{"id":108,"date":"2022-09-26T21:17:49","date_gmt":"2022-09-26T15:47:49","guid":{"rendered":"http:\/\/yuvishere.co.in\/?p=108"},"modified":"2022-10-17T17:58:16","modified_gmt":"2022-10-17T12:28:16","slug":"how-i-started-writing-shell-scripts","status":"publish","type":"post","link":"https:\/\/yuvishere.co.in\/blog\/how-i-started-writing-shell-scripts\/","title":{"rendered":"How I started writing shell scripts"},"content":{"rendered":"\n

As a developer, we use the CLI a lot for various things, and as someone who has to probably learn about computers, it is highly recommended to have some understanding about CLI and it’s capabilities.<\/p>\n\n\n\n

For me, it all started when I was trying to work with ffmpeg<\/a><\/strong> and the bash<\/code> shell to iterate through a number of mp4 files and extract their audio in mp3 files. The idea was, to take each item with the .mp4<\/strong><\/code> file format in the current directory and execute the following code for each of these items: ffmpeg -i filename.mp4 filename.mp3<\/strong><\/code>. In simple words, just take the file with the name filename<\/code> which is of the mp4 format and convert it to mp3 format. I wrote a useful piece of code to do it in the bash shell itself, and it was this: for file in *.mp4; do ffmpeg -i \"$file\" \"${file%???}mp3\"; done;<\/strong><\/code>. The ones who are familiar with bash shell probably know this and are cringing at it, but it simply means this: “for” every item which matches with .mp4 in the end (*.mp4), reference to such an item as “file” and do (or execute) ‘ffmpeg -i “$file” “${file%???}mp3’. Here, “$file” means the item with “.mp4” in the end and “${file%???}mp3” means remove the last three items (or in this case, letters) from “$file” and attach “mp3” to the remaining string.<\/p>\n\n\n\n

For the longest time, I had this command saved somewhere for any rainy day when it would be needed. Just copy pasting it in the CLI. Then I thought about it and figured, there must be a way to automate this. Sometime later, I came across the fact that you can write proper scripts and save them in files, just like writing code in a particular language and saving it in the given file format. I started with some very basic programs and operations, played with the shebang operator. It was good times. In the end, I finally wrote the following script to automate this and saved it as a binary executable in my bash_profile<\/em>.<\/p>\n\n\n\n

\n

#!\/bin\/bash<\/em>
mkdir<\/strong> -p output
for<\/strong> name in<\/strong> *.$1<\/strong>; do<\/strong>
ffmpeg -i \"$name<\/strong>\" \"output\/${name%.*}<\/strong>.$2<\/strong>\"
done<\/strong><\/code><\/p>\n\n\n\n

Definitely an upgrade over the previous one and much more scalable. I replaced a couple of parameters and named this script as ffmpegrec<\/strong><\/code>. So the usage for this script is ffmpegrec <input format> <output format><\/strong><\/code>.<\/p>\n\n\n\n

Without even realising at the moment, this was my very first script which automated some mundane yet essential processes for me. A couple of years down the line, I am a fan of the bash shell trying to work my way through various tasks a developer. Surely, there are a number of benefits to knowing and automating your work using whatever script you use, there are a number of them to choose from with little to no differences.<\/p>\n<\/div>\n\n\n\n

Some references you can use:<\/p>\n\n\n\n