data found in `./data/` such as link:./scripts/create-jobs-from-newpipe-db[a script that is used to generate a jobs database suitable for this task from a NewPipe database].
* link:./default.nix[`./default.nix`], the entry point for this module.
[#integrating-with-newpipe-subscriptions]
== Integrating with Newpipe subscriptions
In this task, I usually just download videos from YouTube.
While I could note every preferred creator manually, I could automate them by getting a list of subscriptions from my Newpipe config which I use surprisingly more often than I thought.
2022-10-03 08:43:26 +00:00
This is done by running the link:./scripts/create-jobs-from-newpipe-db[`./scripts/create-jobs-from-newpipe-db`] script and specifying the exported Newpipe database (as a ZIP file).
[CAUTION]
====
Please don't run the task with all of the subscriptions.
You should select only a few categories and clean them up.
====
[source, sh]
----
./scripts/create-jobs-from-newpipe-db.py ~/Downloads/NewPipeData-20220714_185126.zip
----
You can run the script with the `-h` flag for more information.
There are nifty things you can do with the script.
Such as the following code block which you can interactively select which folders to export.
[source, sh]
----
./scripts/create-jobs-from-newpipe-db.py ~/Downloads/NewPipeData-20220714_185126.zip --list-categories \
| fzf --multi --prompt "Choose which categories to export " \
| ./scripts/create-jobs-from-newpipe-db.py ~/Downloads/NewPipeData-20220714_185126.zip -o ./newpipe-db.json
----
Remember the larger the list, the larger the chances for a throttling.
Thus, it is heavily encouraged that you clean up your list (and/or get good at organizing your categories) before activating the updated version.
== Exporting a jobset with OPML
There is also an easy way to export a jobset from OPML with link:./scripts/create-jobs-from-rss-opml.py[`./scripts/create-jobs-from-rss-opml.py`].
The script should have similar interface with the featured script from <<integrating-with-newpipe-subscriptions>>.
Here's an basic example of using the script.
[source, sh]
----
./scripts/create-jobs-from-rss-opml.py ~/Downloads/MyThunderbirdFeeds-Blogs\ &\ News Feeds.opml
----
Take note there are assumptions to the exported subscription list.
- The folder structure is assumed from the outline.
- Any `<outline>` element with the `title`/`text` attribute is considered valid and as part of the hierarchy.
Otherwise, it is assumed they'll be in a fallback category within the outline.
- The category list is also assumed from the outline, only with the direct children of valid `<outline>` elements. footnote:[The `category` attribute from the RSS nodes is barely taken care of by most of the applications I use. WHY!?!]
- Categories are also extracted from the `category` attribute of each RSS node.
Similarly, it only extracts the head of each category hierarchy (e.g., `Computers` for `/Computers/Science`, `World` for `/World/Countries/Philippines`).
Similar to the Newpipe database script, you can do some nifty things with the script.
Here's the same example from the previous featured script.
[source, sh]
----
./scripts/create-jobs-from-rss-opml.py ~/Downloads/MyThunderbirdFeeds-Blogs\ &\ News Feeds.opml -l \
| fzf --multi --prompt "Choose which categories to export " \
| ./scripts/create-jobs-from-rss-opml.py ~/Downloads/MyThunderbirdFeeds-Blogs\ &\ News Feeds.opml
----