I use it on my Raspberry Pi mini-dlna (aka ReadyMedia) media centre. It’s a node.js app and takes a couple of seconds to start on my tiny, resource-constrained Pi. My normal usage pattern is to look at the file list in put.io, click through the web UI to get the put.io file ID of the
mp4 itself and then call
kaput download <ID>.
If you call
kaput download on the folder ID containing the piece of media you want instead of the media itself, kaput via the put.io API, will download a
zip of the files for you. I do this all the time by mistake and find it incredibly annoying, because minidlna expects to find a folder or a file, not a zip, especially as unzipping on the Pi can be slow too.
The other usage pattern I find myself doing frequently, is downloading a bunch of files in one session. kaput doesn’t have a multiple download option; you can’t say
kaput download <ID1> <ID2> <ID3>.
Initially I’d thought about forking kaput, but I wanted to see what the quickest thing I could do was. There’s a nice wrapper called
with that lets you turn any command into a REPL. The example they quote uses
$ with git git> add . git> commit -a -m "Commited" git> push
So I thought this would be a nice help for the slow start-up issue, plus wanting to download multiple IDs with kaput. You can use subcommand in
with too, so:
$ with kaput download kaput download > 123456789 Gathering file info... done The 39 Steps ░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░ | 2% | Progress: 33.17 MB/1.6 GB | ETA: 22m5s | Speed: 1.72 MB/s
It’s great and just works, but now I’m just staring at the progress bar whilst the file is downloading. What I really want, is to start another download before the first one completes.
I liked the idea of the REPL and once I’d started to think about it, wanted the ability to change and make folders within the REPL so I can organise files as I fetch them. Time for some bash!
I started off with a simple loop:
while true; do read -p "kaslurp> " request # do stuff with $request end
And have ended up with the slightly more complex, but useful (to me)
kaslurp - a REPL wrapper for kaput-cli.
Basic usage is just
kaslurp you’ll get a REPL in the folder that’s defined as
$MEDIADIR (in my case,
/mnt/video). To download a put.io file, enter
d <ID> and it will download in the background. You can continue to use the REPL whilst this is happening, and even if you quit
kaslurp, any running downloads will complete. If you start with a
-d argument, this will override the starting
$ kaslurp /mnt/video kaslurp> d 123456 Downloading 12345 in the background /mnt/video kaslurp> cd Thrillers /mnt/video/Thrillers kaslurp> ls total 263MB -rw-r--r-- 1 pi pi 263M Jan 1 21:31 Dressed_to_Kill_(1946).mp4 /mnt/video/Thrillers kaslurp> mkdir "Sherlock Holmes" /mnt/video/Thrillers kaslurp> cd Sherlock\ Holmes /mnt/video/Thrillers/Sherlock Holmes kaslurp>
As well as downloading, you can see from the example above that you can
mkdir all from within the REPL.
kaslurp also has a non-REPL mode too, where you can quickly background a download in the current directory or in a named one.
Usage: kaslurp Start the REPL kaslurp -d /some/path Start the REPL at /some/path kaslurp <put.io file ID> Start a background download of <ID> to your current pwd and exit kaslurp -d /some/path <put.io file ID> Start a background download of <ID> to /some/path and exit
- Support PCRE2 or PCRE in `build-nginx`
- Tracking remote branches after a `--single-branch` checkout
- TIL: How to check and reset a directory tree of permissions with `chmod` and `namei`
- TIL: How to specify a host or hosts for an ansible-playbook without an inventory file
- Automating Jekyll card generation with ruby's Ferrum gem