I created bash noscripts to automate and simplify getting my homelab up and running.
https://github.com/rishavnandi/Boiler_plates

https://redd.it/wwjchw
@r_bash

Can you add features such as Syntax highlighting and Auto-suggestions to Bash?

Recently I've started to experiment with Bash and I have one question. Is their a way to add Syntax highlighting and Autosuggestions to Bash? The one thing I LOVE about the Fish shell is it's out-of-the-box autosuggestion feature that not only tries to auto-complete commands, but also suggests from your command history, this is the one thing I think Fish dose better than any other shell and I'm hoping that I can replicate it on Bash.

I know it's challenging to implement autosuggestions in Bash because it's a limitation of GNU readline - Bash uses readline for entering and editing text, but this means you're limited to using readline features for dynamic line editing. E.g you cannot dynamically change the color of text a user inputs using readline (limiting the implementation of Fish-like syntax highlighting).

Do any workarounds exist to make syntax highlighting and autosuggestions work in Bash? Any help would be much appreciated.

https://redd.it/wylaoj
@r_bash

How to interpret ANSI escape characters in a file?

Is there an easy way to read a text file (presumably a log of a previously executed command) and interpret any escape sequences so that cat'ing the new log file is equivalent to viewing it in a text editor? For example, if the output of the command sends the letter A, a backspace, and then the letter E, the log file (after being interpreted) should only contain the character E and nothing else.

EDIT: Figured it out

https://redd.it/wzk2ly
@r_bash

Authentication Bearer Token

Is it possible to obtain the auth bearer token of a login session through noscript? I want to POST something on my website via curl POST but can't as it required bearer token to be passed. Can I use my creds to GET token in command line or through bash noscript?

https://redd.it/x05q6v
@r_bash

Brash Cli Trash Manager in Pure Bash

Why Brash

Well, why not. As you see its similar to Trash\_cli. Unlike Trash_cli, Brash don't Depends on any Python libs just pure bash.

So why not ?

https://redd.it/x0m7y4
@r_bash

is there a way to get a command history list view like this in bash ?

​

https://preview.redd.it/m9duxw3cnuk91.png?width=1205&format=png&auto=webp&s=d5708c47a840039b0d59215f65ec4cae9436bab6

https://redd.it/x1ime0
@r_bash

Is there a Bash equivalent to Zsh Named Directories feature?

Hi, I read that Zsh has the Named Directories feature, where you can create an 'alias' to a path.

The neat part is that you can use this alias as part of a command.

So, instead of using something like:

cp .bashrc /very/long/path/name

I could use:

cp .bashrc vlpn

Does Bash has something like that?

https://redd.it/x1nik6
@r_bash

Is it possible to stop a bash a bash noscript during execution and then return back to where you left off the next time you run the noscript

Hopefully the noscript makes sense, for context let's say this is the situation:

- I'm running a bash noscript that runs random commands one by one and in order to proceed to each next command the user has to press the spacebar. there are about 80 items

- Let's say the user presses CTRL+C to stop the noscript from executing halfway through those 80 items

Is there a way to return back to where they were in noscript execution when they cancelled the noscript? Or do they have to re-run the noscript and then start at the very beginning again?

https://redd.it/x225ew
@r_bash

Re-directing outputs from a noscript that uses FFmpeg to recursively transcode video files

Hi all. I have very little experience so far with bash noscripting, so please forgive my rudimentary question... but I've been given a noscript that uses FFmpeg to recursively transcode mkv video files to mp4 derivatives:

