My FeedDiscussionsHeadless CMS
New
Sign in
Log inSign up
Learn more about Hashnode Headless CMSHashnode Headless CMS
Collaborate seamlessly with Hashnode Headless CMS for Enterprise.
Upgrade ✨Learn more
Linux Tips And Tricks

Linux Tips And Tricks

Arpit Mohan's photo
Arpit Mohan
·Sep 1, 2018

It seems that publishing a post of Linux tips & tricks is a rite of passage that all engineers must go through. More importantly, it's a bookmark for my future self to find these eclectic mix of commands easily with minimal googling.

So without further ado, here's my list:

find

This command is very helpful while searching for needles in a haystack. It also has plenty of modifiers (too many to list), so I'll only mention the ones I use almost daily.

The general syntax of the command is:

$ find <starting directory> \
  -type <type of file/directory> \
  -name "<Regex of the file/dir name>" \
  -exec <arbitrary command to execute> {} \;

Basic example looking for java files:

$ find ~/projects -type f -name "*.java"

Example looking for directories called main

$ find ~/projects -type d -name "main"

We can also recursively execute a command on the matching files. This exposes the true power of the find command.

Example: Search all java files for the Main function. The output of this command will only be the matched strings.

$ find ~/projects -type f -name "*.java" -exec grep -i "Main" '{}' \;

Alternatively, if you wish to output the name of the matching file along with the matched string:

$ find ~/projects -type f -name "*.java" -exec grep -i "Main" '{}' +

Notice the difference in how the command ends. The braces '{}' are substituted by each matching file name at runtime. It's similar to how xargs works.

less

For log files (typically large) and files that I wish to only read and NOT edit, I use the less command extensively. One of the major reasons I love this command is, I can tail files without it cluttering my terminal. This command also has the benefit of only loading the file partially into memory, thereby making it a sleek alternative to vi

Example:

$ less /var/log/insanely-large-log-file.log

Once inside, you can use G to go to the end of the file or use F to tail the file.

Alternatively, you can open & tail the file directly using

$ less +F /var/log/insanely-large-log-file.log

To quit and return to the terminal, use the normal vi command

<ESC> :q

entr

I recently came across this command and although it hasn't made it to my most frequently used commands, I think it's pretty useful. It would be a crime to compile such a list and not mention entr

This command allows users to execute an arbitrary script whenever a file changes. It's similar to watchr, guard & nodemon. Since entr is written in C, it's faster and more responsive on larger directories.

Usecases can range from running test cases whenever your source files change

$ ls *.c | entr 'make && make test'

or reloading the browser whenever an HTML file changes.

$ ls *.css *.html | entr reload-browser Firefox

You can also restart server processes using the -r modifier

$ ls *.rb | entr -r ruby main.rb

Check out more details here.

htop

If you are looking at your machine's performance in any way apart from htop, you're doing it wrong. It's what top should have been all along. Although it's not built-in, you can easily install it via:

$ apt install htop

or

$ brew install htop-osx.

The output is very self-explanatory and easier to understand & sort.

jq

Your search for a JSON parser ends here. jq runs on a stream of JSON data. Each input is parsed as a sequence of whitespace-separated JSON values which are passed through each filter of jq. The filters themselves, can be combined in any way by piping the output of one filter to the input of another.

Example:

This example extracts the field 'foo' from the input JSON.

$ jq '.foo' {"foo": 42, "bar": "less interesting data"}
=> 42

This example extracts the 0th element of the JSON array.

$ jq '.[0]' [{"name":"JSON", "good":true}, {"name":"XML", "good":false}]
=> {"name":"JSON", "good":true}

vim commands

By popular opinion, vim is the awesomest editor in town. But for newbies, it can be a little daunting and un-friendly. Once you get vim, you'll never go back to any other editor.

The following vim commands aren't for newbies. It's for more advanced users.

How often do you open a file in vim to edit it and realize you should have opened it as root? You can use the following command to save your changes without exiting vim.

<ESC> :w !sudo tee %

Slightly longer to type but completely worth it.

I think vim commands demand a dedicated post to do justice to the most beloved editor.

Conclusion

I hope this was useful. I'll probably edit this post to reflect any new commands that cross my path/blow my mind.