#!/bin/bash
VIDEOS=/Users/ExampleUser/Desktop/1
cd /Users/ExampleUser/Desktop/1
find "$VIDEOS" -name '*.mkv' -exec sh -c 'ffmpeg -analyzeduration 15M -guess_layout_max 0 -i "$0" -y -pix_fmt yuv420p -codec:v libx264 -fps_mode cfr -r 29.97 -filter:v "yadif,scale=640:480" -aspect 4:3 -preset medium -g 60 -sc_threshold 0 -profile:v main -b:v 1000k -minrate 900k -maxrate 1200k -bufsize 2000k -tune film -pass 1 -an -f mp4 /dev/null ; ffmpeg -i "$0" -y -pass 2 -pix_fmt yuv420p -codec:v libx264 -fps_mode cfr -r 29.97 -filter:v "yadif,scale=640:480" -preset medium -g 60 -sc_threshold 0 -profile:v main -b:v 1000k -minrate 900k -maxrate 1200k -bufsize 2000k -tune film -map 0:0 -map 0:1 -codec:a aac_at -ac 2 -ar 48000 -b:a 128k -filter:a "aresample=async=1:min_hard_comp=0.100000:first_pts=0" -f mp4 -movflags faststart "${0%%.mkv}_sl.mp4"' {} \;
rm *.log.mbtree *.log
exit;

Right now, it outputs the derivative mp4s into the same directory as the source mkv files (specified by the variable VIDEOS). How can I change the output destination for the mp4 files? Ideally, I'd like to re-direct the mp4 outputs to a different disk attached to our workstation. Many thanks for any suggestions!

https://redd.it/x2knt0
@r_bash

How can I write this if condition more efficiently?

if $a -eq 0 && $b -eq 0 && $c -eq 0 ; then

I want to do something like:

if $a && $b && $c -eq 0

i just don't know how to format it correctly

https://redd.it/x37adx
@r_bash

Any tip on optimizing this?

Hi!

I have a waybar module to track Spotify that runs every two seconds.
Since it runs so frequently I want it to be as performant as possible.

This is what I came so far:

#!/bin/sh

player="playerctl -p 'spotify'"
metadata="$player metadata"

playerstatus=$(eval $player status 2> /dev/null)

([ "$playerstatus" = "Playing" ] || "$player_status" = "Paused" ) && \
printf "$(eval $metadata artist) - $(eval $metadata noscript)"

It works, but I figured this is a nice opportunity to learn something new about shell-noscripts.
Does anybody have any tip or idea on how to improve this for runtime footprint?

Thanks in advance :D

EDIT: result thanks to @rustyflavor and @oh5nxo:

player=$(playerctl -p 'spotify' metadata -f "{{ artist }} - {{ noscript }}")

"$player" != "No players found" && printf "$player"

https://redd.it/x3ez5w
@r_bash

Are parentheses wrapped around the primary case statement variable required?

Parentheses:

case "$1" in
esac

Without parentheses:

case $1 in
esac

Would I have to worry about whitespace, and or weird ass symbols in the $1 variable? Or does the case statement simply not care? If parentheses are not required in any stretch of the imagination, then I will write all my dash noscript case statements without parentheses.

https://redd.it/x3vnes
@r_bash

How can I replace this grep command? I'm getting the Argument list too long error.

Hello, I'm very new to bash so this can probably be easily solved.

I'm writing a bash noscript that stores a list of files in a variable, and then I need to grep those files from a much larger list that contains the files from the first list. Don't know if that makes sense. Here's how I did it:

finalList=$(cat <(echo "$completeList") | grep "$partialList")

It works fine most of the time, however, in some cases the "$partialList" has a lot of files, and the "grep: Argument list too long" error pops up. Any workaround I can use? I read I could use the find command but I'm not really sure how to apply it here.

Thanks in advance!

https://redd.it/x42cnc
@r_bash

I've been trying to make a noscript to simplify my very common find command usage, but it's not working in three ways

First, I can't figure out how to use . in a variable. I want to use . so that output from find is like ./dir1/dir2/file, instead of /home/user/project/date/dir1/dir2/file. In line 7 below at the point SEARCH_DIR="$PWD" is what I need help with for using . please, as I've resorted to $PWD.

Second, the find command when executed from my bash noscript is not doing wildcard expansion on the *. I echo the command I'm running and copy and paste it into bash and it works as intended, but when run from my noscript file it says No such file or directory.

Third, it can't actually find the directory I pass to it. So if I do ff *test* -s /, it says No such file or directory.

Here is the noscript:

#!/bin/bash

# File Find

FILENAME="$1"
shift
SEARCHDIR="$PWD"
EXCLUDEDDIRS=

-z "$FILENAME" && echo "No filename pattern supplied."

while "$#" -gt 0 ; do
echo "$#"
case "$1" in
-x | --exclude ) EXCLUDEDDIRS="${EXCLUDEDDIRS} ! -path '${2}'"; shift 2 ;;
-s | --search-dir ) SEARCHDIR="$2"; shift 2 ;;
* ) shift ;;
esac
done

case "$SEARCHDIR" in
"WebModules") EXCLUDEDDIRS="${EXCLUDEDDIRS} ! -path '${PWD}/obj/'" ;;
) ;;
esac

COMMAND="find ${SEARCHDIR} ${EXCLUDEDDIRS} -name ${FILENAME} -print"
echo "$COMMAND"
"$COMMAND"

https://redd.it/x45d70
@r_bash

can someone what did I overlook

what is the purpose of the noscript: unlocking the dataset using the curl command

the code that works:

curl "`https://$host/api/v2.0/pool/dataset/unlock`" -k -X POST -H "accept: */*" -H "Content-Type: application/json" -H "Authorization: Bearer 1-$API_TOKEN" -d '{"id": "the_dataset_hardcoded","unlock_options": {"key_file": false,"recursive": false,"toggle_attachments": true,"datasets": [{"name" : "the_dataset_hardcoded" , "passphrase" : "the_password_hardcoded"}]}}'

now this one works flawlessly, but the moment i try to replace the "the_dataset_hardcoded" and/or the "the_password_hardcoded" as an environmental variable the noscript stops working

my next try is to create the part as a string that i can echo out for debugging so i create:

curl_builder3="'{\"id\": \"$pool\",\"unlock_options\": {\"key_file\": false,\"recursive\": false,\"toggle_attachments\": true,\"datasets\": [{\"name\" : \"$pool\" , \"passphrase\" : \"$pass\"}]}}'"

and run it like this:

curl "`https://$host/api/v2.0/pool/dataset/unlock`" -k -X POST -H "accept: */*" -H "Content-Type: application/json" -H "Authorization: Bearer 1-$API_TOKEN" -d $curl_builder3

but what i get for this is:

curl: (3) unmatched brace in URL position 1:

{"key_file":

^

bash: line 3: Server: command not found

&#x200B;

but when I echo the created env variable i get the correct string that if i copy and run in terminal it does what is supposed to do

&#x200B;

now what did I miss

&#x200B;

&#x200B;

edit:

the curl command is now fixed and it looks like this:

curl "`https://$host/api/v2.0/pool/dataset/unlock`" -k -X POST -H "accept: */*" -H "Content-Type: application/json" -H "Authorization: Bearer 1-$API_TOKEN" -d "{\"id\": \"$pool\",\"unlock_options\": {\"key_file\": false,\"recursive\": false,\"toggle_attachments\": true,\"datasets\": [{\"name\" : \"$pool\" , \"passphrase\" : \"$pass\"}]}}"

&#x200B;

https://redd.it/x5v910
@r_bash

I am once again asking for your noscripting support

How are you doing bash ninjas?

I learned a lot about noscripting in my last post here, and I want to see if there is some other hidden improvement that can be done on a little noscript I created, and hopefully learn something new on the way.

I use wireplumber to control volume, the command I use to increase volume is

wpctl set-volume @DEFAULTAUDIOSINK@ 1%+

but as a downside, this command allows to increase volume well over 100%, which I don't like.
I figured I could check the volume with this command

sicro@sicro ~ $ wpctl get-volume @DEFAULTAUDIOSINK@
Volume: 1.00

so if I were to pipe the output to awk I could: remove the Volume: part, compare the numbers, and pass the corresponding argument to the original command like this

wpctl set-volume @DEFAULTAUDIOSINK@ $(wpctl get-volume @DEFAULTAUDIOSINK@ | awk '{ print ($2 >= 1) ? "100%" : "1%+"}')

Now, this needs two calls to wpctl and awk to do the math.
Does anybody can come up with a better version that can reduce the processes spawned?

Thank you for your time :D

https://redd.it/x7ab4y
@r_bash

How do you create empty collection if $col.json is empty?

ls -1 *.json | sed 's/.json$//' | while read col;
do
mongoimport --host $host --port $port -d $db --collection $col --file $col.json
echo $cmd
done

I have this bash noscript, but the issue is that it doesn't create an empty collection if the file is empty. How do you remedy this situation?

https://redd.it/x7ig30
@r_bash

Using expect to give inputs to bash noscript

Suppose I have a noscript which takes two inputs

`

\#!/bin/bash

echo "enter username: "

read $UNAME

echo "enter port: "

read $PORT

echo $UNAME

echo $PORT

`

I want to use a expect noscript to give input directly for PORT, but allow user to enter username manually and then print both username and port.

https://redd.it/x91ujm
@r_bash

Trigger another noscript when a bin file executed?

Without modifying the executable.. is it possible to add a hook or some type of event on my system (bashrc?) that allows me to trigger a side-effect/another shell noscript when a bin executable has finished being executed

https://redd.it/xatgw3
@r_bash

I Just Want To Show Off Some Scripts I've Been Using

So a couple years ago I finally sat down and decided to get better at bash noscripting. I had a number of things I wanted to automate and was often needing "oddball" utilities no one else had done/I could find.

# CSV2CUE
Probably the most niche of problems. I had exported a very long audio project from a DAW that had embedded CUE points. It was 4 entire CDs in one project. I could split the entire thing out by CUE points. The problem is that the DAW I edited everything on did not follow proper Redbook timecode format of MM:SS:FF (minutes:seconds:frames) and instead used HH:MM:SS:FF. This made the editor I use for splitting completely crash. So I had to make the DAW output a CSV file of CUE points...then process them in to something my other editor liked.

#!/bin/bash
tail -n +2 Markers.csv | tac | awk '{print $3}' | sed -e "s/:/ /g" >> cue.tmp
t=1
cat cue.tmp | while read line;
do
p=($(echo $line))
if [ ${p[0]} -gt 0 ]; then
p[1]=$(echo "(${p[0]} * 60) + ${p[1]}" | bc)
fi
cue=($(echo "${p[1]}:${p[2]}:${p[3]}"))
printf "\nTRACK %02d AUDIO\n" $t >> output.cue
printf "INDEX 01 %s\n" $cue >> output.cue
t=$((t + 1))
done
rm cue.tmp

Now it's not a full cue sheet; but all my editor wanted was TRACK and INDEX fields.

# VPS Backup

I back all my critical VPS stuff up with a bash noscript:

#!/bin/bash
rsync -a /etc/nginx/ /var/www/backups/etc-nginx
rsync -a /etc/asterisk/ /var/www/backups/etc-asterisk
rsync -a /home/dewdude/.znc/ /var/www/backups/znc-conf
rsync -a /home/dewdude/qspbot/ /var/www/backups/qspbot
rsync -a /home/git/ /var/www/backups/home-git-gogs
rsync -arvz -e 'ssh -p [SECRETPORT]' /var/www [HOME PC]:/media/remote/vpsbackup

I used to do a full MYSQL dump as well; but I ditched wordrpess for (semi)static site generation.

# Jekyll Lazy-Compose

Speaking of static site generators; I'm using Jekyll. The way I got it configured I became stupid reliant on front-matter for making everything work. So I decided to hack together a noscript so I can write posts from bash without having to do anything but write, save, and commit to my git so githooks render it.

#!/bin/bash

# usage: ./compose.sh [category] [noscript]
# example: /compose.sh blog MY AWESOME POST TITLE NO YOU DON'T NEED TO ENCLOSE IT!
# run in the root of your site files/repository
# assumes categories are directories in root

# Variables and category argument
category=$1
pd=$(date +'%Y-%m-%d')
pt=$(date +'%T')
file=blog$$.md
# Ditch the category argument
shift 1
# Read everything else as noscript.
noscript=$@
t=${noscript,,}
t=${t// /-}
fd=$(date +'%Y/%^b/%d')
# Let's write the front matter to our temp file.
printf -- "---\nnoscript: $noscript\nlayout: post\ndate: $pd $pt\npermalink: /$category/$fd-$t.php\nexcerpt_separator: <!--more-->\n---\n\n" >> $file
# Write the post in whatever editor you want.
nano + $file

# Move the file to category/_posts replacing spaces with hyphen
mv $file $category/_posts/$pd-${t// /-}.md
# Display some output to verify it's done.
printf "\nPost $noscript created in $category: $category/_posts/$pd-$t.md\n\n"

This one was fun because I had no idea how to make it blindly accept multiple words as arguments. The only issue is if you don't escape characters that require it. This is probably my second most used noscript.

# Asterisk MusicOnHold

Did you know you can do musiconhold from a streaming source with Asterisk? It can call an application and suck input in from stdin. This is fine till you want to use OGG sources since ogg123 doesn't resample and mplayer doesn't support stdout. Good thing noscripts count as executables.

#!/bin/bash

PIPE="/tmp/asterisk-pipe.$$"
mknod $PIPE p
mplayer -playlist http://host:port/playlist.m3u -really-quiet -quiet -ao pcm:file=$PIPE -af resample=8000,pan=1:0.5:0.5,channels=1,format=mulaw 2>/dev/null | cat $PIPE 2>/dev/null
rm $PIPE

I have made ogg123 work with a direct pipe to sox; but holy cow the CPU usage.

# Hacky Auto-Update Page

I've been following this stuff with a particular provider's /16 block relentlessly attacking SIP

accounts. They seem to be doing nothing and the numbers have increased. We're almost to 5% of the /16 being blacklisted.

Anyway...this noscript just plops my iptables output between some pre-rendered HTML/PHP code; along with stuff to count the IPs and keep a list of prior counts.

I have filtered the full IP range and clues just to avoid breaking rules.

#!/bin/bash

date=$(date)
count=$(iptables -S | grep '###\.##\.[0-9]\{1,3\}\.[0-9]\{1,3\}' | wc -l)
count=$(($count - 1))
cp /root/head /root/tmp.tmp
printf "Last updated: $date - Last Count: $count\n<br><pre><code>\n" >> /root/tmp.tmp
iptables -S | grep '###\.##\.[0-9]\{1,3\}\.[0-9]\{1,3\}' >> /root/vypr.tmp
printf "$date : $count\n" >> /root/count
printf "<br>\n" >> /root/tmp.tmp
cat /root/count >> /root/tmp.tmp
printf "</code></pre><br>\n" >> /root/tmp.tmp
cat /root/vfoot >> /root/tmp.tmp
rm [path-to-www-root]/tmp.php
# the file is stored in a different directory because Jekyll wipes root every build
rm [path-to-www-stuff]/tmp.php
mv /root/tmp.tmp /var/www/tmp.php
chmod 777 /var/www/tmp.php
ln -s /var/www/tmp.php /var/www/pickmy/pbx/tmp.php

Since the blacklist only updates every 4 hours; the noscript only has to run every 4 hours. It does so 5 minutes after the blacklist updates.

That's all for now.

https://redd.it/xaxkn1
@r_bash